Balancer - a three-stage inverted pendulum for RL experimentation

Inspired by a student in my reading group, this video, and this paper, I decided to build my own three-stage inverted pendulum system to experiment with different RL algorithms. The plan is to be able to build a system that can run a force controller and then use RL to actually perform the task of lifting all three stages up and keeping them up for as long as possible. I plan on using some of the algorithms I have developed at the lab and testing them out on this system.

Right now, I am planning out how to build the physical frame and mechanical components of the robot, with a lot of help from Karmanyaah Malhotra and others at UT Austin.