Design an Automatic Steering Controller with a Camera-Based System for Lateral Assistance Control in Car Driving

This paper describes an automatic steering controller based on a target and control driver steering model. With the road information provided by the image sensor, the structure revealed that drivers apply a PID-type controller whose look-ahead distances and feedback gains are dependent on the vehicle speed. With these simple gains, the vehicle can perform very well in vehicle lateral dynamics control. Instead of planning and following a desired path according to the traditional trajectory planning concept, drivers use the next lane center as the target points to generate vehicle angle error for steering control. This novel target and control driver model captures driver’s key steering mechanisms. The automatic steering controller was first simulated under the Model-in-the-Loop (MIL) test using the CarSim simulation. The simulation results indicate that lateral offsets are smaller than those from typical driver models. In order words, the results show controller achieved good performances and using the target points as input reference has many advantages over traditional route tracking control methods.