TY - CONF

T1 - Multiple Objective D-Optimal Sensor Management for Group Target Tracking

AU - Dolia, A.

AU - Mihaylova, L.

AU - Page, S.

AU - White, N.

AU - Harris, C.

AU - Bull, D.

N1 - "©2006 IEEE. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists, or to reuse any copyrighted component of this work in other works must be obtained from the IEEE." "This material is presented to ensure timely dissemination of scholarly and technical work. Copyright and all rights therein are retained by authors or by other copyright holders. All persons copying this information are expected to adhere to the terms and constraints invoked by each author's copyright. In most cases, these works may not be reposted without the explicit permission of the copyright holder."

PY - 2006

Y1 - 2006

N2 - A group target is moving in an area well covered by a network of passive sensor nods with known positions. Additionally, there are a number of mobile robots with active sensors. In order to obtain a robust estimate of the position of the target and decrease the amount of energy spent on active sensing and communications by the sensor network and the mobile robots a sensor management system optimises the spatial configuration of the mobile robots over time. A tracking algorithm predicts the position of the target over multiple steps. An estimate for the tracking accuracy for each possible sensor action is calculated based on a function of the expected resulting posterior inverse covariance (information) matrix given the position of the nodes of the sensor networks and the feasible position of the mobile robots in future time instants. We propose a novel approach for active sensor management that combines the Rao-Blackwellised particle filter/predictor and multi-objective D-optimal optimisation. The designed decentralised Rao-Blackwellised particle filter (RBPF) is composed of two parts: a decentralised Information or Kalman filter and a particle filter (PF). The sensor management framework that is based on the generalised D-optimal optimisation with slack variables is proposed.

AB - A group target is moving in an area well covered by a network of passive sensor nods with known positions. Additionally, there are a number of mobile robots with active sensors. In order to obtain a robust estimate of the position of the target and decrease the amount of energy spent on active sensing and communications by the sensor network and the mobile robots a sensor management system optimises the spatial configuration of the mobile robots over time. A tracking algorithm predicts the position of the target over multiple steps. An estimate for the tracking accuracy for each possible sensor action is calculated based on a function of the expected resulting posterior inverse covariance (information) matrix given the position of the nodes of the sensor networks and the feasible position of the mobile robots in future time instants. We propose a novel approach for active sensor management that combines the Rao-Blackwellised particle filter/predictor and multi-objective D-optimal optimisation. The designed decentralised Rao-Blackwellised particle filter (RBPF) is composed of two parts: a decentralised Information or Kalman filter and a particle filter (PF). The sensor management framework that is based on the generalised D-optimal optimisation with slack variables is proposed.

KW - D-optimal optimisation

KW - Rao-Blackwellisation

KW - particle filtering

KW - DCS-publications-id

KW - inproc-459

KW - DCS-publications-credits

KW - dsp

KW - DCS-publications-personnel-id

KW - 121

M3 - Conference paper

T2 - International Conf. on Robotics and Automation

Y2 - 15 May 2006 through 19 May 2006

ER -