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

Our goal is to develop an efficient contact detection algorithm forlarge-scale GPU-based simulation of non-convex objects. Current GPU-basedsimulators such as IsaacGym and Brax must trade-off speed with fidelity,generality, or both when simulating non-convex objects. Their main issue liesin contact detection (CD): existing CD algorithms, such asGilbert-Johnson-Keerthi (GJK), must trade off their computational speed withaccuracy which becomes expensive as the number of collisions among non-convexobjects increases. We propose a data-driven approach for CD, whose accuracydepends only on the quality and quantity of offline dataset rather than onlinecomputation time. Unlike GJK, our method inherently has a uniform computationalflow, which facilitates efficient GPU usage based on advanced compilers such asXLA (Accelerated Linear Algebra). Further, we offer a data-efficient solutionby learning the patterns of colliding local crop object shapes, rather thanglobal object shapes which are harder to learn. We demonstrate our approachimproves the efficiency of existing CD methods by a factor of 5-10 fornon-convex objects with comparable accuracy. Using the previous work on contactresolution for a neural-network-based contact detector, we integrate our CDalgorithm into the open-source GPU-based simulator, Brax, and show that we canimprove the efficiency over IsaacGym and generality over standard Brax. Wehighly recommend the videos of our simulator included in the supplementarymaterials.

Local object crop collision network for efficient simulation of non-convex objects in GPU-based simulators