Particle Filtering for Multisensor Data Fusion With Switching Observation Models: Application to Land Vehicle Positioning

• Published in: IEEE transactions on signal processing Vol. 55; no. 6; pp. 2703 - 2719
• Main Authors: Caron, F., Davy, M., Duflos, E., Vanheeghe, P.
• Format: Journal Article
• Language: English
• Published: New York, NY IEEE 01.06.2007; Institute of Electrical and Electronics Engineers; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Applied sciences; Bayesian methods; Cameras; Data fusion; Detection, estimation, filtering, equalization, prediction; Exact sciences and technology; Failure; Fault detection; Filtering; Global Positioning System; Information theory; Information, signal and communications theory; Land; Land vehicles; Mathematical models; multisensor system; particle filter; Particle filters; Satellite navigation systems; Sensors; sequential Monte Carlo methods; Signal and communications theory; Signal, noise; State estimation; Switches; switching observation model; Telecommunications and information theory; Vehicles; Wheels; Monte Carlo method; Positioning; Multisensor; GPS system; fault detection; Sensor fusion; Radiolocalisation; Data switching; Prior probability; multisensor system; Failures; global positioning system; sequential Monte Carlo methods; switching observation model; Satellite navigation; Sequential method; Information processing; Data fusion; Numerical simulation; Masking; Sequential estimation; Particle filter; Defect detection
• ISSN: 1053-587X; 1941-0476
• DOI: 10.1109/TSP.2007.893914

This paper concerns the sequential estimation of a hidden state vector from noisy observations delivered by several sensors. Different from the standard framework, we assume here that the sensors may switch autonomously between different sensor states, that is, between different observation models. This includes sensor failure or sensor functioning conditions change. In our model, sensor states are represented by discrete latent variables, whose prior probabilities are Markovian. We propose a family of efficient particle filters, for both synchronous and asynchronous sensor observations as well as for important special cases. Moreover, we discuss connections with previous works. Lastly, we study thoroughly a wheel land vehicle positioning problem where the GPS information may be unreliable because of multipath/masking effects