摘要
随着机器人应用范围的扩大,对其能力的要求越来越高,多个机器人协作可以完成单个机器人无法完成的复杂任务。编队控制是一个具有典型和通用性的多机器人协调问题,是多机器人协调问题的基础,对多机器人协作系统的研究有巨大的推动和促进作用。
本文采用基于行为的多机器人编队控制方法,但是在这种编队控制方法中,存在基本行为的有效设置和权重参数值的优化选择两个问题。
针对基本行为的有效设置问题,本文对机器人躲避静态障碍物和躲避机器人碰撞行为做了改进,设计了一个随机器人到障碍物或其他机器人的最短距离减小而增大的偏转角度,使得机器人能够平滑地躲开静态障碍物或其他机器人。对于实体领航机器人的容错性差的问题,本文在机器人队形的几何中心设置一个虚拟机器人作为领航机器人,带领整个机器人群体运动。
其次针对多机器人编队控制中行为控制参数值选择困难的问题,本文采用粒子群优化算法选择更优的控制系数。但是由于优化目标是高维多极值点的函数,标准粒子群算法在优化高维目标时易出现早熟和收敛慢的问题,本文提出了一个改进的自适应粒子群算法,根据粒子的空间分布和进化状态,设置相应的自适应控制参数,同时对陷入全局最优点的粒子采取一个非均匀逃逸跳跃策略,仿真测试表明改进的粒子群算法具有更好的搜索性能。
最后把改进的面向高维优化目标的自适应粒子群优化算法用于多机器人编队控制中,选择出更优的行为权重参数值,在MATLAB平台上对多机器人编队控制做了仿真,通过和优化前的编队控制对比,证明了基于粒子群优化算法的编队控制算法的有效性。
With the expansion of robot application, the demand to robot's ability is more and more high. Multiple robots cooperatively complete more complex task that single robot can not complete. Formation control is a common and typical multi-robot coordination, and it is also the basis for multi-robot coordination. Formation control can greatly promote and facilitate multi-robot cooperative system research.
Here we adopt behavior-based multi-robot formation control method, but there are two problems in the method of formation control, those are the design of effective basic behavior and the choice of better parameters of weight values.
Firstly, for the problem of effective design of basic behavior, this thesis makes improvements on avoid static obstacles and avoid robot. We design a yaw angle that increases with the minimum distance between robot and obstacles or other robots decreases, which can smoothly avoid static obstacles or other robots. For the physical leader robot's poor fault tolerance, we set a virtual leader at the geometric center of the formation, and guide the whole group's moving.
Secondly, for the problem of difficulty in selecting the better control values, we use particle swarm optimization algorithm to select a more appropriate control coefficients. However, due to high-dimensional opt-imization object is a function of multiple extreme points, the standard particle swarm optimization algorithm for high-dimensional target easily gets into the problem of prematurity and has slow convergence, so we proposed an improved adaptive particle swarm optimization, that is, according to the spatial distribution and evolution state of particles, we set the corresponding adaptive control parameters, and the particle which gets into a global optimal point takes non-uniformly jumping out and escaping strategy. The simulation shows that the improved algorithm has better performance.
Finally, the improved oriented high-dimensional adaptive particle swarm optimization is applied to multi-robot formation control, and we get a better parameters. Then we do the simulation of multiple robots formation control on MATLAB platform. The result proved the feasibility of formation control algorithm and better performance.
