Target Localization Using Multiple UAVs with Sensor Fusion via Sigma-Point Kalman Filtering

Abstract

This paper proposes a sensor-fusion methodology based on sigma-point Kalman filtering (SPKF) techniques, where the SPKF is applied to the particular problem of localizing a mobile target using uncertain nonlinear measurements relating to target positions. This SPKF-based sensor fusion is part of our larger research program in which the particular target localization solution of interest must address a number of challenging requirements: (a) covert/passive sensing means must be used; (b) the dynamic characteristics of the target are unknown; (c) the target is episodically mobile; and (d) the target is intermittently occluded from one or more sensors. Our research efforts seek to meet these requirements using a flight of small autonomous unmanned aerial vehicles (UAVs) with heterogeneous sensing capabilities. Each UAV carries one or more sensors, which include: radio frequency sensors, infrared sensors, and image sensors. Due to cost considerations and to payload constraints, it is not desirable for every UAV to have a full complement of sensing capability. In our implementation, the UAVs first search a predefined region for targets. Upon target detection, a small formation of UAVs comprising complementary heterogeneous sensors is autonomously assembled to localize the target. The UAVs then cooperatively locate the target by combining the sensor information collected by heterogeneous sensors onboard the UAVs. The output of the localization process is an estimate of the target's position and velocity as well as error bounds on that estimate. The primary purpose of this paper is to describe SPKF-based sensor-fusion for target localization using sensors that only provide direction-of-arrival information. Results of simulations are compared to prior work where linear Kalman filters were used to localize stationary targets. The SPKF gives much more accurate and consistent results, due mostly to its better handling of the nonlinear measurement relationship. Results are also given for the application of SPKF to mobile targets.

Negotiating between communication and cooperation for multiple unmanned aerial vehicles

“…We desire our system of multiple cooperative UAVs to be autonomous, distributive, and scalable. To that end, we have developed distributive control technologies, local sensor fusion capabilities [2,4], and a unique operating system for unmanned systems. For the communication network, we adopted the IEEE 802.11g multi-hop mesh network system.…”

Formation Control Technology of Fixed-Wing UAV Swarm Based on Distributed Ad Hoc Network

“…The fixed-wing UAV swarm has essential application prospects and has become a current research hotspot. Completing combat missions such as coordinated reconnaissance, early warning, strike, and evaluation in the military field; and realizing disaster emergency, geological survey, and pesticide-spraying tasks in the civilian field [1][2][3][4]. Formation control is one of the key issues to achieve UAV swarm flight [5,6].…”

Cooperative Search and Localization of Ground Moving Targets by a Group of UAVs Considering Fuel Constraint

“…Search, detection, and localization of targets can be done by a group of UAVs, equipped with di erent types of sensors such as downward looking millimeter wave radar [17], vision based [18][19][20], infrared (IR) [21], Radio Frequency (RF) [22][23][24][25][26][27][28][29][30][31][32], etc. In addition, a combination of two or more types of sensors can be utilized [33,34]. Since RF sensors are mentioned in this article, in the following, the research studies contain at least one RF sensor.…”

“…Although each UAV can localize the target separately, it was shown in [32] that the combination of several UAVs' measurements with di erent levels of accuracy improved the precision of localization. In [33], Sigma-Point Kalman Filter (SPKF) was utilized for cooperative localization of a moving target, when each UAV utilized one or more types of sensors between RF, IR, and image sensors.…”