Adaptive human sensor model in sensor networks

This paper presents the design of a probabilistic model of human perception as an integral part of a decentralized data fusion system. The system consists of a team of human operators and robotic platforms, together forming a heterogeneous sensor network. Human operators are regarded as information sources submitting raw observations. The observations are converted into a probabilistic representation suitable for fusion with the system's belief. The conversion is performed by a human sensor model (HSM). The initial HSM is built offline based on an average of multiple human subjects conducting a calibration experiment. Since individual human operators may vary in their performance, an online adaptation of the HSM is required. The network estimate is used for adaptation because the true feature state is unknown at runtime. Results of an outdoor calibration experiment using range and bearing observations are presented. Simulations show the feasibility of efficient online adaptation.

[1]  S S Fukusima,et al.  Stability of Individual Psychophysical Functions for Perceived Distance in Natural Indoor and Outdoor Settings , 1986, Perceptual and motor skills.

[2]  Kurt Konolige,et al.  Centibots: Very Large Scale Distributed Robotic Teams , 2004, AAAI.

[3]  Frank Wittig.  Learning Bayesian networks with hidden variables for user modeling , 1999 .

[4]  Alexei Makarenko,et al.  Human-robot interactions in active sensor networks , 2003, Proceedings 2003 IEEE International Symposium on Computational Intelligence in Robotics and Automation. Computational Intelligence in Robotics and Automation for the New Millennium (Cat. No.03EX694).

[5]  Masayuki Inaba,et al.  User adaptation of human-robot interaction model based on Bayesian network and introspection of interaction experience , 2000, Proceedings. 2000 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2000) (Cat. No.00CH37113).

[6]  David A. Schkade,et al.  WHERE DO THE NUMBERS COME FROM , 1993 .

[7]  Deborah Estrin,et al.  A Collaborative Approach to In-Place Sensor Calibration , 2003, IPSN.

[8]  David E. Culler,et al.  Calibration as parameter estimation in sensor networks , 2002, WSNA '02.

[9]  Eric Horvitz,et al.  The Lumière Project: Bayesian User Modeling for Inferring the Goals and Needs of Software Users , 1998, UAI.

[10]  John C. Baird,et al.  Quantitative functions for size and distance judgments , 1967 .

[11]  Carlos Guestrin,et al.  Robust Probabilistic Inference in Distributed Systems , 2004, UAI.

[12]  Erik Blasch,et al.  JDL level 5 fusion model: user refinement issues and applications in group tracking , 2002, SPIE Defense + Commercial Sensing.

[13]  W.D. Blair,et al.  Estimation of sensor bias in multisensor systems , 1992, Proceedings IEEE Southeastcon '92.

[14]  José-Luis Pérez-de-la-Cruz,et al.  Bayesian student modeling and the problem of parameter specification , 2001, Br. J. Educ. Technol..

[15]  Alexei Makarenko,et al.  Decentralised Data Fusion And Control In Active Sensor Networks , 2004 .

[16]  Ben Upcroft,et al.  Fast re-parameterisation of Gaussian mixture models for robotics applications , 2004 .

[17]  Hugh F. Durrant-Whyte,et al.  A quantitative model for adaptive task allocation in human-computer interfaces , 1993, IEEE Trans. Syst. Man Cybern..

[18]  F. Lantz,et al.  Operator control of shared resources in sensor networks , 2003, Sixth International Conference of Information Fusion, 2003. Proceedings of the.

[19]  Alexei Makarenko,et al.  Implementation of an Indoor Active Sensor Network , 2004, ISER.

[20]  Marek J. Druzdzel,et al.  Building Probabilistic Networks: "Where Do the Numbers Come From?" Guest Editors Introduction , 2000, IEEE Trans. Knowl. Data Eng..