Jason Kim

13.6K posts

Jason Kim banner
Jason Kim

Jason Kim

@jsonbot117

🔥#Legged_Robot_SLAM, State Estimation & Locomotion🔥Open Source Robotics / Now at URobotics, MS. EECS

Katılım Aralık 2016
950 Takip Edilen1.5K Takipçiler
Sabitlenmiş Tweet
Jason Kim
Jason Kim@jsonbot117·
Huge thanks to @spaceandtech_ for featuring E-dongkit! As a research member Spatial Intelligence and System Integration, I’m glad to see our progress being shared. Our team is dedicated to achieving Robust Field Autonomy for legged robots—ensuring they can navigate the most unpredictable, off-road environments with precision. Building robots that can truly "see" and "act" in the wild is a massive challenge, but it's one we tackle with passion every day. Stay tuned as we continue to push the boundaries of what legged robots can do!
Space and Technology@spaceandtech_

South Korea-based URobotics unveiled the E-Dong-Kit, a module for quick robot integration. It enables advanced navigation and stable movement in real-world environments. In demos, humanoid robots move smoothly and run alongside humans, showing practical, human-friendly mobility.

English
2
0
6
1.3K
Jason Kim
Jason Kim@jsonbot117·
@ratygal 케이크만 먹음 ㅠㅠ 하몽은 입도못댐
한국어
1
0
1
16
Jason Kim retweetledi
Dimitrios Kanoulas
Dimitrios Kanoulas@dkanou·
Follow Anything: Robots shouldn’t just follow humans—they should follow everything. Our #ICRA2026 paper introduces: Category-free leader tracking, Goal-aware adaptation, Graph-based planning 📊 +75.1% success💥 −65.1% collisions #Robotics #AI #Autonomy #LeggedRobots
English
1
1
3
664
Jason Kim retweetledi
Claudia Cuttano
Claudia Cuttano@ClaudiaCuttano·
#CVPR2026 Oral ✨ A tale of a failed experiment: what if you fine-tune DINOv2 on sparse keypoints, beat every benchmark, only to discover it performs worse than the original frozen model on novel keypoints? 🚀MARCO closes this gap: a unified model for generalisable correspondences github.com/visinf/MARCO
Claudia Cuttano tweet media
English
7
48
415
23K
Jason Kim retweetledi
Zhengtong Xu
Zhengtong Xu@XuZhengtong·
🤖Manipulation is fundamentally about how robots apply contact and force to the physical world. 💡In our #ICRA2026 paper, Learning Unified Multimodal Extrinsic Contact Estimation, we show that we can ground contact from vision + touch, without prior object knowledge or camera calibration. 😅I unfortunately cannot make it to ICRA this year, 😉but my collaborator @Yuki_Doradora will be there to present the paper. If you are attending ICRA, please talk with Yuki and learn more about the project! Code and dataset: github.com/merlresearch/u… Paper: arxiv.org/abs/2601.04356 🤔Also, please check out our recent #RSS2026 paper, Contact-Grounded Policy (contact-grounded-policy.github.io), where we further explore how to extend contact grounding into dexterous manipulation and achieve extreme dexterity!!
English
3
30
244
26.8K
Jason Kim retweetledi
真広(まひろ)
真広(まひろ)@med_mahiro·
Sim-to-Simの比較(所感)。 ・Choreonoid + ROS 2:Sim-to-Realを見据えた検証に極めて有効。 ・MuJoCo(mujoco_ros2_control):極端な設定をしない限り安定して動作し、最初のSim-to-Sim環境として扱いやすい。 ※ロボの動きにも差があるので比較すると面白い。
真広(まひろ)@med_mahiro

Choreonoid + ROS 2 でKX-01(4.5kg小型犬ロボ)のSim-to-Simに成功した! IsaacSim 5.1 / Lab 2.3.0の環境で学習し、獲得したポリシーをChoreonoidでSim-to-Sim。 諦めなくてよかった!(Sim-to-Realに近い実装が必要であった)

日本語
1
20
126
12.4K
Jason Kim retweetledi
真広(まひろ)
真広(まひろ)@med_mahiro·
小型犬ロボ「KX-01」に火が入った! ・基本姿勢へ移行 ・歩行開始(推論開始) ・各動作確認 まで問題なく動作。 電圧とゲインは下げている状態だけど、このまま歩けそう。 ロボがピクッとなったところが、歩行開始コマンドを送った瞬間。
日本語
2
20
167
9.3K
Jason Kim retweetledi
Robotico
Robotico@roboticomarket·
Hugging Face just released @LeRobotHF a humanoid robot you can build for roughly $2,500! A full open stack with hardware, simulation, training environments, runtime tools, datasets and robot learning workflows designed to take a humanoid from assembly to real world control. Key details below.
Robotico tweet media
English
11
114
716
60K
pepper
pepper@ratygal·
옥상에서 고기 굽기 🤤
pepper tweet media
한국어
2
0
15
202
Jason Kim retweetledi
Ryohei Sasaki@engineer
Ryohei Sasaki@engineer@rsasaki0109·
gnss_ros_standardization github.com/DaikiNiimi/gns… gnss_ros_standardization is an open-source ROS 2 package that standardizes GNSS raw-data handling for robotics and autonomous systems. The goal is to make GNSS raw observations and ephemeris uniformly accessible across receiver brands so that tight-coupling GNSS/IMU methods and multi-sensor fusion frameworks can be developed once and reused everywhere. - Consume GNSS raw observations through standardized ROS 2 topics in robotics applications - Support diverse GNSS receivers and formats (u-blox, Septentrio, NovAtel, RTCM3) - Reproduce experiments across platforms with consistent interfaces
Ryohei Sasaki@engineer tweet media
English
0
15
80
6.7K
Jason Kim retweetledi
AirLab
AirLab@AirLabCMU·
Proud to see Co-Me accepted to CVPR 2026 🎉 Now supporting MapAnything 1.1, Depth Anything 3, and Pi3 - and 2× faster than the original, up to 21.5× speedup on long VGGT sequences. Congrats to the team!
Yutian Chen@YutianChen03

All your favorite 3D models — now faster with Co-Me. 🎉 Accepted to CVPR 2026, Co-Me now supports more 3D foundation models: MapAnything 1.1, Depth Anything 3, and Pi3. Same simple confidence-guided token merging idea — now accelerating even more 3D reasoning models. 👇

English
0
4
11
3K
pepper
pepper@ratygal·
오늘 뭔 날인가 말벌아저씨랑 미팅하면서 큰소리 1개도안내고 서로 네네^^ 좋네요^^ 그렇개할까요^^ 이러고끝낸거 몇달만인지;
한국어
2
0
12
173
Jason Kim retweetledi
Ouster
Ouster@ousterlidar·
We're excited to announce that @geckorobotics is leveraging Ouster’s new Rev8 native color lidar to power next-generation inspection capabilities on its Cantilever operating platform. By integrating inherently colorized, structured 3D data directly into its AI software, Gecko’s flying, climbing and crawling robots can visualize and detect structural anomalies on critical infrastructure with unprecedented certainty. From aircraft to energy assets, we’re proud to help push the boundaries of automated inspection and Physical AI with smarter, actionable data. 🔗 Read the full announcement here: businesswire.com/news/home/2026…
English
2
18
122
7.6K