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.
