Planning Motions with Intention

1
Planning Motions with Intention

• Presented by
• Yan Ke

2
Problem Specification

• Task Generate motions for human or robot arms to
  complete manipulation tasks.
  
• Goal Find a collision-free path in configuration
  space.
  
• Tool Inverse kinematics algorithm.
• Usage Computer animation.

3
(No Transcript)
4
Difficulties

• Motion with Intention Human and robot arms are
  moving with the intention of completing some
  task.
  
• Restriction Neurophysiology.
• Grasping and Regrasping Posture multiple arms.
• PSPACE-hard

5
Manipulation Planning Overview

• Section 1

6
Inputs

• Geometry of the arms
• Movable object
• Obstacles together with their locations
• Initial and goal configuration

7
The Stable Space and Grasp Space

• Stable space The set of all configurations where
  the movable object M is statically stable.
  
• Grasp Space Arms grasping M and moving it
  stably.
  
• Grasp Space Stable Space Free Space of
  the Configuration Space

8
Transit Paths and Transfer Paths

• Transit Paths Arms motions that do not move M
• Transfer Paths Arms motions that move M

9
Planning Result
10
Generating Transfer and Transit Paths

• Section 2

11
Overview

• The entire manipulation planning can be
  accomplished by following
  
• Generate a series of subtasks to achieve the goal
  configuration.
  
• Plan a transit or transfer path for each
  subtasks.
  
• Assumption Transit tasks can be completed by
  transit paths transfer tasks can be completed by
  transfer paths.
  

12
Generating Transfer Tasks

• Grasp set All various possible grasps for a
  certain M.
  
• Grasp assignment A pair associates with an
  element in grasp set and an identity of the
  grasping arm(s).
  
• We first generate the path for M moving alone.
• Secondly, we attach each configuration of M with
  a list of grasp assignment.

13
Generating Transfer Tasks

• The attached list of grasp assignments are
  obtained by pruning out those no longer possible
  in the new configuration from the previous
  configuration.
• If somehow we found the list of grasp assignment
  is empty, then a regrasping is necessary here.
  
• We solve this problem by resetting the list, find
  all possible of grasp again, and associate them
  with arm(s).

14
Assumptions

• An arm can attain a grasp with a finite set of
  different postures.
  
• All arms not involved in the task is placed
  elsewhere without blocking the motions of working
  arms.
  
• If M requires two arms to move, any one of them
  alone, can hold M stably to allow the other one
  to move in a transit path.

15
Result

• A motion planning path for M, each configuration
  is attached with a list of grasp assignment.
  
• The path is partitioned into several subpaths by
  regrasping.
  
• Each subpath is a transfer task.
• It does not guarantee to find the best path.

16
Generating Transit Paths

• Transit paths are the paths moving the arms.
• Connect the initial configuration to the first
  grasp assignment of the first transfer task.
  
• Connect grasp assignments between different
  transfer tasks.
  
• Connect the last grasp assignment to the goal
  configuartion.

17
Human-Arm Kinematics

• Section 3

18
Neurophysiology

• Goal Determine the arm posture for a human arm
  given the position and orientation of its hand.
  
• Two experimental result
• Arm and wrist posture are for the most part
  independent of each other.
  
• Arm posture for pointing is mainly determined by
  an ST model.
  
• ST model Can determine shoulder and elbow joint
  angles given the position of hand.

19
Arm Posture

• What do we have?
• R, ?, X
• What do we want?
• ?, ß,a,?

20
Inverse Kinematics Algorithm
21
Illegal Posture Adjustment

• Claim eis the only one to violate its limits.
• Solution Decrease F.
• Result wrist position unchanged when F decrease.

22
Experimental Result
23
Experimental Result

• Working environment C and UNIX.
• Time used three and a half minutes.
• Identify the transfer tasks one and a half
  minutes.
  
• Different grasp assignments in total 2600.

24
Conclusion

• A novel approach to solve the multi-arm
  manipulation planning problem.
  
• Computation time is unbounded. If no path exist,
  the algorithm may run forever.
  
• Aim to create a task-level animation package for
  human motions.