基于群体智能的多机器人任务分配  

Multi-robot task allocation based on swarm intelligence

作　　者：刘淑华[1] 张嵛[1] 吴洪岩[1] 刘杰[1] 

机构地区：[1]东北师范大学计算机学院,长春130017

出　　处：《吉林大学学报（工学版）》2010年第1期123-129,共7页Journal of Jilin University:Engineering and Technology Edition

基　　金：国家自然科学基金项目(60573067)

摘　　要：针对具有松散和紧密耦合型任务的大规模多机器人系统,研究了基于群体智能的任务分配方法。系统采用层次结构,高层用蚁群算法实现松散耦合型任务分配的寻优,提出逆转分配思想让蚂蚁代表任务,为每个任务选择任务的承担者。底层分别提出了基于蚁群、粒子群蚁群和量子蚁群实现机器人联盟的形成——产生紧耦合型任务解,并进行仿真。仿真结果表明,基本蚁群算法得到的解质量最差;粒子群蚁群算法得到的分配解最好,但是运算时间最长;量子蚁群算法得到的解稍次于粒子群蚁群算法,但分配时间比另两种算法减少了一半。因此,在大规模的多机器人任务分配中,量子蚁群算法具有更强的适用性。The task allocation was studied based on the swarm inteeligence for the large-scale multi- robot system with loose-and tight-coupled tasks adopting the hierarchial architecture. In the high level, the ant colony algorithm was employed to find the optimal allocation of the loose-coupled tasks, namely, based on the reverse distribution idea, taking each ant to form a task, an undertaker was chosen for every task. In the low level, the coalition formation algorithms based on the ant colony optimization（ACO）, the particle swarm and ant colomy optimization（PSACO）, or the quantum-inspised ant colony optimization （QACO） was proposed respectively for performin a tight-coupled task. Simulations were performed and results showed that PSACO provides the best solution, but its running time is the largest; QACO is a little inferior in solution quality to PSACO, however,it needs only a half time of the 2 other methods. Therefore, QACO appears more suitable for the task allocation of the large-scale multi-robot system.

关 键 词：自动控制技术 任务分配 机器人联盟形成 蚁群优化 粒子群蚁群优化 量子蚁群优化 

分 类 号：TP242[自动化与计算机技术—检测技术与自动化装置]