钱堃, 马旭东, 戴先中, 房芳. 预测行人运动的服务机器人POMDP导航[J]. 机器人, 2010, 32(1): 18-24,33..
QIAN Kun, MA Xudong, DAI Xianzhong, FANG fang. POMDP Navigation of Service Robots with Human Motion Prediction. ROBOT, 2010, 32(1): 18-24,33..
Abstract:To improve the natural pedestrian-avoidance skills of service robots in indoor dynamic environments,a method of combining long-term and short-term prediction of pedestrian's motion is introduced on the basis of modeling humans'motion trajectory patterns.In order to accommodate the uncertainties in the perception-control loop of robots,which are mainly caused by sensor noise and time delay in network and other factors,the relative position relation between human and robot is modeled as partially observable Markov state.Partially observable Markov decision process(POMDP) is utilized for probabilistic decision-making under multi-source uncertainties,and the behavior modules of the global path planner,the motion reactor and the speed controller,are coordinated.Experimental results illustrate the performance of safe navigation that can avoid conflicts in advance,as well as the improved robot navigation efficiency by avoiding repeated zigzaging and wandering motion.
[1] Osentoski S,Manfredi V,Mahadevan S.Learning hierarchical models of activity[C]//IEEE/RSJ International Conference on Intelligent Robots and Systems.Piscataway,NJ,USA:IEEE,2004:891-896.
[2] Bermewitz M,Burgard W,Thrun S.Using EM to learn motion behaviors of persons with mobile robots[C]//IEEE/RSJ International Conference on Intelligent Robots and Systems.Piscataway,NJ,USA:IEEE,2002:502-507.
[3] Miura J,Shirai Y.Modeling motion uncertainty of moving obstacles for robot motion planning[C]//IEEE International Conference on Robotics and Automation.Piscataway,N J,USA:IEEE,2000:2258-2263.
[4] Hoeller E Schulz D,Moors M,et al.Accompanying persons with a mobile robot using motion prediction and probabilistic roadmaps[C]//IEEE/RSJ International Conference on Intelligent Robots and Systems.Piscataway,N J,USA:IEEE,2007:1260-1265.
[5] Lin L,Fox D,Hightower J,et al.Voronoi tracking:Location estimation using sparse and noisy sensor data[C]//IEEE/RSJ International Conference on Intelligent Robots and Systems.Piscaraway,NJ,USA:IEEE,2003:723-728.
[6] Schuiz D,Burgard W,Fox D,et al.People tracking with mobile robots using sample-based joint probabilistic data association filters[J].International Journal of Robotics Research,2003,22(2):99-116.
[7] Lopez M E,Bergasa L M,Barea R,et al.A navigation system for assistant robots using visually augmented POMDPs[J].Autonomous Robots,2005,19(1):67-87.
[8] Pineau J,Gordon G,Thrnn S.Anytime point-based approximations for large POMDPs[J].Journal of Artificial Intelligence Research,2006,27:335-380.
[9] Ross S,Chaib-draa B,Pinean J.Bayesian reinforcement learning in continuous POMDPs with application to robot navigation[C]//IEEE International Conference on Robotics and Automation.Piscataway,NJ,USA:IEEE,2008:2845-2851.
[10] Foka A,Trahanias P.Real-time hierarchical POMDPs for autonomous robot navigation[J].Robotics and Autonomous Systems,2007,55(7):561-571.
[11] Qian K,Ma X D,Dai X Z.Simultaneous robot localization and person tracking using Rao-Blackwellised particle filters with multi-modal sensors[C]//IEEE/RSJ International Conference on Intelligent Robots and Systems.Piscataway,NJ,USA:IEEE,2008:3452-3457.
[12] Szczerba R J,Chen D Z,Uhran J J.A grid-based approach for finding conditional shortest paths in an unknown environment[EB/OL].[2009-01-20].http://www.cse.nd.edu/Reports/1994/tr294234.ps,2007.
[13] Minguez J,Montano L.Sensor-based robot motion generation in unknown,dynamic and troublesome scenarios[J].Robotics and Autonomous Systems,2005,52(4):290-311.
[14] Spaan M T J,Viassis N.Perseus:Randomized point-based value iteration for POMDPs[J].Journal of Artificial Intelligence Research,2005,24:195-220.