MartinBraquet / Robot-Learning-UT

Simulation of a neural network model using Deep Deterministic Policy Gradient (DDPG) improved with Hindsight Experience Replay (HER) in the Fetch Reach and Pick and Place environments of Gym Open AI.

Class Project in Robot Learning

CS391R: Robot Learning at the University of Texas at Austin

Final Project (Fall 2021)

Authors: Martin Braquet & Steven Patrick

Simulation of a neural network model using Deep Deterministic Policy Gradient (DDPG) improved with Hindsight Experience Replay (HER) in the Fetch Reach and Pick and Place environments of Gym Open AI.

This repo is a modified version of https://github.com/ARISE-Initiative/robosuite.