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

For autonomous quadruped robot navigation in various complex environments, atypical SOTA system is composed of four main modules -- mapper, global planner,local planner, and command-tracking controller -- in a hierarchical manner. Inthis paper, we build a robust and safe local planner which is designed togenerate a velocity plan to track a coarsely planned path from the globalplanner. Previous works used waypoint-based methods (e.g.Proportional-Differential control and pure pursuit) which simplify the pathtracking problem to local point-goal navigation. However, they suffer fromfrequent collisions in geometrically complex and narrow environments because oftwo reasons; the global planner uses a coarse and inaccurate model and thelocal planner is unable to track the global plan sufficiently well. Currently,deep learning methods are an appealing alternative because they can learnsafety and path feasibility from experience more accurately. However, existingdeep learning methods are not capable of planning for a long horizon. In thiswork, we propose a learning-based fully autonomous navigation frameworkcomposed of three innovative elements: a learned forward dynamics model (FDM),an online sampling-based model-predictive controller, and an informedtrajectory sampler (ITS). Using our framework, a quadruped robot canautonomously navigate in various complex environments without a collision andgenerate a smoother command plan compared to the baseline method. Furthermore,our method can reactively handle unexpected obstacles on the planned path andavoid them. Project pagehttps://awesomericky.github.io/projects/FDM_ITS_navigation/.

Learning Forward Dynamics Model and Informed Trajectory Sampler for Safe Quadruped Navigation