AGV Trajectory Planning Based on Multi-Objective Quantum Particle Swarm Optimization Algorithm in the Sight of Essential Supply Distribution during an Epidemic

The application of autonomous ground vehicles (AGVs) in essential supply distribution during an epidemic have received significant attention due to its high-efficient performance and capability to be operated in line with social separation and quarantine policies. Among the main challenging technical elements of such AGVs, parking and overtaking motions of the systems are of importance to cope with various driving routes and conditions. This paper focuses on AGV parking and overtaking trajectory planning, where a multi-objective quantum particle swarm optimization (MOQPSO) algorithm is the mainly adopted method to achieve a global optimal solution. The overall design adopts the idea of staged optimization. Based on the optimization objectives of a certain task, Quantum Particle Swarm Optimization (QPSO) or MOQPSO are used to generate a near-optimal parking and overtaking trajectory prior to a gradient-based optimization technique is applied to obtain a final optimal parking and overtaking trajectory. The effectiveness and feasibility of the proposed design are verified through simulations.