An indoor path planning and motion planning method based on POMDP

被引：0

作者：

Dong, Wenjie ^{[1
,2
,3
]}

Qi, Xiaozhi ^{[2
,3
]}

Chen, Zhixian ^{[1
,2
,3
,4
]}

Song, Chao ^{[1
,2
,3
]}

Yang, Xiaojun ^{[1
]}

机构：

[1] Harbin Inst Technol, Shenzhen Grad Sch, Shenzhen, Peoples R China

[2] Chinese Acad Sci, Shenzhen Inst Adv Technol, Guangdong Prov Key Lab Robot & Intelligent Syst, Shenzhen, Guangdong, Peoples R China

[3] Chinese Univ Hong Kong, Hong Kong, Peoples R China

[4] Univ Chinese Acad Sci, Beijing, Peoples R China

来源：

2017 IEEE INTERNATIONAL CONFERENCE ON ROBOTICS AND BIOMIMETICS (IEEE ROBIO 2017) | 2017年

关键词：

path planning; partially observable Markov decision; motion planning; mobile robots;

D O I：

暂无

中图分类号：

TP24 [机器人技术];

学科分类号：

080202 ; 1405 ;

摘要：

With the continuous development of robot technology, robots' tasks and environment is becomes more and more complex. And the Mobile Service Robot's need stronger motion planning ability to deal with the uncertainty and dynamic. In this paper, path planning and motion planning of mobile robot in the indoor environment with uncertain pedestrian flows are studied. A dedicated planner including velocity planning layer and direction planning layer based on the pedestrian tracking and observation is designed, then the effective solving method for the planner is studied based on POMDP to calculate the robot linear moving speed and steering angle. What's more, we improve the quality of the output path algorithm, based on the artificial potential field and the A* algorithm. Therefore, the robots can learn from previous experience of path planning by itself to avoid the path which has relatively poor performance, and optimize the motion planning on this basis. Eventually, the robot's adaptability and security in indoor dynamic environment is improved.

引用

页码：1564 / 1570

页数：7

共 20 条

[11]

Roy N, 2012, ROBOTICS SCI SYSTEMS

[12]

Schneider N, 2013, LECT NOTES COMPUT SC, V8142, P174, DOI 10.1007/978-3-642-40602-7_18

[13]

Schulz A T, 2015, INT VEH S IEEE

[14]

Schulz A T, 2015, IEEE INT C INT TRANS

[15] A Controlled Interactive Multiple Model Filter for combined Pedestrian Intention Recognition and Path Prediction [J].

Schulz, Andreas T. ;

Stiefelhagen, Rainer .

2015 IEEE 18TH INTERNATIONAL CONFERENCE ON INTELLIGENT TRANSPORTATION SYSTEMS, 2015, :173-178

[16]

Silver D., 2010, ADV NEURAL INFORM PR, P2164, DOI DOI 10.5555/2997046.2997137

[17] OPTIMAL CONTROL OF PARTIALLY OBSERVABLE MARKOV PROCESSES OVER A FINITE HORIZON [J].

SMALLWOOD, RD ;

SONDIK, EJ .

OPERATIONS RESEARCH, 1973, 21 (05) :1071-1088

[18]

Somani A., 2013, P INT C NEUR INF PRO, P1772

[19]

Thrun S., 2007, SPRINGER TRACTS ADV, P1, DOI DOI 10.1007/978-3-540-73429-1_1

[20] Growing Hidden Markov Models: An Incremental Tool for Learning and Predicting Human and Vehicle Motion [J].

Vasquez, Dizan ;

Fraichard, Thierry ;

Laugier, Christian .

INTERNATIONAL JOURNAL OF ROBOTICS RESEARCH, 2009, 28 (11-12) :1486-1506

← 1 2 →