Human-machine cooperative path planning of an agricultural mobile robot based on a cloud model
-
摘要:目的
实现农业移动机器人在复杂动态的农业环境中实时准确地无碰撞行驶。
方法基于云模型的不确定性在线推理方法,提出一种基于云模型的动态引导A*(CDGA*)算法进行人机合作路径规划,将人的专业知识和喜好等引入DGA* 优化中,实现机器人更快速的路径规划。利用Matlab软件对CDGA* 算法与DGA* 算法进行仿真对比分析。
结果静态路径规划中,DGA* 算法与CDGA* 算法的close的点数分别为158和96,人员规划时间分别为8.8和4.0 s,规划总时间分别为15.6和8.9 s;动态路径规划中,DGA* 算法与CDGA* 算法的人员规划时间分别为12.5和5.8 s,规划总时间分别为23.3和14.6 s。
结论提出的CDGA* 算法能够大大减少产生的节点数,缩短规划时间,提高搜索效率。
Abstract:ObjectiveTo make an agricultural robot accurately find a path without collision in complex and dynamic environment in real time.
MethodUsing online uncertainty reasoning based on a cloud model, a dynamic guidance A* algorithm based on the cloud model (CDGA*) was proposed to realize human-machine cooperative path planning. Human’s expertise and preferences were incorporated into the DGA* optimization process to implement a faster path planning. Matlab software was used to simulate and analyze the CDGA* and DGA* algorithms.
ResultIn static path planning, the numbers of close points of the DGA* and CDGA* algorithms were 158 and 96, human planning time was 8.8 and 4.0 s, the total planning time was 15.6 and 8.9 s, respectively. In dynamic path planning, human planning time of the DGA* and CDGA* algorithms was 12.5 and 5.8 s, the total planning time was 23.3 and 14.6 s, respectively.
ConclusionThe proposed CDGA* algorithm can largely decrease the number of nodes, reduce computation time and improve planning efficiency.
-
-
表 1 静态路径规划中3种算法的性能比较
Table 1 Performance comparison of three algorithms in static path planning
算法 规划总时间/s 人员规划时间/s open点数 close点数 f(i) Jlength/m Jsmooth A* 148.3 0.0 11 706 3 354 591.6 1 124.6 1.12 DGA* 15.6 8.8 2 146 158 594.9 1 127.5 1.27 CDGA* 8.9 4.0 1 568 96 596.2 1 129.7 1.45 表 2 动态路径规划中DGA*与CDGA*算法的性能比较
Table 2 Performance comparisons of DGA* and CDGA* algorithms in dynamic path planning
算法 规划总
时间/s人员规划
时间/sf(i) Jlength/m Jsmooth DGA* 23.3 12.5 661.4 1 276.7 2.16 CDGA* 14.6 5.8 664.9 1 279.1 2.27 -
[1] NOGUCHI N, TERAO H. Path planning of an agricultural mobile robot by neural network and genetic algorithm[J]. Comput Electron Agr, 1997, 18(2/3): 187-204.
[2] 马建光, 方敦原. 一种基于概率方法的车型机器人路径规划方法[J]. 计算机工程与应用, 2003, 39(34): 93-95. [3] 孙凤池, 黄亚楼, 康叶伟, 等. 车型移动机器人SPRM路径规划[J]. 机器人, 2005, 27(4): 325-329. [4] 赵百轶, 张立军, 贾鹤鸣. 基于四叉树和改进蚁群算法的全局路径规划[J]. 应用科技, 2011, 38(10): 23-28. [5] 史恩秀, 陈敏敏, 李俊, 等. 基于蚁群算法的移动机器人全局路径规划方法研究[J]. 农业机械学报, 2014, 45(6): 53-57. [6] 李擎, 张超, 韩彩卫, 等. 动态环境下基于模糊逻辑算法的移动机器人路径规划[J]. 中南大学学报(自然科学版), 2013, 44: 104-108. [7] 孟蕊, 苏维均, 连晓峰. 基于动态模糊人工势场法的移动机器人路径规划[J]. 计算机工程与设计, 2010, 31(7): 1558-1561. [8] 王殿君. 基于改进A*算法的室内移动机器人路径规划[J]. 清华大学学报(自然科学版), 2012, 52(8): 1085-1089. [9] DONG Z, CHEN Z J, ZHOU R, et al. A hybrid approach of virtual force and A* search algorithm for UAV path re-planning[J]. Industrial Electronics & Applications, 2011, 49(20): 1140-1145.
[10] HSU W Y. Brain-computer interface: The next frontier of telemedicine in human-computer interaction[J]. Telemat Inform, 2015, 32(1): 180-192.
[11] WINCK C R, ELTON M, BOOK J W. A practical interface for coordinated position control of an excavator arm[J]. Automat Constr, 2015, 51: 46-58.
[12] ZHONG H, WACHS J P, NOF S Y. Telerobot-enabled HUB-CI model for collaborative lifecycle management of design and prototyping[J]. Comput Ind, 2014, 65(4): 550-562.
[13] SPARC. Farming with robots[EB/OL]. [2016-05-04]. http://robohub.org/farming-with-robots/.
[14] 苗夺谦. 不确定性与粒计算[M]. 北京: 科学出版社, 2011:1-6. [15] 谭雁英, 胡淼, 祝小平, 等. 基于人机合作策略下SAS算法的多无人及路径再规划[J]. 西北工业大学学报, 2014, 32(5): 688-692. [16] 马飞, 杨皞屾, 顾青,等. 基于改进A*算法的地下无人铲运机导航路径规划[J]. 农业机械学报, 2015, 47(10): 303-309. [17] ZHENG C W, XU F J, HU X H, et al. Online route planner for unmanned air vehicle navigation in unknown battlefield environment[C]//IMACS multiconference on computational engineering in systems applications, Oct 4-6, 2006, Beijing. New York: IEEE, 2006, 1:814-818.
[18] DUAN H B, YU Y X, ZHANG X Y, et al. Three-dimension path planning for UCAV using hybrid meta-heuristic ACO-DE algorithm[J]. Simul Model Pract Th, 2010, 18(8): 1104-1115.
[19] BESADA-PORTAS E, DE LA TORRE L, DE LA CRUZ J M, et al. Evolutionary trajectory planner for multiple UAVs in realistic scenarios[J]. IEEE T Robot, 2010, 26(4): 619-634.
[20] DECHTER R, PEARL J. Generalized best-first search strategies and the optimality of A*[J]. J Acm, 1985, 32(3): 505-536.
[21] SUN X, CAI C, SHEN X. A new cloud model based human-machine cooperative path planning method[J]. J Intell Robot Syst, 2015, 79(1): 3-19.
[22] MARLER R T, ARORA J S. The weighted sum method for multi-objective optimization: New insights[J]. Struct Multidiscip O, 2010(41): 853-862.
[23] BOSKOVIC J, KNOEBEL N, MOSHTAGH N, et al. Collaborative mission planning & autonomous control technology (CoMPACT) system employing swarms of UAVs[C]//AIAA guidance, navigation, and control conference, August 10-13, 2009, Chicago, Illinois. Reston: AIAA. doi.org/10.2514/6.2009-5653.