Trigonometry Supplemental Material

for Biology Majors

This manual is the result of collaborative efforts between

faculty from the Dept. of Mathematics and Life Sciences at
Los Angeles Mission College and Mathematics faculty at
University of California, Los Angeles

This manual is best used in conjunction with Academic Excellence Workshop manual.
Website: https://bit.ly/2QCe0ld

January 2020

Funding for this publication was provided by NSF HSI Award number 1832348
 Trigonometry Supplemental Material for Biology Majors
Table of Contents

Content Worksheets………………………………………………………………………………………………………………………..…1
Discovering Radians Activity……………………………………………………………………………………………………….2

Applications of Trigonometric Functions…………………………………………………………………………………….6

Introduction to Trigonometric Functions and the Unit Circle………………………………………………………7

Graphing Trigonometric Functions……..…………………………………………………………………………………….11

Inverse Trigonometric Functions………..…………………………………………………………………………………….16

Graphing Trigonometric Functions and Inverse Trigonometric Functions………………………………….19

Trigonometric Identities…………………….……………………………………………………………………………………..23

Trigonometric Equations…………………………………………………………………………………………………………..25

Law of Sines and Law of Cosines……………………………………………………………………………………………….27

Vectors and Polar Coordinates………………………………………………………………………………………………….29

Polar Form of Complex Numbers………………………………….…………………………………………………………..32

Test Reviews……………………………………………………………………………………………………………………………………..34
Trigonometric Identities and Equations Review………………………………………………….…………………….35

Polar Forms and Vectors Review…………………………………..………………………………………………………….38

Cumulative Review………………………………………………………………….………………………………………………..42

Articles……………………………...……………………………………………………………………………………………………………..48
Angular Momentum in Human Walking…………………………………………………………………………………….49
Acknowledgments & Credits....…………………………………………………………………………………………………….……65
 Content
Worksheets

Page 1
 Discovering Radians Activity
Supplies:

• Pre-cut circle from construction paper, or a paper plate

• String or heavy thread
• Ruler and protractor
• Scissors

Steps:

1. Using the string, measure the circumference of the circle (or paper plate). Record this length.

Circumference = ______________

2. Fold the circle (or paper plate) in half. Crease the fold line so that it can be clearly seen.

3. Fold the circle again into quarters. Crease the fold lines.

4. Open the circle. Using the ruler, draw line segments along the fold lines forming four
quadrants. Label the points on the edge of the circle that correspond to 0º, 90º, 180º and 270º.

5. The folding process has located the center of the circle. Use your string to measure the radius
of the circle. Cut the string to this length. Record this length.

Radius = ___________

6. Hold one end of the radius length string at the edge of the circle at 0º. Wrap the string around
the edge of the circle and mark its ending location. Connect this point to the center of the circle.

7. Using your protractor, find the number of degrees in the central angle formed from 0º to the
segment you drew in step 6. In terms of radians, this angle has a measure of one radian.
Record this answer.

1 Radian = __________˚

8. Using your radius length string, continue to wrap the string around the edge of the circle
marking its ending locations. Record the number of radian angles that will fit in the circle.

How many radian angles are in your circle? ___________

9. Ponder:

If the central angle has a radian measure of 2, what is the number of degrees in the angle?

2 radians = __________˚

Page 2
 Conversion between degrees and radian

For example:

Ex 1) 225˚=______(radians)

a. To convert to radians you should multiply by ______.

b. You should only focus on ________ and write  after the fraction is reduced.
c. _________ = _________.
d. The answer is ________.

Ex 2) = ______(degrees)

a. To covert to degrees you should multiply by _______.

b. Since the ’s cancel each other, we should only focus on the ________.
c. ________ = _______.
d. The answer is _______.

Ex 3) An angle representing one complete revolution of the unit circle measures 2 

radians, formerly ______º.

1. Change the following radians to 2. Change the following degrees to

degrees if 2  =360º, radians if 360º= 2  ,

a)  = ____________ a) 270º = __________

 b) 60º = __________
b)  ____________
2

c)  ____________ c) 150º = __________
4
3
d)  ___________
4 d) 30º = __________
11
e)  __________
6 e) 240º = __________

Page 3
 Work with partners to find the angles in degrees and radians.

1. 2. 3.

Degree_____ Radian_____ Degree____ Radian____ Degree____ Radian____

4. 5. 6.

Degree____ Radian____ Degree____ Radian____ Degree____ Radian____

7. 8. 9.

Degree____ Radian____ Degree____ Radian____ Degree____ Radian____

10. 11. 12.

Degree____ Radian____ Degree____ Radian____ Degree____ Radian____

13. 14. 15.

Degree____ Radian____ Degree____ Radian____ Degree____ Radian____

Page 4
Page 5
 Applications of Trigonometric Functions

1. The root system for some native Caribbean plants requires 5 m2 of land area to collect the required
amount of nutrients.
a. If this land area is circular, what is the area?
b. If this land is a 35o sector of a circle between two rocks, what is the radius?

2. The equation P = 20 sin(2πt) + 100 models the blood pressure, P, where t represents time in seconds.
a. Find the blood pressure after 15 seconds.
b. What are the maximum and minimum blood pressures?

3. The amount of nutrients in plants depends on the amount of sunlight they receive. The amount of
 1 
sunlight in a certain region can be modeled by the function h = 15cos  d  , where h represents the
 600 
hours of sunlight, and d is the day of the year. Use the equation to find how many hours of sunlight
there are on February 10, the 42nd day of the year.

4. At Mauna Loa, Hawaii, atmospheric Carbon Dioxide levels in parts per million (ppm) have been
) .022 x 2 + .55 x + 316 + 3.5sin ( 2π x ) can
measured regularly since 1958. The function defined by L( x=
be used to model levels, where x is in years and x = 0 corresponds to 1960.
a. Calculate the Carbon Dioxide levels in 1970.
b. Calculate the Carbon Dioxide levels in 2017.

Page 6
 Introduction to Trigonometric Functions and the Unit Circle

1) Find the angle of smallest possible positive measure coterminal with the given angle: -172°

2) Find the angle of smallest possible positive measure coterminal with the given angle: 438°

3) Suppose that θ is in standard position and the given point is on the terminal side of θ. Give the exact value
of the indicated trig function for θ: (6, 8); Find csc θ.

4) Convert 330° to radians. Leave answer as a multiple of π.

5) Convert t 0.2521 to degrees. Give answer using decimal degrees to the nearest hundredth. Use 3.1416 for π.

6) Find the exact value: sin(-180°)

7) Find the exact value: sec 270°

8) Find the exact value: cot 90° + 2 cos 180° + 6 sec2 360°

9) Find the exact value: cos 30°

-5π
10) Find the exact value: tan
6

4π
11) Find the exact value: csc
3

π
12) Find the length of an arc intercepted by a central angle θ = radians in a circle of radius r = 38.81 ft; . Round
3
your answer to 1 decimal place.

13) Assume that the cities lie on the same north -south line and that the radius of the earth is 6400 km. Find the
latitude of Spokane, WA if Spokane and Jordan Valley, OR, 43.15° N, are 486 km apart.

2π
14) Find the area of a sector of a circle having radius r = 15.0 ft, and central angle θ = radians. Express the
3
answer to the nearest tenth

15) Each tire of an automobile has a radius of 2 feet. How many revolutions per minute (rpm) does a tire make
when the automobile is traveling at a speed of 79 feet per sec? Round your answer to the nearest tenth.

1
16) Find sec θ if cos θ = and sin θ > 0.
4

Page 7
 37
17) Find cot θ if csc θ = and θ is in quadrant I.
6

18) Find cot θ.

7 24
,-
25 25

19) Find the exact value of x in the figure.

34

20) Find the exact value of x in the figure.

38

Page 8
21) Find a formula for the area of the figure in terms of s.

22) Find h as indicated in the figure. Round your answer to the hundredths place.

25.3° 58.4°

147 ft

23) An airplane travels at 160 km/h for 4 hr in a direction of 306° from St. Louis. At the end of this time, how far
west of St. Louis is the plane (to the nearest kilometer)?

24) The angle of elevation from a point on the ground to the top of a tower is 37.87°. The angle of elevation from a
point 106 feet farther back from the tower is 24.15°. Find the height of the tower. Round your answer to the
hundredths place.

Page 9
Answer Key

1) 188°
2) 78°
5
3)
4
11π
4)
6
5) 14.44°
6) 0
7) Undefined
8) 4
3
9)
2
3
10)
3
2 3
11) -
3
12) 40.6 ft
13) 47.50 °N
14) 235.6 ft2
15) 377.2 rpm
16) 4
1
17)
6
7
18) -
24
19) 17 3
76 6
20)
3
3 2
21) s
6
22) 97.98 ft
23) 518 km
24) 112.26 ft

Page 10
 Graphing Trigonometric Functions

For textbook reference you can use the free opnestax Precalculus text: https://openstax.org/details/books/precalculus
Sections 6.1 and 6.2

1) Match the function with its graph.

1) y = sin 3x 2) y = 3 cos x
3) y = 3 sin x 4) y = cos 3x

A) B)

C) D)

1
2) Give the amplitude: y = -2 cos x
3

3) Give the period: y = cos 5x

4) Find the phase shift of the function. y = -3 + 2 sin 3x -

6

1
5) Graph the function over a one-period interval: y = 4 + sin (2x - )
3

1 2
6) Graph the function over a one-period interval. y = + cos 2x -
2 3

Page 11
 7) The population size of most insects is dependent on the temperature of their habitats. The temperature in Fairbanks is
2
approximated by T(x) = 37 sin (x - 101) + 25, where T(x) is the temperature on day x, with x = 1 corresponding to
365
Jan. 1 and x = 365 corresponding to Dec. 31. Estimate the temperature on day 10.

4 1
8) Graph the function.: y = tan x -
5 2 6

4 1
9) Graph the function. y=- cot x -
5 2 6

10) Graph the function. y = csc x -

3

11) Graph the function. y = 2 + 4 sec x +

5

12) The average person’s blood pressure is modeled by the function f(t) = 20 sin(160 t) + 100, where f(t) represents the
blood pressure at time t, measured in minutes.

a) Graph the function.

b) Find the highest and lowest values for the average blood pressure and the time at which they occur.

13) Circadian rhythms are approximately 24-hour internally controlled biological changes that occur in the absence of
environmental cues- although they can be altered by the environment. The first example that might pop in your mind when
you hear circadian rhythms is the sleep-wake cycle. Other examples include daily fluctuations in fluctuations in body
temperature, hormones, behavior, and heart rate. Individual neurons in the suprachiasmatic nucleus, a cluster of cells in the
region of the brain called the hypothalamus, generates this "biological clock" in mammals. But you don't need to have a
brain to have a biological clock. Circadian rhythms are found in a wide variety of organisms- from single-celled yeast to
plants.

For this module, we will consider a hypothetical example. Suppose a particular species exhibits daily regular
fluctuations in body temperature that can be approximated by the equation,

T(t) = 36.8 - 1.3sin ( (t + 2))

12
where T represents temperature in°C, t represents time (in hours), and t = 0 corresponds to 12 o'clock midnight (12:00 am)

a) Find tte period of the function. Does the period make sense? Why?
b) What tiem of the day, does the body temperature reach the maximum? What is the temperature at that time?
c) Approximate the body temperature at 10 am.

Page 12
Answer Key
Testname: GRAPHING TRIG FUNCTIONS

1) 1B, 2D, 3C, 4A

2) 2
2
3)
5

4) units to the right

18
5)

6)

7) -12°
8)

Page 13
Answer Key
Testname: GRAPHING TRIG FUNCTIONS

9)

10)

11)

Page 14
Answer Key
Testname: GRAPHING TRIG FUNCTIONS

12)

13)

Page 15
 Inverse Trigonometric Functions

For textbook reference you can use the free opnestax Precalculus text: https://openstax.org/details/books/precalculus
Section 6.3

2
1) Find the exact value of the real number y. y = cos-1
2

2) Find the exact value of the real number y. y = arctan 1

3) Find the exact value of the real number y. y = sin-1 (0.5)

1
4) Graph: y = arccsc x
3

Evaluate the expression.

1
5) cos arcsin
4

6) arccos cos
2

7) sin (arctan 2)

Write the following as an algebraic expression in u, u > 0.

8) cos(arcsin u)

u
9) tan cos-1
3

10) cos(arctan u)

u
11) sin arctan
2

12) True or false? The statement cos(cos-1 x) = x for all real numbers in the interval 0 x .

Page 16
13) True or false? The statement tan-1 (tan x) = x for all real numbers in the interval -∞ x .

14) A crystal is an array of atoms that forms atomic layers known as atomic planes. When an x-ray is passed through a crystal,
the x-ray beam is diffracted according the crystal's atomic structure. Using a technique called x-ray crystallography, one can
construct the three dimensional atomic structure based upon the diffraction pattern.
X-ray crystallography has been used to uncover the atomic structure of thousands of macromolecules ranging from vitamins
to protein complexes. X-ray crystallography was a critical technique in many discoveries that were honored with the Nobel Prize.
Perhaps the most famous structure revealed by x-ray crystallography is the double helical structure of DNA.
The structure of a crystal can be experimentally determined by Bragg's equation, =2d sin
where is the wavelength of x-rays, d is the distance between atomic planes, is the angle of reflection (in degrees), and
n is a positive integer.

a) Assume that for a given crystal n=1, find the angle of reflection, if the wavelength adn teh distance are equal.
b) Assume that n=2, and the distance is three times the wavelength, use your calculator to approximate the angle.

15) A 5.8-ft fence is 12.052 ft away from a plant in the direction of the sun. It is observed that the shadow of the
fence extends exactly to the bottom of the plant. (See drawing) Find , the angle of elevation of the sun at that
time. Round the measure of the angle to the nearest tenth of a degree.

5.8 ft

12.052 ft

Page 17
Answer Key
Testname: INVERSE TRIG FUNCTIONS

1)
4

2)
4

3)
6
4)

15
5)
4

6)
2
2 5
7)
5
8) 1 - u2
9 - u2
9)
u

u2 + 1
10)
u2 + 1
u u2 + 2
11)
u2 + 2
12) True
13) False
14)
15) = 25.7°

Page 18
 Graphing Trigonometric Functions and Inverse Trigonometric Functions

For textbook reference you can use the free opnestax Precalculus text: https://openstax.org/details/books/precalculus
Chapter 6
Graph the function.
1) y = sin x +
4

2) y = 2 sin x -
4

2
3) y = cos x +
3 3

4) y = 3 + sin(2x- )

1 2
5) y = + cos 2x -
2 3

1
6) y = tan 2x
3

2 1
7) y = tan x -
3 3 3

1 1
8) y = cot x +
2 2 5

2
9) y = csc x-
5 2

10) y = 2 + 4 sec x +
5

11) y = sin-1 x

12) A generator produces an alternating current according to the equation I = 80 sin 106 t, where t is time in seconds and I is
the current in amperes. What is the smallest time t such that I = 40?

