Deep Learning Enabled Joint Sensor Localization and Automatic ID Assignment Methodology for Tire Pressure Monitoring Systems

Due to the rapidly increasing demand for road safety and improved driving experiences, as well as the more recent introduction of self-driving cars into the automotive industry, the global market for tire pressure monitoring systems (TPMS) is set to hit 24 Billion USD by 2027. However, one of the major challenges with tire pressure monitoring systems is the limited capability of the system to provide the relative positioning of individual tires on demand, particularly, in the case where frequent shuffling of tires happen, either within the same vehicle, or between multiple vehicles. This challenge becomes exacerbated for vehicle fleet managers, like bus rapid transit operators, who require to shuffle tires, more often than not. This project, therefore, seeks to develop new deep learning methodologies for enabling joint localization and automatic ID assignment of TPMS sensors, within tires, for accurate prediction of the relative position of tires, on demand.

Faculty Supervisor:

Thomas Johnson

Student:

Partner:

SST Wireless

Discipline:

Engineering

Sector:

Manufacturing

University:

The University of British Columbia - Okanagan

Program: