A sensor data fusion approach for the integration of negative information

Authors:

1 July 2007

conference paper
Published by Institute of Electrical and Electronics Engineers (IEEE)

p. 1-7
DOI:10.1109/icif.2007.4408160

Abstract

Negative information provides important additional knowledge that is not exploited for sensor data fusion tasks by default. This paper presents a new approach to incorporate such information about unoccupied, observed areas or missing measurements in the Kalman filtering process. For this purpose, a combination with a grid-based method is proposed to generate a visibility map. This enables a plausibility check and an enhanced understanding for the collaborative perception of the environment with multiple cognitive vehicles. Thereby, the reliability and consistency of the joint description of the environment are increased. Results from a realistic traffic simulation are presented.

Keywords

SENSOR FUSION
AREA MEASUREMENT
KALMAN FILTERS
INFORMATION FILTERING
INFORMATION FILTERS
FUSION POWER GENERATION
MESH GENERATION
COLLABORATION
VEHICLES
TRAFFIC CONTROL

This publication has 5 references indexed in Scilit:

Grid Based Fusion & Tracking
2006
On Combining The Hough Transform And Occupancy Grid Methods For Detection Of Moving Objects
2005
High resolution maps from wide angle sonar
2005
Enhanced environmental perception by inter-vehicle data exchange
2005
Making use of what you don't see: negative information in Markov localization
2005