基于蚁群算法求路径规划问题的新方法及仿真

该文提出了一种基于蚁群算法求解路径规划问题的新方法及其仿真，蚁群算法就是对自然界中蚂蚁的寻食过程进行模拟而得出的一种模拟进化算法。与传统的算法相比，该算法的主要特点是正反馈和并行性，正反馈使得该算法能很快发现较好解，并行性使得该算法易于实现并行计算。虽然蚁群算法在时间复杂度上可能不如传统的算法，但是理论研究表明该方法是一种基于种群的鲁棒性较强的模拟进化算法。最后，利用Java语言对蚁群算法和改进的Dijkstra算法进行了仿真，并进行了比较。

A New Method and Simulation for Path Planning Problem Based on Ant Colony Algorithm

This paper puts forward a new method and simulation for path planning problem (PPP) based on ant colony algorithm (ACA). ACA is a simulated evolutionary algorithm (SEA) of simulating to seek food of ants in nature. Comparing with conventional algorithm, it has several primary characteristics such as positive feedback and parallelism. Positive feedback makes it faster to find better solutions. Parallelism makes it easier to realize parallel computing. Although ACA is not as good as conventional algorithm on the complexity of time, studies on the theory demonstrate that it is a robust SEA based on population. At last, it is simulated by Java and compared with Improved Dijkstra Algorithm.