The work aims to analyze and model multi-batch collaborative tasks, and to study the algorithm of task planning. In this work, vehicle-mounted equipment multi-batch collaborative task was analyzed and modeled. In the modeling process, the constraints of time collaboration, task area collaboration and replenishment area collaboration were integrated, and the minimum exposure time was taken as the objective function. The simulation results showed that the improved strategy could avoid the situation that some neighborhoods could not find the optimal solution for a long time in the later stage of the standard variable neighborhood search algorithm, and improve the efficiency of the algorithm without reducing the quality of the optimal solution. It is concluded that the variable neighborhood search algorithm can solve the multi-batch task planning problem. The improved algorithm reduces the search times of the neighborhood with weak optimization ability in the later stage, and effectively improves the efficiency of the algorithm.