News

GJK++: Leveraging Acceleration Methods for Faster Collision Detection

GJK++: Leveraging Acceleration Methods for Faster Collision Detection

Discover our partner’s, CIIRC CTU, publication, titled “GJK++: Leveraging Acceleration Methods for Faster Collision Detection”.

The paper delves into the investigation of collision detection, a crucial aspect in fields like robotics, computational physics, and computer graphics. It focuses on enhancing the efficiency of the Gilbert, Johnson, and Keerthi (GJK) algorithm, a prominent solution for computing collision and distance between 3D convex geometries. By treating these tasks as optimization problems and adapting Polyak and Nesterov accelerations to Frank-Wolfe (FW) methods, the study introduces accelerated GJK versions. These advancements hold promise for modern robotic applications, offering substantial computational speed-ups.

Read the paper here.

Find AGIMUS publications here.

Newsletter

ABOUT
TESTING ZONES & INDUSTRIAL PILOTS
NEWS & EVENTS
RESOURCES

This work is supported by the AGIMUS project funded by the European Union under GA no.101070165. Views and opinions expressed are, however, those of the authors only and do not necessarily reflect those of the European Union or the European Commission. Neither the European Union nor the European Commission can be held responsible for them.

Image