Adaptive Space Expansion for Fast Motion Planning  

作　　者：Shenglei Shi Jiankui Chen 

机构地区：[1]State Key Laboratory of Digital Manufacturing Equipment and Technology,School of Mechanical Science and Engineering,Huazhong University of Science and Technology,Wuhan 430074,China

出　　处：《IEEE/CAA Journal of Automatica Sinica》2024年第6期1499-1514,共16页自动化学报（英文版）

基　　金：supported in part by the National Natural Science Foun-dation of China(51975236);the National Key Research and Development Program of China(2018YFA0703203);the Innovation Project of Optics Valley Laboratory(OVL2021BG007)。

摘　　要：The sampling process is very inefficient for sam-pling-based motion planning algorithms that excess random sam-ples are generated in the planning space.In this paper,we pro-pose an adaptive space expansion(ASE)approach which belongs to the informed sampling category to improve the sampling effi-ciency for quickly finding a feasible path.The ASE method enlarges the search space gradually and restrains the sampling process in a sequence of small hyper-ellipsoid ring subsets to avoid exploring the unnecessary space.Specifically,for a con-structed small hyper-ellipsoid ring subset,if the algorithm cannot find a feasible path in it,then the subset is expanded.Thus,the ASE method successively does space exploring and space expan-sion until the final path has been found.Besides,we present a particular construction method of the hyper-ellipsoid ring that uniform random samples can be directly generated in it.At last,we present a feasible motion planner BiASE and an asymptoti-cally optimal motion planner BiASE*using the bidirectional exploring method and the ASE strategy.Simulations demon-strate that the computation speed is much faster than that of the state-of-the-art algorithms.The source codes are available at https://github.com/shshlei/ompl.

关 键 词：Adaptive space expansion(ASE) hyper-ellipsoid ring informed sampling motion planning. 

分 类 号：TP242[自动化与计算机技术—检测技术与自动化装置]