This study proposes an intelligent local path planning approach for mobile robots by integrating a neural network into the Dynamic Window Approach. The neural network is trained to predict optimal velocity commands based on real-time sensor data and robot state information, with the goal of effectively replacing the traditional handcrafted cost function. The proposed neural network based on Dynamic Window Approach (NNDWA), built upon the traditional Dynamic Window Approach, is implemented in the Robot Operating System and evaluated through both simulation and real-world experiments. The results demonstrate that the proposed NNDWA significantly reduces computation time during local obstacle avoidance, while maintaining stable motion, thereby improving the responsiveness in the dynamic environment of the mobile robot compared with the traditional DWA algorithm.