Abstract: The existing ant colony algorithm is applied to the path planning of automated?guided vehicle, which is slow in the convergence and easily fails into the local optimum. To solve these problems, this paper proposes a Hyper Heuristic Ant Colony Optimization algorithm based on pheromone negative feedback（ACONhh） for path planning of mobile robots. The algorithm makes full use of historical search information and continues to gain error experience, thus further leads ant colony to explore optimal path. Hierarchical selection of feasible nodes is adopted to accelerate the initial convergence rate of the algorithm. Meanwhile, the volatility factor changes constantly with an analogous?parabola, and pheromone update mechanism is adjusted to improve the randomness of global search. The convergence of ACONhh algorithm is strictly proved. Simulation and experimental results show that the convergence speed and global search performance of the proposed algorithm are outperform those of popular ACO, ACOhh and ACOihh algorithms.

Key words: automated guided vehicle, path planning, ant colony optimization, pheromone negative feedback mechanism, hierarchical selection

摘要： 针对蚁群算法应用于自动导引小车路径规划收敛速度慢、极易陷入局部最优的缺点，提出一种基于信息素负反馈的超启发式蚁群优化（ACONhh）算法。该算法充分利用历史搜索信息和持续获得错误经验，较快引导蚁群探索最优路径；分层化选择可行节点，加快算法初期收敛速度；设置挥发因子呈类抛物线变化以及调整信息素更新机制，改善路径全局的随机搜索特性。通过严格的数学方式证明了ACONhh算法具有收敛性。仿真和实验结果表明，该算法的收敛速度以及全局搜索性能显著优于目前流行的ACO、ACOhh和ACOihh算法。

关键词: 自动导引小车, 路径规划, 蚁群优化算法, 信息素负反馈, 分层化选择