Assignment 6: Week 6 Solution

If not explained all symbols have same meaning as in the lectures

You can find the codes Subjective – Q3 and Q4 at this link:
https://github.com/roboticslabiisc/NPTEL-Robotics-Basics-and-Selected-Advanced-Concepts

MCQ/MSQ/NM
1. The linear velocity of the origin Oi of a coordinate system {i} with respect to {i} is (2 marks)
(A) is always zero.
(B) need not be zero.
0
Oi (t + ∆t) −0 Oi (t)
 
(C) is defined as i0 [R] lim
∆t→0 ∆t
(D) is not defined in coordinate system {i}.
Answer : (B) & (C)

2. The space fixed angular velocity vector of a rigid body (2 marks)

(A) is same as the time derivatives of three Euler angles used to describe the orientation of a rigid body in 3D
space.
(B) can be derived from the time derivative of a rotation matrix [R].
(C) can be extracted from a skew symmetric matrix [Ṙ]T [R] where [Ṙ], [R] are the time derivative of the rotation
matrix and the rotation matrix of a rigid body in 3D space, respectively.
(D) can be extracted from a skew symmetric matrix [Ṙ] [R]T where [Ṙ], [R] are the time derivative of the rotation
matrix and the rotation matrix of a rigid body in 3D space, respectively.

Answer : (D)

3. The space fixed angular velocity vector (2 marks)

(A) is independent of the choice of the orientation of the body coordinate system.
(B) is independent of the choice of the orientation of the fixed coordinate system.
(C) is dependent on the origin of fixed coordinate system.
(D) is dependent on the origin of body coordinate system.

Answer : (A)

4. The propagation of linear and angular velocities (2 marks)

(A) can be used to obtain the linear and angular velocities of links in a serial robot with rotary or prismatic joint.
(B) is from the fixed coordinate system {0} to the free end {n} in a serial robot.
(C) is from the fixed base to the moving platform in a parallel robot.
(D) can be used in parallel robot with multi-degree-of-freedom joints.
Answer : (A) & (B)

1
5. The Jacobian matrix for a spacial manipulator moving in 3D space (2 marks)

(A) cannot be obtained by differentiation of a vector valued function.

(B) is not a proper matrix as elements have different units.
(C) does not have a well-defined determinant.
(D) cannot be inverted always.

Answer : (A) & (B)

6. The Jacobian matrix for a serial robot (2 marks)

(A) contains elements which are linear functions of the joint variables.
(B) is a 6 × n matrix where n is the number of joints.
(C) relates the linear and angular velocities of the end-effector with the joint rates.
(D) is the same for all coordinate systems.

Answer : (B) & (C)

7. The Jacobian matrix for a parallel manipulator (2 marks)

(A) cannot be obtained using velocity propagation formulas.
(B) is not defined since there are passive joints.
(C) is not defined since passive joints cannot be eliminated.
(D) requires that the loop-closure equations are solvable.
Answer : (A) & (D)

8. In a four-bar mechanism, the equivalent Jacobian matrix for the linear velocity of a chosen point on the coupler is
(2 marks)
(A) is a 2 × 2 matrix.
(B) is a 3 × 3 matrix.
(C) is a 2 × 1 matrix .
(D) is a scalar.
Answer : (C)

P3 ˙2
9. In the case of the 3-RPS parallel manipulator with i li = 1, the tip of the linear velocity vector of the centroid
of the moving platform (2 marks)

(A) lies in a plane.

(B) lies on an ellipse in 3D space.
(C) lies on an ellipsoid.
(D) lies on an ellipsoid if the the 3-RPS is not at a singular configuration.
Answer : (C)

2
10. At a singular configuration, a serial manipulator (2 marks)
(A) looses one or more degrees of freedom.
(B) can loose all its degrees of freedom.
(C) can move in some directions only.
(D) cannot apply force in any direction.
Answer : (A) & (C)

11. At a gain singular configuration, a parallel manipulator (2 marks)

(A) gains one or more degrees of freedom instantaneously.
(B) becomes instantaneously locked completely.
(C) can instantaneously move in some direction only.
(D) cannot resist force in any direction.
Answer : (A) & (C)

