Name: Kyle Morgenstein
Type: User
Company: UT Austin
Bio: PhD Student @ UT Austin, Boston Dynamics AI Institute. Working on RL + control for safe, friendly, social robots. Previously MIT, NASA JPL
Twitter: KyleMorgenstein
Blog: www.kylemorgenstein.com
Kyle Morgenstein's Projects
Reinforcement Algos
RL implementations
minimal node for performing inference on onnx policies in ros
Fast and simple implementation of RL algorithms, designed to run fully on GPU.
Taka's WFC terrain generator
socially aware locomotion for legged robots
learn dynamics with transformers
VMAS is a vectorized framework designed for efficient Multi-Agent Reinforcement Learning benchmarking. It is comprised of a vectorized 2D physics engine written in PyTorch and a set of challenging multi-robot scenarios. Additional scenarios can be implemented through a simple and modular interface.
minimal example to set up wandb