知识卡片
高斯噪声是模型选择而非必然
高斯便于传播与融合,有些小扰动的叠加也能支持这一近似,但足够多样本不自动让任意原始噪声变成高斯。偏置、离群值、打滑和多种模式可能违反假设,中央极限定理也有条件,需用观测核对。
这张卡片看不懂?
这张卡片已通过人工审查,不再接受失败反馈。
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
Experiments in robotics rarely involve true statistical randomness because of their scale and design. Instead, there are two primary sources of uncertainty: sensors and physical interactions. Sensors are intrinsically noisy because of the physical phenomena associated with them. These sources of uncertainty are often modeled using Gaussian (or “normal”) probability distribution functions, as they accurately model measurement uncertainty at large samples, per the central limit
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:摘录中关于高斯噪声是模型选择而非必然的定义、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:Experiments in robotics rarely involve true statistical randomness because of their scale and design. Instead, there are two primary sources of uncertainty: sensors and physical interactions. Sensors are intrinsically noisy because of the physical phenomena associated with them. These sources of uncertainty are often modeled using Gaussian (or “normal”) probability distribution functions, as they accurately model measurement uncertainty at large samples, per the central limit