首页|无人自主系统能力边界参数自适应判别方法

无人自主系统能力边界参数自适应判别方法

扫码查看
原文链接
  • NETL
  • NSTL
  • 万方数据
为有效应对仿真测试面临的维度灾难问题,降低传统全参数空间遍历中所需的仿真次数,需要获取针对性的仿真数据以准确反映实验数据建模特征,以较少的仿真次数获得信息量丰富且代表原始数据特征的样本.提出一种面向无人自主系统能力边界参数自适应判别的数字化仿真测试模型,采用多权重结构的佳点集进行初始构建,结合自适应核函数边界点判别算法,通过高斯过程回归对模型进行迭代优化,自适应地判别无人自主系统的能力边界.实验结果表明:该方法能够降低建模所需数据量,提高自适应参数边界判别的效率,为提升智能无人系统试验的效率提供了高效途径.
Adaptive Recognition Method of Capability Boundary Parameters for Unmanned Autonomous Systems
To effectively cope with the dimension curse in simulation testing and reduce the number of simulations times needed in the traditional full-space parameter traversal,it is necessary to obtain specific simulation data to accurately reflect the modeling characteristics of the test data to obtain the informative and representative samples of the original data with a smaller number of simulations.A digital simulation test model for adaptive recognition;/of capability boundary parameters for UAS is proposed.The model is initially constructed with a good point set with a multi-weight structure;In combination with an adaptive kernel function boundary point recognition,the model is iteratively optimized by Gaussian process regression,so as to adaptively detect the capability boundary of UAS.The experimental results show that the method can reduce the amount of data required for modeling and improve the efficiency of adaptive parameter boundary recognition,which provides an approach to enhance the efficiency of intelligent UAS testing.

unmanned autonomous systemadaptive recognition of boundary parametersgaussian process regression modeladaptive kernel functiongood points setsimulation testing

李锦文、王鹏、潘优美、惠新遥

展开 >

中国科学院软件研究所,北京 100190

中国科学院大学,北京 100049

无人自主系统 边界参数自适应判别 高斯过程回归模型 自适应核函数 佳点集

国家自然科学基金面上项目中国科学院软件研究所基础研究重点项目

61973313ISCAS-JCZD-202309

2024

10.16182/j.issn1004731x.joss.23-0775

系统仿真学报
北京仿真中心 中国系统仿真学会

系统仿真学报

CSTPCD北大核心
影响因子：0.551
ISSN：1004-731X
年,卷(期)：2024.36(10)
  • 2