Path Following of Autonomous Vehicles with an Optimized Brain Emotional Learning-Based Intelligent Controller

This article proposes a control framework which combines the longitudinal and lateral motion control of the path-following task for Autonomous Ground Vehicles (AGVs). In terms of lateral motion control, a modified kinematics model is introduced to improve the performance of path following, and Brain Emotional Learning-Based Intelligent Controller (BELBIC) is applied to control the heading direction. In terms of longitudinal motion control, a safe speed is derived from the road condition, and a Proportional-Integral (PI) controller is implemented to force the AGV to drive at the desired speed. In addition, for a better performance of path-following and driving stability, Particle Swarm Optimization (PSO) algorithm is used to tune the parameters of BELBIC. In this article, a Carsim and Simulink joint simulation is provided to verify the effectiveness of the modified model and the control framework. The simulation result indicates that, in the scenario of the modified kinematics model, the AGV could follow the desired path with a smalle lateral offset than the conventional model, except that the modified model is less sensitive to preview time. Compared with the Proportional- Integral-Derivative (PID) controller, the BELBIC allows the AGV to follow the desired path with a smaller lateral offset. Specifically, the maximum lateral offset with the BELBIC controller is 0.18 m, while it is up to 1.37 m with the PID controller.