知识卡片
Markov状态要足够完整
递推假设下一状态在给定当前状态和动作后不再依赖更早历史,观测在给定当前状态后也不再依赖其他历史。若偏置或动态障碍有持续影响,可能需扩充状态或改模型,不能仅因使用滤波器就认为这些假设自然成立。
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
- · 定位
原书出处 · 1 处
- · 查看原书依据
By leveraging the Markov assumption, that our current state only depends on our previous state xt−1 and action ut, we can greatly simplify the computation required as such: For example, if we wanted to calculate the probability of an observation zt, we know that the only term that actually matters is the robot’s current state (since the other terms don’t affect what sensor readings we’d
参考来源
- 位置:《Introduction to Autonomous Robots Mechanisms, Sensors, Actuators, and Algorithms (Nikolaus Correll, Bradley Hayes etc.)》第16章《定位》(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_16.xhtml)
- 结论依据:摘录中关于Markov状态要足够完整的定义、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:By leveraging the Markov assumption, that our current state only depends on our previous state xt−1 and action ut, we can greatly simplify the computation required as such: For example, if we wanted to calculate the probability of an observation zt, we know that the only term that actually matters is the robot’s current state (since the other terms don’t affect what sensor readings we’d