Artificial Intelligence in Mass Transit

Transportation systems are evolving towards intelligent transportation systems and ISR Transit is a leading provider of these systems providing solutions in fleet management. In these systems, one of the enabling technologies is wireless sensor networks in which sensors are used to obtain information about the fleets. For example, sensors are deployed on motor, brake modules, doors, emergency buttons and passenger stop request. The information captured by these sensors is transmitted to a central controller to optimize productivity by tracking, monitoring, and managing mass transit elements and static operational data such as vehicle numbers, drivers, routes, schedules, timetables, and so on.

One of the main challenges in these systems is that since sensors have limited power resources, they cannot perform monitoring tasks over a long period of time. To address this issue, the energy consumption at sensors should be reduced which can be achieved by minimizing the number of data transmissions. More specifically, As sensory observations are highly correlated in time and space domain, some of the collected readings might be redundant. The objective of this project is to propose data reduction schemes that while they minimize the energy consumption of sensors, the quality of data is preserved.

Faculty Supervisor:

Tho Le Ngoc

Student:

Atoosa Dalili Shoaei

Partner:

BusPas Inc

Discipline:

Engineering - computer / electrical

Sector:

Information and cultural industries

University:

McGill University