Skip to main content
eScholarship
Open Access Publications from the University of California

Scholarly Works (4 results)

  • Thesis
  • Peer Reviewed
UC Davis Electronic Theses and Dissertations (2022)

Rear-wheel steering (RWS) as we know it today is an idea that has been around for decades. Almost all literature discusses the technology though the conventional use, where the rear steers opposite the front at low speeds and in phase at high speeds. The work in this dissertation explores additional uses and benefits for using RWS.

The first idea explores decoupling yaw motion and lateral motion in order to allow the vehicle to behave with much more freedom than its front steering only counterpart. The handling characteristics are then controlled by a novel model reference concept, a virtual wheel base.

The second use explores how rear-wheel steering can enhance safety features in the vehicle. Rear-wheel steering is coupled with antilock braking and electronic stability control. These concepts show promising results and improvements to the vehicle.

    Cover page: Rear Wheel Steering: Applications Beyond the Conventional
    • Article
    • Peer Reviewed
    UC Davis Previously Published Works (2021)
    In an application of rear-wheel steering for an automobile, an electromechanical steering actuator can be used for positioning the rear wheel’s angle. The wheel may experience large forces not only from the motor, but also from the road. However, depending on the configuration of the system, the motor typically cannot be backdriven. If this is not considered in the modeling process, a model of the system may allow backdriving and give erroneous results. This paper explores different methods of implementing the non-backdriving behavior with the aid bond graph modeling. The goal is to compare different ideas of implementation and find an appropriate method for the system presented.
      Cover page: Modeling Non-Backdriving Behavior in an Electromechanical Steering Actuator Using Bond Graphs
      • Article
      • Peer Reviewed
      UC Davis Previously Published Works (2018)
      Additional sensors on a passenger vehicle can give more options for control but add cost to the manufacturer. Using existing on-board sensors to create an observer in place of installing a new sensor is a valuable option when possible. A model of a ground vehicle is created with the use of bond graphs for the design of a rack force output observer, which can be designed with two common on-board sensors. To design an observer that does not require information about the road condition, one of the sensors is fed forward, eliminating a system state. A strategy for reducing the bond graph is proposed for a controller-output-observer design and is tested. The accuracy of the observer is validated against real data, and shows promising results.
        Cover page: Bond graph reduction for an automotive rack force observer application
        • Article
        • Peer Reviewed
        UC Davis Previously Published Works (2021)
        Powertrains for vehicles are transitioning from the internal combustion engine to electric motors. Electric motors have the ability to both output traction and generative braking torque. Furthermore, electric motor powertrains have a smaller size compared to the internal combustion engine powertrain. This opens up thepossibility of different configurations of powertrain layouts. This paper explores an electric vehicle model with motors on both the front and the rear axles. The planar vehicle model with varying motor torque distribution is simulated to validate its handling dynamics.
          Cover page: Bond Graph Modeling of a Planar Vehicle with Electric Traction Motors for Assisted Torque-Vectoring
          Top