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

Robotic manipulation holds the potential to replace humans in the executionof tedious or dangerous tasks. However, control-based approaches are notsuitable due to the difficulty of formally describing open-world manipulationin reality, and the inefficiency of existing learning methods. Thus, applyingmanipulation in a wide range of scenarios presents significant challenges. Inthis study, we propose a novel method for skill learning in roboticmanipulation called Tactile Active Inference Reinforcement Learning(Tactile-AIRL), aimed at achieving efficient training. To enhance theperformance of reinforcement learning (RL), we introduce active inference,which integrates model-based techniques and intrinsic curiosity into the RLprocess. This integration improves the algorithm's training efficiency andadaptability to sparse rewards. Additionally, we utilize a vision-based tactilesensor to provide detailed perception for manipulation tasks. Finally, weemploy a model-based approach to imagine and plan appropriate actions throughfree energy minimization. Simulation results demonstrate that our methodachieves significantly high training efficiency in non-prehensile objectspushing tasks. It enables agents to excel in both dense and sparse reward taskswith just a few interaction episodes, surpassing the SAC baseline. Furthermore,we conduct physical experiments on a gripper screwing task using our method,which showcases the algorithm's rapid learning capability and its potential forpractical applications.

Tactile Active Inference Reinforcement Learning for Efficient Robotic Manipulation Skill Acquisition