Path planning for localization of an RF source by multiple UAVs on the Crammer-Rao Lower Bound

This paper presents a path planning approach of several UAVs for Radio Frequency (RF) source localization in None Line Of Sight (NLOS) propagation condition using the Received Signal Strength Indication (RSSI). The paths are planned such that the lower bound of standard deviation of localization error, which is equivalent to the Cramer-Rao Lower Bound (CRLB) of the estimation, minimized at any time. Due to the complexity of Jacobin calculation to perform global CRLB optimization, the local values of CRLBs in the current waypoint and next probable waypoints are used in the steepest decent approach to determine the best path. Furthermore, the complexity is reduced by discretizing the space for UAVs to make the computation feasible. The effect of NLOS propagation on the RSSI measurements is simulated by a log-normal distribution and the last estimation of radio source location is used to calculate the local CRLBs. The proposed approach has been simulated and compared with the basic bio-inspired approach of going toward the sensed direction of the source. The result shows better performance than the basic approach.

[1]  Patric Jensfelt,et al.  An optimality analysis of sensor-target geometries for signal strength based localization , 2009, 2009 International Conference on Intelligent Sensors, Sensor Networks and Information Processing (ISSNIP).

[2]  D. Pack,et al.  Comparative study on time-varying target localization methods using multiple unmanned aerial vehicles: Kalman estimation and triangulation techniques , 2005, Proceedings. 2005 IEEE Networking, Sensing and Control, 2005..

[3]  Kai Sheng Zhang,et al.  Improved Localization Algorithm Based on Proportion of Differential RSSI , 2012 .

[4]  Eric Thomas Brewer,et al.  Autonomous Localization of 1/R^2 Sources Using an Aerial Platform , 2009 .

[5]  Maciej Stachura,et al.  Cooperative Target Localization with a Communication-Aware Unmanned Aircraft System , 2010 .

[6]  Kwanho You,et al.  TDoA Based UAV Localization Using Dual-EKF Algorithm , 2009, ICCA 2009.

[7]  K. Doganay.  Online Optimization of Receiver Trajectories for Scan-Based Emitter Localization , 2007, IEEE Transactions on Aerospace and Electronic Systems.

[8]  Neeti Wagle,et al.  A Particle Filter Approach to WiFi Target Localization , 2010 .

[9]  Dezhen Song,et al.  Localization of multiple unknown transient radio sources using multiple paired mobile robots with limited sensing ranges , 2011, 2011 IEEE International Conference on Robotics and Automation.

[10]  Mauro Birattari,et al.  Lazy Learning Meets the Recursive Least Squares Algorithm , 1998, NIPS.

[11]  Dirk van der Walle,et al.  Hierarchical UAV formation control for cooperative surveillance , 2008 .

[12]  Tolga Eren.  Using Angle of Arrival (Bearing) Information for Localization in Robot Networks , 2007 .

[13]  Yilong Lu,et al.  Angle-of-arrival estimation for localization and communication in wireless networks , 2008, 2008 16th European Signal Processing Conference.

[14]  Guoqiang Mao,et al.  Collinearity problems in passive target localization using direction finding sensors , 2009, 2009 International Conference on Intelligent Sensors, Sensor Networks and Information Processing (ISSNIP).

[15]  K. Dogancay.  Optimal Receiver Trajectories for Scan-Based Radar Localization , 2007, 2007 Information, Decision and Control.

[16]  Hadi Moradi,et al.  Aerial localization of an RF source in NLOS condition , 2012, 2012 IEEE International Conference on Robotics and Biomimetics (ROBIO).

[17]  Kutluyil Dogançay.  Optimized path planning for UAVS with AOA/scan based sensors , 2007, 2007 15th European Signal Processing Conference.

[18]  Paul Scerri,et al.  Geolocation of RF Emitters by Many UAVs , 2007 .

[19]  K. Brown,et al.  Autonomous Control of Multiple UAVs for the Passive Location of Radars , 2005, 2005 International Conference on Intelligent Sensors, Sensor Networks and Information Processing.