Multi-UAV Path Planning for Persistent Monitoring Under Uncertain Flight Time

Unmanned aerial vehicles (UAVs) have been widely used in persistent monitoring tasks because of their high flexibility and ease of deployment. However, the time consumed by a UAV to fly on a fixed route can be uncertain due to factors such as weather changes, human influences, and the condition of the UAV itself. This study addresses the path planning problem for multi-UAV persistent monitoring under uncertain flight time, in which UAVs are required to persistently monitor nodes with strict revisit constraints. The objectives are to minimize the number of used UAVs without violating monitoring constraints as well as to ensure balanced task allocation among them. By introducing chance constraints, the stochastic optimization problem is converted into a deterministic one. A weighted approach is then employed to tackle the dual-objective optimization. The study proposes the ant colony initial genetic algorithm (ACIGA) to obtain the minimum number of used UAVs and their corresponding paths. Simulation results demonstrate that this algorithm could effectively find viable solutions for persistent monitoring tasks, with UAVs’ paths meeting monitoring requirements even under uncertain flight time. Furthermore, this algorithm has good scalability and significant improvement in computational cost.