12. At a gain singular configuration in a parallel manipulator, the constraint equations (2 marks)
(A) used to obtain the passive joint variables are independent.
(B) used to obtain the passive joint variables are dependent.
(C) cannot be used to find the passive joint rates.
(D) cannot be differentiated.
Answer : (B) & (C)

3
Subjective Questions
1. The space fixed angular velocity vector, 0 ω si , for X − Y − Z Euler angle, in terms of three Euler angles θi , i = 1, 2, 3
and their derivatives are (4 marks)

θ˙3 sin θ2 + θ˙1

 

ω i = −θ˙3 sin θ1 cos θ2 + θ˙2 cos θ1 

0 s

θ˙3 cos θ1 cos θ2 + θ˙2 sin θ1

Obtain expression for θ̇1 , θ̇2 and θ̇3 in terms of the components of 0 ω si . What happens when θ2 = π/2?

θ˙3 sin θ2 + θ˙1

   
ωx
ω i = ωy  = −θ˙3 sin θ1 cos θ2 + θ˙2 cos θ1 
0 s

ωz θ˙3 cos θ1 cos θ2 + θ˙2 sin θ1

Solving the above 3 equations for θ̇1 , θ̇2 and θ̇3

sin (θ1 ) sin (θ2 ) ωy − cos (θ1 ) sin (θ2 ) ωz + ωx cos (θ2 )
θ̇1 =
cos (θ2 )
θ̇2 = sin (θ1 ) ωz + cos (θ1 ) ωy
cos (θ1 ) ωz − sin (θ1 ) ωy
θ̇3 =
cos (θ2 )
if θ2 = π/2
   ˙
θ3 + θ˙1

ωx
ω i = ωy  = θ˙2 cos θ1 
0 s

ωz θ˙2 sin θ1

∴ θ̇2 = ωy cos θ1 + ωz sin θ1 and θ̇1 + θ̇3 = ωx

∴ θ̇1 and θ̇3 cannot be found out uniquely. This is a singular configuration. So, one needs to assume a value for
either θ̇1 or θ̇3 and find out the other.

2. Show that the maximum and minimum |V|2 in a planar 2R manipulator, subject to constraint θ̇12 + θ̇22 = 1, is
equivalent to solving the eigenvalue problem (as shown in equation (25) in Lecture and given below)

[ g ]Θ̇ − λΘ̇ = 0

where |V|2 is the magnitude of the velocity of tip of a planar 2R manipulator and [ g ] is the matrix with elements
given in equation (24) in the Lecture. (4 marks).
The optimization problem can be stated as:

min |V|2 = (l12 + l22 + 2l1 l2 c2 )θ̇12 + 2(l22 + l1 l2 c2 )θ̇1 θ̇2 + l22 θ̇22
θ̇1 ,θ̇2

Subject to : θ̇12 + θ̇22 = 1

∴ L = (l12 + l22 + 2l1 l2 c2 )θ̇12 + 2(l22 + l1 l2 c2 )θ̇1 θ̇2 + l22 θ̇22 − λ(θ̇12 + θ̇22 − 1)
∂L ∂L
Solving = 0 and = 0 and rearranging we get:
∂ θ̇1 ∂ θ̇2
2
l1 + l22 + 2l1 l2 c2 l22 + l1 l2 c2 θ̇1
   
θ̇
−λ 1 =0
l22 + l1 l2 c2 l22 θ̇2 θ̇2
 
θ̇
which is same as solving the eigenvalue problem [ g ]Θ̇ − λΘ̇ = 0, where Θ̇ = 1
θ̇2

4
3. For a planar 2R manipulator, plot the velocity ellipse for l1 = 1.0 and l2 = 0.5 at θ1 = θ2 = 45◦ . The plot should
be pasted as an image in the PDF you submit. (4 marks).
For the MATLAB code and plots, visit the GitHub link provided at the beginning of the document.

4. For the 3-RPS parallel manipulator, for a = 1/2, b = 1 and (l1 , l2 , l3 ) = (1.0, 1.0, 1.0) meters, solve the direct
kinematics problem and obtain θ1 , θ2 and θ3 numerically using the expressions given in the Lecture. Compute [K]
and [K ∗ ] for the above values of the passive and active joint variables for the 3-RPS manipulator. (4 marks)
For the MATLAB code, visit the GitHub link provided at the beginning of the document.