Small-Scale Autonomous Racing Vehicles

This project was funded by the lab CAVREL under the supervision of Dr. Yaser Fallah.

Our Goal

Our goal was to build two small-scale autonomous race cars capable of competitively navigating an indoor racetrack in the presence of other race cars with the same goal

Project Overview

Mechanical

The vehicle features a lightweight and durable chassis optimized for high-speed racing. The chassis is made from materials that provide an ideal balance between strength and weight, enhancing speed and handling. Aerodynamics is a key focus, reducing drag to achieve higher top speeds and improving overall efficiency. The drivetrain is designed for rapid acceleration and high speeds, while the finely tuned suspension system ensures optimal handling through various cornering scenarios, allowing the vehicle to adapt to a variety of racing conditions.

Hardware

The vehicle is equipped with advanced sensor systems to support autonomous operation. Key sensors include LiDAR for proximity sensing and collision avoidance, an RGBD camera to capture color and depth information for enhanced environmental awareness, and an Inertial Measurement Unit (IMU) for precise orientation and motion data. These sensors work together to feed data into the SLAM algorithm for mapping and navigation. The vehicle also have an external remote control system for manual operation and a Status Indicator Board to monitor critical metrics like power distribution and speed during testing and races.

Software

The vehicle’s main software manages autonomous navigation, path planning, and real-time decision-making. SLAM algorithms use data from the sensor suite to simultaneously create a map of the environment and localize the vehicle within it, enabling dynamic obstacle avoidance and real-time path adjustments. Path-following algorithms calculates the optimal trajectory based on the vehicle’s dynamic capabilities, adjusting speed and steering in real-time. The software systems run on a high-performance computer with a GPU, ensuring low-latency processing for fast and accurate decision-making during races.

Motivation

Cutting-edge Challenge

Autonomous racing combines various complex domains such as computer vision, real-time decision-making, and control systems. The fast-paced environment of a race demands the highest level of precision and optimization, which pushes us to challenge our technical skills and innovate.

Real-World Applications

Autonomous vehicles are shaping the future of transportation and robotics. By working on this project, we are contributing to the development of technologies that have the potential to revolutionize industries.

Interdisciplinary Collaboration and Professional Growth

This project allowed us to collaborate across disciplines, combining expertise from electrical, mechanical, computer engineering, and computer science. Practicing this kind of complex collaboration now puts us a step ahead in preparing to work in the industry.

Community Contribution

This project is available for anyone to use and learn from. We want to contribute to bridging the gap of knowledge when it comes to autonomous vehicles. We created simplified, detailed, step-by-step instructions on how to replicate our vehicles.