为在加快算法收敛速度的同时又能避免停滞现象，提出一种基于混合行为的蚁群算法．首先就蚂蚁行为对算法性能的影响进行了分析，在此基础上提出了该算法的模型；然后定义了蚂蚁行为，并为该算法设计了4种具体的蚂蚁行为，根据模型实现了该算法．实验结果表明，该算法在性能上远优于蚂蚁系统．-To accelerate the convergence speed while avoiding stagnation behavior is proposed based on hybrid ant colony algorithm. First to ant behavior on algorithm performance are analyzed on the basis of the algorithm proposed model　and then define the ant behavior, and 4 for the algorithm design of the specific behavior of ants, according to the model realizes the algorithm. Experimental results show that the algorithm performance is much better than ants in the system.