A Coordinated Scheduling Approach for Task Assignment and Multi-AGV Path Planning in Aircraft Assembly Lines

Aircraft pulsating assembly lines require highly efficient and timely material delivery. The coordination of Automated Guided Vehicles (AGVs) for task assignment and path planning is therefore critical to reducing delivery delays and improving logistics efficiency. This paper proposes a coordinated scheduling framework integrating task allocation, single-AGV path planning, and multi-AGV conflict resolution. At the task allocation layer, an Improved Adaptive Genetic Algorithm (IAGA) with segmented population initialization and entropy-based adaptive crossover and mutation is developed to minimize earliness penalties and total AGV running time under time-window constraints. At the path planning layer, an enhanced TOA-star algorithm introduces turning cost and occupation cost into the heuristic function to reduce path inflection points and path overlap. At the multi-AGV coordination layer, an improved Conflict-Based Search (CBS) method adopts TOA-star as the low-level planner, and a delay-aware extension, TCBS, further embeds deadline information into high-level search. Extensive simulations and real-vehicle experiments verify the effectiveness of the proposed approach. In a real aircraft assembly line case, IAGA reduces total AGV travel time from 6203 to 4892 and eliminates delayed deliveries. TOA-star reduces inflection points by up to 71% and path overlap by up to 84%. On the aircraft assembly line map, the improved CBS reduces inflection points from 17 to 9 and search time from 0.412 s to 0.132 s. Compared with sequential scheduling, TCBS reduces delay occurrences from 10 to 2 and total delay duration from 42 s to 8 s. These results demonstrate the engineering value of the proposed method in aircraft material handling systems.