Event
Xbotics Lecture 6 releases tutorial on robot pose estimation and grasping
2 reports · 1 sources
Yu Fan of the Xbotics Embodied Intelligence Lab released Lecture 6 of the "Twenty-One Lectures" series, explaining the complete perception pipeline from RGB-D observations to object poses and then to grasp poses, with a modular pipeline approach as the main thread. The experiment code is located in the Visual_Perception directory of the GitHub repository Xbotics-Embodied-AI-club/Xbotics-Embodied-AI-Handbook, and a sample input can be downloaded as a frame of 640×480 RGB-D data from the University of Washington RGB-D Object Dataset. The subsequently released second part further explains how to generate grasp candidates from object poses and complete the coordinate transformation from the camera frame to the robot base frame. The article breaks down the five elements of grasp point, approach direction, opening width, safety offset, and orientation matching, compares the two hand-eye calibration topologies of Eye-in-Hand and Eye-to-Hand, and provides reachability verification, observation degradation fallback, and perception failure troubleshooting procedures.
Developments
- Xbotics Lecture 6, Part 2: The full pipeline from object pose to grasp pose, with hand-eye calibration practice · 1 reports
- Xbotics Lecture 6: The Complete Pipeline from Robot Perception to Pose Estimation · 1 reports
Report timeline
- Xbotics Lecture 6, Part 2: The full pipeline from object pose to grasp pose, with hand-eye calibration practice
The Xbotics Embodied Intelligence Lab has released Part 2 of Lecture 6 in its "Twenty-One Lectures" series, explaining how to generate grasp candidates from object poses and complete the coordinate transformation from the camera frame to the robot base frame. The article breaks down five elements: grasp point, approach direction, opening width, safety offset, and orientation matching. It compares the Eye-in-Hand and Eye-to-Hand hand-eye calibration topologies and provides procedures for reachability checks, degradation under observation failure, and troubleshooting perception failures.
Xbotics具身智能实验室 · 微信公众号 ·
- Xbotics Lecture 6: The Complete Pipeline from Robot Perception to Pose Estimation
The Xbotics Embodied AI Lab has released Lecture 6 of its 21-lecture series, explaining the complete perception pipeline from RGB-D observations to grasp poses, with accompanying code in the GitHub repository Xbotics-Embodied-AI-club/Xbotics-Embodied-AI-Handbook.
Xbotics具身智能实验室 · 微信公众号 ·