Path Planning Efficiency Maximization for Ball-Picking Robot Using Machine Learning Algorithm

This paper aims to find a ball-picking optimal path and drives the robots to collect all the tennis in the shortest time, according to the optimal path. Thus, the work to be completed for us includes: establish agent model for robot working environment in tennis yard, analyze the advantage and shortcoming of ACO and provide an improved ACO. Our scheme improves the pheromone updating strategy. The global and local updates are integrated to strength the pheromone strength of optimal ant. Then we add the crossover and mutation operation of GA to speed the convergence of algorithm. By the simulation results analysis we find that the improved ACO has stronger optimizing ability and stability, which further improves the performance of ball-picking path.