基于改进合同网算法的异构多AUV协同任务分配

Heterogeneous Multi-AUV Cooperative Task Allocation Based on Improved Contract Net Algorithm

传统合同网算法应用在异构多自主式水下航行器(AUV)协同任务分配时,存在招标过程中多种招标者并存的情况,不易产生有效招标者;在投标过程中,潜在投标者增加了无效投标数量及其招标者对投标结果的评估负担,极易产生任务不合理的情况。针对以上2种问题,文中提出了一种基于改进合同网算法的异构多AUV任务分配策略。该方法将任务负载率指标和令牌环网概念结合起来,有效解决选择招标者及其任务不合理的问题。基于MATLAB的三维任务环境的仿真实验表明,对于异构型多AUV进行任务分配,文中提出的改进合同网算法能够有效提高整体效能并进行合理的任务分配。

When traditional contract net algorithm is applied to heterogeneous multi-AUV collaborative task allocation, co-existence of a variety of bid inviters occurs in bid winning process, leading to difficulty for producing effective inviter, while in bidding process, potential bidders raise the number of invalid bids, and hence increase burden on the inviter for evaluating the bid, which is easy to result in unreasonable tasks. Aiming at the above two problems, this paper proposes a heterogeneous rnulti-AUV task allocation strategy based on improved contract net algorithm. This strategy combines the task load rate and the token ring network to solve the problems of selecting bid inviter and its unreasonable task. Three-dimensional environment simulation based on MATLAB shows that the improved contract net algorithm can effectively enhance the overall performance and make reasonable task allocation scheme for task allocation of heterogeneous multi-AUV.