# Xbotics Lecture 7: A Robot Manipulation Loop from Target Pose to Grasping, Placement, and Failure Recovery

- Source: Xbotics具身智能实验室 · 微信公众号
- Published: 2026-11-02T04:12:56.000Z
- RoboHot: https://robohot.news/en/items/mep2myh69jxpvfobr8j44177x
- Original: https://mp.weixin.qq.com/s?__biz=MzkyMTc1NTI3Ng==&mid=2247509754&idx=1&sn=6469f21ac1e7c1a4ad4a79e9e033d6eb&chksm=c073b6f30b11f89e0d84be53cab11b3dcce2605158dc7e45d441e809a0e5881bce1d6a577389&scene=126&sessionid=1793622627#rd

## Summary

In Lecture 7 of the "Twenty-One Lectures" series, the Xbotics Embodied Intelligence Lab converts the target object pose obtained in Lecture 6 into gripper actions, linking the grasp-and-place loop with five poses—pre-grasp, grasp, lift, place, and retreat—and a state machine.