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

Path planning is a fundamental capability for autonomous navigation ofrobotic wheelchairs. With the impressive development of deep-learningtechnologies, imitation learning-based path planning approaches have achievedeffective results in recent years. However, the disadvantages of theseapproaches are twofold: 1) they may need extensive time and labor to recordexpert demonstrations as training data; and 2) existing approaches could onlyreceive high-level commands, such as turning left/right. These commands couldbe less sufficient for the navigation of mobile robots (e.g., roboticwheelchairs), which usually require exact poses of goals. We contribute asolution to this problem by proposing S2P2, a self-supervised goal-directedpath planning approach. Specifically, we develop a pipeline to automaticallygenerate planned path labels given as input RGB-D images and poses of goals.Then, we present a best-fit regression plane loss to train our data-driven pathplanning model based on the generated labels. Our S2P2 does not need pre-builtmaps, but it can be integrated into existing map-based navigation systemsthrough our framework. Experimental results show that our S2P2 outperformstraditional path planning algorithms, and increases the robustness of existingmap-based navigation systems. Our project page is available athttps://sites.google.com/view/s2p2.

S2P2: Self-Supervised Goal-Directed Path Planning Using RGB-D Data for Robotic Wheelchairs