3
13) Find the exact value of the real number y. y = arcsin
2

1
14) Find the exact value of the real number y. cos arcsin
4

Page 19
 u
15) Write the following as an algebraic expression in u, u > 0. tan cos-1
3

Page 20
Answer Key
Testname: CHAPTER 6 REVIEW

1) 4)

2) 5)

3) 6)

Page 21
Answer Key

7) 10)

8) 11)

9) 1
12) sec
636

13)
3
15
14)
4

9 - u2
15)
u

Page 22
 Trigonometric Identities

For textbook reference you can use the free opnestax Precalculus text: https://openstax.org/details/books/precalculus
Sections 7.1 and 7.2

Use the fundamental identities to simplify the expression.

1
1) + sec cos 1)
cot2

2) sin2 + tan2 + cos2 2)

3) cos x (csc x - sec x) - cot x 3)

Verify that each equation is an identity.

4) tan x(csc x - sin x) = cos x 4)

sec - 1 tan
5) = 5)
tan sec + 1

6) (sec - tan )(sec + tan ) = 1 6)

7) csc s - sin s = cos s cot s 7)

1 + sin
8) (sec + tan )2 = 8)
1 - sin

Use Identities to find the exact value.

9) cos 165° 9)

10) cos 10)

12

Find the exact value of the expression using the provided information.
1 3
11) Find cos(s - t) given that cos s = - , with s in quadrant III, and cos t = - , with t in 11)
2 5
quadrant III.

Find the exact value by using a sum or difference identity.

12) tan 75° 12)

13) sin 15° 13)

Use a sum or difference identity to find the exact value.

7 7
14) sin cos - cos sin 14)
24 8 24 8

Page 23
Answer Key
Testname: TRIG IDENTITIES

1) sec2
2) sec2
3) -1
sin x 1 sin x 1 sin2 x 1 - sin2 x cos2 x
4) tan x(csc x - sin x) = tan x · csc x - tan x · sin x = · - · sin x = - = = =
cos x sin x cos x cos x cos x cos x cos x
cos x
sec - 1 sec - 1 sec +1 sec2 - 1 tan2 tan
5) = · = = =
tan tan sec + 1 tan (sec + 1) tan (sec + 1) sec + 1
6) (sec - tan )(sec + tan ) = sec2 - tan2 = 1
1 1 - sin2 s cos2 s cos s
7) csc s - sin s = - sin s = = = cos s · = cos s cot s
sin s sin s sin s sin s
1 2 sin sin2 1 + 2 sin + sin2 (1 + sin )2
8) (sec + tan )2 = sec2 + 2 sec tan + tan2 = + + = = =
cos2 cos2 cos2 cos2 1 - sin2
(1 + sin )2 1 + sin
=
(1 - sin )(1 + sin ) 1 - sin
- 6- 2
9)
4
6+ 2
10)
4
3+4 3
11)
10
12) 3+2
6- 2
13)
4
1
14)
2

Page 24
 Trigonometric Equations

For textbook reference and videos you can use section 7.5 of the free openstax Precalculus text:
https://openstax.org/details/books/precalculus

To access the section and videos,you can use use:

https://cnx.org/contents/[email protected]:aeVxcRIM@12/7-5-Solving-Trigonometric-Equations

Solve the equation for the interval [0, 2 π).

1) cos2 x + 2 cos x + 1 = 0

2) 2 sin2 x = sin x

3) cos x = sin x

4) sin2 x - cos2 x = 0

Determine the solution set of each equation in radians (for x) or degrees (for θ) to the nearest tenth as appropriate.
5) 2 sin2 x + sin x = 1

4 tan θ
6) =1
5 - tan2 θ

Solve the equation for solutions in the interval [0, 2 π).

3
7) sin 4x =
2

1
8) sin x cos x =
2

9) sin 2x + sin x = 0

Determine the solution set of each equation in radians (for x) or degrees (for θ) to the nearest tenth as appropriate.
10) 3 cos2 θ + 2 cos θ = 1

Page 25
Answer Key:

1) {π}
π 5π
2) 0, π, ,
6 6
π 5π
3) ,
4 4
π 3π 5π 7π
4) , , ,
4 4 4 4
π 5π 3π
5) + 2nπ, + 2nπ, + 2nπ
6 6 2
6) {45° + 180°n, 101.3° + 180°n}
π π 2π 7π 7π 13π 5π 19π
7) , , , , , , ,
12 6 3 12 6 12 3 12
π 5π
8) ,
4 4
2π 4π
9) 0, , π,
3 3
10) {70.5° + 360°n, 180° + 360°n, 289.5° + 360°n}

Page 26
 Law of Sines and Law of Cosines
For reference, you can use chapters 8.1 and 8.2 of the openstax Precalculus text: https://openstax.org/details/books/precalculus

Solve the triangle.

1)

26 m

Solve the problem.

2) To find the distance AB across a river, a distance BC = 1053 m is laid off on one side of the river. It is found that B =
101.3° and C = 17.0°. Find AB rounded to the nearest meter.

Find the area of triangle ABC with the given parts. Round to the nearest whole number.
3) A = 26.4°
b = 12.3 in.
c = 7.7 in.

Find the missing parts of the triangle.

4) C = 35°30'
a = 18.76
c = 16.15

5) A = 79°
a = 32 yd
b = 65 yd

6) B = 63°30'
a = 12.20 ft
c = 7.80 ft

Find the missing parts of the triangle. (Find angles to the nearest hundredth of a degree.)
7) a = 27 ft
b = 32 ft
c = 41 ft

Find the area of triangle ABC with the given parts. Round to the nearest whole number.
8) a = 17.4 cm
b = 15.0 cm
c = 13.4 cm

9) Explain, in your own words, the situation called "the ambiguous case of the law of sines."

10) What happens if C = 90° when the law of cosines is applied in the form C2 = A2 + B 2 - 2ab cos C?

Page 27
Answer Key

1) C = 103°, a = 11.7 m, b = 20.7 m

2) 350 m
3) 21.1 in. 2
4) A1 = 42°25', B1 = 102°05', b 1 = 27.20;
A2 = 137°35', B2 = 6°55', b 2 = 3.35
5) no such triangle
6) b = 11.17 ft, A = 77°49', C = 38°41'
7) A = 41.14°, B = 51.24°, C = 87.62°
8) 97 cm2
9) Answers will vary
10) Answers will vary

Page 28
 Vectors and Polar Coordinates

For reference, you can use chapters 8.1 and 8.2 of the openstax Precalculus text:
https://openstax.org/details/books/precalculus
Find the component form of the indicated vector.
1) Let u = -9, -9 , v = -3, 3 . Find -u + 9v.

Find the magnitude and direction angle (to the nearest tenth) for each vector. Give the measure of the direction angle as an angle
in [0,360°].
2) -3, -4

Vector v has the given magnitude and direction. Find the magnitude of the indicated component of v.
3) α = 38.3°, ∣v ∣ = 281
Find the vertical component of v.

Write the vector in the form <a, b>.

4)

Two forces act at a point in the plane. The angle between the two forces is given. Find the magnitude of the resultant force.
5) forces of 25.0 and 31.8 lb, forming an angle of 162.8°

Find the dot product for the pair of vectors.

6) -16, 8 , 0, 15

Find the angle between the pair of vectors to the nearest tenth of a degree.
7) 5, 5 , -3, 8

8) Starting at point A, a ship sails 57 km on a bearing of 188°, then turns and sails 37 km on a bearing of 330°. Find the
distance of the ship from point A.

9) Suppose you would like to cross a 209-foot wide river in a boat. Assume that the boat can travel 32 mph relative to the
water and that the current is flowing west at the rate of 6 mph. If the bearing is chosen so that the boat will land at a point
exactly across from its starting point, how long will it take for the boat to make the crossing? Give your answer to the
nearest second.

Page 29
The rectangular coordinates of a point are given. Express the point in polar coordinates with r ≥ 0 and 0° ≤ θ < 360°.
10) (2, -2)

Give the rectangular coordinates for the point.

11) (6, 225°)

For the given rectangular equation, give its equivalent polar equation.
12) 2x + 3y = 6

13) x2 + y2 = 64

Plot the point.

9π
14) 4,
4

-5 5

-5

Graph the polar equation for θ in [0°, 360°).

15) r = 4 + 4 sin θ

Page 30
Answer Key

1) -18, 36
2) 5; 233.1°
3) 174.2
4) ≈ 6.73, 25.11
5) 11 lb
6) 120
7) 65.6°
8) 36 km
9) 5 sec
10) (2 2, 315°)
11) (-3 2, -3 2)
6
12) r =
2 cos θ + 3 sin θ
13) r = 8
14)
5

-5 5

-5

15)

16
12

8
4

-16 -12 -8 -4 4 8 12 16 r
-4

-8

-12

-16

Page 31
 Polar Form of Complex Numbers

For reference, you can use chapters 8.5 of the openstax Precalculus text:
https://openstax.org/details/books/precalculus

1) Graph the complex number. -5 - 4i

Find the following quotient, and write the quotient in rectangular form, using exact values.
12cis 158°
2)
3cis 38°

3) Graph the complex number. -4i

π π
4) Write the complex number in rectangular form. 8 cos + i sin
6 6

5) Write the complex number in rectangular form. 9(cos 180° + i sin 180°)

Write the complex number in trigonometric form r(cos θ + i sin θ), with θ in the interval [0°, 360°).
6) 5 3 + 5i

Find the product. Write the product in rectangular form, using exact values.
7) [8 cis 300°] [6 cis 330°]

8) Find the given power. Write answer in rectangular form: (2 - 2i)5

9) Find the given power. Write answer in rectangular form.: (- 3 + i)6

10) Find all cube roots of the complex number. Leave answers in trigonometric form: -125i

11) Find all cube roots of the complex number. Leave answers in trigonometric form: 3 + 3i 3

12) Find all solutions of the equation. Leave answers in trigonometric form.: x3 - 8 = 0

13) Find all solutions of the equation. Leave answers in trigonometric form.: x5 - 32 = 0

Use a table of values to graph the plane curve defined by the following parametric equations. Find a rectangular
equation for the curve.
14) x = 2t, y = t + 1, for t in [-2, 3]

15) Find a rectangular equation for the plane curve defined by the parametric equations: x = sin t, y = 3 cos t

16) Find a rectangular equation for the plane curve defined by the parametric equations.: x = sec t, y = tan t

Page 32
Answer Key

1) 14)
y y
6
10

-6 -4 -2 2 4 6 x -10 10 x
-2

-4

-10
-6

2) - 2 + 2i 3 y=
1
x + 1, for x in [-4, 6]
3) 2
y 15) 9x2 + y 2 = 9
6
16) x2 - y 2 = 1
4

-6 -4 -2 2 4 6 x
-2

-4

-6

4) 4 3 + 4i
5) -9
6) 10(cos 30° + i sin 30°)
7) -48i
8) -128 + 128i
9) -64
10) 5 cis 90°, 5 cis °210, 5 cis 330°
3 3 3
11) 6 cis 20°, 6 cis 140°, 6 cis 260°
12) {2, 2 cis 120°, 2 cis 240°}
2π 4π 6π 8π
13) 2 cis 0, 2 cis , 2 cis , 2 cis , 2 cis
5 5 5 5

Page 33
Test Reviews

Page 34
 Trigonometric Identities and Equations Review
You can use chapter 7 in the free Precalculus text at openstax for reference: https://openstax.org/details/books/precalculus

Use the fundamental identities to simplify the expression.

cos2 θ
1) + csc θ sin θ
sin2 θ

2) sin2 θ + tan2 θ + cos2 θ

Verify that each equation is an identity.

3) cot2 x = (csc x - 1)(csc x + 1)

cos β
4) sec β + tan β =
1 - sin β

5) (sec α - tan α)(sec α + tan α) = 1

6) csc s - sin s = cos s cot s

Use Identities to find the exact value.

7) cos 255°

Find the exact value by using a sum or difference identity.

8) tan 105°

Find the exact value of the expression using the provided information.
1 1
9) Find tan(s + t) given that sin s = , with s in quadrant II, and sin t = - , with t in quadrant IV.
4 2

Find the exact value by using a sum or difference identity.

10) sin 15°

Use trigonometric identities to find the exact value.

11) sin 25° cos 35° + cos 25° sin 35°

Use identities to find the indicated value for each angle measure.
2
12) cos 2θ = and θ terminates in quadrant I Find sin θ.
3

1
13) cos 2θ = and θ terminates in quadrant III Find cos θ.
4

Express the function as a trigonometric function of x.

14) cos 4x

Verify that each equation is an identity.

15) cos(4u) = 2 cos2 (2u) - 1

Page 35
Write the product as a sum or difference of trigonometric functions.
16) 8 cos 14° cos 7°

Find the exact value by using a half-angle identity.

17) sin 75°

Determine all solutions of the equation in radians.

x 1 π
18) Find sin , given that sin x = and x terminates in 0 < x < .
2 4 2

Solve the equation for the interval [0, 2π).

19) cos2 x + 2 cos x + 1 = 0

20) 2 sin2 x = sin x

21) cos x = sin x

22) sin2 x - cos2 x = 0

Determine the solution set of each equation in radians (for x) or degrees (for θ) to the nearest tenth as appropriate.
23) 2 sin2 x + sin x = 1

Solve the equation for solutions in the interval [0, 2 π).

3
24) sin 4x =
2

1
25) sin x cos x =
2

26) sin 2x + sin x = 0

Page 36
Answer Key

1) csc2 θ
2) sec2 θ
3) cot2 x = csc 2 x - 1 = (csc x - 1)(csc x + 1).
1 sin β 1 + sin β 1 + sin β 1 - sin β 1 - sin2 β cos2 β cos β
4) sec β + tan β = + = = ∙ = = =
cos β cos β cos β cos β 1 - sin β cos β(1 - sin β) cos β(1 - sin β) 1 - sin β
5) (sec α - tan α)(sec α + tan α) = sec 2 α - tan2 α = 1
1 1 - sin2 s cos2 s cos s
6) csc s - sin s = - sin s = = = cos s ∙ = cos s cot s
sin s sin s sin s sin s
2- 6
7)
4
8) -2 - 3
4 3 + 15
9)
-11
6- 2
10)
4
3
11)
2
6
12) sin θ =
6
101
13) cos θ = -
4

14) cos4 x - 6 sin2 x cos2 x + sin 4 x

15) cos(4u) = cos[2(2u)] = 2 cos2 (2u) - 1
16) 4(cos 21° + cos 7°)
1
17) 2+ 3
2
8-2 15
18)
4
19) {π}
π 5π
20) 0, π, ,
6 6
π 5π
21) ,
4 4
π 3π 5π 7π
22) , , ,
4 4 4 4
π 5π 3π
23) + 2nπ, + 2nπ, + 2nπ
6 6 2
π π 2π 7π 7π 13π 5π 19π
24) , , , , , , ,
12 6 3 12 6 12 3 12
π 5π
25) ,
4 4
2π 4π
26) 0, , π,
3 3

Page 37
 Polar Forms and Vectors Review

For reference, you can use chapters8 of the openstax Precalculus text: https://openstax.org/details/

books/precalculus

1) x = 5(t - sin t), y= 5(1 - cos t), 0 ≤ t ≤ 4π

Graph the cycloid for t in the indicated interval.
y
12

10 20 30 40 50 60 x

2) x = t - sin t, y = 1 - cos t, -4π ≤ t ≤ 4π

2.5 y

2
1.5
1
0.5

-16 -12 -8 -4 4 8 12 16 x
-0.5
-1
-1.5
-2
-2.5

Find the missing parts of the triangle.

