← 返回刚才阅读的位置

知识卡片

普通读书笔记卡 · 9163.c · 生成模型:GPT-6

分裂拟线利用观测的空间顺序

理解等级:中级

先拟合一组有序扫描点,找最大残差点,超过阈值则在该处拆分并递归拟合。顺序决定如何分段,不能直接对任意无序点云按数组索引分裂;阈值过小或孤立异常会制造许多伪线段。

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

      It is often unclear how many lines there are and where a line starts and ends. This creates a challenge for the matching-and-estimation strategy discussed previously. Looking through the camera, for example, we will see vertical lines corresponding to wall corners and horizontal lines that correspond to wall-floor intersections and the horizon; using a distance sensor, the robot might detect a corner. We therefore need

    参考来源

    - 位置:《Introduction to Autonomous Robots Mechanisms, Sensors, Actuators, and Algorithms (Nikolaus Correll, Bradley Hayes etc.)》第09章《特征提取》(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_9.xhtml) - 结论依据:摘录中关于分裂拟线利用观测的空间顺序的定义、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:It is often unclear how many lines there are and where a line starts and ends. This creates a challenge for the matching-and-estimation strategy discussed previously. Looking through the camera, for example, we will see vertical lines corresponding to wall corners and horizontal lines that correspond to wall-floor intersections and the horizon; using a distance sensor, the robot might detect a corner. We therefore need