Research on Dynamic Obstacle Avoidance and Path Planning Methods in Emergency Scenarios

With the rapid development of autonomous driving, much attention has been paid to the development of collision avoidance systems in emergency scenarios that have a critical impact on driving safety. In this paper, a multi-constraint collision avoidance path planning method that considers the drivable area of a vehicle is proposed to meet the safety and real-time requirements of collision avoidance for autonomous vehicles. The algorithm defines the collision hazard moment, calculates the relationship between the collision hazard moment and the vehicle position constraint, makes corrections to the drivable area in the vehicle collision avoidance process, and finally obtains the vehicle emergency collision avoidance path by solving an optimization problem that satisfies the vehicle dynamics constraints and the position constraints. In this paper, the Matlab/Carsim/Prescan simulation sets up a high-speed driving condition in front of which there is an obstacle in the same direction of motion to prove that the multi-constraint optimized path planning algorithm considering the vehicle's travelable area can obtain a smooth and collision-free path, and the steering wheel changes smoothly in the tracking control, and the tracking control accuracy is below 0.07m. The experimental results show that the planning algorithm achieves the accurate lane change process The experimental results show that the planning algorithm achieves precise constraints on the lane change process, meets the requirements of the collision avoidance process on the computational efficiency, and takes into account the comfort of the driver.