3) A = 98°
b = 15.2 ft
a = 43.4 ft

4)

5 6 10

Page 38
 5) C = 35°30ʹ
a = 18.76
c = 16.15

Find the missing parts of the triangle. (Find angles to the nearest hundredth of a degree.)
6) a = 27 ft
b = 32 ft
c = 41 ft

Find the missing parts of the triangle.

7) C = 112.5°
a = 5.30 m
b = 9.66 m

Solve the problem.

8) Two ships leave a harbor together traveling on courses that have an angle of 125° between them. If they each
travel 501 miles, how far apart are they (to the nearest mile)?

9) If u = -5, 7 , v = -7, 6 , and w = -11, 2 , evaluate u ∙ (v - w).

Find the angle between the pair of vectors to the nearest tenth of a degree.
10) 5i - 3j, 5i - 6j

Find the magnitude and direction angle (to the nearest tenth) for each vector. Give the measure of the direction angle as
an angle in [0,360°].
11) -3, -4

Two forces act at a point in the plane. The angle between the two forces is given. Find the magnitude of the resultant
force.
12) forces of 52 and 54 newtons, forming an angle of 90°

Solve the problem.

13) A hot-air balloon is rising vertically 12 ft/sec while the wind is blowing horizontally at 5 ft/sec. Find the angle
that the balloon makes with the horizontal.

14) A pilot wants to fly on a bearing of 65.3°. By flying due east, he finds that a 50-mph wind, blowing from the
south, puts him on course. Find the ground speed of the plane.

15) A box weighing 80 lb is hanging from the end of a rope. The box is pulled sideways by a horizontal rope with
a force of 24 lb. What angle, to the nearest degree, does the first rope make with the vertical?

Find the product. Write the answer in standard form.

16) 3i(3 + 6i)2

Simplify the power of i.

17) i79

Page 39
Find the following quotient, and write the quotient in rectangular form, using exact values.
12cis 158°
18)
3cis 38°

Find the given power. Write answer in rectangular form.

19) (1 + i)20

Find all specified roots.

20) Fifth roots of 1.

Find all solutions of the equation. Leave answers in trigonometric form.

21) x5 - 32 = 0

For the given rectangular equation, give its equivalent polar equation.
22) x - y = 10

23) x2 + y2 = 64

Find a rectangular equation for the plane curve defined by the parametric equations.
24) x = t + 4, y = t 2

25) x = 5 tan t, y = 4 cot t

Page 40
Answer Key

1) 10
22) r =
y cos θ - sin θ
12
23) r = 8
24) y = x2 - 8x + 16
8 20
25) y =
x

10 20 30 40 50 60 x

2)
2.5 y

2
1.5
1
0.5

-16 -12 -8 -4 4 8 12 16 x
-0.5
-1
-1.5
-2
-2.5
3) B = 20.3°, C = 61.7°, c = 38.6 ft
4) A = 45°, C = 75°, c = 5 3 + 5
5) A1 = 42°25ʹ, B1 = 102°05ʹ, b1 = 27.20;
A2 = 137°35ʹ, B2 = 6°55ʹ, b2 = 3.35
6) A = 41.14°, B = 51.24°, C = 87.62°
7) c = 12.7 m, A = 22.7°, B = 44.8°
8) 889 mi
9) 8
10) 19.2°
11) 5; 233.1°
12) 75 newtons
13) 67.4°
14) 120 mph
15) 17°
16) -108 - 81i
17) -i
18) - 2 + 2i 3
19) -1024
2π 4π 6π 8π
20) 1, cis , cis , cis , cis
5 5 5 5
2π 4π 6π 8π
21) 2 cis 0, 2 cis , 2 cis , 2 cis , 2 cis
5 5 5 5

Page 41
 Cumulative Review of Trigonometry
Final review
1) Find the supplement of an angle whose measure is 37°45 2

2) 139°47 + 108°48

Convert the angle to decimal degrees and round to the nearest hundredth of a degree.
3) 45°31 46

Convert the angle to degrees, minutes, and seconds.

4) 28.34°

Find the angle of smallest possible positive measure coterminal with the given angle.
5) 840°

Suppose that is in standard position and the given point is on the terminal side of .
6) (6, 8); Find cos .

The triangles are similar. Find the missing side, angle or value of the variable.
7) x

a = 30 b = 90 c = 52

Evaluate the expression.

8) sin(-180°)

9) sec 270°

Identify the quadrant for the angle satisfying the following conditions.
10) sin > 0 and cos < 0

11) sec < 0 and tan <0

Use the fundamental identities to find the value of the trigonometric function.
2
12) Find csc , given that sin = - and is in quadrant IV.
3

Without using a calculator, give the exact trigonometric function value with rational denominator.
13) cos 60°

14) sec 45°

Page 42
Suppose ABC is a right triangle with sides of lengths a, b, and c and right angle at C. Find the unknown side length .
15) Find csc A when b = 8 and c = 17

Find a solution for the equation. Assume that all angles are acute angles.
16) sin(2 + 15°) = cos(3 - 25°)

17) A fire is sighted due west of lookout A. The bearing of the fire from lookout B, 5.1 miles due south of A, is N 48°
22'W. How far is the fire from B (to the nearest tenth of a mile)?

Convert the degree measure to radians. Leave answer as a multiple of .

18) 330°

Convert the radian measure to degrees. Round to the nearest hundredth if necessary.
19) -
5

Find the exact value without using a calculator.

3
20) sin
4

21) Find h as indicated in the figure. Round your answer to the hundredths place.

26° 55 55° 53

86 m

Solve the problem.

22) Two wheels are rotating in such a way that the rotation of the smaller wheel causes the larger wheel to rotate.
The radius of the smaller wheel is 6.5 centimeters and the radius of the larger wheel is 17.0 centimeters.
Through how many degrees will the larger wheel rotate if the smaller one rotates 120°?

Graph the function over a one-period interval.

1
23) y = cos 4 x -
2 3

24) y =sin(2x-180)

Graph the function.

1
25) y =tan x -
2 6

Page 43
 26) y = 2 csc 3x + 60

Solve the problem.

27) The temperature in Fairbanks is approximated by

2
T(x) = 37 sin (x - 101) + 25,
365

where T(x) is the temperature on day x, with x = 1 corresponding to Jan. 1 and x = 365 corresponding to Dec. 31.
Estimate the temperature on day 10.

The function graphed is of the form y = a sin bx or y = a cos bx, where b > 0. Determine the equation of the graph.
28)

29)

Verify that each equation is an identity.

30) tan x(csc x - sin x) = cos x

31) (sec - tan )(sec + tan ) = 1

cos
32) sec + tan =
1 - sin

Page 44
 33) A wheel is rotating at 3 radians/sec, and the wheel has a 54-inch diameter. To the nearest foot, what is the speed
of a point on the rim in ft/min?

Identify the quadrant for the angle satisfying the following conditions.
34) sec < 0 and tan < 0

1 1
35) Find cos(s + t) given that cos s = , with s in quadrant I, and sin t = - , with t in quadrant IV.
3 2

Use trigonometric identities to find the exact value.

36) sin 100° cos 40° - cos 100° sin 40°

Find the exact value by using a sum or difference identity.

37) sin 15°

3 13 10
38) Find tan(s - t) given that sin s = - , with s in quadrant IV, and sin t = - , with t in quadrant IV.
13 10

12
39) cos = , sin <0 Find sin(2 ).
13

2
40) cos 2 = and terminates in quadrant III Find cos .
3

Determine all solutions of the equation in radians.

3
41) Find cos , given that cos = - and terminates in 90° < < 180°.
2 5

Find the exact value

1
42) cos arcsin
4

3
43) cos 2arcsin
5

Determine the solution set of each equation in radians (for x) or degrees (for ) to the nearest tenth as appropriate.
44) 2 sin2 x + sin x = 1

Solve the equation for the interval [0, 2 ).

45) sin2 x - cos2 x = 0

Determine the solution set of each equation in radians (for x) or degrees (for ) to the nearest tenth as appropriate.
46) 3 cos2 + 2 cos = 1

Solve the equation for solutions in the interval [0°, 360°). Round to the nearest degree.
47) sin 2 = cos

Page 45
Solve the triangle.
48)

21 m

Find all th epossible values for the missing parts of the triangle.
49) A = 23° a = 35 km b = 55 km

Find the missing parts of the triangle. (Find angles to the nearest hundredth of a degree.)
50) a = 22 ft b = 32 ft c = 43 ft

51) wo airplanes leave an airport at the same time, one going northwest (bearing 135°) at 407 mph and the other
going east at 345 mph. How far apart are the planes after 3 hours (to the nearest mile)?

Find the magnitude and direction angle (to the nearest tenth) for each vector. Give the measure of the direction angle as
an angle in [0,360°].
52) -5, 12

Vector v has the given magnitude and direction. Find the magnitude of the indicated component of v.
53) = 25.9°, v = 85.6 Find the horizontal component of v.

54) Two forces, of 45.2 and 17.0 lb, forming an angle of 141.9°, act at a point in the plane. Find the magnitude of the
resultant force.

Find the quotient. Write the answer in standard form.

9 - 3i
55)
5 - 7i

Write the complex number in trigonometric form r(cos + i sin ), with in the interval [0°, 360°).
56) -6 - 6i 3

Find the given power. Write answer in rectangular form.

57) (1 + i)20

Find all solutions of the equation. Leave answers in trigonometric form.

58) x5 - 243 = 0

59) Find an equivalent equation in rectangular coordinates. r = cos

Find a rectangular equation for the plane curve defined by the parametric equations.
60) x = t2 + 1, y = t2 - 1

Page 46
 Answers:

1) 142°14 58 24)
2) 248°35
3) 45.53°
4) 28°20 24
5) 120°
3
6)
5
7) x = 39
8) 0
9) Undefined
10) Quadrant II
11) Quadrant II
3
12) - 25)
2
1
13)
2
14) 2
17
15)
15
16) 20°
17) 7.7 mi
11
18)
6
19) -36°
2
20) 26)
2
21) 66.55 m
22) 45.88°
23)

27) -12°
1
28) y = 4 sin x
3
1
29) y = -4 cos x
2

Page 47
 sin x 53) 77.0
30) tan x(csc x - sin x) = tan x · csc x - tan x · sin x =
cos x 54) 33.5 lb
33 24
1 sin x 1 sin2 x 1 - sin2 x 55) + i
· - · sin x = - = = 37 37
sin x cos x cos x cos x cos x
56) 12(cos 240° + i sin 240°)
cos2 x 57) -1024
= cos x
cos x 2 4 6 8
58) 3 cis 0, 3 cis , 3 cis , 3 cis , 3 cis
31) (sec - tan )(sec + tan ) = sec2 - tan2 = 1 5 5 5 5
1 sin 1 + sin 1 + sin 59) x2 + y2 = x
32) sec + tan = + = = ·
cos cos cos cos 60) y = x - 2, x 1
1 - sin 1 - sin2 cos2
= = =
1 - sin cos (1 - sin ) cos (1 - sin )
cos
1 - sin
33) 405 ft/min
34) Quadrant II
3+2 2
35)
6
3
36)
2
6- 2
37)
4
7
38) -
9
120
39) -
169
30
40) cos =-
6
5
41)
5
15
42)
4
4 3-3
43)
10
5 3
44) + 2n , + 2n , + 2n
6 6 2
3 5 7
45) ,,,
4 4 4 4
46) {70.5° + 360°n, 180° + 360°n, 289.5° + 360°n}
47) {30°, 90°, 150°, 270°}
48) C = 103°, a = 9.4 m, b = 16.7 m
49) B1 = 38°, C1 = 119°, c1 = 78 km
B2 = 142°, C2 = 15°, c2 = 23 km
50) A = 29.76°, B = 46.22°, C = 104.02°
51) 2085 mi
52) 13; 112.6°
Page 48
Articles

Page 49
The Journal of Experimental Biology 211, 467-481
Published by The Company of Biologists 2008
doi:10.1242/jeb.008573

Angular momentum in human walking

Hugh Herr1,2,* and Marko Popovic1
1 2
The MIT Media Laboratory and The Harvard-MIT Division of Health Sciences and Technology, 20 Ames Street, Cambridge,
MA 02139, USA
*Author for correspondence (e-mail: [email protected])

Accepted 2 December 2007

SUMMARY
Angular momentum is a conserved physical quantity for isolated systems where no external moments act about a bodyʼs center
of mass (CM). However, in the case of legged locomotion, where the body interacts with the environment (ground reaction forces),
there is no a priori reason for this relationship to hold. A key hypothesis in this paper is that angular momentum is highly
regulated throughout the walking cycle about all three spatial directions [兩L(t)兩≈0], and therefore horizontal ground reaction forces
and the center of pressure trajectory can be explained predominantly through an analysis that assumes zero net moment about
the bodyʼs CM. Using a 16-segment human model and gait data for 10 study participants, we found that calculated zero-moment
forces closely match experimental values (Rx2=0.91; Ry2=0.90). Additionally, the centroidal moment pivot (point where a line parallel
to the ground reaction force, passing through the CM, intersects the ground) never leaves the ground support base, highlighting
how closely the body regulates angular momentum. Principal component analysis was used to examine segmental contributions
to whole-body angular momentum. We found that whole-body angular momentum is small, despite substantial segmental
momenta, indicating large segment-to-segment cancellations (~95% medio-lateral, ~70% anterior–posterior and ~80% vertical).
Specifically, we show that adjacent leg-segment momenta are balanced in the medio-lateral direction (left foot momentum cancels
right foot momentum, etc.). Further, pelvis and abdomen momenta are balanced by leg, chest and head momenta in the
anterior–posterior direction, and leg momentum is balanced by upper-body momentum in the vertical direction. Finally, we
discuss the determinants of gait in the context of these segment-to-segment cancellations of angular momentum.
Key words: biomechanics, biped, locomotion, angular momentum, human.

