Symbolic Model-Based Design of a Semi-Autonomous Vehicle Prototype Implementing Independent Wheel Torque Vectoring for Training an Advanced Driver Assistance System

There is a strong belief that autonomous vehicles will play a vital role in the future of the global transportation economy. There, however, exists many open challenges which need to be overcome to realize this future vision. One such challenge is the acceptance from the driver to relinquish full control of a vehicle and ultimately putting one’s safety in the hands of a computer. This foreseen inertia to change has prompted the development of a staged transition strategy by the US National Highway Traffic Safety Administration with the extremes on a spectrum represented by i) no automation, to ii) fully autonomous vehicles. Potential Motors™ seeks to support this gradual transition by developing semi-autonomous vehicle technology that allows the driver to still be in control of bulk driving decisions while their Advanced Driver Assist System, RallyAI™, responds to events in the environment at a speed faster than any human could ever respond, augmenting the driver’s reaction time and ultimately making them safer. This project seeks to support Potential Motors™ in their mission, through developing an advanced computer simulation of a vehicle to test and train RallyAI™ to enable that safer future.

Faculty Supervisor:

Kush Bubbar

Student:

Meaghan Charest-Finn

Partner:

Potential Motors

Discipline:

Engineering - mechanical

Sector:

Transportation and warehousing

University:

University of New Brunswick