Research on Intelligent Vehicle Trajectory Tracking and Lateral Stability Control Algorithm Based on Improved Long Short-Term Memory

With the aim to tackle the challenges of poor trajectory tracking accuracy and the risks of skidding and instability of intelligent vehicles under low-adoption extreme high conditions, a hierarchical control architecture for trajectory tracking algorithms based on distributed electric drive intelligent vehicles is proposed. This algorithm employs a 3-DOF single-track vehicle dynamics model and an improved LSTM to still concentrate the constrained problem in a rolling, obtaining the optimal front wheel steering angle. Simultaneously, the proportional-integral-derivative (PID) longitudinal driver model outputs the desired torque for speed tracking. By establishing a co-simulation platform using Carsim and Matlab/Simulink, we verified the effectiveness of the designed trajectory tracking control algorithm through simulations. The results demonstrate that the proposed trajectory tracking control strategy significantly enhances the trajectory tracking accuracy and lateral stability of distributed electric drive intelligent vehicles on high-speed high-adoption and high-speed low-adoption roads.