INTRODUCTION Based on pilot data from a single human participant, he argued that
The advancement of a comprehensive model of human walking the arms reduced both angular momentum and rotation about both
is a formidable task and a critical research objective in the fields vertical and medio-lateral (left–right) axes.
of biomechanics, neural science and legged machine control. Following Elftman’s findings in the late 1930s, it was not until
Although many walking studies have put forth experimental and the turn of the century that additional research was conducted in
theoretical descriptions of center of mass (CM) mechanics and the area of human walking angular momentum behaviors. Xu and
energetics (Saunders et al., 1953; Bekker, 1956; Alexander, Wang (Xu and Wang, 1998) quantified angular momenta for lower-
1976; Cavagna et al., 1976; Margaria, 1976; Mochon and extremity segments for altering direction during walking, and
McMahon, 1980a; Mochon and McMahon, 1980b; McGeer, Simoneau and Krebs (Simoneau and Krebs, 2000) studied whole-
1990; Lee and Farley, 1998; Croce et al., 2001; Kuo, 2002; body angular momentum in elderly participants in an attempt to
Ortega and Farley, 2005; Geyer et al., 2006; Srinivasan and quantify balance deficiencies in the elderly population. More
Ruina, 2006), surprisingly few investigations have specifically recently, a pilot study on a single study participant found that
focused upon whole-body rotational behavior. Clearly, a whole-body angular momentum is highly regulated about all three
comprehensive understanding of human walking would require spatial directions in walking, not deviating substantially from zero
descriptions of not only global body translations but also throughout each phase of gait (Popovic et al., 2002; Gu, 2003;
rotations. The objective of this investigation was to study the Popovic et al., 2004a).
rotational behavior of human steady-state walking through the Although angular momentum behaviors have been studied for
characterization of whole-body angular momentum, as well as human walking, the studies have been limited to a single study
body segment momenta, computed about the body’s CM. participant and often a single walking step. In this study we
The preponderance of research into human angular momentum examined angular momentum behaviors of 10 study participants
behaviors has focused not on walking but on other movement tasks walking at self-selected speeds. Motivated by the findings of
such as sit-to-stand maneuvers (Riley et al., 1997), running previous pilot investigations that showed a relatively small whole-
(Hinrichs et al., 1983; Hinrichs, 1982; Hinrichs, 1987; Hinrichs, body angular momentum, we hypothesized that horizontal ground
1992) and various sporting activities (Frohlich, 1979; Dapena and reaction forces and the center of pressure (CP) trajectory in steady-
McDonald, 1989; Dapena, 1978; Dapena, 1993; LeBlanc and state walking can be explained predominantly through an analysis
Dapena, 1996; King 1999). Specific to walking maneuvers, that assumes zero net moment about the body’s CM. To test the
Elftman (Elftman, 1939) calculated the angular momenta of all hypothesis, we first derived what the horizontal ground reaction
body segments across one walking step, from heel strike to toe-off. force, and CP location, would be if no moments were to act about

THE JOURNAL OF EXPERIMENTAL BIOLOGY Page 50

 H. Herr and M. Popovic

the body’s CM. Using a 16-segment human model and gait data During the walking trials, ground reaction forces were measured
from the 10 study participants, we tested the hypothesis by synchronously with the kinematic data at a sampling rate of
comparing the calculated zero-moment forces and CP trajectory 1080·Hz using two staggered force platforms (model no. 2222 or
with measured values from a force platform. OR6-5-1, Advanced Mechanical Technology Inc., Watertown,
We also examined segmental contributions to whole-body MA, USA) embedded in the walkway. The platforms measured
angular momentum. Motivated by Elftman (Elftman, 1939), we ground reaction force and CP location at a precision of ~0.1·N and
hypothesized that whole-body angular momentum is small ~2·mm, respectively.
throughout the walking gait cycle, despite substantial segmental
momenta, indicating large segment-to-segment cancellations. Human model
Specifically, since the arms and legs alternately protract and retract A human model was constructed in order to calculate physical
within the sagittal plane, we anticipated that adjacent limb segment quantities such as CM position and angular momentum. The model
contributions are effectively balanced in the medio-lateral direction. and coordinate system used in the study are shown in Fig.·1. The
Furthermore, due to pelvic obliquity, where the leg hip that is model comprises 16 rigid body segments: feet, tibias, femurs,
entering the swing phase drops lower than the adjacent leg hip hands, forearms, arms, pelvis-abdomen, chest, neck and head. The
(Saunders et al., 1953), we hypothesized that angular momenta feet and hands were modeled as rectangular boxes. The tibia
contributions of the pelvis and abdomen are balanced by segments, femur segments, forearm segments and arm segments
contributions from the rest of the body in the anterior–posterior were modeled as truncated cones. The pelvis-abdomen and chest
(front–back) direction. Still further, due to pelvic rotation where the segments were modeled as elliptical slabs [ellipses in the horizontal
pelvis and upper body rotate about the vertical axis over the stance (x–y) plane and extruded in the vertical (z) direction]. The neck was
leg in walking (Saunders et al., 1953), we anticipated that leg modeled as a cylinder, and the head was modeled as a sphere. The
angular momentum is balanced by upper-body momentum in the following 28 anthropometric measurements were taken for each
vertical direction. To test these hypotheses, we once again employed study participant to accurately construct a representative model: (1)
the 16-segment human model and gait data measured from the body weight, height, and total leg length measured from the medial
10 study participants. Principal component (PC) analysis was malleolus to the anterior superior iliac spine; (2) lengths, widths
performed on all 16 body segments’ angular momenta to produce and thicknesses of foot and hand segments; (3) segment lengths and
PCs for each of three orthogonal directions. We then calculated their proximal/distal base radii of tibia, femur, forearm and arm; (4)
respective time-dependent weighting coefficients, or tuning heights, widths and thicknesses of chest and pelvis-abdomen
coefficients. Finally, we obtained the amount and source of segments; and (5) radius of the head. The neck radius was set equal
segmental momentum cancellation for all three spatial directions. to half the head radius. The human model had a total of 38 degrees
of freedom, or 32 internal degrees of freedom (12 for the legs, 14
MATERIALS AND METHODS for the arms, and six for the head, neck and trunk) and six external
Experimental procedures degrees of freedom.
Kinetic and kinematic walking data were collected at the Gait For acceptance of the human model, we required that each
Laboratory of Spaulding Rehabilitation Hospital, Harvard Medical segment’s relative mass and density were in reasonable agreement
School, in a study approved by the Spaulding committee on the Use with human morphological data from the literature (Winter, 1990).
of Humans as Experimental Subjects. Ten healthy adult Relative mass was defined as segment mass divided by total body
participants, five male and five female, with an age range from 20 mass, and density as segment mass divided by segment volume. We
to 38·years, volunteered for the study. The participants walked at a accepted a segment design if both its relative mass and density fell
self-selected speed across a 10·m walkway in the Motion Analysis within one standard deviation of the segment’s mean experimental
Laboratory. Participants were timed between two fixed points to values from the literature. When the relative mass of each model
ensure that the same walking speed was used between experimental segment was set equal to each segment’s mean experimental value
trials. Walking speeds within a ±5% interval from the self-selected
speed were accepted. For each study participant, a total of seven
walking trials were collected.
The data collection procedures were based on standard
techniques (Kadaba et al., 1989; Winter, 1990; Kadaba et al., 1990; z z
Kerrigan et al., 2000; Kerrigan et al., 2001). An infrared camera
system (eight cameras, VICON 512 motion analysis system,
Oxford Metrics, Oxford, UK) was used to measure the three- x y
dimensional locations of reflective markers at 120·frames·s–1. A
total of 33 markers were placed on various parts of a participant’s
body: 16 lower-body markers, five trunk markers, eight upper-limb
markers and four head markers. The markers were attached to the
following bony landmarks: bilateral anterior superior iliac spines,
posterior superior iliac spines, lateral femoral condyles, lateral
malleoli, forefeet and heels. Additional markers were rigidly
attached to wands over the mid-femur and mid-shaft of the tibia. Fig.·1. Human model and coordinate frame. The human model has 16
The kinematics of the upper body were also collected with markers segments with 32 internal degrees of freedom. Using human morphological
data from the literature, mass is distributed throughout the model in a
placed on the following locations: sternum, clavicle, C7 vertebra, realistic manner. The coordinate frame is oriented by the right-hand rule
T10 vertebra, head, and bilaterally on the shoulder, elbow and wrist. with the z-axis directed vertically, the y-axis pointing in the direction of the
The VICON 512 system was able to detect marker position with a walking motion (anterior–posterior direction), and the x-axis pointing to the
precision of ~1·mm. right of the participant (medio-lateral direction).

THE JOURNAL OF EXPERIMENTAL BIOLOGY Page 51

 Walking angular momentum
from the literature, model segment density often became abnormal, The projection of the CM onto the horizontal ground surface, or
falling beyond two standard deviations from the experimental mean. xCM and yCM (see Fig.·1), was computed from the human model
In distinction, when the density of each model segment was set equal using Eqn 3, and then compared with the CP location measured
to each segment’s mean experimental value from the literature, directly from the force platform. The separation distance between
model relative mass then became abnormal. As a resolution to this the CM projection on the ground and the CP was ~3·mm. To
difficulty, we performed an optimization where model relative mass determine whether the error changed appreciably for a different
was varied until the error between model and experimental density static pose, we repeated the experiment with one leg retracted
values were minimized. We then confirmed that each segment’s rearward and the second leg protracted forward (comparable to the
relative mass and density fell within one standard deviation of their body’s posture during the double-support phase of walking). Using
experimental means reported in Winter (Winter, 1990). this second pose, the CM model error was still small (~3·mm). At
In detail, the relative mass distribution throughout the model, a self-selected gait speed, the body’s CM oscillates with a peak-to-
MR, described by a 16-component vector corresponding to the 16 peak amplitude of between 4 and 5·cm in the medio-lateral (x)
segments of the model, was modeled as a function of a single direction (Crowe et al., 1995). Thus, the estimated CM model error
parameter ␣ such that: was less than 10% of these oscillations.
MR(␣) = (MRExp+␣VR) / (1+␣) . (1)
Whole-body angular momentum and moment
Here MRExp is a 16-component vector of mean relative mass values Whole-body angular momentum was estimated using the human
obtained from the literature (Winter, 1990), and VR is a 16- model and kinematic gait data. Angular momentum, L, was
component vector of relative volumes computed directly from the calculated as the sum of individual segment angular momenta about
human model. The relative volume of the i-th segment, VRi , was the body’s CM, or:
defined as the ratio of the segment’s volume, V i, over the total 16

⌺ [(r
,
body volume, V, or V Ri =V i/V. By using Eqn 1, total body mass and L= i
CM–rCM) ⫻ mi(vi–vCM) + Iiωi] . (4)
individual segment volumes computed from the model, model i=1
segment densities were computed and represented by a 16-
component vector D(␣). Here the density of the i-th segment was The first term within the square brackets is the angular momentum
defined as Di(␣)=MsubjectM Ri (␣)/V i, where Msubject is total body due to the i-th segment’s CM movement. Here rCM is the CM
mass and V i is the volume of the i-th segment. The final relative position of the entire body defined in Eqn 3, and vCM is the whole-
mass distribution was obtained as MR=MR(␣min) where ␣min body CM velocity in the lab frame. Further, r iCM and vi are the i-th
,
minimized the absolute error between the distribution of segment segment’s CM position and velocity in the lab frame, respectively,
densities, D(␣), and the mean distribution of segment densities and mi is the i-th segment’s mass. The second term within the square
from the literature, DExp. This analysis procedure may be brackets is the angular momentum of the i-th segment about its CM
}
expressed as: position. Here I i and i are the i-th segment’s inertia tensor (3·⫻ 3)
and angular velocity (3·⫻ 1) about the segment’s CM, respectively.
min兩D(␣) – DExp兩 = min
⌺ [D (␣) – D
i
i Exp,i]2 In order to reduce data variance across study participants,
angular momentum was represented in dimensionless form using a
MExp
R + ␣minVR
normalization constant Nsubject, equal to the product of the
c ␣min c MR = . (2) participant’s mass Msubject, CM height Hsubject, and the mean self-
1 + ␣min
selected gait speed Vsubject across seven gait trials, or:
Nsubject = MsubjectVsubjectHsubject . (5)
Whole-body center of mass
The body’s CM location was estimated using the human model and For each participant, the CM height was estimated during upright
joint position data from the motion capture measurements. The CM standing using the human model, the motion capture data, and Eqn
position, rCM, of the entire 16-segment model was calculated as a 3. Angular momentum was computed using Eqn 4 for each gait
sum of the products of the segments’ relative masses and CM cycle and then put into dimensionless form by dividing by the
locations, or: normalization constant Nsubject defined in Eqn 5. Dimensionless
16 angular momentum was then plotted versus percentage gait cycle
rCM =
⌺M r
i=1
i i
R CM . (3) equal to gait time divided by total cycle time. At each percentage
cycle time, the mean and standard deviation of the dimensionless
angular momentum were computed over a total of 70 walking trials
Here M Ri is the relative mass of the i-th body segment, and r CM
i
is (10 participants, 7 gait trials per participant).
the CM location of the i-th body segment relative to the lab frame. An alternative method for computing angular momentum is by
integration of the moment about the CM. We computed angular
CM error estimate momentum in this manner and compared the result to the angular
To estimate the error in the CM calculation, we first collected momentum estimate of Eqn 4. We found little difference between
kinematic data from the aerial phase of running and then, using Eqn these two estimates (R2 values of 0.97, 0.96 and 0.98 for Lx, Ly and
3, estimated the body’s aerial phase CM trajectory. We found good Lz, respectively). We preferred computing angular momentum
agreement between this estimated CM trajectory and a ballistic directly from kinematics data because a single methodology could
trajectory (R2=0.99; see Eqn 11 for R2 definition). It was also noted then be used when estimating both whole-body angular momentum
that, during the aerial phase, the maximal distance error between and individual segment momenta in walking. The topic of
these trajectories was less than 2·mm. As an additional check of individual segment angular momenta is addressed in the subsequent
CM error, we first collected kinetic and kinematic data while a Materials and methods section entitled ‘Segmental contributions to
participant stood on the force platform in a static standing pose. whole-body angular momentum’.

