Unverified70% confidenceFactExact time
卡尔曼滤波是一种递归贝叶斯估计算法，在线性高斯假设下可给出最优闭合解
2
Sources
70%
Confidence
Long-term
Relevance
7/17/2026
First Seen
Sources
Related Claims
Unverified卡尔曼滤波是线性高斯假设下的贝叶斯最优估计器，其预测-更新两步迭代对应先验与后验概率的递推计算84% similarUnverified卡尔曼滤波是一种递归贝叶斯估计算法，能够在含噪声的传感器数据中实时估计系统状态74% similarUnverified卡尔曼滤波核心思想是将预测值与观测值按各自的不确定性（协方差）加权融合，不确定性越低的来源获得越高权重70% similarUnverifiedFrom a Bayesian inference perspective, Kalman filtering is the exact computation of the Bayesian posterior distribution under Gaussian linear assumptions, with the prediction step propagating the prior and the update step computing the posterior using observation data68% similarUnverified贝叶斯优化通过构建目标函数的概率代理模型（通常是高斯过程），利用采集函数（如Expected Improvement或Upper Confidence Bound）决定下一个采样点65% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/544385API
curl https://kongchang.com/api/v1/knowledge/claims/544385MCP
get_claim(id=544385)