Improving Robotic Exploration via RL Graph Pruning             06/2025 - present
--------------------------------------------------------------------------------

This work aims to extend an existing approach to safety-guaranteed reinforcement
learning control of multi-agent systems to tasks with more complex control
dynamics. It specifically aims to extend the DGPPO algorithm developed by Zhang
et al to a custom simulation based on the F16 control dynamics. The goal is to
control a formation of F16s as they follow a pre-designated path with various
safety conditions that must be met at all times during flight.

I am currently pursuing this project under the mentorship of a professor at MIT.

<<< Go back