Learning Visual Tracking and Reaching with Deep Reinforcement Learning on a UR10e Robotic Arm

TL;DR

This paper explores applying deep reinforcement learning to enable a UR10e robotic arm to learn visual tracking and reaching tasks, demonstrating that proximal policy optimization outperforms deep Q-learning in stability and data efficiency.

Contribution

The study develops specialized reinforcement learning environments for the UR10e robot and compares deep Q-learning with proximal policy optimization, highlighting the latter's advantages.

Findings

Proximal policy optimization learns more stable policies.

PPO requires less training data than DQN.

Initial results show promise for RL in industrial robotics.

Abstract

As technology progresses, industrial and scientific robots are increasingly being used in diverse settings. In many cases, however, programming the robot to perform such tasks is technically complex and costly. To maximize the utility of robots in industrial and scientific settings, they require the ability to quickly shift from one task to another. Reinforcement learning algorithms provide the potential to enable robots to learn optimal solutions to complete new tasks without directly reprogramming them. The current state-of-the-art in reinforcement learning, however, generally relies on fast simulations and parallelization to achieve optimal performance. These are often not possible in robotics applications. Thus, a significant amount of research is required to facilitate the efficient and safe, training and deployment of industrial and scientific reinforcement learning robots. This technical report outlines our initial research into the application of deep reinforcement learning on an industrial UR10e robot. The report describes the reinforcement learning environments created to facilitate policy learning with the UR10e, a robotic arm from Universal Robots, and presents our initial results in training deep Q-learning and proximal policy optimization agents on the developed reinforcement learning environments. Our results show that proximal policy optimization learns a better, more stable policy with less data than deep Q-learning. The corresponding code for this work is available at \url{https://github.com/cbellinger27/bendRL_reacher_tracker}