知识卡片

普通读书笔记卡 · 9201.a · 生成模型:GPT-6

创新是观测与预测观测之差

理解等级:中级

滤波把先验状态通过观测模型映到测量空间,再以实际观测减预测得到创新。它不是未知真值与估计的直接误差,只有模型和数据关联正确时才能作为合理校正依据;量纲和角度周期也须一致。

这张卡片看不懂?

已有反馈 · 0 条

    还没有人提交反馈。

    在整书笔记中的位置

    原书出处 · 1 处

    • · 查看原书依据

      Although we have introduced the problem above as fusing two observations of the same quantity and weighting them by their variance, we can also interpret the equation above as an update step that calculates a new estimate of an observation based on its old estimate and a new measurement. Specifically, we can interpret the expression from equation (15.18) as the difference between what the robot

    参考来源

    - 位置:《Introduction to Autonomous Robots Mechanisms, Sensors, Actuators, and Algorithms (Nikolaus Correll, Bradley Hayes etc.)》第15章《不确定性与误差传播》(来源项目:a19c7933-4eda-4176-aebd-30fe7e1ad292;源文件:extracted/Introduction to Autonomous Robots Mechanisms, Sensors, Actuators, and Algorithms (Nikolaus Correll, Bradley Hayes etc.)/_epub-src/OEBPS/xhtml/chapter_15.xhtml) - 结论依据:摘录中关于创新是观测与预测观测之差的定义、比较或条件支撑卡片;原书表述不严谨的范围已在覆盖记录逐项限定,不把案例结果外推为普遍结论。 - 原始内容:Although we have introduced the problem above as fusing two observations of the same quantity and weighting them by their variance, we can also interpret the equation above as an update step that calculates a new estimate of an observation based on its old estimate and a new measurement. Specifically, we can interpret the expression from equation (15.18) as the difference between what the robot