In this paper, a new identification method of linear discrete-time stochastic systems is proposed. We assume that some entries of the system matrix are unknown and propose a new method which identifies those unknown entries and the state vector of the system simultaneously. The key idea of the proposed method is the use of pseudomeasurement which was first introduced by Whitecombe in tracking of maneuvering targets for obtaining high-accurate estimates of the system states from noisy observation data and has been used by several researchers for various purposes. We utilize the pseudomeasurement as a ficticious observation process on the unknown entries of the system matrix for obtaining better estimates of them. Augmenting the pseudomeasurement with the original observation process, we derive the new identification method by applying the extended Kalman filter.
By restricting `what-to-be-observed' in naturally complex scenes to `easy-to-compute' patterns, this paper presents a stochastic systems approach to the stabilization of early visual perception. In this approach, attentional landmarks are identified with saliency images latently specified in terms of a set of generic parameters. To stabilize the EM algorithm for the estimation of the latent parameters,a truncated version of innovation process is visualized via the multiplexing of residual distribution spanning entire the scene images; resulted visualization, simultaneously, yields an effective cue to confine the images of not-yet-identified landmarks. Based on experimental studies, it has been demonstrated that the focus control algorithm is available in the analysis of the first visit scenes; in this algorithm, it is sufficient to a priori define the attentional landmarks in terms of generic rules governing the perspective of the scene images; in contrast with conventional algorithms, no adhoc pre-processing is required to extract the saliency images under not-yet-identified photographing conditions.
This paper describes a state space model for estimating the location of a mobile wireless client based on sensor data fusion. Data pertaining to the physical position of personal electronic devices or mobile robots are important for information services and robotic applications. We use the strength of signals received by wireless LAN access points to locate mobile clients in multistory buildings. We also use the acceleration of the mobile client as sensor data. The output of an accelerometer is prone to uncertainty; we use a probabilistic model of the location estimated by wireless LAN signals and the acceleration acquired by a sensor to create an observation model. In addition, a simplified measurement method is proposed for use by a fingerprinting-type estimation system based on human footsteps. Experimental results show the feasibility of the proposed method.
In this paper we study the H∞ tracking problems with preview by state feedback for a class of linear continuous-time systems with impulsive effects and stochastic uncertainties on the finite time interval. The systems include linear stochastic discrete-time systems and linear systems with stochastic uncertainties and an input realized through a zero-order hold. In this paper we focus on the direct derivation method of noncausal compensator dynamics from the point of view of dynamics constraint. We derive the pair of noncausal compensator dynamics and impulsive Riccati equations by calculating the stochastic first variation of the performance index under the dynamics constraint.
Sensor fusion is a very powerful tool in various applications. Contrary to conventional methods using a single sensor, sensor fusions with multiple nodes (multiple sensors) give a state estimate from the distinct perspective of the nodes. By letting each node exchange data with its neighbors, a group of nodes form a sensor network, sensor fusion on each node can achieve more accurate estimation.However, in large-scale applications, a centralized sensor fusion which requires data from all nodes at once is not practical. On the other hand, using distributed or decentralized sensor fusion with conventional consensus algorithms which do not consider cross-covariance terms among nodes is not considerably efficient. We propose Ensemble Kalman Filter (EnKF) storing an estimation as a group of distinct particles to determine correlation between estimations. A cross-covariance can be easily obtained by simply approximating a sample second central cross-moment between estimations. Such a property of EnKF enables us to directly update a current estimation with the estimation from nearby nodes, even when the measurement is delayed. The proposed algorithm is proven to be globally optimal and iteratively stable even if redundancy of the estimation between nodes occurs.
The H∞ state feedback control problem for discrete-time single input-delay systems is studied based on the min-max optimization and J-spectral factorization theories. The focus is on efficient construction of the H∞ state feedback law despite the augmented state space due to the input delay. By the min-max optimization approach, the feasibility of the H∞ disturbance attenuation is characterized in terms of a Riccati difference equation, and the stabilizing solution of the standard KYP equation for the augmented system is constructed from that for the delay-free case. The J-spectral factorization approach is outlined based on the author’s previous results. This approach yields another kind of feasibility conditions characterized in terms of a symplectic matrix. The relationship between the first and second approaches is addressed via finite-horizon ℓ2-gain analysis.