Incorporation of time delayed measurements in a discrete-time Kalman filter

Abstract

In many practical systems there is a delay in some of the sensor devices, for instance vision measurements that m a y have a long processing time. How to fuse these measurements in a Kalman filter is not a trivial problem if the computational delay is critical. Depending on how much time there is at hand, the designer has to make trade offs between optimality and computational burden of the filter. In this paper various methods in the literature along with a new method proposed by the authors will be presented and compared. The nem method is based on "extrapolaiing" the measurem.ent to present time using past and present estimates of the Kalman filter and calculating an optimal gain for this extrapolated m.easurement.

Quadrature filters for one-step randomly delayed measurements

“…In the literature, the problem appears with different names such as out of sequence measurement (OOSM) [16], filtering with random sample delay [17], filtering with random time delayed measurements [18] etc.…”

“…However, the same is not true for randomly delayed measurement problems except a few notable publications on linear systems [18][19][20][21], and nonlinear systems [22][23][24]. The literature on the described problem began with the work of Ray et al, where the authors developed a randomly delayed filtering method for the linear systems.…”

Nonlinear Observers based on the Functional Mockup Interface with Applications to Electric Vehicles

“…The presented example of a battery state estimation and its results make us confident that this framework can be used for many control system tasks in the future, especially in the ROboMObil project. Furthermore, the estimation algorithms will be extended to handle constraints in a recursive way, [Sim09] [Kan08] and to take "out of sequence measurements" into account, [Lar98] [Mer04] .…”

Visual Target Tracking and Relative Navigation for Unmanned Aerial Vehicles in a GPS-Denied Environment

“…That is, the visual measurement arriving at time k was extracted fr captured at time s . It cannot be fused using the normal Kalman filter equatio Thus, the filter requires some modifications in the structure [10]. …”

“…That is, the visual measurement arriving at time k was extracted fr captured at time s . It cannot be fused using the normal Kalman filter equatio Thus, the filter requires some modifications in the structure [10]. The filter without update of delayed measurement Typically, the acceleration and attitude measurements from IMU are delivered at a higher frequency than the pixel measurements from the image processing system.…”