Sensor Data Fusion
Lecture for the 2nd
Summer School 2008 on "ADVANCED TECHNOLOGIES FOR NEURO-MOTOR ASSESSMENT AND REHABILITATION". Promoted by:
Dept. of Electronics Computer Science & Systems (DEIS) University of Bologna - Italy.
Lecture Abstract: The talk concerns a problem, which is basic
to perception: the integration of perceptual information into a
coherent description of the world. In this talk the perception is
presented as a process of dynamically maintaining a model of the local
external environment. Fusion of perceptual information is at the heart
of this process. After a brief introduction, the background of the
problem of fusion in machine vision is reviewed. Then data fusion is
presented as part of the process of dynamic world modelling, and a set
of principles is postulated for the ~fusion~ of independent
observations. These principles lead to techniques, which permit
perceptual fusion with qualitatively different forms of data, treating
each source of information as constraints. For numerical information,
these principles lead to specific well-known tools such as various
forms of Kalman filter. For symbolic information, these principles
suggest representing objects and their relations as a conjunction of
properties encoded as schema. Dynamic world modelling is a cyclic
process composed of the phases: predict, match and update. These
phases provide a framework with which perceptual systems can be
organised and designed. It is shown that in the case of numerical
measurements, this framework leads to the use of a Kalman filter for
the prediction and update phases, while other tools can be used for
matching. In the case of symbolic information, elements of the
framework can be constructed with schema and production rules. The
framework for perceptual information is illustrated with the
architectures of several systems.
- 1. Introduction
- 1.1 Perception and Sensing
- 1.2 Background and State of the Art in Sensor Fusion
- 2. Sensor Fusion and Dynamic World Modelling
- 2.1 A General Framework for Dynamic World Modelling
- 2.2 Principles for Integrating Perceptual Information
- 3. Techniques for Fusion of Numerical Properties
- 3.1 State Representation
- 3.2 Prediction: Discrete State Transition Equations
- 3.3 Matching Observation to Prediction
- 3.4 The Kalman Filter Update Equations
- 4. Example Systems and Practical Applications
- 4.1 Dynamic World Modelling
- 4.2 An Integrated Supervision System
- 5. Conclusions
- K. Brammer, G. Siffling, Kalman Bucy Filters, Artech House Inc., Norwood MA, USA, 1989.
- J. Jazwinski, Stochastic Processes and Filtering Theory. Academic Press, New York, 1970.
- R. Kalman, A new approach to Linear Filtering and Prediction Problems, Transactions of the ASME,
Series D. J. Basic Eng., Vol 82, 1960.
- R. Kalman, R. Bucy, New Results in Linear Filtering and Prediction Theory, Transaction of the ASME,
Series D. J. Basic Eng., Vol 83, 1961.
- A. Papoulis, Probability, random variables, and stochastic processes. New York: McGraw-Hill, 1991.
- D. L. Hall, Mathematical techniques in multisensor data fusion. Boston: Artech House, 1992.
- Brian D. O. Anderson, John B. Moore, Optimal Filtering. Information and System Science Series.
Thomas Kailath Editor, 2005.
"Lecture Slides" (in English): (PDF file, KB).
(PDF file, KB, 2 slides per page).
(PDF file, KB, 4 slides per page).
Movies used for the Presentation: (WMF and MPG files in zipped format, 24MB).
- Matlab and Simulink files and software used for the presentation:
(Matlab and Simulink files in zipped format, 185KB).
- James L Crowley and Yves Demazeau, "Principles and Techniques for Sensor Data Fusion". LIFIA (IMAG)
46 avenue Felix Viallet. F-38031 Grenoble CÚdex, FRANCE. (PDF file, 90KB).
- RON ROTH, "Trends in Sensor and Data Fusion". 'Photogrammetric
Week 05', Dieter Fritsch, Ed. Wichmann Verlag, Heidelberg 2005. (PDF file, 3.74MB).
- G. GIRIJA, J. R. RAOL, R. APPAVU RAJ and SUDESH KASHYAP, "Tracking filter and multi-sensor data fusion".
Sadhana, Vol. 25, Part 2, April 2000, pp. 159-167. Printed in India, 2000.
(PDF file, 117KB).
- SHRABANI BHATTACHARYA and R APPAVU RAJ, "Performance evaluation of multi-sensor data fusion
technique for test range application", Sadhana Vol. 29, Part 2, April 2004, pp. 237-247.
Printed in India, 2004. (PDF file, 257KB).
Introduction paper: "Kalman Filtering: Whence, What and Whither?"
by B. D. O. Anderson and J. B. Moore. (PDF file).
The book by BRIAN D. O. ANDERSON and JOHN B. MOORE, "Optimal
Filtering". INFORMATION AND SYSTEM SCIENCES SERIES. Thomas Kailath
Editor. (PDF file).
R. E. KALMAN, Research Institute for Advanced Study, Baltimore, Md. "A New Approach to Linear Filtering
and Prediction Problems" (1960). Transactions of the ASME Journal of Basic Engineering, 82 (Series D): 35-45.
"An Introduction to the Kalman Filter", by Greg Welch and Gary Bishop. Department of Computer Science
University of North Carolina at Chapel Hill Chapel Hill, NC 27599-3175. April 5, 2004
Pattern Recognition and Machine Vision: Kalman Filter. By Dr. Simon J.D. Prince
Computer Science University College London. Gower Street, London, WC1E 6BT:
"Least-Squares Estimation: From Gauss to Kalman". By H.W. Sorenson. University of California.
San Diego. IEEE Spectrum, vol. 7, pp. 63-68. July, 1970. (PDF file).
Kalman filter Main Home Page: . Some tutorials, references,
and research on the Kalman filter. The site is maintained by Greg Welch
and Gary Bishop, faculty members of the
Department of Computer Science at the University of North Carolina at Chapel