Abstract
This paper derives a square-root information-type filtering algorithm for nonlinear multi-sensor fusion problems using the cubature Kalman filter theory. The resulting filter is called the square-root cubature Information filter (SCIF). The SCIF propagates the square-root information matrices derived from numerically stable matrix operations and is therefore numerically robust. The SCIF is applied to a highly maneuvering target tracking problem in a distributed sensor network with feedback. The SCIF’s performance is finally compared with the regular cubature information filter and the traditional extended information filter. The results, presented herein, indicate that the SCIF is the most reliable of all three filters and yields a more accurate estimate than the extended information filter.
Highlights
Sensor fusion is generally defined as the use of techniques that combine data from multiple sensors such that the resulting information is more accurate and more reliable than that from a single sensor
For multiple sensor fusion in a linear Gaussian environment, the information filter, which can be considered as the dual of the Kalman filter, has been a viable solution [1,2,3]
The Kalman filter and the information filter are algebraically equivalent, the Kalman filter propagates a state vector and its error covariance whereas the information filter uses an information vector and an information matrix. This difference makes the information filter to be superior to the Kalman filter in fusion problems because computations are straight-forward and simple
Summary
Sensor fusion is generally defined as the use of techniques that combine data from multiple sensors such that the resulting information is more accurate and more reliable than that from a single sensor. For multiple sensor fusion in a linear Gaussian environment, the information filter, which can be considered as the dual of the Kalman filter, has been a viable solution [1,2,3]. The Kalman filter and the information filter are algebraically equivalent, the Kalman filter propagates a state vector and its error covariance whereas the information filter uses an information vector and an information matrix This difference makes the information filter to be superior to the Kalman filter in fusion problems because computations are straight-forward and simple. Vercauteren et al derived the sigma-point information filter using the statistical linear regression theory and the unscented transformation [4] whereas Kim et al derived a similar set of steps using a minimum mean squared-error criterion [5].
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have