Unified GPU voxel collision detection for mobile manipulation planning

Hermann, A. M.; Drews, Florian; Bauer, Joerg; Klemm, Sebastian; Roennau, Arne; Dillmann, Ruediger

doi:10.1109/iros.2014.6943148

Cited by 60 publications

(36 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…By similar reasoning, we can also see that the infimum contributes the same values as in (4). Therefore, the total enlargement is the sum of the contributions from the supremum and the infimum:…”

Section: B Amount Of Rectangle Enlargement For Each Dimension 1) Dimmentioning

confidence: 53%

See 1 more Smart Citation

On time-memory trade-off for collision detection

Rizaldi

Sontges

Althoff

2015

2015 IEEE Intelligent Vehicles Symposium (IV)

View full text Add to dashboard Cite

Abstract-Future collision avoidance systems, which are capable of fully controlling the vehicle, have to make critical decisions in a very short time. To do this, they need to check constantly if their own vehicle's occupancy collides with the other traffic participants' occupancy. Those collision checks consume a substantial amount of time and consequently, the collision avoidance systems could fail to intervene in complex scenarios. We propose a new approach to reduce the computation time for collision checks significantly. Instead of using geometric methods, we store finitely many possible collision scenarios between two objects in a table and thus collision checks become a matter of lookup table queries. To ensure that the finite number of configurations cover all possible scenarios, we use a novel abstraction technique which guarantees that every collision will be detected. The approach works for arbitrarily many traffic participants by applying the approach pairwise (own vehicle and other object) to each traffic participant. Randomly generated scenarios show that the new approach can be several times faster than geometric intersection techniques thanks to the trade-off between memory consumption and computation time.

show abstract

Section: B Amount Of Rectangle Enlargement For Each Dimension 1) Dimmentioning

confidence: 53%

“…The first approach uses sensor data, which has only been processed to a small extent and is gathered in occupancy grids, see e.g. [3] for the two-dimensional case and [4] for the three-dimensional case. In order to improve access to grid-based occupancy information, tree-based data structures are often used [5].…”

Section: Introductionmentioning

confidence: 99%

On time-memory trade-off for collision detection

Rizaldi

Sontges

Althoff

2015

2015 IEEE Intelligent Vehicles Symposium (IV)

View full text Add to dashboard Cite

show abstract

“…Collision detection between point clouds was for example researched by Klein and Zachmann (Klein and Zachmann, 2004) who use the implicit surface created by a point cloud to calculate intersections. Another example is the recent work by Hermann et al (Hermann et al, 2014) who use voxels to check for spatial occupancy for robot motion planning.…”

Section: Related Workmentioning

confidence: 99%

“…One method is to apply a spatial hierarchical partitioning of the input geometry using octrees (Jung and Gupta, 1996), BSP-trees (Ar et al, 2000) or k-d trees (Held et al, 1995) . Other solutions apply regular partitioning using voxels (Garcia-Alonso et al, 1994, Hermann et al, 2014. The goal of any partitioning is to be able to quickly search and check only the relevant geometries in the same or neighboring cells.…”

Section: Related Workmentioning

confidence: 99%

Efficient point cloud collision detection and analysis in a tunnel environment using kinematic laser scanning and K-D Tree search

Schauer

Nüchter

2014

Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci.

View full text Add to dashboard Cite

ABSTRACT:Measuring the structure gauge of tunnels and other narrow passages has so far been the only way to evaluate whether large vehicles can pass through them. But especially for very long vehicles like train wagons and their cargo, the structure gauge is an insufficient measure because the center part of the vehicle between two bogies will inevitably leave the swept volume of its cross section when moving along any other trajectory than a straight line perpendicular to its cross section. In addition, the vehicle as well as the cargo must keep a minimum safety margin from the environment at all points of its trajectory. This paper explores an automated method to check for possible collisions of a model represented by a 3D point cloud moving through the 3D point cloud of an environment. We were given environment data of a train track through a narrow tunnel where simply relying on the structure gauge would indicate that a given wagon would pass through without any collision even though in reality, the train wagon would collide with the inner tunnel wall inside a sharp turn of the tracks. The k-d tree based collision detection method presented in this paper is able to correctly highlight these collisions and indicate the penetration depth of each colliding point of the environment into the model of the train wagon. It can be generalized for any setup where two static point clouds have to be tested for intersection along a trajectory.

show abstract

“…There are methods to subdivide a 3D space into small cells by means of the octree or the kd-tree (Figueiredo et al, 2010, Hermann et al, 2014, Klein et al, 2004, Pan et al, 2013a, Shauer et al, 2014. This type of method is suitable for precise collision detection.…”

Section: Introductionmentioning

confidence: 99%

Collision Visualization of a Laser-Scanned Point Cloud Of Streets and a Festival Float Model Used for the Revival Of A Traditional Procession Route

Shigeta

Hasegawa

et al. 2017

Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci.

View full text Add to dashboard Cite

ABSTRACT:Recently, laser-scanning technology, especially mobile mapping systems (MMSs), has been applied to measure 3D urban scenes. Thus, it has become possible to simulate a traditional cultural event in a virtual space constructed using measured point clouds. In this paper, we take the festival float procession in the Gion Festival that has a long history in Kyoto City, Japan. The city government plans to revive the original procession route that is narrow and not used at present. For the revival, it is important to know whether a festival float collides with houses, billboards, electric wires or other objects along the original route. Therefore, in this paper, we propose a method for visualizing the collisions of point cloud objects. The advantageous features of our method are (1) a see-through visualization with a correct depth feel that is helpful to robustly determine the collision areas, (2) the ability to visualize areas of high collision risk as well as real collision areas, and (3) the ability to highlight target visualized areas by increasing the point densities there.

show abstract

Unified GPU voxel collision detection for mobile manipulation planning

Cited by 60 publications

References 13 publications

On time-memory trade-off for collision detection

On time-memory trade-off for collision detection

Efficient point cloud collision detection and analysis in a tunnel environment using kinematic laser scanning and K-D Tree search

Collision Visualization of a Laser-Scanned Point Cloud Of Streets and a Festival Float Model Used for the Revival Of A Traditional Procession Route

Contact Info

Product

Resources

About