برنامه ریزی مسیر موبایل رباتی بر اساس پارامتر بهینه سازی الگوریتم مورچگان

The basic ant colony algorithm for mobile robot path planning exists many problems, such as lack of stability, algorithm premature convergence, more difficult to find optimal solution for complex problems and so on. This paper proposes improvement measures. Apply genetic algorithm to optimization and configuration parameters of the basic ant colony algorithm. Simulation results show that the improved optimal path length significantly less than the basic ant colony algorithm and volatility is smaller, stability significantly improves. The stability of improved ant colony algorithm is superior to the basic ant colony algorithm, verify the effectiveness of the improvement measures.

Mobile robot path planning is an important research field of robotics. It refers to that, the mobile robot in a work environment with obstacles, based on one or some optimization criterion, search for a motion path from the initial state to the target , state and the path is the optimal or near optimal, safe, obstacle avoidance[1]. Robot movement environment which is studied in this paper is known two-dimensional flat space, and don’t take obstacles and the robot height information into consider. In the process of environment description, all the obstacles in the environment have done pretreatment which extend out each obstaclesof the maximum radius of a robot. This allows considering the robot as a particle, thus ensuring the safety and greatly reducing the complexity of path planning algorithms [2-3]. This study aims to: • In the known static environment, find a collision-free path connect the start and the end. • Obstacle avoidance, meanwhile make the length of the path as short as possible • Algorithm's time complexity is as low as possible, good stability.

Improved ant colony algorithm’s search optimization capability has greatly improved than the basic ant colony algorithm. As can be seen in fig 3, improved ant colony algorithm’s three parameters: the optimal path length, the worst path length and average path length are all smaller than the basic ant colony algorithm, this means, we have found a shorter path. It is proved by simulation experiments, This paper, through the optimization for basic ant colony algorithm, make path planning ant colony algorithm for mobile robot has stronger searching optimization ability, greater stability and better comprehensive performance. The final path is more optimized. The algorithm for solving the problem of mobile robot path planning has some theoretical innovation and practical value.