Autonomous racing at the limits of control
Located at the University of Virginia, the Cavalier Autonomous Racing team is a mix of faculty and students driven by the mission of building the fastest fully autonomous racing car. We believe that autonomous racing is the next grand challenge for safe self-driving vehicles. Motorsport racing has always been the proving grounds for new automotive technologies; and autonomous ‘battle of algorithms’ racing will play the same role for self-driving software and hardware. In motorsport racing there is a saying that “If everything seems under control, then you are not going fast enough” - we are building an autonomous racing AI with this as the objective function.
Our team has a rich history in autonomous racing as majority of the team members have been involved in the F1/10 autonomous racing development for the past 5 years. One of our biggest strengths is in autonomous racing algorithms. We have strong expertise in robust perception, planning, and control at high speeds, overtaking algorithms, mapping, localization, head-to-head racing, and end-to-end autonomous driving. Another strength of our group is experience with high-fidelity racing simulation. We have developed and released several open-source autonomous racing simulators: e.g. ROS F1/10 Simulator (f1tenth.dev) and the DeepRacing AI (deepracing.ai) framework. Finally, the team is also very strong in rapid prototyping and racecar design on all scales. The team is also located in proximity of the Virginia International Raceway.
Our team is involved in the development of the F1/10 autonomous racing platform and in organizing the Internaitonal F1/10 Autonomous Racing Compeitions
We have access to more than 16 1/10 scale autonomous racing platforms.
We have developed and launched the ROS F1/10 Autonomous Racecar Simulator - a Gazebo based virtual racing environment which includes a realistic model of the F1/10 autonomous racecar and associated race controllers. Otimize autonomous racing algorithms and design and visualize head-to-head racing with multiple autonomous racecars.
Racing and driving fast, and driving safely may seem as two very contradictory objectives, but the idea is not to drive fast all times, but enhance the autonomous vehicle with the ability to be able to brake aggressively and maneuver aggressively, when it encounters a safety critical situation. We want to turn the unexpected situations and edge cases, to hedge cases, which you can bet your lives on. In formula 1 racing and motorsports in general, there is a saying that if everything seems under control, then you are not going fast enough. In a racing setting, we counter unusual situations, wheel to wheel action, and edge cases more often than regular driving. We are developing algorithms for structured deep neural networks that can learn an agile vehicle controller from expert driver behavior by watching annotated videos of how racing drivers, drive at the limits of control. We aim to demonstrate that a vehicle equipped with this agile controller leads to increased overall safety
Coming Soon - Under Construcution