Method for cooperative task assignment and path planning of intelligent mine trucks with different capacities in underground mines

To address issues such as route constraints, time-window constraints, and frequent spatiotemporal conflicts in the cooperative transportation of intelligent mine trucks with different capacities in complex underground mine roadway networks, this paper investigates the task assignment and path planning problem for two types of mine trucks with different capacities under time-window constraints. An optimization model is established with the objectives of minimizing the maximum system completion time and the total travel distance, while comprehensively considering constraints such as vehicle capacity, time windows, and spatiotemporal conflicts. To address the tendency of heuristic algorithms to become trapped in local optima under strongly coupled constraints, this paper proposes a Hybrid Variable Neighborhood Search-Particle Swarm Optimization (HVNS-PSO) algorithm. The proposed algorithm is improved in three aspects: (1) high-quality initial solutions generated based on task assignment and conflict resolution mechanisms; (2) the introduction of six types of variable neighborhood search to enhance global search capabilities; and (3) the proposal of three repair mechanisms to ensure solution feasibility. Finally, experiments were conducted on two rule-based scenarios of different scales and a simulated underground mining scenario, followed by analyses of task scale and speed perturbations. The results demonstrate that the proposed method can obtain feasible solutions with zero conflicts and within the time window for all test instances. It exhibits superior performance in terms of solution feasibility, stability, and overall fitness.