AWS DeepRacer Student is a program that introduces students to the exciting field of reinforcement learning and autonomous vehicle technology. Designed by Amazon Web Services (AWS), this educational initiative empowers students to develop their skills in machine learning and artificial intelligence. Participants have the opportunity to use the DeepRacer, a 1/18th scale autonomous race car, to apply their knowledge and compete in virtual racing leagues.

The program encourages hands-on learning, allowing students to experiment with reinforcement learning algorithms and gain practical experience in training autonomous models. With the support of AWS resources and community engagement, AWS DeepRacer Student aims to inspire the next generation of AI enthusiasts and equip them with the skills needed for the future of technology.

The goal of this project is to compete in said racing simulation and get to the top of national and/or international leaderboards. That is possible by making a good reward function for the reinforcement learning agent to find the fastest way to race in a lap.

After iterating through 8 different reward functions using the PPO (Proximal Policy Optimization) model, I was able to achieve the desired results. I also got an extra bonus which is the eligibility of qualifying for the AWS AI & ML Scholarship Program!