Hierarchical Planning for Dual Mechanical Arms Cooperative Fruit Picking

Aiming at the problems of high labor cost in large-scale orchards and low picking efficiency of single mechanical arms, this paper proposes a hierarchical planning method for dual mechanical arm cooperative picking systems. A task allocation model based on an improved Multi-Traveling Salesman Problem (MTSP) is constructed, and a genetic algorithm is used to optimize the picking sequence of the dual arms. For the end-effector trajectory of the mechanical arms, an improved APF-RRT* path planning algorithm is designed by integrating the target directivity of the Artificial Potential Field (APF) method and the global search capability of the RRT* algorithm. Simulation and experimental studies are conducted for the above research. The results show that the proposed task allocation strategy increases picking efficiency by more than 20%, and the improved algorithm increases the picking success rate by 10%, verifying the effectiveness of the planning method in trellised orchard environments.