Accurate prediction of interception positions for catching thrown objects in production systems

Barteit, D.; Frank, Heinz; Kupzog, F.

doi:10.1109/indin.2008.4618228

Cited by 14 publications

(10 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It is shortly introduced in [4]. There, one camera triggers another camera when a thrown object has passed a certain plane.…”

Section: Related Workmentioning

confidence: 99%

“…Also the performance of an implementation of a tracking and prediction algorithm for robotic catching presented in [4] is studied again with the light grid approach. The position where a tennis ball that is launched with approximately 9 m/s will intersects a plane in 3 m distance can be predicted with an accuracy better than 16.7 mm in 90% of the throws.…”

Section: Positionsmentioning

confidence: 99%

“…Further three subtasks of the catching part can be distinguished (see also Figure 1): during flight the object's trajectory has to be observed and estimated for a prediction of an interception position [4]. Then a robot has to move a gripping-device to the predicted catching position [6].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Measuring the intersection of a thrown object with a vertical plane

Barteit

Frank

Pongratz

et al. 2009

2009 7th IEEE International Conference on Industrial Informatics

Self Cite

View full text Add to dashboard Cite

This work is part of a project that investigates a new approach for material transport within manufacturing systems by throwing or shooting. Such material transport systems includes both throwing and catching of workpieces by robots. In production processes objects can be thrown with high velocity over distances of several meters. High accuracy in throwing and catching is important. Here, two alternative methods are presented for measuring the accuracy of predictions of interception positions used by the catching robot. The first measurement method presented uses a single triggered gray-value camera to determine the actual interception position. The second method bases on a Dispersive Signal Technology (DST) touch kit. Measurement techniques and the accuracies of these systems are presented in this paper.

show abstract

“…It is shortly introduced in [4]. There, one camera triggers another camera when a thrown object has passed a certain plane.…”

Section: Related Workmentioning

confidence: 99%

Section: Positionsmentioning

confidence: 99%

See 1 more Smart Citation

Measuring the intersection of a thrown object with a vertical plane

Barteit

Frank

Pongratz

et al. 2009

2009 7th IEEE International Conference on Industrial Informatics

Self Cite

View full text Add to dashboard Cite

show abstract

“…Similar to most work regarding the topic of catching [1] [4] also this approach is dealing with a tennis ball as thrown object. Approaches based on a monocular vision system [4] and based on binocular vision system [1] [5] have already been done. In this context usage of binocular vision systems equal a bio-inspired approach as most predators in nature feature two eyes.…”

Section: Introductionmentioning

confidence: 99%

“…In contrast to small scale catching [1] [5] no gripper is used to verify the quality of the prediction system. An impact position verification system (compare [4] and [6]) based on a touch-kit is used to continuously meter the quality of the prediction. This systems detects the position of the tennis ball in the interception plane enabling better evaluation of the prediction quality than the binary result of a successful/unsuccessful catch.…”

Section: Introductionmentioning

confidence: 99%

Transport by throwing - A bio-inspired approach

Pongratz

Kupzog

Frank³

et al. 2010

2010 8th IEEE International Conference on Industrial Informatics

Self Cite

View full text Add to dashboard Cite

Abstract-In modern industrial production fast and easily reconfigurable transportation systems are necessary. A viable bioinspired approach to this is throwing and catching of transportation goods. In order to catch a thrown object the catching device has to be moved to the right position on time. This requires a fast and accurate acquisition of flight position and a prediction system for the interception point. The main topic of this paper is the development and comparison of two prediction models for the flight trajectory of a thrown tennis ball. The position acquisition, that is the base for the prediction, is based on a binocular vision system similar to two-eyed humans. The impact of the vision systems frame rate on the error of the prediction is reviewed as well. Future prediction is planned to be done based on a bio-inspired approach using a small set of reference throws.

show abstract