The ant colony algorithm is slow in convergence and easy to fall into local optimal value in complex environment. To solve these problems, an improved ant colony optimization algorithm is proposed. The position information of the starting point and the target point are utilized to select the global favorable region. In this way, the initial pheromone concentration is increased and the efficiency of early ant search is improved. The obstacle avoidance strategy is added to avoid ant blind search. A large number of cross paths are generated and the number of ant deadlocks is effectively reduced. Based on the pseudorandom transfer strategy of dynamic parameter control, the global performance of the algorithm is improved. The updating principle of high quality ant pheromone and adjusting the volatility coefficient adaptively are proposed. The second path planning is carried out to optimize the path and reduce the loss of energy consumption of mobile robots. Experimental results show that the algorithm has the feature of higher global searching ability, faster convergence speed and higher working efficiency of mobile robot. The proposed algorithm is verified to be effective and superior.