Path planning refers to the planning of path points for mobile robots to move from the starting point to the target point, and the planned path points need to satisfy motion constraints and ensure that the robot can safely reach the destination. However, the current path planning methods have problems such as the difficulty in balancing the computational efficiency of the algorithm in complex environments and the performance in practical applications, and the existing path planning strategies often lack sufficient adaptability and robustness in dynamic and uncertain environments. To address the above problems, this paper proposes a reinforcement learning path planning method based on the Transformer model. Deep reinforcement learning is an adaptive learning method that continuously optimizes paths through interaction with the environment, improving the robot’s environmental adaptability and task execution efficiency. The Transformer neural network model has strong feature extraction capabilities and can better handle complex environmental information. This paper combines the Q-learning algorithm with the Transformer model, achieving the fusion of deep reinforcement learning and effectively improving the performance of the path planning algorithm. The experimental results show that after using the deep reinforcement learning method, the number of learning iterations for the robot to complete the path planning task is reduced by about 8%, and the cost of completing the task is also reduced by about 12%. These significant improvements demonstrate the superior performance of deep reinforcement learning in path planning.