晶格式模块机器人系统自修复研究与仿真  

Research and Simulation of Lattice Modular Robot Self-Repairing

作　　者：俞港[1] 付庄[1] 管恩广[1] 

机构地区：[1]上海交通大学,上海200240

出　　处：《计算机仿真》2016年第11期384-389,共6页Computer Simulation

基　　金：国家自然科学基金资助项目(61473192)

摘　　要：模块机器人系统由多个自主的智能模块机器人(以下简称"模块")组成。当模块出现故障时,会被系统移除并在原位置留下"空穴"。通过模块间的运动协作改变系统内部结构,可将"空穴"移至位于系统边缘的备用模块处,以实现自修复。自修复在保证系统的可靠性和鲁棒性上起到了关键的作用。模块机器人系统自修复的难点在于"空穴"路径规划,由于系统动态的故障情况和模块机器人特殊的运动特性,路径规划算法不仅需要具有动态性,且相比于自由体路径规划算法具有更多的约束。基于数字激素模型,提出一种晶格式模块机器人系统自修复算法。上述算法参考生物的激素作用机理,引入活化激素和抑制激素的概念,模块通过其周围激素浓度进行动态路径规划。仿真结果表明,对于晶格式模块机器人系统自修复问题,改进算法表现出可行的、良好的结果,可适用于大规模、多故障模块、故障动态变化的系统。Modular robot system consists of many independent and intelligent modules. Serf-repairing means the system can replace the fault module with the good one by cooperative motion between modules; it makes the system more reliable and robust. The key of modular robot system self-repairing is path planning algorithm. It is different from the traditional one, because of the dynamic failure situations and the special kinematics characteristic of modular robot. A self-repairing algorithm based on Digital Hormone Model （DHM） was presented, and planning path was conducted with the surrounding concentration of hormone. Simulation results show that the algorithm can guide the self-repairing path planning of lattice modular robot system, especially for the large-scale, muhi-fault module, dynamic changing system.

关 键 词：模块机器人 自修复 数字激素 

分 类 号：TP242[自动化与计算机技术—检测技术与自动化装置]