THE JOURNAL OF EXPERIMENTAL BIOLOGY Page 52

 H. Herr and M. Popovic

Angular momentum error estimate where Fz is the measured vertical ground reaction force, and Mx
To estimate the error in the angular momentum calculation, we first and My are horizontal moments measured about that same lab
collected kinematic data from the aerial phase of running where reference point.
angular momentum is a conserved quantity (assuming air drag Eqn 6 can be solved for the horizontal ground reaction forces,
exerts a negligible moment). From the flight phase kinematic data or:
and Eqn 4, the angular momentum vector for the aerial phase was
obtained, and one standard deviation about the mean value was ⎧ Fz ⎫ ⎧ Ty ⎫
assigned to be the model error for each spatial direction. To Fx = ⎨ (xCM–xCP) ⎬ + ⎨ – ⎬, (9)
quantify its relative size, model error was then compared with the ⎩ zCM ⎭ ⎩ zCM ⎭

⎧
⎨
⎩
⎧
⎨
⎩
maximum angular momentum value found during the walking
cycle about each spatial direction. Using walking data from the FxZero-moment FxMoment
same study participant that participated in the running experiments, ⎧ Fz ⎫ ⎧ Tx ⎫
we first calculated the mean angular momentum curve for each Fy = ⎨ (yCM–yCP) ⎬ + ⎨ ⎬ , (10)
spatial direction (n=7 walking trials). The maximum angular ⎩ zCM ⎭ ⎩ zCM ⎭
momentum values from the mean curves were then compared with

⎧
⎨
⎩
⎧
⎨
⎩
the model errors for the three orthogonal directions. We found the FyZero-moment FyMoment
angular momentum errors were 1.7%, 4.2% and 10% of the
maximum angular momentum values in the medio-lateral (x), where Tx and Ty are the CM moments in the medio-lateral (x) and
anterior–posterior (y) and vertical (z) directions, respectively (see anterior–posterior (y) directions, respectively. Throughout this
Fig.·1 for coordinate frame specifications). manuscript we refer to the first and second terms on the right-hand
In addition to angular momentum, CM moment T was sides of Eqns 9 and 10 as the zero-moment and moment force
estimated by taking the rate of change of angular momentum contributions1 to the horizontal ground reaction forces, respectively
at each percentage cycle time. Moment was then put into (Popovic et al., 2005). To evaluate the hypothesis that horizontal
dimensionless form using the scaling factor MsubjectGHsubject, ground reaction forces can be explained predominantly through a
where G is the gravitational constant. Similar to the angular zero-moment analysis, zero-moment forces FxZero-moment and
momentum data analysis procedure, dimensionless CM FyZero-moment were compared with the actual horizontal ground
moment was plotted versus percentage gait cycle, and at each reaction forces measured from a force platform. As defined by Eqns
percentage cycle time the mean and standard deviation were 9 and 10, these zero-moment forces were obtained using the
computed over a total of 70 walking trials. calculated position of body CM (Eqn 3), the experimentally
measured CP, and the experimentally determined vertical ground
Horizontal ground reaction force predictions reaction force.
A key hypothesis in this paper is that angular momentum is highly To assess the amount of agreement between zero-moment model
regulated in steady-state human walking about all three orthogonal forces and experimentally measured horizontal forces, we used the
directions [兩L(t)兩≈0], and therefore horizontal ground reaction forces coefficient of determination, R2, where R2=1 only if there is a
can be explained predominantly through an analysis that assumes perfect fit and R2=0 indicates that the model’s estimate is worse
zero net moment about the body’s CM. To test this hypothesis, we than using the mean experimental value as an estimate. More
first derived a relationship between horizontal ground reaction force, specifically, R2 was defined as:
whole-body CM, and CP consistent with zero net moment. We then NTrial NPercent
compared the predicted zero-moment forces with ground reaction
forces measured directly from a force platform. ⌺ ⌺(F
i=1 j=1
ij
Exp–FMod)
ij 2

The horizontal component (hor) of the total moment about 1– , (11)

r r NTrial NPercent

⌺ ⌺ (F
the CM [T兩hor=(Tx,Ty)=(T CM·i ,TCM·j )] may be expressed as: ij 2
Exp–FExp)
dL i=1 j=1
T 兩hor = [(rCP–rCM) ⫻ F]hor = 兩hor , (6) ij ij
dt where F Exp and F Mod are the forces taken at the j-th percentage gait
cycle of the i-th trial for the experimental data and model-predicted
where F is the ground reaction force, and rCP is the CP location on data, respectively. Before computing R2 values for each spatial
the ground surface. The CP ground reference point is frequently direction and study participant, both experimental and zero-
used in the study of human gait and postural balance (Winter, 1990; moment forces for the medio-lateral (x) and anterior–posterior (y)
Rose and Gamble, 1994). For a body in contact with the force directions were plotted versus percentage gait cycle (equal to gait
platform, the position of the CP, measured relative to a lab frame time divided by total cycle time). We then computed medio-lateral
reference point located on the force platform walking surface, is (x) and anterior–posterior (y) R2 values for each participant by
calculated as: summing over all walking trials (NTrial=7) and gait percentage times
analyzed (NPercent=100).
My
xCP = – (7) In Eqn 11, experimental mean FExp is the grand mean over all
Fz walking trials and gait percentage times analyzed, or:
NTrial NPercent

⌺ ⌺F
and 1
FExp = ij
Exp . (12)
NTrialNPercent i=1 j=1
Mx
yCP = , (8)
Fz
1
It is noted that moment as used here refers to horizontal moment and not
vertical moment.

THE JOURNAL OF EXPERIMENTAL BIOLOGY Page 53

 Walking angular momentum
We used a non-parametric Wilcoxon matched-pairs test (Wilcoxon, When the CMP departs from the CP, there exist non-zero
1945) for comparing the two R2 values corresponding to the medio- horizontal CM moments, causing variations in whole-body angular
lateral (x) and anterior–posterior (y) directions. This statistical momentum (see Fig.·2A). While by definition the CP cannot leave
significance test was performed to look for differences in the the ground support base2, the CMP can – but only when horizontal
capacity of the zero-moment force model to predict experimental moments act about the CM. In this investigation, the CMP was
force equally well in the two horizontal directions. calculated using Eqn 14 and 15, the calculated CM position from
the human model, and the ground reaction force measured from
Center of pressure predictions the force platforms. We then computed the mean separation
An alternative strategy for quantifying the degree to which whole- distance between the CMP and CP points, normalized by foot
body angular momentum is regulated in walking is to estimate a length, across the entire gait cycle. If angular momentum is highly
zero-moment CP position. This ground reference point, previously regulated in walking and CM moments are small, we anticipate
defined in the robotics literature, is called the centroidal moment that the CMP location should coincide with the experimentally
pivot (CMP) (Herr et al., 2003; Hofmann, 2003; Popovic et al., measured CP position (see Fig.·2B).
2004a; Popovic et al., 2005; Goswami and Kallem, 2004). The
CMP location, rCMP, is defined as the point where a line parallel Segmental contributions to whole-body angular momentum
to the ground reaction force, passing through the CM, intersects Motivated by Elftman (Elftman, 1939), in this paper we tested the
with the ground surface. As its name implies, when the CMP hypothesis that whole-body angular momentum is small throughout
coincides with the CP, no horizontal moments act about the body’s the walking cycle, despite substantial segmental momenta,
CM. In distinction, when these ground reference points diverge, indicating large segment-to-segment cancellations. To investigate
non-zero horizontal CM moments exist. To further quantify the segmental movement correlations in connection with angular
whole-body rotational dynamics, we compare a calculated CMP momentum, we used principal component (PC) analysis. We first
trajectory with an experimentally measured CP trajectory from the obtained the segmental angular momentum PCs together with the
force platforms. amount of data explained by each PC. We then calculated their
The CMP can be expressed mathematically by requiring that the respective weighting coefficients, or tuning coefficients. Finally,
cross-product of the CMP–CM position vector and the ground we obtained the level of momentum cancellation between body
reaction force vector vanishes, or: segments for all three spatial directions, and the strategy employed
by the body to achieve that level of cancellation.
[(rCMP–rCM) ⫻ F]hor = 0 . (13)
By expanding this cross-product, the CMP location can be written Principal component analysis
in terms of the CM location and the ground reaction force, or: PC analysis (e.g. Jackson, 1991) was performed on all segmental
angular momenta, for each of the three spatial components, to
Fx produce PCs. Each PC was a 16-component unit vector, Pji,
xCMP = xCM – zCM (14)
Fz corresponding to the 16 body segments of the human model. Here
Pji represents the i-th PC in the j-th direction. Vector components Pjiq
and denoted the relative contributions of the q-th body segment to Pji.
As is customary, each PC was assigned a value for the percentage
Fy of data explained, DEij, where j=1…3 denotes the number of spatial
yCMP = yCM – zCM . (15)
Fz components, and i=1…N corresponds to the total number of PCs
(equal to the number of human model segments, i.e. N=16).
The PC vectors and the corresponding percentages of data
explained were obtained using the MATLAB statistical toolbox
(MathsWorks Inc., Natick, MA, USA). The 16-component angular
A B momentum vector was compactly represented as:
N

⌺ C (t) · P ,
r
Λj(t) = i
j j
i
(16)
CM CM i=1

F where Cji(t), t苸(0%, 100%) are time-dependent tuning coefficients.

The components of the momentum vector defined in Eqn 16
F correspond to the 16 segments of the human model.

Normalized tuning coefficients

In order to extract directional dependence, we introduced the
CP CMP CP=CMP normalized tuning coefficients cji(t), such that:

Fig.·2. Centroidal moment pivot (CMP). The CMP is the point where the ⎧ N ⎫ N

⌺ ⌺ c (t) · P ,
r
ground reaction force would have to act to keep the horizontal component Λj(t) = ⎨ [Cji(t)]2 ⎬ i
j j
i
(17)
of the whole-body angular momentum constant. When the moment about ⎩ ⎭
i=1 i=1
the center of mass (CM) is zero (B), the CMP coincides with the center of
pressure (CP). However, when the CM moment is non-zero (A), the extent 2
When in single support, the support base is the outline of the part of the stance
of separation between the CMP and CP is equal to the magnitude of the foot that is actually in contact with the ground. When in double support, where
horizontal component of moment about the CM, divided by the normal both feet are on the ground, the support base is the smallest convex shape that
component of the ground reaction force. includes all points where both feet are in contact with the ground.

THE JOURNAL OF EXPERIMENTAL BIOLOGY Page 54

 H. Herr and M. Popovic

where the expression in the brackets represents the magnitude of the To provide the reader with a better understanding of the relative
angular momentum vector in the N-dimensional space. It is noted size of these measured human values, we computed the angular
here that the time-dependent normalized tuning coefficients satisfy: momentum about the CM of single-segment, rigid-body models. In
N
the vertical (z) direction, we computed the angular momentum

⌺ 兩c (t)兩 = 1 ,
i
j = 1...3 , t 苸 (0%,100%) . about the CM of a rigid body rotating about a stationary vertical
2 A
j (18)
axis passing through the stance foot with an angular velocity equal
i=1
to Vsubject/W, where W is half the foot separation distance in the
Analysis was performed to find the smallest number of angular medio-lateral (x) direction during quiet standing. The angular
momentum PCs necessary to explain greater than 90% of the momentum, scaled by MsubjectVsubjectHsubject, for this simple
segmental angular momentum data. To determine the minimum comparison case is then equal to Izz/(MsubjectHsubjectW). Using the
number of PCs to capture both inter- and intrasubject variability, human model and kinematic data from the 10 study participants,
both participant-dependent and participant-independent analysis we computed the average Izz value during the single-support phase
methods were performed. For the participant-dependent method, PC for all 10 participants. Using this value, the normalized angular
analysis was performed for each participant across seven walking momentum for the rotating rigid body was equal to ~0.05, 5-fold
trials, whereas for the participant-independent method, PC analysis larger than our measured human angular momentum value of 0.01
was conducted across all participants and walking trials (70 trials). in the vertical (z) direction.
In the medio-lateral (x) direction, the normalized angular
Segmental angular momentum cancellation momentum about the CM of a physical inverted pendulum falling
The participant-dependent PCs were used to estimate the amount forward while rotating about a stationary rotational axis at the
2
of segmental momentum cancellation for each participant and j-th ground surface is Ixx/(MsubjectH subject), assuming an angular velocity
spatial direction, or: equal to Vsubject/Hsubject and a moment of inertia about the CM equal
to Ixx. Once again, using the human model and kinematic data from

⎡⎢ ⌺ ⎤⎥
2
16 the 10 study participants, we computed the average Ixx value during
NE Pjiq the single-support phase for all 10 participants. Using this value,
⌺ ⎢⎢ ⌺ ⎥⎥
Sj = 1 – DEji q=1
, (19) the normalized angular momentum for the physical inverted

⎣ ⎦
i=1
16 pendulum falling forward was equal to ~0.2, 4-fold larger than our
兩Pjiq兩 measured human angular momentum value of 0.05 in the medio-
q=1 lateral (x) direction.
The authors cannot think of a simple comparison case for the
where q=1…N and where NE=4 sufficed for the level of precision peak angular momentum in the anterior–posterior (y) direction.
in our study. The net cancellation was therefore expressed as a sum Thus, the normalized peak human value of 0.03 in the
of squares (i.e. treated as independent variables or orthogonal anterior–posterior (y) direction can be compared with both the
directions) of cancellations, across the largest PCs, weighted by medio-lateral (x) and the vertical (z) single-segment, rigid-body
their respective data explained. values; the vertical (z) rigid body value of 0.05 is 1.7-fold larger,
To test whether the amount of angular momentum cancellation and the medio-lateral (x) value of 0.2 is over 6-fold larger, than the
for all 10 participants across the three spatial directions was human value of 0.03.
sampled from the same distribution, we used a non-parametric
Friedman ANOVA test (Friedman, 1937; Friedman, 1940). This CM moment estimations
statistical significance test was performed to look for differences in Moment curves, scaled by MsubjectGHsubject, are shown in Fig.·3B
the amount of angular momentum cancellation across the three versus percentage gait cycle. Throughout the gait cycle, the
orthogonal directions, or vertical (z), anterior–posterior (y) and absolute value of the normalized CM moment mean, plus one
medio-lateral (x). Two types of non-parametric post hoc tests were standard deviation, remains smaller than 0.07, 0.03 and 0.014
independently performed to compare cancellation for pairs of dimensionless units in the medio-lateral (x), anterior–posterior (y)
spatial directions. These tests were the Dunn procedure with and vertical (z) directions, respectively.
Wilcoxon test (Wilcoxon, 1945; Dunn, 1964) and the minimum
significant difference (Portney and Watkins, 2000). Horizontal ground reaction force and CP predictions
In Fig.·4A,B, zero-moment and experimental forces are plotted
RESULTS versus percentage gait cycle for the medio-lateral (x) and
Whole-body angular momentum and moment anterior–posterior (y) directions, respectively. Plotted data are for
Angular momentum estimations one representative participant (participant no. 1 in Table·1) and
To quantify whole-body rotational behavior during steady-state experimental trial. Mean R2 values for each participant, and across
walking, angular momentum was computed from kinematic gait all participants, are listed in Table·1. Across all study participants,
data, as defined in Eqn 4. Angular momentum curves, scaled by the mean R2 value is 0.91±0.06 in the medio-lateral (x) direction
MsubjectVsubjectHsubject, are shown in Fig.·3A versus percentage gait and 0.90±0.03 in the anterior–posterior (y) direction. The high R2
cycle. Throughout the gait cycle, the absolute value of the values indicate good agreement between zero-moment force
normalized angular momentum mean, plus one standard deviation, predictions and experimental force values. No significant
remains smaller than 0.05, 0.03 and 0.01 in the medio-lateral (x), difference was observed between the distributions of R2 values for
anterior–posterior (y) and vertical (z) directions, respectively3. these two spatial directions (P=0.267), indicating that the zero-
moment model has an equal capacity to predict horizontal ground
3
The angular momentum curves shown in Fig. 3A agree well with the reaction forces independent of horizontal direction.
measurements of Elftman (Elftman, 1939), in terms of overall curve shape, as
well as peak momentum values in the medio-lateral (x), anterior–posterior (y) and Additionally we calculated the CMP using Eqn 14 and 15, and
vertical (z) directions. then compared the values with the experimentally measured CP

THE JOURNAL OF EXPERIMENTAL BIOLOGY Page 55

 Walking angular momentum

A B

0.02 0.05

Tx/(MGH)
0
Lx/(MVH)

–0.02
–0.05
–0.04

0.02
0.02

Ty/(MGH)
Ly/(MVH)

0 0

–0.02 –0.02

0.01 0.01

Tz/(MGH)
Lz/(MVH)

0 0

–0.01 –0.01
0 20 40 60 80 100 0 20 40 60 80 100
Gait cycle (%) Gait cycle (%)

