一种基于预测滤波器的自适应卡尔曼滤波算法

针对系统过程噪声统计特性不确切或未知的条件下,研究了一种基于预测滤波器的自适应卡尔曼滤波算法。由预测滤波器实时估计系统模型误差及其协方差矩阵,再用其修正系统状态预测值及预测误差协方差矩阵,从而自适应调节卡尔曼增益。将该算法应用于弹载SINS/GPS紧耦合组合导航系统并与普通卡尔曼滤波、基于新息的移动开窗自适应卡尔曼滤波进行了对比,仿真结果说明该自适应滤波算法具有更高的可靠性和精度。     

对我国农村环境污染防治对策的系统分析

分析了现阶段我国农村环境污染的主要特点，并对农村环境污染防治的宏观对策进行了系统性分析。由手农村污染严重、防治形势严峻且比城市环境污染情况复杂，在制定农村环境污染防治对策时，需要运用系统工程原理全方位、全过程分析实际情况，制定整体效果最优的系统化对策组合，并注意对策措施的系统性、协调性、连续性和针对性。     

A note on a paper by houston and huffman

This paper calls attention to an experimental design which enhances the efficiency of the technique described by B. F. Houston and R. A. Huffman [3]. This alternative design is a modification of the simplex lattice design developed by Scheffé [6] for the mixture problem. While not possessing the optimal statistical properties of the central composite design proposed in [3], such as minimum bias and minimum variance, this modified design is optimum with respect to the number of design points employed; hence, it is the minimum cost design. Further, the design proposed in this note is excellent for use in multiple-block experimentation, which is very often required with constrained systems.     

An optimum t-test for the scale parameter of an extreme-value distribution

A Student's t-test proposed by Ogawa is considered for the hypothesis Ho: σ=σo against the alternative hypothesis H₁: σ ≠ σo, where σ is the scale parameter of the Extremevalue distribution of smallest values with known location parameter μ. The test is based on a few sample quantiles chosen from a large sample so as to give asymptotically maximum power to the test when the number of sample quantiles is fixed. A table which facilitates the computation of the test statistic is given. Several schemes for determining the ranks of the sample quantiles by the optimal spacings are compared and the effect of the bias of the estimate of σ on the test is investigated through a Monte Carlo study.     

Inventory control of by-products

The system to be controlled produces n products simultaneously in fixed proportions every time it is activated. Demands for the products in any period are components of an n dimensional vector random variable with known distribution function. Cases of excess demands backlogged and excess demands lost are considered. In the former the notion of k convexity can be generalized to guarantee relatively simple form for the optimal policy in an n decision problem. In the latter, this generalization was not successful although when there is no setup cost, a convexity argument can be used to show that the optimal policy has a simple form.