Only a few publications exist at present on heterogeneous track-to-track fusion (T2TF). A common limitation of current work on heterogeneous T2TF is that the cross-covariance due to common process noise cannot be computed. This is due to the fact that two local trackers use different dynamic models and hence it is difficult to account for the common process noise. We consider a heterogeneous T2TF problem in 3D using a passive infrared search and track (IRST) sensor and an active air moving target indicator (AMTI) radar with the nearly constant velocity (NCV) motion of the target.

Accurate 3D position and velocity estimates of an aircraft are important for air traffic control (ATC) systems. An ATC 2D radar measures the slant range and azimuth of an aircraft. Thus, a single measurement from a 2D radar is not sufficient to calculate the 3D position of an aircraft. Previous researchers have used the multiple model based heightparametrized (HP) extended Kalman filter with Cartesian state vector (HP-CEKF) with one or two 2D radars for an aircraft with nearly constant velocity and altitude. However, the filter initialization algorithms contain errors. In this paper, in addition to the HP-CEKF, we present the HP Cartesian unscented Kalman filter (HP-CUKF) and HP Cartesian cubature Kalman filter (HPCCKF). We also present two new nonlinear filters for the tworadar problem. The first filter uses modified spherical coordinates based HP UKF (HP-MSCUKF) where the range and azimuth are components of the target state. The second filter uses a cubature Kalman filter with filter initialization by the bias-compensated pseudo linear estimator. We consider the climbing motion of an aircraft with nearly constant climbing rate which has not been studied before.

The use of soft computing techniques such as NNs, GAs, etc. in the financial market has become one of the most exciting and promising application areas. We propose a decision support system (DSS) for dealing in the TOPIX (Tokyo Stock Exchange Prices Indexes), which utilizes neural networks and genetic algorithms. In the proposed system, the neural network is utilized in order to make a forecast of the TOPIX four weeks in the future. The genetic algorithm is utilized in order to find an effective way of dealing. Several computer simulations have been carried out in order to compare the proposed DSS with the other approaches such as the DSS using traditional technical analysis and a buy-and-hold method. These simulations confirm the effectiveness of the proposed DSS.

This paper presents a novel slider crank power take-off system and wave tracking control methodology for efficiently converting the energy of open ocean waves into electrical energy. The Wave Energy Converter (WEC) and Power Take-off System (PTOS) are modeled and a non-parametric control strategy to maximize energy extraction under regular sinusoidal wave condition is introduced. Energy extraction results from simulation are compared with theoretical optimums, and a number of factors that influence energy extraction are discussed. The study shows that a suboptimal wave energy conversion technique can be achieved.