Collaborative Vehicle and UAV Inspection Path Planning Considering Fault Point Differentiation

With rising labor costs and increasing safety requirements, the economics of traditional inspection methods are further reduced. Utilizing UAV inspection provides new ideas to address the limitations of traditional inspection methods. The aim of this thesis is to consider a two-stage hybrid heuristic algorithm (QWMA) based on the whale migration optimization algorithm for inspecting and repairing faulty nodes of different levels using the working mode of collaborative inspection by vehicles and UAVs. In the first stage, considering the inspection node variability and size, a clustering algorithm is utilized to assign different nodes to be inspected to different vehicles for inspection. In the second stage, the QWMA algorithm is utilized to solve the scheduling problem of vehicles and UAVs by adding seven local optimization operators, which are combined with the Q-learning selection strategy to finally obtain the route with the least costly vehicle and UAV collaboration. Finally, the efficiency and stability of the QWMA algorithm in solving the vehicle and UAV cooperative inspection problem is demonstrated by comparing and analyzing it with other five different types of algorithms under 16 instances of different sizes.