Name: Kyle Morgenstein
Type: User
Company: UT Austin
Bio: PhD Candidate @ UT Austin. Working on RL + control for safe, friendly, social robots. Previously MIT, NASA JPL, Boston Dynamics AI Institute
Twitter: KyleMorgenstein
Blog: www.kylemorgenstein.com
Kyle Morgenstein's Projects
AHHH: a programming language for the dreadful
Projects from TavTech Fellowship winter 2017
IL for Bumpybot
Competency Conditioned Multi Agent Search
Optimization Library
class for creating urdf files from occupancy grids. also supports generating A* trajectories with collision constraints
a repository for minimal implementations of common estimators
a repository for minimal implementations of common generative models
repo to manage repos for running isaac sim orbit environments for robot learning
Repo for models of robots operated by the UT Austin Human Centered Robotics Lab
We let two agents approach each other, collide and generate contact impulses in the simulation engine Bullet (via pybullet).
LWR research
Imitation Learning Algorithms
Repo for HCRL Orbit Environments
Isaac Gym Reinforcement Learning Environments
Data for LiDAR Informed Visual Search (LIVES) with Learned Scan Classification
LiDAR Informed Visual Search (LIVES) with Learned Scan Classification
multi agent reinforcement learning search with information gain rewards
quick CV project to identify mask-like objects
multi-robot RL search with information gains
N-Body Dynamics Simulator
YOLO Object detection with OpenCV and Python.
notes for optimal control theory
Unified framework for robot learning built on NVIDIA Isaac Sim
classic game of pong
pronto running on the Unitree A1 quadruped
Simplifying reinforcement learning for complex game environments