Minimum distance estimation and collision prediction under uncertainty for on-line robotic motion planning

K. J. Kyriakopoulos, G. N. Saridis

Research output: Contribution to conferencePaperpeer-review

Abstract

An efficient method for computing the minimum distance and predicting collisions between moving objects is presented. This problem is incorporated in the framework of an in-line motion planning algorithm to satisfy collision avoidance between a robot and moving objects modeled as convex polyhedra. In the beginning the deterministic problem, where the information about the objects is assumed to be certain is examined. If instead of the Euclidean norm, L1 or L norms are used to represent distance, the problem becomes a linear programming problem. The stochastic problem is formulated, where the uncertainty is induced by the visual feedback and the unknown dynamics of the moving obstacles. Two problems are considered: First, an optimal filtering of the minimum distance between the robot and the moving object, at the present time. Second, an optimal prediction of the minimum distance in the future, in order to predict possible collisions with the moving obstacles and estimate the collision time.

Original languageEnglish (US)
Pages93-98
Number of pages6
StatePublished - 1991
EventProceedings of the 11th Triennial World Congress of the International Federation of Automatic Control - Tallinn, USSR
Duration: Aug 13 1990Aug 17 1990

Conference

ConferenceProceedings of the 11th Triennial World Congress of the International Federation of Automatic Control
CityTallinn, USSR
Period8/13/908/17/90

ASJC Scopus subject areas

  • Engineering(all)

Fingerprint

Dive into the research topics of 'Minimum distance estimation and collision prediction under uncertainty for on-line robotic motion planning'. Together they form a unique fingerprint.

Cite this