Task Allocation Optimization for Multi-UAV Collaborative Power Inspection at Single Parking Point

Abstract

This paper focuses on task allocation for multi-UAV collaborative power inspection at a single parking point, proposing an optimization method to minimize the total inspection time per parking. Key contributions include: 1) establishing a dynamic model for multi-rotor UAVs and a collaborative inspection model with clear assumptions; 2) developing a technical route covering task allocation, path planning, and conflict detection; 3) designing an immune algorithm-based parking point selection method and a UAV-operation vehicle collaborative inspection model; 4) implementing a genetic algorithm for sparse tower scenarios and a variable neighborhood search (VNS) algorithm for dense scenarios; 5) studying anti-collision path planning and optimizing takeoff/landing strategies. Case simulations verify the algorithms’ effectiveness in different scenarios, providing a practical UAV allocation scheme to enhance inspection efficiency.