Simulation code for testing RoboCup Strategies (Especially more complicated passing plays). Future PPO robot model to come! Released as the 2nd day in the Open Source Week.
THIS CODE IS VERY EXPERIMENTAL AND BUGS ARE EXTREMELY COMMON. This is more for fun and learning (especially in physics) rather than for actual usecase.
The robots always get stuck onto eachother. There is no work being done to fix this bug. But if you know a solution, PLEASE feel free to open a pull request.
Implementing PPO is a very difficult task that will require a lot of time. I will eventually try to integrate PPO into this project to make strategies that rival human built ones.
While the code is still in very early developmental stages, I hope this project encourages you to dive into Open Source!