基于人工免疫改进的搬运机器人蚁群路径规划

针对搬运机器人在障碍环境下的路径寻优问题,提出一种基于人工免疫改进的蚁群路径规划算法(AI-ACA)。蚁群算法(ACA)的规划依赖于信息素挥发系数、期望启发因子和信息启发因子等参数的选取,传统ACA通过经验来设定这三个参数,但路径寻优中的最优参数因障碍环境而异,为解决经验参数对不同环境路径寻优结果的影响,引入人工免疫算法(AIA),对ACA的相关参数进行迭代优化,以此改善路径寻优结果。仿真结果及在自制机器人平台上测试表明,AI-ACA对于不同障碍环境可以准确地进行路径规划,在同样环境下较所参考的定参数蚁群路径规划效果有明显提升,提高了整个系统的运输效率。

ANT PATH PLANNING FOR HANDLING ROBOT BASED ON ARTIFICAL IMMUNE ALGORITHM

Aiming at the optimization path problem of handling robot in the obstacle environment, an novel ant colony algorithm for path planning based on artificial immune algorithm (AI-ACA) is presented. The planning result of ant colony algorithm (ACA) depends on the selection of key parameters,such as pheromone volatilization coefficients, expected arouse factor and information arouse factor. The parameters of traditional ACA are set through experience. But optimal parameter values of path optimization vary remarkably in different obstacle environments. To solve the effect of experiential parameters on the path planning results in different environments, artificial immune algorithm (AIA) is introduced, and the algorithm is used to achieve the iterative optimization of associated parameters of ant colony algorithm and then to improve path optimizing result of ant colony algorithm. Simulation results and test results on self-made robot platform show AI-ACA can be used to effectively make path planning in different obstacle environments. AI-ACA is superior to invariable parameter ant colony path planning referred in the same environment and can raise transportation efficiency of the whole system.