Automatic steering control system model and simulation to analyze performance of off-road vehicles subject to sloped terrain and vehicle parameter changes
GPS guided navigation systems have been a topic of great interest and much research in the area of precision agriculture. A successful guidance system yields desirable response on varied terrain conditions and for a variety of vehicle weights, speeds, and center of gravity positions. The objective of this study is to provide a simulation that enables fast and robust controller design and quick evaluation of the controller and vehicle performance for hillside applications. A linear yaw-plane bicycle model that incorporates lateral tire forces is used for the vehicle dynamics. PID steering controllers command the steer angle to minimize the vehicle heading and position errors. Generalized root locus techniques for the individual controller gains help find a set of optimized gain values for an appropriate controller design. Steady state solutions for the equations of motion yield show that the steady state steer angle and steady state sideslip angle are functions of tire parameters and slope angle; they are controller independent. Furthermore, for a front steer only or rear steer only vehicle, a non-zero steady state sideslip angle will always exist. The simulation studies straight line tracking of a John Deere STS combine on sloped terrain. The simulation focuses on the response of a combine moving at a constant forward velocity of 10 mph and subjected to a five degree side slope. The desired path is a straight line track directly across the slope. The combine has an initial positive lateral error of 10 feet downhill from the track and no initial heading error. Results show the steering control system successfully guides the vehicle along the desired path and is robust over realistic vehicle parameter changes.