Get the latest tech news
ReKep: Spatio-Temporal Reasoning of Relational Keypoint Constraints for Robots
ReKep: Spatio-Temporal Reasoning of Relational Keypoint Constraints for Robotic Manipulation
However, it remains unclear how to formulate the constraints such that they are 1) versatile to diverse tasks, 2) free of manual labeling, and 3) optimizable by off-the-shelf solvers to produce robot actions in real-time. Finally, a constrained optimization solver is used to obtain a dense sequence of end-effector actions in SE(3), subject to the generated constraints. We would like to thank the anonymous reviewers, Albert Wu, Yifan Hou, Adrien Gaidon, Adam Harley, Christopher Agia, Edward Schmerling, Marco Pavone, Yunfan Jiang, Yixuan Wang, Sirui Chen, Chengshu Li, Josiah Wong, Wensi Ai, Weiyu Liu, Mengdi Xu, Yihe Tang, Chelsea Ye, Mijiu Mili, and the members of the Stanford Vision and Learning Lab for fruitful discussions, helps on experiments, and support.
Or read this on Hacker News