Fig.·3. Whole-body angular momentum and moment. (A) A normalized angular momentum for walking is plotted about three orthogonal directions versus
percentage gait cycle. The angular momentum is normalized by the product of each participantʼs mass, CM height and self-selected gait speed (MVH; see
Table·1 for values). (B) Normalized CM moment is plotted about three orthogonal directions versus percentage gait cycle. Moment is normalized by the
product of each participantʼs weight and CM height (MGH). For both A and B, the solid line is the mean normalized value, and the dashed lines are one
standard deviation about the mean (10 participants and seven walking trials per participant). In addition, 0% and 100% gait cycles correspond to
consecutive heel strikes of the same foot.

from a force platform. In Fig.·4C, the CP, CMP and CM ground CMP and CP points, normalized by foot length, across the entire
projection locations are plotted. Again, plotted data are for one gait cycle, or ␤%. The mean normalized distance across
representative participant (participant no. 1 in Table·1) and participants is small (␤=14±2%), indicating good agreement
experimental trial. Table·1 includes the mean distance between the between the CMP and CP points in steady-state walking. Finally,
for all participants and walking trials, the CMP remains within the
Table 1. Body mass Msubject, CM height Hsubject and self-selected ground support base throughout the walking cycle, indicating how
walking speed Vsubject for each study participant closely the human body regulates angular momentum in walking.
Msubject Hsubject Vsubject
Segmental contributions to whole-body angular momentum
Participant (kg) (m) (m·s–1) Rx2 Ry2 ␤%
For the participant-dependent analysis approach, the data explained
1 50.1 0.88 1.30 0.97 0.94 16 by the first PC, as well as the first three PCs combined, are shown
2 62.7 0.94 1.38 0.97 0.90 13 in Table·2. On average across all study participants, the first three
3 81.9 1.08 1.32 0.96 0.92 12
4 73.9 1.01 1.18 0.95 0.85 16
PCs accounted for 98±1%, 92±2% and 95±1% in the medio-lateral
5 49.9 0.88 1.06 0.95 0.90 15 (x), anterior–posterior (y) and vertical (z) directions, respectively.
6 57.2 0.92 1.40 0.95 0.87 10 Thus, the participant-dependent PC analysis performed on each
7 82.6 1.08 1.24 0.91 0.88 12 participant’s trial data reveals that only three PCs are necessary to
8 64.6 1.03 1.41 0.82 0.84 12 explain greater than 90% of segmental momentum data. In Fig.·5,
9 65.3 0.99 1.40 0.83 0.95 15 the average participant-dependent first PC is shown for all three
10 76.8 1.06 1.33 0.81 0.91 15
spatial directions. Standard deviation error bars are included to
Mean ± s.d. 66±12 1.0±0.1 1.3±0.1 0.91±0.06 0.90±0.03 14±2 quantify the level of variability in the first PC across study
2
Also listed are the mean coefficients of determination R between the zero- participants. The greatest variability in the segmental momentum
moment force curve and the experimentally measured horizontal ground distribution was found to occur in the coronal plane (x–z plane in
reaction force for medio-lateral (Rx2) and anterior-posterior (Ry2) directions Fig.·1).
(see Eqn 11), and the mean distance between the CMP and CP points The results of participant-independent PC analysis performed
across the entire gait cycle normalized by foot length for each study simultaneously on all participants and trials (total of 70 trials) are
participant (␤%) is given.
shown in Figs·6 and 7. In Fig.·6, the angular momentum PCs, that

THE JOURNAL OF EXPERIMENTAL BIOLOGY Page 56

 H. Herr and M. Popovic

found that whole-body angular momentum is small, despite

substantial segmental momenta, indicating large segment-to-
segment cancellations: 95% medio-lateral (x) cancellation, 69%
A C –100 anterior–posterior (y) cancellation and 77% vertical (z)
40
cancellation. We found no significant difference between the
–200
20 amount of momentum cancellation in the anterior–posterior (y) and
vertical (z) directions (P=0.19). However, we did find a significant
Fx (N)

Anterior–posterior direction (mm)

0 –300
difference between the amount of momentum cancellation in the
–20 –400 medio-lateral (x) direction compared with the anterior–posterior (y)
direction (P=0.001) and the vertical (z) direction (P=0.002).
–40
–500 In Fig.·7, the mean normalized tuning coefficients, as defined in
Eqn 17, are shown for three spatial directions. The tuning coefficients
B 150 –600 correspond to the PCs plotted in Fig.·6, and define the relative
100 dominance of each PC as a function of gait cycle. In the subsequent
–700 sections, we use the PCs, and their respective tuning coefficients, to
50
Fy (N)

0
describe segment-to-segment momentum cancellations.
–800
–50 Segmental cancellation in the medio-lateral direction
–900
–100 The most dominant first PC in the medio-lateral (x) direction shows
–150 –1000 that adjacent leg-segment momenta oppose one another (the left
0 10 20 30 40 50 0 100 200 300
Gait cycle (%) Medio-lateral direction (mm)
foot momentum cancels the right foot momentum, etc.; see Fig.·6).
Further, the first PC shows that arm, abdomen, pelvis, chest, neck
Fig.·4. Horizontal ground reaction force and CP predictions. (A,B) The and head momenta are negligible. However, as the first tuning
horizontal ground reaction forces in walking are plotted versus percentage coefficient in the medio-lateral (x) direction shows in Fig.·7, the
gait cycle in the medio-lateral (x) and anterior–posterior (y) directions, first PC becomes less dominant during the powered plantar flexion
respectively. The thick red line is the calculated zero-moment force (see phase, from 43% to 62% gait cycle, and the second PC increases
Eqns 9 and 10), and the thin blue line is the force measured experimentally in dominance.
using force platforms. (C) Plotted are the CP (blue dashed line), CMP (red
For the second PC, cancellation occurs within each leg, in
solid line) and CM ground projection (green dash-dotted line) trajectories
and corresponding footprints. The two circles on each line denote the contrast to the first PC where cancellation occurs from leg segment
transition from single to double support, and vice versa. In all plots, only to adjacent leg segment. For the trailing limb, foot and calf
half the gait cycle is shown. Data span from the middle of a single-support momenta oppose thigh momentum during ankle-powered plantar
phase (0% gait cycle) to the middle of the next single-support phase of the flexion, pre-swing knee flexion, and pre-swing hip flexion. For the
opposite limb (50% gait cycle). Data shown are for one representative forward limb, foot and calf momenta oppose thigh momentum
participant and trial (participant no. 1 in Table·1).
during ankle-controlled plantar flexion, early stance knee flexion
and early stance hip flexion.

when combined account for more than 90% of the experimental Segmental cancellation in the anterior–posterior direction
data, are shown for three spatial directions. While only three PCs The most dominant first PC in the anterior–posterior (y) direction
are needed to explain more than 90% of the data in sagittal (y–z) shows that foot, calf, chest and head momenta oppose the abdomen
and transverse (x–y) planes, four PCs are needed in the coronal and pelvis momenta, and further that arm, thigh and neck momenta
(x–z) plane. are negligible (see Fig.·6). However, as the first tuning coefficient
The percentage of segmental momentum cancellation (see Eqn in the anterior–posterior (y) direction shows in Fig.·7, the first PC
19) per participant and spatial direction is listed in Table·2. We is most dominant during the double-support phase of walking, from
0% to 12% gait cycle and from 50% to
Table 2. The data explained (DE) by the first principal component (PC), as well as the first 62% gait cycle. During single support, the
three PCs combined, listed for each participant and spatial direction first PC is not as dominant, and the second,
Participant DEx1
DEy1
DEz1
DEx III
DEyIII
DEzIII
Sx Sy Sz
third and fourth PCs increase in
dominance. It is noted here that for the
1 87 73 86 99 93 94 96 92 73 second PC, explaining 29% of the data, no
2 89 75 82 98 92 94 96 92 74
3 86 49 85 98 91 95 93 64 63
dominant segmental cancellation occurs.
4 86 66 89 99 94 95 93 52 80
5 87 67 89 99 90 95 93 89 81 Segmental cancellation in the vertical
6 90 65 90 99 94 95 97 74 79 direction
7 86 57 87 98 93 95 95 71 79 The first PC in the vertical (z) direction
8 83 78 85 97 92 93 93 84 94 shows that leg angular momenta oppose
9 88 65 89 99 90 97 96 44 69
the remaining body segment momenta of
10 87 85 88 98 95 97 95 21 78
the arms, pelvis, abdomen, chest, neck and
Mean ± s.d. 87±2 68±10 87±2 98±1 92±2 95±1 95±2 69±23 77±8 head. However, the first PC becomes less
DEx1 (%) denotes the percentage of data explained by the first PC in the x-direction, and (%) DEyIII dominant during the powered plantar
denotes the data explained by the first three PCs combined in the x-direction. Also listed is the flexion phase, from 43% to 62% gait cycle,
percentage of angular momentum cancellation, S (%) defined in Eqn 19, for each participant and and the second and third PCs increase in
spatial direction. dominance (see Fig.·7). However, the

THE JOURNAL OF EXPERIMENTAL BIOLOGY Page 57

 Walking angular momentum

1 Left foot (1) Fig.·5. The mean participant-

dependent first PC about three spatial
0.5 Right foot (2) directions. Here the participant-
PC1 (x)

dependent first PCs were averaged

0 Left calf (3) across the 10 study participants. Error
bars are one standard deviation about
–0.5 Right calf (4)
the mean. The abscissa numbers and
–1 Left thigh (5) human model segments are paired to
the right of the figure. In the
1 Right thigh (6) anterior–posterior (y) direction, large
variations in the relative contribution of
0.5 Left hand (7)
angular momentum are observed for
PC1 (y)

Right hand (8) the pelvis and abdomen (segment 13),

0 chest (segment 14) and head
Left forearm (9) (segment 16) [see large standard
–0.5
Right forearm (10) deviations in Fig.·5 for PC1 (y)].
–1
Left upper arm (11)
1 Right upper arm (12)
0.5
Pelvis and abdomen (13)
PC1 (z)

0 Chest (14)
–0.5 Neck (15)
–1 Head (16)
2 4 6 8 10 12 14 16
Link number

second and third PCs only explain ~4% of the data, and thus the these two spatial directions (P=0.267), indicating that zero-moment
rotational dynamics that they explain are not discussed here. forces are equally dominant in the two horizontal directions.
Not all assumptions of the inverted pendulum model are
DISCUSSION supported by this investigation. The fact that the model assumes
Would simple inverted pendulum mechanics also predict the that the CP acts as a fixed point limits its capacity to predict
observed horizontal ground reaction forces? horizontal ground reaction force and CM dynamics. For the zero-
The inverted pendulum model has been used by researchers to moment force predictions shown in Fig.·4A,B, the CP was not
describe sagittal-plane CM dynamics during the single support represented as a fixed point, but rather experimental CP trajectories
phase of walking (Bekker, 1956; Cavagna et al., 1976; Alexander, were used as inputs, as dictated by Eqns 9 and 10. As noted earlier,
1976; Margaria, 1976; Cavagna et al., 1977). The model comprises zero-moment forces closely match horizontal ground reaction
a point mass attached to a mass-less rigid leg that intersects with forces during single support (Rx2=0.97±0.02; Ry2=0.94±0.02). In
the ground surface at a single point. The model makes three critical contrast, using a fixed CP point as required by the inverted
assumptions. First, that the body moves as if all its mass is located pendulum model4, the calculated horizontal forces agree well in the
at the CM, requiring that the resultant ground reaction force vector medio-lateral (x) direction (Rx2=0.88±0.08) but the agreement is
always points at the CM (zero-moment condition). Second, that the very poor in the anterior–posterior (y) direction (Ry2=–0.33±0.44).
CP is a stationary, time-invariant point. And finally, that the leg Not surprisingly, for double support a fixed CP analysis results in
during single support behaves as a rigid strut. extremely poor agreement between zero-moment and experimental
This study directly addresses the first two assumptions of the forces (Rx2=–0.28±0.22; Ry2=–0.11±0.53). In summary, for the
inverted pendulum model. Since the entire mass of the body is advancement of bipedal walking models that accurately predict
represented as a point mass, the model will, by definition, always steady-state CM dynamics, we feel a point mass representation can
move with a constant angular momentum and with no moments be assumed, but the mechanical behavior of the model’s legs and
acting about its CM. The data presented in this study support the their interaction with the ground surface must be sufficiently
point mass representation, assumed by many recent walking models human-like so as to achieve realistic CP dynamics.
(Kuo, 2002; Geyer et al., 2006; Srinivasan and Ruina, 2006). In
fact, we further generalize the point mass representation to include Is angular momentum always regulated during human
single- and double-support phases. We found that zero-moment movement tasks?
forces agree remarkably well with experimental values for single Whole-body angular momentum regulation is not a general feature
support (Rx2=0.97±0.02; Ry2=0.94±0.02), and agree reasonably well across all human movement tasks. For some movement patterns,
for double support (Rx2=0.72±0.14; Ry2=0.78±0.06). Still further, humans purposefully generate angular momentum to enhance
whereas the inverted pendulum model is a 2-D sagittal plane model, stability and maneuverability (Popovic et al., 2004b; Hofmann et
the results of this study further generalize the point mass al., 2007). By actively rotating body segments (arms, torso, legs),
representation to include the medio-lateral (x) direction. Across all CM moments can be generated that cause horizontal moment forces
study participants and including the entire gait cycle, the mean R2
parameter was 0.91±0.06 in the medio-lateral (x) direction and 4
For the stationary CP analysis, the zero-moment force components of Eqns 9
and 10 were used where the CP position was assumed to be at a fixed, lab
0.90±0.03 in the anterior–posterior (y) direction. No significant frame location, corresponding to a point at the foot center for single support and
difference was observed between the distributions of R2 values for halfway between both feet for double support.

