新型混合式多任务规划优化算法在服务机器人仿真中的应用  

New Hybrid Multi-task Planning Optimization Algorithm in Service Robot Simulation

作　　者：胡晨潇 陈万米[1] 刘璇 

机构地区：[1]上海大学机电工程与自动化学院,上海200072

出　　处：《工业控制计算机》2020年第5期89-91,共3页Industrial Control Computer

摘　　要：服务机器人仿真属于非经典规划的范畴,主要研究基于真实场景中服务机器人在完成多个任务时的最优行动路径和行动序列规划。将全局规划算法与局部最优算法相结合,提出了一种新型混合式多任务规划优化算法,在保证全局规划任务序列为优的同时,将补充信息作为启发式信息及时反馈处理,加强局部搜索规划能力,并通过仿真实验证明了该算法的可行性和任务执行的高效性。The simulation of service robots belongs to the category of non-classical planning.It mainly studies the optimal action path and action sequence planning based on the real scene when the robot completes multiple tasks.In this paper,a new hybrid multi task planning optimization algorithm is proposed,which combines the global planning algorithm with the local optimization algorithm.While ensuring the global planning task sequence to be optimal,the supplementary information is used as heuristic information to feed back in time to strengthen the local search planning ability.The feasibility of the algorithm and the efficiency of task execution are proved by simulation experiments.

关 键 词：非经典规划问题 局部搜索 服务机器人仿真 启发式信息 全局规划 

分 类 号：O224[理学—运筹学与控制论] TP242[理学—数学]