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

This paper presents a Deep Reinforcement Learning based navigation approachin which we define the occupancy observations as heuristic evaluations ofmotion primitives, rather than using raw sensor data. Our method enables fastmapping of the occupancy data, generated by multi-sensor fusion, intotrajectory values in 3D workspace. The computationally efficient trajectoryevaluation allows dense sampling of the action space. We utilize our occupancyobservations in different data structures to analyze their effects on bothtraining process and navigation performance. We train and test our methodologyon two different robots within challenging physics-based simulationenvironments including static and dynamic obstacles. We benchmark our occupancyrepresentations with other conventional data structures from state-of-the-artmethods. The trained navigation policies are also validated successfully withphysical robots in dynamic environments. The results show that our method notonly decreases the required training time but also improves the navigationperformance as compared to other occupancy representations. The open-sourceimplementation of our work and all related info are available at\url{https://github.com/RIVeR-Lab/tentabot}.

Deep Reinforcement Learning based Robot Navigation in Dynamic Environments using Occupancy Values of Motion Primitives