引文
[1]谭明.机器人群体协作与控制的研究.机器人,2001,23(2)：178～182
[2]王越超,谈大龙.协作机器人学的研究现状和发展.机器人,2002,20(6)：69～75
[3]陈卫龙,董胜龙,席裕庚.基于开放式多智能体结构的分布式自主机器人系统.机器人,2001,23(1)：45～50
[4]董胜龙,陈卫东,席裕庚.多移动机器人编队的分布式控制系统.机器人,2000,22(6)：433～438
[5]Y.Chen, Z.Wang, Formation control:A Review and Consideration. IEEE/-RSJ Int Conf on Intelligent Robots and Systems, Alberta, Canada,2005.3181-3186
[6]韩学东,洪炳熔,孟伟.多机器人任意队形分布式控制研究.机器人,2003,25(1)：66～73
[7]Bahceci E, Soysal O, Sahin E. A review, pattern formation and adaptation in multi-robot systems:[CMU technical report]. CMU-RI-TR-03-43,2003
[8]R.Carellie, C.De la Cruz, F Roberti. Centralized formation control of non-holonomic mobile robots. In:Latin American Applied,2006
[9]Li Wang, Zengqiang Chen, Zhongxin Liu. Formation Control of Multi-agent System Based on Potential function. Asia Simulation Conference 7th Intern-ational Conference on System simulating and scientific computing,2008. 101-105
[10]曹志强,王硕,张斌.基于Motor Schema的多移动机器人队形优化控制.控制与决策,2001,16(增刊)：763～765
[11]王醒策,张汝波,顾国昌.基于强化学习的多机器人编队方法研究.计算机工程,2002,28(6)：15～16
[12]F.Kobayashi, N.Tomita, F.Kojima. Re-formation of Mobile Robots using Genetic Algorithm and Reinforcement Learning. In:SICE Annual Conference in Fukui,2003
[13]Goldberg D, Mataric M J. Maximizing reward in a non-stationary mobile robot environment, invited submission to the Best of Agents-2000, special issue of Autonomous Agents and Multi-Agent Systems,2003,6(3):67-83
[14]A.K.Das, R.Fierro,V.Kumar, etal. A vision-based formation control framework. IEEE Transactions on Robotics and Automation,2002,18(5):813-825
[15]Naffin D J, Sukhatme G S. Negotiated formations. In:Proceedings of Internat-ional Conference on Intelligent Autonomous Systems (IAS),2004,181-190
[16]Desai J P. Modeling multiple teams of mobile robots:a graph theoretic approach. In:Proceedings of IEEE/RSJ International Conference on Intelligent Robots and Systems,2001
[17]Bahceci E, Soysal O, Sahin E. A review:pattern formation and adaptation in multi-robot systems:[CMU technical report]. CMU-RI-TR-03-43,2003
[18]Lynne E Parker. ALLIANCE:an architecture for fault tolerant multi-robot cooperation. IEEE Robotics &Automation Magazine,1998,5(1):36-47
[19]Jorg P Muller. The Design of Intelligent Agents:a Layered Approch. New York: Springer-Verlag,1996
[20]赵忆文.移动机器人体系结构及多机器人协调规划研究：[硕士学位论文].沈阳：中国科学院沈阳自动化研究所,2000
[21]Jacques Ferber. Multi-Agent Systems:An Introduction to Distributed Artificial Intelligence. New York:ADDISON-WESLEY,1999
[22]王越超.多机器人协作系统研究：[硕士学位论文].哈尔滨：哈尔滨工业大学,1999
[23]丁永生.计算智能理论技术与应用.北京：科学出版社,2004.24～29
[24]王树禾.图论及其算法.北京：中国科学技术出版社,1990.54～58
[25]T. Balch, and R.C. Arkin. Behavior based formation control for multi-robot teams, IEEE Trans on Robotics and Automation,1998,14 (6):926-939
[26]S G Goodridge. A fuzzy behavior-based nervous system for an autonomous mobile robot:[Master thesis]. North Carolina State:North Carolina State University,1994
[27]Kostelnik P, Samulka M, Janosik M. Scalable multi-robot formations using local sensing and communication. Proceeding of the 3rd International Workshop on Robot Motion and Control, Poznan,2002
