A Voronoi-Based Isotemporal Task Allocation System for Autonomous Tractor Fleet

This study proposes an isotemporal task allocation system for autonomous tractor vehicles to improve agricultural task efficiency. The proposed system integrates Voronoi-based workspace partitioning and isotemporal task allocation. The method performs isotemporal tasks by considering the performance and state (including distance, velocity, and fuel and battery capacity) of each tractor by adopting an optimal workspace partitioning method. Based on these factors, the system optimizes the sub-workspace allocation to minimize the task time deviation and ensure balanced workload distribution among heterogeneous robots. The proposed system is evaluated through numerical verification and field evaluation in an agricultural environment. The results of the field evaluation show that the task efficiency is significantly improved, such as a 25.88% reduction in total task time and a 92.89% reduction in task time deviation under optimized conditions. In addition, the similar results of the two evaluations indicate high consistency and performance maintenance of the proposed system performance. Through the proposed system, it can be easily applied to various tractor-based vehicle cooperative task models, and efficient task performance can be expected by reducing idle time and allowing tractors to perform the next task.