Abstract: To ensure the active safety of self-driving cars steering to avoid collisions in the front vehicle cut-in scenario, an active obstacle avoidance path planning method was proposed based on the improved artificial potential field method. Based on the traditional artificial potential field method, the repulsive potential field function was changed to ensure that the vehicle drived in a straight line after the end of collision avoidance by increasing the road boundary repulsive potential field. The speed adjustment factor was introduced to establish dynamic obstacle potential field to solve the artificial potential field for collision avoidance path planning under dynamic obstacles. The proposed path planning method was simulated using the joint simulation of MATLAB and Carsim in the dynamic scenario for the front vehicle cut-in. The results show that the smooth and safe localized collision avoidance path can be obtained by the improved algorithm, and the vehicle dynamics requirements are met when the self-driving car travels along this path.