Distance Metrics and Algorithms for Task Space Path Optimization - - PowerPoint PPT Presentation

distance metrics and algorithms for task space path
SMART_READER_LITE
LIVE PREVIEW

Distance Metrics and Algorithms for Task Space Path Optimization - - PowerPoint PPT Presentation

Distance Metrics and Algorithms for Task Space Path Optimization Rachel Holladay Siddhartha Srinivasa The Robotics Institute Carnegie Mellon University Goal: Follow End-Effector Path in Task Space Goal: Follow End-Effector Path in Task


slide-1
SLIDE 1

Distance Metrics and Algorithms for Task Space Path Optimization

The Robotics Institute Carnegie Mellon University

Rachel Holladay Siddhartha Srinivasa

slide-2
SLIDE 2
slide-3
SLIDE 3
slide-4
SLIDE 4

Goal: Follow End-Effector Path in Task Space

slide-5
SLIDE 5

Goal: Follow End-Effector Path in Task Space subject to constraints.

slide-6
SLIDE 6
slide-7
SLIDE 7
slide-8
SLIDE 8
slide-9
SLIDE 9
slide-10
SLIDE 10
slide-11
SLIDE 11

Goal: Follow a Reference Path

slide-12
SLIDE 12

Goal: Follow a Reference Path by leveraging motion planning.

slide-13
SLIDE 13
slide-14
SLIDE 14
slide-15
SLIDE 15
slide-16
SLIDE 16

Key Insight Use trajectory optimization to

  • ptimize our path to be close

to our reference path.

slide-17
SLIDE 17

Key Insight Use trajectory optimization to

  • ptimize our path to be close

to our reference path.

slide-18
SLIDE 18
slide-19
SLIDE 19
slide-20
SLIDE 20
slide-21
SLIDE 21

Task Space

slide-22
SLIDE 22

Task Space

slide-23
SLIDE 23

Task Space

slide-24
SLIDE 24

Task Space Configuration Space

slide-25
SLIDE 25

Task Space Configuration Space

slide-26
SLIDE 26

Task Space Configuration Space IK FK

slide-27
SLIDE 27

How to compare the distance between task space paths?

slide-28
SLIDE 28

How to compare the distance between task space paths? Borrow from computational geometry.

slide-29
SLIDE 29
slide-30
SLIDE 30
slide-31
SLIDE 31
slide-32
SLIDE 32
slide-33
SLIDE 33

One-way Hausdorff Distance

slide-34
SLIDE 34
slide-35
SLIDE 35
slide-36
SLIDE 36

One-way Hausdorff Distance

slide-37
SLIDE 37

Two-way Hausdorff Distance

slide-38
SLIDE 38

Follow Balls in order.

slide-39
SLIDE 39
slide-40
SLIDE 40

Fréchet Distance

slide-41
SLIDE 41

Fréchet Distance

slide-42
SLIDE 42

Fréchet Distance

slide-43
SLIDE 43

Fréchet Distance

slide-44
SLIDE 44

Fréchet Distance

slide-45
SLIDE 45

Fréchet Distance

slide-46
SLIDE 46

Use the Fréchet Distance to capture the task space distance between paths.

slide-47
SLIDE 47

Gradient of Distance Function

slide-48
SLIDE 48

Gradient of Distance Function

slide-49
SLIDE 49

Gradient of Distance Function

slide-50
SLIDE 50
slide-51
SLIDE 51

Discrete Fréchet

slide-52
SLIDE 52
slide-53
SLIDE 53
slide-54
SLIDE 54
slide-55
SLIDE 55
slide-56
SLIDE 56

How can we assist our optimizer?

slide-57
SLIDE 57
slide-58
SLIDE 58
slide-59
SLIDE 59

Use our distance metrics to find the areas that need additional constraints.

slide-60
SLIDE 60
slide-61
SLIDE 61
slide-62
SLIDE 62

Key Insight Use trajectory optimization to

  • ptimize our path to be close

to our reference path.

slide-63
SLIDE 63

Use computational geometry techniques to measure task space distance between paths.

slide-64
SLIDE 64

Distance Metrics and Algorithms for Task Space Path Optimization

https://www.personalrobotics.ri.cmu.edu http://www.andrew.cmu.edu/user/rmh/

Rachel Holladay Siddhartha Srinivasa