THE JOURNAL OF EXPERIMENTAL BIOLOGY Page 58

 H. Herr and M. Popovic

1 1
PC1 (x) 86.25% B2 (x)=0.26
0 0.5
B3 (x)=0.32
–1 0
1
9.33%
PC2 (x)

–0.5
0 B1 (x)=0.14
–1

Normalized tuning coefficients

–1 1
1
2.20% 0.5
PC3 (x)

B4 (y)=1.03
0
0
–1
0 2 4 6 8 10 12 14 16
–0.5 B3 (y)=1.20
1 B1 (y)=0.52 B2 (y)=1.45
51.83%
PC1 (y)

0 –1
1
–1
1 B2 (z)=1.00
29.21% 0.5
PC2 (y)

B3 (z)=0.64
0
–1 0
1
7.62%
PC3 (y)

–0.5
0
B1 (z)=0.19
–1 –1
1 0 20 40 60 80 100
7.20%
PC4 (y)

Gait cycle (%)

0
–1 Fig.·7. Tuning coefficients. The mean values, over all participants and trials,
0 2 4 6 8 10 12 14 16 of the normalized tuning coefficients are plotted for three spatial directions.
1 The tuning coefficients correspond to the PCs shown in Fig.·6. Here the
83.45% normalized tuning coefficients were computed using Eqn 17, and were
PC1 (z)

0 obtained as gait percentage averages over all participants and walking

trials. To provide information on the variability of each normalized tuning
–1 coefficient, a variability number Bi(j) is assigned to each curve, where
j=1…3 (spatial directions) and i=1…N (model segments or PCs). Each
1 number was computed by first estimating the area between plus and minus
3.92%
PC2 (z)

one standard deviation about the tuning coefficient mean, and then dividing
0
by the total area beneath the tuning coefficient mean – from 0% to 100%
gait cycle.
–1
1
3.85%
PC3 (z)

0 2007). Such a stability feature cannot be achieved using a bipedal

control scheme that simply applies a zero-moment, constant
–1
0 2 4 6 8 10 12 14 16 angular momentum control. Clearly, if the ground CM projection
Link number falls outside the support envelope and the CM velocity is zero, the
zero-moment force does not act to restore CM position, but rather
Fig.·6. The participant-independent PCs about three spatial directions.
Plotted are the PCs that when combined account for more than 90% of
continually accelerates the CM away from the stance foot. This fact
experimental data. The abscissa numbers correspond to the same human can be easily verified by reviewing the zero-moment force
model segments as defined in Fig.·5. While only three PCs are needed to component from Eqn 9, or FxZero–moment=Fz(xCM–xCP)/zCM. Imagine
explain 90% of the data in the sagittal and transverse planes, four PCs are the case where a person is trying to balance on his right leg, and a
required in the coronal (x–z) plane. laterally directed force disturbance causes xCM to move beyond the
foot envelope in a lateral, positive direction (positive x-direction as
defined in Fig.·1). Since xCP cannot extend beyond the foot’s lateral
to act on the CM, as defined by Eqns 9 and 10. This strategy allows edge, xCM–xCP is positive, making the CM zero-moment force
humans to perform movement tasks that would not otherwise be positive, and causing the CM to move farther from the stance foot.
possible. For example, while balancing on one leg, humans are The only way a person can successfully balance on one leg from
capable of repositioning the CM just above the stance foot from an these initial conditions is to actively generate angular momentum.
initial body state where CM velocity is zero, and the ground CM By rotating arms, trunk, head and swing leg, a CM moment in the
projection is outside the foot support envelope5 (Hofmann et al., positive y-direction can be generated, causing a negative moment
force, FxMoment=–Ty/zCM, that can restore the CM back to a position
directly over the stance foot (Hofmann et al., 2007). This behavior
5
The support envelope is the support base when the foot is flat on the ground
during single support, or when both feet are flat on the ground during double
can be observed in tightrope walking. Here body segments are
support. See Materials and methods for a definition of support base. accelerated to generate angular momentum about the CM and to

THE JOURNAL OF EXPERIMENTAL BIOLOGY Page 59

 Walking angular momentum

Fig.·8. Hula-hoop body motions. Angular momentum, horizontal

A B 0.2 ground reaction force and ground references points are plotted in
the medio-lateral (x) and anterior–posterior (y) directions. In this
0.1

Lx,y/(MVH)
experiment a participant rotated his hips while standing in double
support, similar to how one twirls a hula hoop, at an increasing and
0
then decreasing rotational speed for approximately 10·s (see A for
–0.1 one representative cycle). (B) The horizontal components of
normalized angular momentum are plotted versus time. For ease of
–0.2 comparison with walking values shown in Fig.·3A, the angular
7.2 7.4 7.6 7.8 8.0 0 5 10 momentum is normalized by the product of the participantʼs mass,
Time (s) CM height and self-selected gait speed (MVH; see participant no. 1
Time (s)
C D in Table·1). (C,D) The horizontal ground reaction forces measured
2 2 experimentally (thin blue line) are plotted along with the calculated
Anterior–posterior

zero-moment forces (thick red line) versus time for the same
Medio-lateral
1 1
participant and trial as in B. Both experimental and calculated zero-
Fy /kylimit

Fx /kxlimit

0 moment forces are normalized by the stiffness term, Fz/zCM, and

0
the radius of the ground support base in the appropriate direction
–1 –1 (see Eqns 9 and 10). In the medio-lateral (x) direction, the radius
was measured while standing in double support, and was equal to
–2 –2 one-half the distance from the lateral edge of the right foot to the
0 5 10 0 5 10 lateral edge of the left foot. In the anterior–posterior (y) direction,
Time (s) Time (s) the radius was equal to one-half the participantʼs foot length.
E 400
(E) Plotted are the CP (blue dashed line), CMP (red solid line) and
CMP CM ground projection (green dash-dotted line) trajectories and
CP
corresponding footprints. As in A, only one hula-hoop cycle is
CM
Anterior–posterior direction (mm)

shown from 7.2 to 8·s. The ground CM projection remains within

300 the support envelope while the CMP often falls outside the region.

200

100

0
400 500 600 700 800
Medio-lateral direction (mm)

create a moment force that restores the CM position over the stance Fig.·8E, the CMP ground reference point often moves beyond the
foot. ground support envelope, diverging from the CP and the ground
To further investigate movement tasks where the moment force CM projection. This dynamical behavior is distinct from that of
is dominant, we conducted two pilot investigations. We first steady-state normal walking where the CMP never leaves the
analyzed a particular movement task while standing in double ground support base, as indicated in Fig.·4C.
support. Here a study participant rotated his hips, similar to how For the hula-hoop motion, angular momentum becomes
one twirls a hula-hoop, at an increasing and then decreasing speed sufficiently large that moment forces become dominant over zero-
for approximately 10·s (see Fig.·8A). Kinetic and kinematic data moment forces, and the CMP moves beyond the support envelope.
were collected using the same experimental setup employed in the Between t=6·s and t=8·s in Fig.·8C,D, the moment force is often as
walking study, and the same human model was then used for large as, or larger than, the zero-moment force. Even with the
physical modeling. existence of large CM moments, the participant still remains
In Fig.·8B, the horizontal components of normalized angular upright and stable. The regulation of angular momentum [兩L(t)兩≈0]
momentum are plotted versus time, showing angular momentum is therefore not a necessary condition for human stability. This
values that are approximately an order of magnitude larger than the finding is in direct disagreement with the arguments of Morasso
steady-state walking values shown in Fig.·3A. Further, in and Schieppati (Morasso and Schieppati, 1999) and Morasso et al.
Fig.·8C,D, we show zero-moment forces, as well as experimental (Morasso et al., 1999), who stated that the ‘CP–CM phase-lock’, a
forces measured from a force platform, for the hula-hoop twirling relation similar to the zero-moment force component of Eqns 9 and
motion. Here the difference between the two curves is equal to the 10, is a pure physics consequence of stability. In fact, the generation
moment force component of the horizontal ground reaction forces, of angular momentum and CM moments is a key strategy for the
as defined by Eqns 9 and 10. Beyond t=5·s, there is no longer good enhancement of bipedal maneuverability and stability (Popovic et
agreement between zero-moment forces and experimental values al., 2004b; Hofmann et al., 2007). Clearly, the CM motions found
as CM moment, or the rate of change of angular momentum, in the hula-hoop task could not be achieved using only a constant
becomes dominant. During this same time period, as is shown in angular moment, zero-moment control approach.

THE JOURNAL OF EXPERIMENTAL BIOLOGY Page 60

 H. Herr and M. Popovic

50
0.2
A C E
800

0
Lx /(MVH)

Fx (N)
600

Anterior–posterior direction (mm)

–50
–0.2
400
–100
–0.4

200 200
B D
0.05
Ly /(MVH)

0
Fy (N)

0
–200 –200

–0.05 –400
0 20 40 60 80 100 0 10 20 30 40 50 –200 –100 0 100
Gait cycle (%) Gait cycle (%) Medio-lateral
direction (mm)

Fig.·9. Exaggerated walking gait. Angular momentum, horizontal ground reaction force and ground reference points are plotted in the medio-lateral (x) and
anterior–posterior (y) directions. In this experiment a participant walked with exaggerated leg protraction and retraction movements, similar to a military
marching gait, at a forward speed of 1.3 m·s–1 (see whole-body sketches above plots). (A,B) The horizontal components of normalized angular momentum
are plotted versus percentage gait cycle. For ease of comparison with walking values shown in Fig.·3A, the angular momentum is normalized by the product
of the participantʼs mass, CM height, and self-selected gait speed (MVH; see participant no. 1 in Table·1). Here 0% and 100% gait cycles correspond to
consecutive heel strikes of the same foot. (C,D) The horizontal ground reaction forces measured experimentally (thin blue line) are plotted along with the
calculated zero-moment forces (thick red line) versus percentage gait cycle for the same participant and trial as in A and B. Here 0% to 50% gait cycle
spans from the middle of a single-support phase to the middle of the next single-support phase of the opposite limb. (E) Plotted are the CP (blue dashed
line), CMP (red solid line) and CM ground projection (green dash-dotted line) trajectories and corresponding footprints. As in C and D, only 50% of the gait
cycle is shown. The two circles on each line denote the transition from single to double support, and vice versa. For this exaggerated gait, the CMP often
falls outside the ground support envelope.

What would be the benefits of a generalized strategy of The exaggerated walking plots of Fig.·9 clearly show that it is
angular momentum regulation in normal walking? possible to walk with large CM moments. The dominant source of
As highlighted in the last section, active generation of angular CM moment in this particular walking pattern is due to rapid body
momentum is a key strategy for bipedal maneuverability and movements within the sagittal plane, as indicated by the large
stability. Why then is momentum highly regulated during normal angular momentum values in the medio-lateral (x) direction (Fig.
steady-state walking? To address this question, we conducted a 9A) compared with the anterior–posterior (y) direction (Fig. 9B).
second pilot study where a participant walked with exaggerated leg Throughout much of the single-support phase, the swing leg
protraction and retraction movements, similar to a military protracts rapidly forward, often generating a positive CM moment
marching gait, at a forward speed of 1.3·m·s–1 (see whole-body in the medio-lateral (x) direction. As described by Eqn 9, this
sketches above plots in Fig.·9). Kinetic and kinematic data were moment causes a positive moment force to act in the
collected using the same experimental setup employed in the anterior–posterior (y) direction, accelerating the CM forward
normal walking study, and the same human model was used for during early to mid-swing phase (see Fig. 9D). During terminal
physical modeling. In Fig.·9, the angular momentum (Fig.·9A,B), swing (18% to 30% gait cycle in Fig. 9D), the swing leg retracts
horizontal ground reaction forces (Fig.·9C,D) and ground towards the walking surface, generating a negative CM moment in
references points (Fig.·9E) are plotted in the medio-lateral (x) and the medio-lateral (x) direction and causing a negative moment force
anterior–posterior (y) directions. As with the hula-hoop motion, to act in the anterior–posterior (y) direction. This exaggerated
angular momentum is an order of magnitude larger than in normal retraction motion tends to decelerate the CM in the
walking (see Fig.·3A). Further, zero-moment forces do not agree anterior–posterior (y) direction just prior to heel strike. Thus,
well with experimental values (Rx2=0.01±0.12; Ry2=–1.6±0.3; throughout the single-support phase, the moment force tends to
mean±s.d. across seven gait trials), and the CMP diverges from the accelerate the CM forward as the result of swing-leg protraction,
CP, often moving outside the ground support envelope and then rapidly decelerate the CM as the swing leg retracts just
(␤=50±6%). prior to the double-support phase.

THE JOURNAL OF EXPERIMENTAL BIOLOGY Page 61

 Walking angular momentum

Walking in this manner, although possible from a stability required for a participant-dependent analysis, highlights a more
standpoint, is nonetheless energetically expensive. Generating dominant rotational variability in the coronal (x–z) plane, across
substantial CM moments by driving the swing leg in rapid study participants, compared with sagittal and transverse planes.
protraction and retraction movements is likely to increase muscle This result suggests that body dynamics that may be used to
work. The total kinetic energy increment resulting from body distinguish individual gait patterns in an angular momentum
movements relative to the CM6 (Willems et al., 1995) is 123±16·J representation are mainly present in the coronal (x–z) plane. The
(mean±s.d. across seven gait trials) for the exaggerated walk, a specific source of the observed variability is predominantly due to
value that is 5-fold larger than the normal walking value from the large variations in the relative contribution of angular momentum
same participant (22±2·J). Further, using the individual limbs in the abdomen and pelvis (segment 13), chest (segment 14) and
method (Donelan et al., 2002), the positive and negative external head (segment 16) [see large standard deviations in Fig.·5 for PC1
work performed by the legs on the CM is 53±3·J and –40±6·J for (y)].
the exaggerated walk, respectively. For normal walking from the In an analysis of passive dynamic walking stability, Kuo (Kuo,
same participant, the positive and negative external work is 28±2·J 1999) argued that bipedal walking is inherently unstable in the
and –25±1·J, respectively. These preliminary results suggest that medio-lateral direction, and thus body movements within the
perhaps CM moments are kept small in normal steady-state walking coronal (x–z) plane would need to be actively controlled in order
in order to lower the body’s work requirements, allowing for less for the body to remain upright and stable. One interpretation of why
muscle work and a more economical walking pattern. there is a more dominant rotational variability in the coronal (x–z)
plane is that distinct participant-dependent strategies are being
The determinants of gait and segmental angular momentum expressed to achieve stable bipedal gait due to perhaps
cancellations morphological variations between study participants. Although
The gait determinants of pelvic rotation, controlled plantar flexion beyond the scope of the present study, the precise reason for the
and powered plantar flexion are thought to be important more dominant rotational variability in the coronal (x–z) plane is
mechanisms for making the CM trajectory flatter and smoother in an interesting area for future research.
walking (Saunders et al., 1953; Kerrigan et al., 2000; Kerrigan et
al., 2001). Although pelvic obliquity and early stance knee flexion Future work
were also believed to produce flatter CM motions (Saunders et al., An understanding of angular momentum behaviors in human
1953), recent evidence suggests otherwise (Gard and Childress, walking and other movement tasks may have important
1997; Gard and Childress, 1999). We now discuss these particular implications for several fields of study. In clinical gait research, the
determinants of gait in the context of segment-to-segment moment and zero-moment force components, as well as the CMP
momentum cancellations. We found that feet and calf momenta are ground reference point, may potentially serve as valuable
balanced by thigh momenta as a result of controlled plantar flexion, identification metrics for the diagnosis of pathological walking
powered plantar flexion and early stance knee flexion, resulting in patterns such as was shown in Fig.·9. Further, in legged robotics
the relatively small medio-lateral (x) component of whole-body research, an understanding of human angular momentum behaviors
angular momentum. Further, we found that pelvis and abdomen is likely to motivate improvements in humanoid controllers that
momenta are balanced by segmental momenta from the rest of the effectively exploit both moment and zero-moment CM force
body through the action of pelvic obliquity, resulting in the components to improve robotic stability and maneuverability
relatively small anterior–posterior (y) component of whole-body (Hofmann et al., 2004; Popovic et al., 2004b; Hofmann et al.,
momentum. Finally, we show that leg angular momenta are 2007). It is our hope that this work will motivate further studies
balanced by segmental momenta from the rest of the body during related to the biomechanics and control of human rotational
pelvic rotation, producing the relatively small vertical (z) behavior.
component of whole-body momentum.
LIST OF SYMBOLS AND ABBREVIATIONS
Coronal plane rotational variability in normal human walking Bi(j) PC variability number: ratio between the area of the one sigma
Since steady-state walking comprises a series of coupled and confidence interval of the normalized tuning coefficient
cji(t) to the total area beneath the mean value curve
alternating movement patterns, a reduced-order PC representation
cji(t) Normalized tuning coefficients
naturally exists for the high dimensional space of segmental angular Cji(t) Time-dependent tuning coefficients
momenta. For participant-independent PC analyses (using data D(␣) A 16-component vector of segmental model densities
from all participants and gait trials), we found that only three PCs DExp A 16-component vector of mean segmental densities obtained
are required to explain greater than 90% of angular momentum data from the literature
in the medio-lateral (x) and vertical (z) directions, whereas four PCs DE ji Percentage of data (variance) explained with PC vector Pji
Di(␣) Density of the i-th segment
are necessary in the anterior–posterior (y) direction. Furthermore,
F Ground reaction force vector
for participant-dependent analyses (using only individual FExp Grand mean, over all trials and gait percentages, of measured
participant data), we found that only three PCs are needed to ground reaction force (x or y)
ij
explain greater than 90% of data about all three spatial directions FExp j-th gait percentage of the i-th trial measured ground reaction
(see Table·2). force (x or y component)
ij
In the anterior–posterior (y) direction, the fact that four PCs are F Mod j-th gait percentage of the i-th trial model-predicted ground
required for a participant-independent analysis, and only three are reaction force (x or y)
Moment
Fx,y Horizontal components of ground reaction force corresponding
to CM moment
6
This kinetic energy quantity was computed by taking the sum of the increments Zero–moment
Fx,y Horizontal components of ground reaction force corresponding
in the segment kinetic energy versus time curves of all the body segments to zero CM moment
relative to the CM. Willems et al. (Willems et al., 1995) argues that this kinetic
energy quantity relates to the upper limit of internal mechanical work necessary
G Gravitational constant
to accelerate the limbs relative to the CM. Hsubject CM height

