Stability Control via In-Wheel Motors of a Solar-Electric Vehicle

This thesis investigates the principle and engineering application of dynamic yaw moment control through simulation and real-time testing of the Australian Technology of Networks (ATN) solar car. The ATN solar car competed in the Bridgestone World Solar Challenge (BWSC), 2019; an Australian international competition, where teams drive 3,000 km in custom-designed solar-electric vehicles. Drivers are exposed to long driving stints in vehicles with generally poorer handling and steering performance, owing to the need for lightweight, high performing designs. The design features, particularly being rear-wheel drive and very lightweight, impact the vehicles controllability and dynamic behaviour. Investigating vehicles susceptible to extreme handling, handling safety can be improved, within solar racing sports and for the development of future lightweight road vehicles, which is why this research is important. To undertake this investigation a simulation-based approach was developed. Co-simulation of the vehicle model, a nonlinear 15-DOF model using Siemens Amesim, which incorporates the load transfer effects and nonlinear tyre characteristics, and the control, where control algorithms were developed in MATLAB/Simulink. This thesis presents four control methods that can be applied to the rear in-wheel motors; Dynamic Curvature Control (DCC), Proportional-Integral Control (PI), Sliding Mode Control (SMC) and Model Predictive Control (MPC). Using this simulation-based approach the dynamics of the vehicle was studied. Large variations load-to-curb weight ratios are linked to significant changes in parameters critical to control design for vehicle stability control system. Unique and highly customised vehicles, such as the lightweight solar car in this research, are more susceptible to the impact of such variations when developing control methods. As such the influence of variation in loading condition and the effect of ignoring changes in inertial parameters is studied. The study demonstrated that by ignoring the change in the inertial parameters in simulation environments can produce an incorrect translation of the control performance. Finally, to verify the applicability and performance of the simulations, open-loop real-time testing on the real vehicle. This was done by implementing the control to the vehicles Control Area Network (CAN), via dSPACE MicroAutoBox II. The evaluation was performed by comparing a slow speed baseline vehicle to tests with higher velocity, addition of passenger, low tyre pressure and cases of uneven tyre pressures. It was found that despite significant sensor and estimation errors due to compromises caused by COVID-19, the SMC and MPC both have vigorous performance capabilities and are safe for future closed-loop testing.

Author dc:creator

• Lidfors Lindqvist, Anna

Statement dc:rights

• The author owns the copyright in this thesis including all reproduction and reuse rights for the work. The work may not be altered without the permission of the copyright owner. Attribution is essential when quoting or paraphrasing from this thesis.
• au.edu.uts.lib/ppc
• info:eu-repo/semantics/embargoedAccess
• © 2021 Anna Lidfors Lindqvist

Language dc:language.iso

en_US

Handle dc:identifier.uri

http://hdl.handle.net/10453/158175

OAI identifier oai:identifier

oai:opus.lib.uts.edu.au:10453/158175