DRIVER ASSISTED CONTROL STRATEGIES: THEORY and EXPERIMENT

This work investigates the use of feedback controller augmentation of driver inputs to achieve a desired vehicle performance. The vehicle performance is specified as a Reference Model. The driver maintains nominal control of the vehicle by direct actuation of the front steering inputs. The controller then determines the appropriate rear steer inputs necessary for tracking the reference model. As a consequence the driver is able to specify, within limits, the type of handling behavior required of the vehicle. A strategy based on yaw rate control is presented. An appropriate vehicle model is developed and a polynomial pole placement technique is used to control the vehicle. To account for vehicle model changes due to variations in forward velocity, a continuous time Recursive Least Squares approach is examined for on-line identification and adaptive control. The strategy and control designs are implemented experimentally on the Illinois Roadway Simulator (IRS), a scale vehicle testbed for vehicle dynamics and controls. Results and limitations are discussed.