• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文会议>Saint Petersburg International Conference on Integrated Navigation Systems; 20030526-28; St.Petersburg(RU) >SYNTHESIS AND RESEARCH INTO THE ACCURACY OF THE NONLINEAR AND NEURAL NETWORK FILTERS FOR ESTIMATING THE PARAMETERS OF MOVING OBJECTS
【24h】

SYNTHESIS AND RESEARCH INTO THE ACCURACY OF THE NONLINEAR AND NEURAL NETWORK FILTERS FOR ESTIMATING THE PARAMETERS OF MOVING OBJECTS

机译：估计运动物体参数的非线性和神经网络滤波器的合成与研究

获取原文
获取原文并翻译 | 示例
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

We offer a solution for the problem of the optimum and adaptive nonlinear discrete filtering of the parameters of moving objects based on the Bayesian approach and neural networks technology. We use nonlinear models of monitoring for the Radar measuring the range, azimuth, and angle of fire as well as the range and two direction cosines or bearing for range and/or bearing measurement methods. We synthesize nonlinear filters for 3 models of movement: an average fixed object at random speed, an even rectilinear driving at random acceleration, and a maneuvering object. We look into the possibilities of reducing the order of magnitude of the multiple integrals for the nonlinear filters, synthesized on the Bayesian basis. We studied the parallel programming of the nonlinear filters at a supercomputer. On the basis of the neural networks we solve the problem of nonlinear and linear filtering of the parameters of the moving objects. We discuss the application of different architectures of the neural networks and the technology of their usage. We offer the results of the comparison between the accuracy of the synthesized filters: nonlinear Bayesian, neural network and Kalman.
机译：我们基于贝叶斯方法和神经网络技术为运动对象的参数进行最优和自适应非线性离散滤波的问题提供了解决方案。我们使用用于监视雷达的非线性模型来测量距离，方位角和射击角度，以及用于距离和/或方位角测量方法的距离和两个方向余弦或方位角。我们为3种运动模型合成了非线性滤波器：随机速度下的平均固定物体，随机加速度下的均匀直线驱动以及机动物体。我们研究了减少在贝叶斯基础上合成的非线性滤波器的多个积分的数量级的可能性。我们在超级计算机上研究了非线性滤波器的并行编程。在神经网络的基础上，我们解决了运动对象参数的非线性和线性滤波问题。我们讨论了神经网络的不同体系结构的应用及其使用技术。我们提供了合成滤波器的精度之间的比较结果：非线性贝叶斯，神经网络和卡尔曼。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号