Maxar is excited to announce that two team members earned the top spots at the AWS DeepRacer event at the AWS National Security Partner Day on February 26, 2020. Dan Foley placed first with his DeepRacer finishing the course in 11.98 seconds (left in image below) and Evan Gay placed second (right in image below), completing the race in 13.07 seconds.
AWS DeepRacer is an autonomous 1/18th scale race car (seen below). In corresponding racing competitions, engineers compete against one another to see who can build the best AI deep reinforcement learning (RL) algorithms to pilot the DeepRacer car around racing tracks in the quickest time. RL is an advanced artificial intelligence (AI)/machine learning (ML) technique that enables complex behaviors based on incentives and rewards while optimizing and balancing between short-term and long-term goals, and all without masses of labelled training data like what is required with traditional AI neural network approaches.
AWS hosts races for the DeepRacer League at their conferences around the world. The National Security Partner Day DeepRacer event invited AI and ML experts from the U.S. defense, intelligence and law enforcement communities to go toe-to-toe with one another to see who would come out on top.
Only a few weeks old, the Maxar DeepRacer Team has already garnered interest from over 70 Maxar engineers spread across five office locations. By participating, these engineers are combining cutting-edge AI technology with the thrill of racing and competition. DeepRacer brings engineers together to collaborate and innovate virtually via the AWS DeepRacer console and virtual racing circuits. At Maxar, the engineers are also gathering in person to build their own tracks at the office to test the AI models and compete against each other.
If you’re interested in joining Maxar’s team of developers and engineers working on leading space technologies, please check out our Careers webpage.