Search

Coning effects must be accurately compensated in the attitude updating of the classical strapdown inertial navigation integration algorithm. These effects are not involved in the velocity updating, though the attitude change is being taken into it. The novel velocity error term induced by the coning correction is studied in this paper. These effects as the gyro angular increment data have the influence on the velocity updating in the same way. For the assumed linear ramp model, a two-sample compensation algorithm for the novel velocity error term is deduced in detail by using the two successive integrated angular rate increment and specific force increment measurements. The weighting coefficients are then optimized for best average performance in a specific environment; i. e. the body is exposed to a coning angular vibration about the x axis of the body frame combined with a linear vibration at the same frequency and in phase along the z axis. The simulation results validate that the two-sample optimization algorithm can compensate the integrated term of the novel velocity error effectively. [ABSTRACT FROM AUTHOR]

In traditional Kalman filtering algorithms, optimization estimation cannot be achieved for practical problems because the dynamics system model and noises are unknown. Firstly the impacts of various error sources and interval matrix operations in the traditional Kalman filter are analyzed. The analysis shows that the interval operation can ensure the completeness of set mapping but can not be the optimization. Based on the analysis, a new interval Kalman filtering algorithm is proposed in this paper. In the algorithm, various error sources are reduced to a priori estimation interval and a posterior estimation interval and the interval intersection set operation is adopted to achieve the optimization. This algorithm has comparable computational complexities with the traditional one. Simulations show that the proposed algorithm can achieve a 2dB gain over the traditional Kalman filter in high dynamic navigation and positioning with time-varying noises. [ABSTRACT FROM AUTHOR]

Under the condition of a short sampling period, limited resource and computational capacity, a method for designing reduced-order fault detection filtering algorithm is proposed in this paper. Small perturbation theory and a Taylor series expansion about the balance point are used to linearize the nonlinear dynamics equations of space vehicles. Then the reduced-order fault detection filter is designed for the linear state-space equation. The problem of feasible solution of the reduced-order fault detection filtering algorithm is transformed into a nonconvex optimization problem expressed in terms of linear matrix inequalities. The local optimal solution of the nonconvex optimization is achieved by an alternate mapping algorithm. The reduced-order fault detection filtering algorithm is obtained under the condition that the feasible solution exists. Finally, the reduced-order fault detection algorithm is applied to the fault detection of the yaw control system in a small-scale autonomous helicopter. Simulation results are presented to show effectiveness of the proposed approach. [ABSTRACT FROM AUTHOR]

Low-thrust phasing maneuver design of Earth circle-orbit satellites is researched in this paper. A semi-analytical phasing parameter analysis method is proposed, and based on that an efficiently accurate design method is developed. First, according to the influence law of thrust direction on phase change rate, functional relationship between phasing parameters are derived by using orbit averaging technique and thrust direction assumption. The key phasing parameters can be obtained by using simple differential correction. Then, the fuel-optimal phasing orbit design under the condition of complex constraints based on a perturbation model is transformed into a multi-parameter optimization problem by discretizing control law. With the initial guess obtained by the analysis method, the problem can be solved rapidly by using a sequential quadratic programming algorithm. Finally, the simulation results of Geosynchronous orbit low-thrust phasing transfer is presented, it is shown that feasible initial guess can be provided with the analytical method for phasing orbit design, and by using the robust design method, low-thrust phasing trajectory design problem can be solved effectively under the perturbation model. [ABSTRACT FROM AUTHOR]

Signals of interest (SOIs) extraction is a vital issue in the field of communication signal processing. In this paper, the cyclostationary property of communication signals is incorporated into the constrained ICA (cICA) framework to form a new constrained optimization problem. The approach incorporates cyclostationarity constraints in the ICA process to perform separation and selection of the SOIs simultaneously. Numerical simulations demonstrate the effectiveness and higher performance of the proposed algorithm compared with the conventional ICA algorithms such as FastICA. [ABSTRACT FROM AUTHOR]