In this final year project, study and develop an Artificial Intelligence (AI) model which can learn how to walk or run by itself. We have two different robots for our study, called IT Dog and Minikame.
we customize a suitable AI model for our robots, and reinforcement learning is our major research direction. There is a lot of learning algorithm which can be classified into reinforcement learning. We focus on PPO and DQN in this project. Our target is to find the best algorithm and successfully applying our customized AI to our robots. In the end, it is positive to transform our project results and experiences into STEM material. Our study finds out PPO is the best option when we develop and apply it to our robots and we have a lot of data that will be transferred as STEM material for reference.