<p>With the development of artificial intelligence technology, mobile robots have become a popular research direction. As one of the basic technologies for robot navigation, path planning occupies an important position in the field of robot research. The traditional ant colony algorithm (ACO) is one of the most widely used methods in solving path planning. However, ACO still has some disadvantages, such as low search efficiency and easily falling into the local optimum. To address these shortcomings of ACO, the learning strategy of the Q-learning algorithm is introduced to improve the convergence speed and global optimization of ACO. Therefore, an improved ant colony Q-learning algorithm (IQLACO) is proposed. In IQLACO, firstly, the turn times is introduced into heuristic information that improves the smoothness of planned paths. An angular guidance factor is introduced into the state transfer probability, which improves the search efficiency of ants. An adaptive parameter pseudo-random search strategy is introduced, which improves the global search ability of ACO. Secondly, in order to improve the convergence ability of ACO, a new pheromone update rule is proposed. Then, the Q-learning algorithm is used for pre-training pheromones to provide some direction for ants. Finally, the three indicators of optimal path length, convergence speed, and turn times are analyzed. To demonstrate the performance of IQLACO, IQLACO is compared with three algorithms. The experimental results show that the three indicators are considered comprehensively, and IQLACO has a more obvious advantage than the other three algorithms in finding the optimal paths. Both ‘Std’ and ‘Mean’ are smaller than the other three algorithms, which proves the stability of IQLACO. In real life, the autonomous navigation ability of robots is improved in complex environments by optimum path planning algorithms, which enable them to complete their tasks accurately.</p>