THE JOURNAL OF EXPERIMENTAL BIOLOGY Page 62

 H. Herr and M. Popovic
,i
I Segment i moment of inertia tensor about segment’s CM Friedman, M. (1937). The use of ranks to avoid the assumption of normality implicit in
L Angular momentum vector about the whole-body CM the analysis of variance. J. Am. Stat. Assoc. 32, 675-701.
Friedman, M. (1940). A comparison of alternative tests of significance for the problem
mi Mass of the i-th segment
of m rankings. Ann. Math. Stat. 11, 86-92.
MR A 16-component vector of relative mass distribution Frohlich, C. (1979). Do spring board divers violate angular momentum conservation?
throughout the model Am. J. Phys. 47, 583-592.
MRExp A 16-component vector of mean relative mass values obtained Gard, S. A. and Childress, D. S. (1997). The effect of pelvic list on the vertical
from the literature displacement of the trunk during normal walking. Gait Posture 5, 233-238.
Gard, S. A. and Childress, D. S. (1999). The influence of stance-phase knee flexion
MiR(␣) Relative mass of the i-th segment on the vertical displacement of the trunk during normal walking. Arch. Phys. Med.
Msubject Total body mass Rehabil. 80, 26-32.
Mx,y Horizontal moments measured relative to a fixed reference Geyer, H., Seyfarth, A. and Blickhan, R. (2006). Compliant leg behaviour explains
point on the force platform basic dynamics of walking and running. Proc. R. Soc. Lond. B: Biol. Sci. 273, 2861-
2867.
N Number of segments of human model, N=16 Goswami, A. and Kallem, V. (2004). Rate of change of angular momentum and
NE Number of PCs used for angular momentum cancellation balance maintenance of biped robots. Proceedings of the IEEE International
estimate Conference on Robotics and Automation, New Orleans, LA, USA, pp. 3785-3790.
NPercent Number of gait percentage times analyzed doi: 10.1109/ROBOT.2004.1308858, www.ieee.org.
Gu, W. (2003). The regulation of angular momentum during human walking.
Nsubject Angular momentum normalization constant
Undergraduate thesis, Physics Department, MIT, USA.
NTrial Number of trials Herr, H., Hofmann, A. and Popovic, M. (2003). New horizons for orthotic and
Pji Principal component (PC) basis vectors, with j=1…3 (spatial prosthetic technology: merging body and machine. Presented at the ZIF International
directions) and i=1…N (model segments or PCs) Conference on Walking Machines, Bielefeld, Germany,
Pjiq The angular momentum contribution of the q-th body segment http://biomech.media.mit.edu/publications/ArtificialMuscle_O&P.pdf.
Hinrichs, R. (1982). Upper extremity function in running. PhD thesis, The
to Pji Pennsylvania State University, University Park, USA.
rCM Whole-body CM position relative to lab frame Hinrichs, R. (1987). Upper extremity function in running. II. Angular momentum
r iCM Segment i CM position relative to lab frame considerations. Int. J. Sport Biomech. 3, 242-263.
rCMP Centroidal moment pivot (CMP) receptor Hinrichs, R. (1992). Case studies of asymmetrical arm action in running. Int. J. Sport
Biomech. 8, 111-128.
rCP Center of pressure location Hinrichs, R., Cavanagh, P. and Williams, K. (1983). Upper extremity contributions to
R2 Coefficient of determination angular momentum in running. Biomechanics VIII-B. Champaign, IL, 641-647.
Sj Segmental angular momentum cancellation in the j-th spatial Hofmann, A. (2003). Control rules for biomimetic human bipedal locomotion based
direction on biomechanical principles. PhD Thesis Proposal, Submitted to the Computer
Science and Electrical Engineering Department, Massachusetts Institute of
T Moment vector about the whole-body CM
Technology.
vCM Whole-body CM velocity relative to lab frame Hofmann, A., Massaquoi, S., Popovic, M. and Herr, H. (2004). A sliding controller
vi The i-th segment’s CM velocity relative to lab frame for bipedal balancing using integrated movement of non-contact limbs. IEEE/RSJ
V Total body volume International Conference on Intelligent Robots and Systems, October, Sendai,
Vi Volume of segment i Japan, pp. 1952-1959. doi: 10.1109/IROS.2004.1389683, www.ieee.org.
Hofmann, A., Popovic, M. and Herr, H. (2007). Exploiting angular momentum to
V Ri Relative volume of the i-th segment enhance bipedal center-of-mass control. IEEE Trans. Rob. Autom. In press.
VR A 16-component vector of relative volumes computed directly Jackson, J. (1991). A User’s Guide to Principal Components. New York: John Wiley &
from the human model Sons.
Vsubject Walking speed Kadaba, M. P., Ramakrishnan, H. K., Wootten, M. E., Gainey, J., Gorton, G. and
Cochran, G. V. (1989). Repeatability of kinematic, kinetic, and electromyographic
␣ Parameter used in optimization of segment mass properties data in normal adult gait. J. Orthop. Res. 7, 849-860.
␣min Value of parameter ␣ minimizing the absolute error Kadaba, M. P., Ramakrishnan, H. K. and Wootten, M. E. (1990). Measurement of
兩D(␣)–DExp兩 lower extremity kinematics during level walking. J. Orthop. Res. 8, 383-392.
␤% Mean CP–CMP separation distance normalized by foot length Kerrigan, D. C., Croce, U. D., Marciello, M. and Riley, P. O. (2000). A refined view
of the determinants of gait: significance of heel rise. Arch. Phys. Med. Rehab. 81,
j A 16-component vector of segmental angular momenta in j-th
1077-1080.
spatial direction Kerrigan, D. C., Riley, P. O., Lelas, J. L. and Della Croce, U. (2001).
i
Segment i angular velocity vector about segment’s CM Quantification of pelvic rotation as a determinant of gait. Arch. Phys. Med. Rehab.
82, 217-220.
King, D. (1999). Generating vertical velocity and angular momentum during skating
We thank the Michael and Helen Schaffer Foundation of Boston, Massachusetts
jumps. Paper presented at 23rd annual meeting of the American Society of
for their generous support of this research. We also thank Paolo Bonato, Tod
Biomechanics, University of Pittsburgh, October, 21-23, http://asbweb.org/
Farrell, Hartmut Geyer and Jennifer Johansson for their helpful suggestions and conferences/1990s/1999/ACROBAT/086.PDF.
support. Kuo, A. (1999). Stabilization of lateral motion in passive dynamic walking. Int. J.
Robot. Res. 18, 917-930.
Kuo, A. D. (2002). Energetics of actively powered locomotion using the simplest
REFERENCES walking model. J. Biomech. Eng. 124, 113–120.
Alexander, R. (1976). Mechanics of bipedal locomotion. In Perspectives in LeBlanc, M. and Dapena, J. (1996). Generation and transfer of angular momentum in
Experimental Biology (ed. P. S. Davies), pp. 493-504. Oxford, UK: Pergamon Press. the javelin throw. Proceedings of the 20th annual meeting of the American Society of
Bekker M. G. (1956). Theory of Land Locomotion. Ann Arbor: University of Michigan Biomechanics. Atlanta, Georgia, USA, October 17-19, http://www.elitetrack.com/
Press. asbjavelin.pdf.
Cavagna, G. A., Thys, H. and Zamboni, A. (1976). The sources of external work in Lee, C. R. and Farley, C. T. (1998). Determinants of the center of mass trajectory in
level walking and running. J. Physiol. Lond. 262, 639-657.
human walking and running. J. Exp. Biol. 201, 2935-2944.
Cavagna, G. A., Heglund, N. C. and Taylor, C. R. (1977). Mechanical work in
Margaria, R. (1976). Biomechanics and Energetics of Muscular Exercise. Oxford, UK:
terrestrial locomotion: two basic mechanisms for minimizing energy expenditure. Am.
Clarendon Press.
J. Physiol. 233, R243-R261.
McGeer, T. (1990). Passive dynamic walking. Int. J. Robot. Res. 9, 62- 82.
Croce, U. D., Riley, P. O., Lelas, J. L., and Kerrigan, D. C. (2001). A refined view of
the determinants of gait. Gait Post 14, 79-84. Mochon, S. and McMahon, T. A. (1980a). Ballistic walking. J. Biomech. 13, 49-57.
Crowe, A., Schiereck, P., Boer, R. W. and Keessen, W. (1995). Characterization of Mochon, S. and McMahon, T. A. (1980b). Ballistic walking: an improved model. Math.
human gait by means of body center of mass oscillations derived from ground Biosci. 52, 241-260.
reaction forces. IEEE Trans. Biomed. Eng. 42, 293-303. Morasso, P. and Schieppati, M. (1999). Can muscle stiffness alone stabilize upright
Dapena, J. (1978). A method to determine the angular momentum of a human body standing? J. Neurophysiol. 82, 1622-1626.
about three orthogonal axes passing through its center of gravity. J. Biomech. 11, Morasso, P., Baratto, L., Capra, R. and Spada, G. (1999). Internal models in the
251-256. control of posture. Neural Netw. 12, 1173-1180.
Dapena, J. (1993). An analysis of angular momentum in the discus throw. Paper Ortega, J. D. and Farley, C. T. (2005). Minimizing center of mass vertical movement
presented at 14th International Congress of Biomechanics, Paris, France. increases metabolic cost in walking. J. Appl. Physiol. 99, 2099-2107.
Dapena, J. and McDonald, C. (1989). A three-dimensional analysis of angular Popovic, M., Gu, W. and Herr, H. (2002). Conservation of angular momentum during
momentum in the hammer throw. Med. Sci. Sports Exerc. 21, 206-220. human locomotion. MIT, Artificial Intelligence Laboratory, Research Abstracts.
Donelan, M., Kram, R. and Kuo, A. (2002). Simultaneous positive and negative September, 231-232, http://biomech.media.mit.edu/research/Popovic_Gu_Herr.pdf.
external mechanical work in human walking. J. Biomech. 35, 117-124. Popovic, M., Hofmann, A. and Herr, H. (2004a). Angular momentum regulation
Dunn, O. J. (1964). Multiple comparisons using rank sums. Technometrics 6, 241- during human walking: biomechanics and control. Proceedings of the IEEE
252. International Conference on Robotics and Automation. New Orleans, LA, USA, April,
Elftman, H. (1939). The function of the arms in walking. Hum. Biol. 11, 529-535. pp. 2405-2411. doi: 10.1109/ROBOT.2004.1307421, www.ieee.org.

THE JOURNAL OF EXPERIMENTAL BIOLOGY Page 63

 Walking angular momentum
Popovic, M., Hofmann, A. and Herr, H. (2004b). Zero spin angular momentum Simoneau, G. and Krebs, D. (2000). Whole-body angular momentum during gait: a
control: definition and applicability. IEEE-RAS/RSJ International Conference on preliminary study of non-fallers and frequent fallers. J. Appl. Biomech. 16, 1-13.
Humanoid Robots. Santa Monica, CA, USA, November, www.ieee.org. Srinivasan, M. and Ruina, A. (2006). Computer optimization of a minimal biped model
Popovic, M., Goswami, A. and Herr, H. (2005). Ground reference points in legged discovers walking and running. Nature 439, 72-75.
locomotion: definitions, biological trajectories and control implications. Int. J. Robot. Wilcoxon, F. (1945). Individual comparisons by ranking methods. Biometrics 1, 80-
Res. 24, 1013-1032. 83.
Portney, L. G. and Watkins, M. P. (2000). Foundations of Clinical Research: Willems, A., Cavagna, A. and Heglund, N. C. (1995). External, internal and total
Applications to Practice (2nd edn). Upper Saddle River: Prentice Hall. work in human locomotion. J. Exp. Biol. 198, 379-393.
Riley, P., Krebs, D. and Popat, R. (1997). Biomechanical analysis of failed sit-to- Winter, D. A. (1990). Biomechanics and Motor Control of Human Movement. New
stand. IEEE Trans. Rehabil. Eng. 5, 353-359. York: John Wiley & Sons.
Rose, J. and Gamble, J. G. (1994). Human Walking (2nd edn). Baltimore, MD: Xu, D. and Wang, T. (1998). Lower extremity contributions to altering direction during
Williams and Wilkins. walking: analysis of angular momentum. Proceedings of the North American
Saunders, J. B., Inman, V. T. and Eberhart, H. D. (1953). The major determinants in Congress on Biomechanics. University of Waterloo, Waterloo, Ontario, Canada,
normal and pathological gait. J. Bone Joint Surg. Am. 35, 543-558. August. http://asbweb.org/conferences/1990s/1998/160/index.html.

THE JOURNAL OF EXPERIMENTAL BIOLOGY Page 64

 Acknowledgements & Credits

Funding for this publication was provided by NSF HSI Award number 1832348

 Some of the content in this manual was excerpted from the following:
• Mathematics for the Life Sciences by Erin N. Bodine, Suzanne Lenhart & Louis J. Gross
• The University of Arizona’s Biology Project website: http://www.biology.arizona.edu/
• The Discovering Radians Activity comes from:
http://mrsnicoleburns.weebly.com/uploads/8/6/7/0/8670276/_1__discovering_radians_acti
vity__page_1_.pdf

 Content was developed for Los Angeles Mission College Trigonometry classes which use the
Pearson textbook Trigonometry 11th Edition, by Lial/Hornsby/Schneider/Daniels, and some of
the content was derived from Pearson’s TestGen testbanks for this textbook.

 The Open Educational Resource Textbook from OpenStax Precalculus by Abramson is

referenced for students as a resource for content review.

 The article is from the Journal of Experimental Biology