A Bayes-Maximum Entropy method for multi-sensor data fusion

Conference · Tue Jan 01 00:00:00 EST 1991

OSTI ID:5617107

Beckerman, M

In this paper we introduce a Bayes-Maximum Entropy formalism for multi-sensor data fusion, and present an application of this methodology to the fusion of ultrasound and visual sensor data as acquired by a mobile robot. In our approach the principle of maximum entropy is applied to the construction of priors and likelihoods from the data. Distances between ultrasound and visual points of interest in a dual representation are used to define Gibbs likelihood distributions. Both one- and two-dimensional likelihoods are presented, and cast into a form which makes explicit their dependence upon the mean. The Bayesian posterior distributions are used to test a null hypothesis, and Maximum Entropy Maps used for navigation are updated using the resulting information from the dual representation. 14 refs., 9 figs.

• Research Organization: Oak Ridge National Lab., TN (United States)
• Sponsoring Organization: USDOE; USDOE, Washington, DC (United States)
• DOE Contract Number: AC05-84OR21400
• OSTI ID: 5617107
• Report Number(s): CONF-920540-13; ON: DE92000680
• Resource Relation: Conference: 1992 Institute of Electrical and Electronics Engineers (IEEE) international conference on robotics and automation, Nice (France), 10-15 May 1992
• Country of Publication: United States
• Language: English

Related Subjects

42 ENGINEERING
99 GENERAL AND MISCELLANEOUS//MATHEMATICS, COMPUTING, AND INFORMATION SCIENCE
ROBOTS
TASK SCHEDULING
ARTIFICIAL INTELLIGENCE
AUTOMATION
DATA ACQUISITION
ENTROPY
FEEDBACK
MANIPULATORS
MATHEMATICS
NAVIGATION
PROBABILITY
REMOTE SENSING
DATA PROCESSING
EQUIPMENT
LABORATORY EQUIPMENT
MATERIALS HANDLING EQUIPMENT
PHYSICAL PROPERTIES
PROCESSING
REMOTE HANDLING EQUIPMENT
THERMODYNAMIC PROPERTIES
420200* - Engineering- Facilities
Equipment
& Techniques
990200 - Mathematics & Computers