[28]J. Lawton, B.Young, R.Beard. A decentralized approach to elementary formation manuevers. In:Proc. IEEE Int. Conf. Robotics and Automation, San Francisco, CA,2000.2728-2733
[29]J. R. T. Lawton, B. J. Young, R.W. Beard. A decentralized approach to formation maneuvers. IEEE Trans.On Robotics and Automation,2003,19(6):933-941
[30]Jaydev P Desai, Jim Ostrowski, Vijay Kumar. Controlling formations of multiple mobile robots. IEEE Int Conf on Robotics and Automation Belgium, 1998.2864-2869
[31]Hiroacki Yamaguchi, Joel w Burdick. Asymptotic stabilization of multiple nonholonomic mobile robots forming group formations. IEEE Int Conf on Robotics and Automation. Belgium,1998.3573-3580
[32]Pereira G A S, Das A K, Kumar V, etal. Formation control with configuration space constraints. Proc of the IEEE/RJS hit Conf on Intelligent Robots and Systems. Las Vegas,2003.2755-760
[33]王娟,胡玉兰.基于条件反馈机制Leader-follower多机器人队形控制.大众科技,2008年7月：42～46
[34]李波,王祥凤.基于动态Leader多机器人队形控制.长春工业大学学报,2009,39(2)：12～15
[35]M. A. Lewis, K.H. Tan. High precision formation control of mobile robots using virtual structures. AutoRobots,2001,4:387-403
[36]Qin Li, Zhongping Jiang. Formation Tracking Control of Unicycle Teams with Collision Avoidance.47th IEEE Conference on Decision and Control,2008. 496-501
[37]Khatib O. Real-time obstacle avoidence for manipulators and mobile robots. International Journal of Robotic Research,1986,5(1):90-98
[38]刘佰龙,张汝波.基于Amigobot平台的多机器人编队控制方法研究：[硕士学位论文].哈尔滨：哈尔滨工程大学,2006
[39]胡玮韬,李团结.多机器人编队及运动控制研究：[博士学位论文].西安：西安电子科技大学,2010
[40]Kennedy J, Eberhan R C. Particle Swarm Optimization. Proceeding IEEE Int Conf on Neural Networks,1995.1942-1948
[41]Shi Y, Eberhart R.C. Parameter selection in particle swarm optimization. Proceedings of Evolutionary Programming VII,1998.591-600
[42]Lovbjerg M, Krink T. The Life Cycle Model:combining particle swarm optimisation, genetic algorithms and hillclimbers. Proceedings of Parallel Problem Solving from Nature VII (PPSN),2002.621-630
[43]Niknam, T, Amiri, B. An efficient hybrid approach based on PSO, ACO and k-means for cluster analysis. Applied Soft Computing,2010,10(1):183-197
[44]J.Kennedy, R. Mendes. Population structure and particle swarm performance. In: Proc. IEEE Congr. Evol. Compute, Honolulu, HI,2002.1671-1676
[45]J.Kennedy, R.Mendes. Neighborhood topologies in fully informed and best-of-neighborhood particle swarms. IEEE Trans on Syst, Man, Cybern.2006 36(4):515-519
[46]姜海明.具有随机停滞行为的粒子群优化算法.四川大学学报(工程科学版),2006,38(4)：117～121
[47]夏桂梅,曾建潮.双群体随机微粒群算法册.计算机工程与应用,2006,42(24)：46～48
[48]Riget.J, Vestterstrom.J. A diversity-guided particle Swarm optimizer-the ARPSO. Department of Computer Science, University of Aal-hUs, Denmark,2002
[49]Zhan Z, Zhang J, Chung H. Adaptive Particle Swarm Optimization. IEEE Transactions on Systems, Man, and Cybernetics,2009,39 (6):1362-1381
[50]赵海,宋纯贺.面向高维度目标函数的微粒群优化算法.东北大学学报(自然科学版),2008,29(5)：6～10
[51]Thrun S, Bucken A. Interating grid-based and topological maps for mobile robot navigation. In:Proceedings of the Thirteeth National Conference on Artifical Intelligence, Portland,1996.944-950