Multi heuristic_ACO

This paper proposes a mutual learning and adaptive ant colony optimization algorithm (MuL-ACO) for path planning of mobile robots in complex uneven environments. MuL-ACO is composed of two independent and continuous algorithms, which complete path generation and optimization in turn. First, based on the de-temperature function of the simulated annealing algorithm (SA), the pheromone volatilization factor is adaptively adjusted, thus accelerating the convergence of the ant colony algorithm. Secondly, a mutual learning trajectory optimization algorithm is proposed to generate the node characteristics of the initial path. Each node learns from other nodes to generate the optimal path node, so as to optimize the smoothness and minimize the path length. In addition, in order to adapt to the uneven outdoor environment, a 2D map with high characteristics is modeled, and the height information is introduced as an important consideration factor into MuL-ACO. The simulation results show that this method can quickly produce the high comprehensive quality of path without collisions.