English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

As technology progresses, industrial and scientific robots are increasinglybeing used in diverse settings. In many cases, however, programming the robotto perform such tasks is technically complex and costly. To maximize theutility of robots in industrial and scientific settings, they require theability to quickly shift from one task to another. Reinforcement learningalgorithms provide the potential to enable robots to learn optimal solutions tocomplete new tasks without directly reprogramming them. The currentstate-of-the-art in reinforcement learning, however, generally relies on fastsimulations and parallelization to achieve optimal performance. These are oftennot possible in robotics applications. Thus, a significant amount of researchis required to facilitate the efficient and safe, training and deployment ofindustrial and scientific reinforcement learning robots. This technical reportoutlines our initial research into the application of deep reinforcementlearning on an industrial UR10e robot. The report describes the reinforcementlearning environments created to facilitate policy learning with the UR10e, arobotic arm from Universal Robots, and presents our initial results in trainingdeep Q-learning and proximal policy optimization agents on the developedreinforcement learning environments. Our results show that proximal policyoptimization learns a better, more stable policy with less data than deepQ-learning. The corresponding code for this work is available at\url{https://github.com/cbellinger27/bendRL_reacher_tracker}

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