Dynamic modeling and bio-inspired LQR approach for off-road robotic vehicle path tracking

Ground robotic vehicles have received special research attention in recent years. Application scenarios range from advanced driver assistance systems (ADAS) in urban vehicles up to autonomous navigation strategies on field environments. This work focuses on common subjects for these applications: Dynamic modeling and path tracking control of off-road robotic vehicles. Firstly, a nonlinear 3D dynamic model of a four wheels vehicle was conceived and used to simulate the vehicle motion behavior. An LQR path tracking controller was obtained from a linearized 2D vehicle model. Finally, a bio-inspired reference shaping approach is introduced to enhance the lateral tracking performance. The overall methodology and simulation results are presented.