Lunar Exploration and Missile Target Optimization
It consists of coursework of an 8-week Project based training course on "Artificial Intelligence for Lunar Exploration".
And a final project on Missile Target Optimization.
In this project, a custom environment was created using OpenGymAI and a Reinforcement Learning agent to solve missile target optimization.
Developed a custom environment using OpenAI gym that looks similar to an Atari game environment. Objects like Missile, Targets, and Obstacles are set under a Canvas and assigned action to each of them using OpenAI gym, this will simulate our required environment. The projectiles for Missile moments are drawn using OpenCV functions that are included in that OpenAI gym environment created.
The Missile needs to fetch optimal parameters to hit the target over a dynamic environment with obstacles and random environment changes, just like an angry bird's game. This cannot be done using traditional problem-solving approaches model need to act according to the environment, learning should also be done through interacting with the environment.