PROT: Productive Regions-Oriented Task space path planning for hyper-redundant manipulators

by Junghwan Lee and Sung-Eui Yoon
IEEE International Conference on Robotics and Automation, 2014

Approximately constructed Maximum Reachable Area (MRA) A test environment with a manipulator and experiment results. Initial and goal poses are shown in blue and a solution trajectory of the end-effector is shown in red. This benchmark contains many obstacles (black rectangles) and gives rise to many potential local optima cases.

Abstract

Sampling-based planners can generate a probabilistically complete solution for manipulators with moderate degrees of freedom (dofs), but they have suffered from high dimensionality of hyper-redundant manipulators that have high dofs. In this paper we propose a novel efficient sampling bias technique to improve the performance of the task space trajectory planner for hyper-redundant manipulators. Our method defines productive regions in the task space as a set of states that can lead effectively to the goal state. To check whether a node of the random tree is in the productive regions or not, we construct maximum reachable area (MRA) for the node in the task space, where the manipulator can reach from the node by using the employed local planner without any collisions. When MRA of a node contains the goal state, we call it promising and bias our sampling to cover promising, MRA. When the MRA does not contain the goal state, we call it unpromising and construct a detour sampling domain for detouring operations from obstacles constraining the manipulator. The union of promising MRA and the detour sampling domain approximates our productive regions, and we bias our sampling to cover these domains more.We have applied to our Productive Regions-Oriented Task space planner (PROT) to 8- and 20-dofs manipulators. We have observed that our method achieves up to 2.88 times improvement over the prior task-space planner.

Contents

Paper: PDF (0.23MB)
@inproceedings{PROT14,
  author    = {Junghwan Lee and Sung-eui Yoon},
  title     = {PROT: Productive Regions-Oriented Task space path planning for
hyper-redundant manipulators},
  booktitle = {ICRA},
  year      = {2014},
 }

Earlier version (technical report): PROT: Productive Regions-Oriented Task space path planning for hyper-redundant manipulators 

Junghwan Lee and Sung-Eui Yoon
KAIST Tech. Report, CS-TR-2013-375
April, 2013

Dept. of Computer Science
KAIST
373-1 Guseong-dong, Yuseong-gu, Daejeon, 305-701
South Korea
sglabkaist at gmail dot com