Innovative. Intelligent. Imminent.
University of Waterloo Autonomous Vehicle Design Team
University of Waterloo Student Design Team Building a Self-Driving Car
WATonomous is a student design team at the University of Waterloo at the forefront of design, creation, and neural training for autonomous self-driving vehicles. We are proud to have finished 2nd place in the SAE AutoDrive Challenge, an international competition to build a Level 4 autonomous vehicle in four years.
Following the competition, we continued our path towards perfect autonomy, completing research papers on various autonomous vehicle technologies including action classification, control, and environment modeling. We have a track record of submitting to prestigious research conferences like ICRA, ICVES, and ITS.
Looking ahead, we are excited to pursue more opportunities, including several research projects, the Indy Autonomous Challenge as a part of the MPRW joint team, and the EcoCAR EV Challenge, in collaboration with the UWAFT student design team.
The EV Bolt model with our designed competing roof rack
We are currently in the process of developing multiple research projects. They are all ongoing and will be ready by early fall. Check out the topics here!
Environment modeling is the backbone of how autonomous agents understand the world.
We present a novel method for extending the priori relationship graph to include online observations, creating a unified environment model of both a priori and online data sources.
We present an optimal control problem
(OCP) formulation for the autonomous dynamic driving task
(DDT), and a model predictive control (MPC) solution method.
In autonomous driving applications, understanding the world around the ego vehicle is paramount for decision-making. In this project, we explore the problem of understanding the surrounding environment of an ego vehicle in a semantically meaningful way using the ROAD dataset.
Lane Detection and Classification
We explore Wayve's sim2real model to perform perception tasks such as performing detection and classification on the latent space.
The Journey So Far
Check out everything we have accomplished in five years