Path planning for mobile robots in complex environments based on improved ant colony algorithm.

Abstract

Aiming at the problems of the basic ant colony algorithm in path planning, such as long convergence time, poor global path quality and not being suitable for dynamic environments and unknown environments, this paper proposes a path planning method for mobile robots in complex environments based on an improved ant colony (CBIACO) algorithm. First, a new probability transfer function is designed for an ant colony algorithm, the weights of each component in the function are adaptively adjusted to optimize the convergence speed of the algorithm, and the global path is re-optimized by using the detection and optimization mechanism of diagonal obstacles. Second, a new unknown environment path exploration strategy (UPES) is designed to solve the problem of poor path exploration ability of the ant colony algorithm in unknown environment. Finally, a collision classification model is proposed for a dynamic environment, and the corresponding dynamic obstacle avoidance strategy is given. The experimental results show that CBIACO algorithm can not only rapidly generate high-quality global paths in known environments but also enable mobile robots to reach the specified target points safely and quickly in a variety of unknown environments. The new dynamic obstacle avoidance strategy enables the mobile robot to avoid dynamic obstacles in different directions at a lower cost.