On October 4, Intelliware held our Fall 2018 Hackathon. This time the topic was Robocode.
Robocode is a programming game, where the goal is to develop a robot battle tank to battle against other tanks in Java or .NET. The robot battles are running in real-time and on-screen.
Playing robocode involves writing a robot tank by extending a framework-provided base tank class. The robot can sense via a radar, and then act via movement and shooting.
This event was significantly different from previous versions of Intelliware’s hackathons. Instead of having the teams create a product with a given technology or domain constraint we chose to spend the day playing a game. Using sensors and actuators crosses over very directly to programming in the Internet of Things. There was a chance that teams would take a Machine Learning approach to controlling their tank. We were sure that coding the robot tanks and the subsequent demonstration to the rest of the company would be engaging and fun. And from a logistical perspective, this event was straightforward to run and the judging criteria were purely automated and objective.
Team Alice produced DreadNot as in “do not dread this bot”. True to its name, DreadNot finished in last place. However, to their credit Team Alice chose to work on their bot from scratch and used little to no code from other published robots. They chose Camilo Bernal as their Most Valuable Player.
Team Carol produced an early leader that gave the ultimate winner a run for its money. Their MVP was Matthew Short, who has only been with Intelliware for a month.
The winners were Team Bob. They produced a bot that had a good balance of offensive and defensive strategies. Caleb Powell, MVP from the winning team, said that the problem would lend itself to Reinforcement Learning. However, given the time constraints of the hackathon, it was bit optimistic to commit to a “moon shot” like that.
Want to do fun things at work? Intelliware’s hiring – Click here to see our current opportunities!