基于分层置信规则库的惯导系统性能评估方法

为解决惯导系统（INS）性能评估所面临的高价值样本缺失、评估指标多、系统复杂等问题，提出一种基于分层置信规则库（Hierarchical BRB）的惯导系统性能评估方法。将专家知识与监测数据进行有效融合，提高了惯导系统的性能评估精度。首先，针对惯导系统结构构建分层BRB模型，同时将系统内部器件产生组合误差考虑在模型中。其次，为降低专家知识不确定性对初始模型评估精度的影响，采用基于投影算子的协方差矩阵自适应优化策略（P-CMA-ES）构建优化模型，通过监测数据对模型参数进行微调。最后，以某型捷联惯导系统的性能评估为例，验证了所提方法的有效性。

Performance evaluation method for inertial system based on hierarchical belief rule base

To solve the problems of high-value sample shortage, multiple evaluation indicators, and system complexity in the performance evaluation of the Inertial Navigation System (INS), we propose a performance evaluation method for the INS based on the hierarchical Belief Rule Base (BRB). By integrating the expert knowledge and the monitoring data, the performance evaluation accuracy of the INS is significantly improved. Firstly, a hierarchical BRB model is established for the INS structure, considering the combined errors generated by the internal components of the system. Then, to reduce the influence of expert knowledge uncertainty on the evaluation accuracy of the initial model, the Projection operator-based Covariance Matrix Adaptive optimization Strategy (P-CMA-ES) is employed to construct the optimization model, where the model parameters are fine-tuned using the monitoring data. Finally, a certain type of strapdown INS is taken as an example, verifying the effectiveness of the proposed method.