Pinned Tweet
馃殌 Introducing N2M!
In mobile manipulation, the performance of a manipulation policy is very sensitive to the robot鈥檚 initial pose. N2M guides the robot to a suitable pose for executing the manipulation policy.
N2M comes with 5 key features - Check them out in the posts below!
Details of our approach, along with further experiments and analysis, are presented below.
馃殌 Project page: clvrai.github.io/N2M/
馃摑 Paper: arxiv.org/abs/2509.18671
馃捇 Code: github.com/clvrai/N2M/
1
3
139
Huge thanks to my co-author @Hyunjun_Lee1013, and valuable guidance from @JosephLim_AI, as well as to everyone who supported us throughout this project. @jng_jun, Junseung_Lee , @ksunw0209, @doohyun22, @MinhoHeo, RAINBOW_ROBOTICS馃槉
3
220
Very practical work馃コ馃コ Traditional remote controllers are not intuitive and hard to make robots carry out human-like movements. VR-based demonstrations will undoubtedly help collect mobile manipulation data in a more comfortable and easier way.
馃ソ Want to use your new Apple Vision Pro to control your robot? Want to record how you navigate / manipulate the world to train a policy?
I developed an app for VisionOS that can stream your head / wrist / finger movements over WiFi, which you can subscribe on any machines using a simple python library. 馃У[1/6]
222

