Scribd
Upload
5 views

Lecture 06

The document discusses inverse kinematics, focusing on the challenges of solving for joint variables in robotic systems. It highlights the distinction between analytic and numerical solutions, emphasizing the optimization approach to achieve desired end-effector poses. Additionally, it introduces the concept of a controller with corrective behavior, specifically the PID control method for managing errors in system outputs.

Uploaded by

melat
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
5 views

Lecture 06

The document discusses inverse kinematics, focusing on the challenges of solving for joint variables in robotic systems. It highlights the distinction between analytic and numerical solutions, emphasizing the optimization approach to achieve desired end-effector poses. Additionally, it introduces the concept of a controller with corrective behavior, specifically the PID control method for managing errors in system outputs.

Uploaded by

melat
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 26

Inverse Kinematics 1

(x,y)

q2

q1

x
2
Top view x0,y0

a3 a3?
a2

y
a2?
a1
z q1
y
x

3
x
Side view
a3
a3
a2 q3
a2

q2 z0
a1 z
z
y

x 4
x
1 θ
1 tan-1(1/2) = 260 tan-1(1/-2) = -260
θ
-2
2

tan-1(-1/-2) = 260 2 tan-1(-1/2) = -260


-2
θ -1
-1
θ
5
▪ Get around the sign ambiguity problem

6
y

x
7
▪Create a model of the robot joints and links

▪Write down lots of equations and attempt to

solve for the unknown joint variables

▪Solutions are compact and fast to execute

8
What if the analytic solution
▪ does not exist?

▪ is too hard to figure out?

9
Numerical solutions are trial-and-error procedures
that result in approximate solutions

10
▪ easy to figure out the forward kinematics
ξ(K(q))

▪ we know the pose we want the end-effector to


reach ξ*

▪adjust q until ξ → ξ*
11
This is posed as an optimization problem.

In mathematics, computer science and economics,


an optimization problem is the problem of finding the
best solution from all feasible solutions.

12
This is posed as an optimization problem.

q* = argminq |ξ(K(q)) ⊖ ξ*|

13
(x,y)

q2

q1

x
14
15
Applying input to cause system variables to conform
to desired values
▪ Cruise-control

▪ flying a quadcopter at a certain height

16
To do this, a controller with the corrective
behavior is required.

17
Input Output
Process

18
error Input Output
Process

change the error into a


command

19
error

time

20
21
error Input Output
PID Process

▪ P – Proportional
▪ I – Integral
▪ D – Derivative

22
error = set_point – process_value
output = proportional_gain * error

23
v

24
v

25
Proportional Integral Derivative

26

You might also like