Task Planning Method for Multi-inspection Robot Collaboration in Substation
Inspection is essential for ensuring the safety of substations.At present,inspection robots have been widely used in substation inspection.In order to furtherly integrate the inspection resources,improve the efficiency,and achieve the transformation from single robot inspection to collaborative inspection,this paper studies the collaborative planning of multi-robot inspection tasks.Firstly,it constructs a multi-dimensional intelligent inspection system of substation without information barriers.Then,it proposes an optimization model of task point allocation and traversal,and designs a comprehensive index considering energy consumption and time cost,which is solved by improved genetic algorithm as the goal function.Finally,it uses a deep Q learning network for training to realize the collaborative path planning of multi-robots.An actual circumstance of 500 kV substation in operation is taken as an example to generate a collaborative inspection scheme,which verifies the feasibility of the proposed method.The result indicates this inspection scheme has lower energy-time cost compared with the traditional inspection schemes.
万方数据
维普
