Ant colony algorithm is often used to solve the problem of UAV path planning, but the traditional ant colony algorithm has many defects such as slow iteration speed and easy to fall into local optimization. A series of improvement measures are proposed to address these issues. For the blind search problem of ant colony algorithm in the early stage of route planning, pheromones are distributed unevenly and guidingly in the task space, which makes ants explore along the line from the starting point to the target point, and the exploration of ant colony is more directional. At the same time, the influence of rotation angle on track smoothness is considered in the heuristic function to improve the quality of track planning. In addition, the adaptive volatilization coefficient is used to dynamically adjust the pheromone volatilization rate, so as to avoid premature convergence to the local optimum, and to ensure the accelerated convergence in the later period, thus to prevent the algorithm from falling into endless calculation, and a redundant node elimination strategy is used to further optimize the track.