Trajectory Generation and Optimization Using the Mutual Learning and Adaptive Ant Colony Algorithm in Uneven Environments

Aiming at the trajectory generation and optimization of mobile robots in complex uneven environments, a hybrid scheme using mutual learning adaptive ant colony (MuL-ACO) is proposed this paper. In order to describe environment with various obstacles, 2D-H map introduced Then an algorithm based on simulated annealing (SA) generate initial trajectories robots, where de-temperature function algorithm, pheromone volatilization factor adaptively adjusted accelerate convergence algorithm. Moreover, length factor, height smooth are considered comprehensive heuristic ACO adapt environments. Finally, designed further shorten trajectories, which different node sequences learn from each other acquire shortest sequence optimize trajectory. verify effectiveness scheme, MuL-ACO compared several well-known novel algorithms terms running time, length, height, smoothness. The experimental results show that can collision-free high quality