gProximity: Hierarchical GPU‐based Operations for Collision and Distance Queries

Abstract

We present novel parallel algorithms for collision detection and separation distance computation for rigid and deformable models that exploit the computational capabilities of many-core GPUs. Our approach uses thread and data parallelism to perform fast hierarchy construction, updating, and traversal using tight- fitting bounding volumes such as oriented bounding boxes (OBB) and rectangular swept spheres (RSS). We also describe efficient algorithms to compute a linear bounding volume hierarchy (LBVH) and update them using refitting methods. Moreover, we show that tight-fitting bounding volume hierarchies offer improved performance on GPU-like throughput architectures. We use our algorithms to perform discrete and continuous collision detection including self-collisions, as well as separation distance computation between non-overlapping models. In practice, our approach (gProximity) can perform these queries in a few milliseconds on a PC with NVIDIA GTX 285 card on models composed of tens or hundreds of thousands of triangles used in cloth simulation

A pilot study to assess the feasibility and accuracy of using haptic technology to occlude digital dental models

“…A GPU-based collision detection method was used to determine the contact triangle pairs between the upper and lower dental models. 19,20 As a consequence of contact, a force along the normal direction was applied to prevent interpenetration between the two models. Two colliding forms, the collision form or the contact form, were distinguished from the module of the normal component of the relative velocity.…”

Parallel GPU-based collision detection of irregular vessel wall for massive particles

“…We calculated error rate as formula (7). For example, when the grid length is 0.2, compared to using one-triangle, using two-triangles will improve accuracy from 0.963049 to 0.993950, and the error rate reduces from 3.11% to 0.0075%.…”

“…This algorithm recursively constructs bounding volumes and treats each object as a leaf node. It makes self-collision detection within the same BVH to detect intersecting objects [5][6][7].…”

GPU-Accelerated Minimum Distance and Clearance Queries

“…Lauterbach et al have recently developed a GPU algorithm where the hierarchy traversal and primitive queries are also performed on the GPU [25]. Even though such an exclusive GPU algorithm overcomes the CPU/GPU bottleneck, it requires newer hardware to perform the atomic operations on the GPU.…”