知识卡片

普通读书笔记卡 · 9193.f · 生成模型:GPT-6

随机规划比较需要重复实验

理解等级:中级

单次运行受采样随机性影响,不能只用一次成功速度比较算法。应固定相同任务和有效性检查,报告多次成功率、时间与路径成本分布,并区分首次可行解和后续改进;基础RRT也不保证练习题所说“最短”。

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

      b) What do you need to take into account when using randomized planners? Is comparing single-shot experiments sufficient? 15. Implement a coverage path planner for a single robot on a grid map using depth-first search. Evaluate the amount of redundancy for different starting locations.

    参考来源

    - 位置:《Introduction to Autonomous Robots Mechanisms, Sensors, Actuators, and Algorithms (Nikolaus Correll, Bradley Hayes etc.)》第13章《路径规划》(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_13.xhtml) - 结论依据:摘录中关于随机规划比较需要重复实验的定义、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:b) What do you need to take into account when using randomized planners? Is comparing single-shot experiments sufficient? 15. Implement a coverage path planner for a single robot on a grid map using depth-first search. Evaluate the amount of redundancy for different starting locations.