F1TENTH is a complete, ready-to-race autonomous race car that is 1/10th-scale and 1/100th the cost of a real self-driving car. The platform is powerful and versatile enough to be used for a variety of research that includes and is not limited to autonomous racing, reinforcement learning, robotics, communication systems, and much more. In this talk we will demonstrate how F1TENTH is an easy-to-use high-performance platform for machine learning engineering for perception, planning, control and coordination for future safe and connected autonomous systems.
F1TENTH has a growing community of over 60 universities, 7 international autonomous racing competitions and hands-on course offerings in over a dozen institutions. We'll detail the platform’s hardware, autonomous vehicle software stack, simulators, and systems infrastructure. We highlight three specific capabilities for streamlined algorithm development, testing and validation: a set of simulators, control and verification, and efficient machine-learning algorithm development.