

Reinforcement Learning: Sim-to-Real Part II
Hello YoungAI friends! This is part 2 of our sim-to-real workshop. We're going from simulation to a real car.
In part 1 we implemented PPO and trained agents in Gymnasium. Now we train a driving agent in the DonkeyCar simulator, deploy it on a PiRacer Pro, and see how it handles a real track.
Program:
Mingling & snacks
Intro to sim-to-real transfer, what makes it hard?
Train a driving agent in the DonkeyCar simulator
Deploy on the PiRacer Pro, groups take turns at the track
Mini race & wrap-up
You need:
Laptop with Python
DonkeyCar simulator installed (instructions shared beforehand)
We have one car, so groups take turns deploying and testing on the track. While waiting, you keep improving your model in the simulator.
Attended part 1? Great, you're all set. Missed it? You can still join, we'll provide pretrained models so everyone can participate.
Big thanks to Visito for sponsoring the workshop and providing the PiRacer Pro.
This event is mainly for our members, but if you're new, you're welcome to join one meetup for free. Interested in becoming a member? Check out our website: https://www.youngai.no.