Fast Continuous Collision Detection for Articulated Models

[+] Author and Article Information
Stephane Redon

Department of Computer Science,  University of North Carolina at Chapel Hill, Chapel Hill, North Carolina 27599-3175redon@cs.unc.edu

Ming C. Lin

Department of Computer Science,  University of North Carolina at Chapel Hill, Chapel Hill, North Carolina 27599-3175lin@cs.unc.edu

Dinesh Manocha

Department of Computer Science,  University of North Carolina at Chapel Hill, Chapel Hill, North Carolina 27599-3175dm@cs.unc.edu

Young J. Kim

Computer Science and Engineering,  Ewha Womans University, Koreakimy@ewha.ac.kr

J. Comput. Inf. Sci. Eng 5(2), 126-137 (Feb 06, 2005) (12 pages) doi:10.1115/1.1884133 History: Received September 03, 2004; Revised February 06, 2005

We present a novel algorithm to perform continuous collision detection for articulated models. Given two discrete configurations of the links of an articulated model, we use an “arbitrary in-between motion” to interpolate its motion between two successive time steps and check the resulting trajectory for collisions. Our approach uses a three-stage pipeline: (1) dynamic bounding-volume hierarchy (D-BVH) culling based on interval arithmetic; (2) culling refinement using the swept volume of line swept spheres (LSS’) and graphics hardware accelerated queries; (3) exact contact computation using OBB trees and continuous collision detection between triangular primitives. The overall algorithm computes the time of collision and contact locations, and prevents any interpenetration between the articulated model and the environment. We have implemented the algorithm and tested its performance on a 2.4GHz Pentium PC with 1Gbyte of RAM and a NVIDIA GeForce FX 5800 graphics card. In practice, our algorithm is able to perform accurate and continuous collision detection between articulated models and modestly complex environments at nearly interactive rates.

Copyright © 2005 by American Society of Mechanical Engineers
Your Session has timed out. Please sign back in to continue.



Grahic Jump Location
Figure 1

Continuous collision detection between a Puma robot and a CAD /CAM model of pipes

Grahic Jump Location
Figure 2

Link i is moving in the reference frame of its parent. The initial and final positions of the link as well as the motion trajectory have been outlined.

Grahic Jump Location
Figure 3

The pipeline of our continuous collision detection algorithm

Grahic Jump Location
Figure 4

The swept volume (d) of the LSS consists of LSS’ at initial and final configurations (a), the offsets of the ruled surface (b) and the pipe surfaces (c)

Grahic Jump Location
Figure 5

(a) The LSS’ bounding the Puma robot. (b) The right image shows the swept volume of the LSS bounding the end-effector of the Puma robot.

Grahic Jump Location
Figure 6

Dynamic swept volume culling based on graphics hardware applied to the last two links of a Puma robot model. In this example, the swept LSS of the last link does not collide with the environment, and thus the link is culled away. However, the swept LSS of its parent link (shown in red) does intersect the environment, and the parent link has to undergo the third stage of the collision detection pipeline.

Grahic Jump Location
Figure 7

Pipes and Puma robot

Grahic Jump Location
Figure 8

The Puma robot in the AMR environment (11,80 objects, 187,000 triangles)

Grahic Jump Location
Figure 9

The star-shaped robot (5 branches of 10 links each, and one root link) in the AMR environment (1180 objects, 187000 triangles). An edge∕edge contact has just been detected.




Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging and repositioning the boxes below.

Related Journal Articles
Related eBook Content
Topic Collections

Sorry! You do not have access to this content. For assistance or to subscribe, please contact us:

  • TELEPHONE: 1-800-843-2763 (Toll-free in the USA)
  • EMAIL: asmedigitalcollection@asme.org
Sign In