An improved ant colony algorithm for mobile robot path planning strategy

Ant colony optimization algorithm often be used to path planning for mobile robots, but they have the following problems: 1. slow convergence speed, 2. hard to search real optima and easy to affected by deadlocks, 3. path curve is not smooth. To overcome these shortcomings, an improvement strategy was proposed: 1. An adaptive heuristic function based on the iteration times was applied, the impact of the heuristics function on the probability of state transition will decrease as the iteration time getting bigger, thereby the convergence speed of the heuristics function was accelerated in the early stage and a stable value can be quicker obtained in the later stage. 2. Adaptive pheromone volatilization factor and adaptive pheromone concentration were applied to solve the unreasonable pheromone allocation and pheromone updating concentration, which further improves the convergence speed and falling into local optima can be avoided; 3. The Monte Carlo algorithm is used to adjust the motion direction of the robot so that the robot's motion direction points directly to the next target point, which the tortuous motion can be avoided when approaching the target point, and the problem of tortuous path can be solved. We can see in the experiments that compared with three algorithms, our strategy can obtain optimal and smooth path with fewer ants, iterations times and lower time cost.