
Changyi Lin
@changyi_lin1 • 1,055 subscribers
Researcher @OpenAI PhD @CarnegieMellon | Prev @GoogleDeepMind LLM Reasoning | Robotics
Videos

Manipulation happens through surfaces. To understand contact-rich dexterous interaction, motion alone is not enough. We also need to know surface properties and contact state. Excited to share ART-Glove, an articulated tactile glove that captures contact-grounded information while preserving human dexterity. It provides: - Known Geometry: 16 rigid functional surfaces - Surface Motion: 22 anatomically aligned joints - Tactile Contact: 2048 piezoresistive taxels Huge thanks to my advisor Ding Ding Zhao, and to Yuxiang Yuxiang Yang, Maria Maria Bauza Villalonga, Marissa, and Peide Peide Huang for the valuable advice and discussions. Paper:
Changyi Lin90,269 Aufrufe • vor 5 Monaten

Contact does not always deform surfaces, especially when interacting with liquids and soft objects. However, most existing tactile sensors rely on surface deformation to infer contact. Excited to share LightTact, a visual-tactile fingertip sensor that makes contact directly visible. It provides: - Deformation-Independent Sensing: contact is detected optically, not inferred from surface deformation. - High-Contrast Raw Images: non-contact pixels stay near-black. - Pixel-Level Contact Segmentation: robust across materials, forces, appearances, and external lighting conditions. This project was co-led with Boda ( Grateful to all co-authors for their contributions, and many thanks to Yuxiang Yang for the insightful discussions. We will present LightTact at RSS 2026 in Sydney next week! Paper: Open-source:
Changyi Lin65,715 Aufrufe • vor 4 Monaten

Introducing LocoTouch: Quadrupedal robots equipped with tactile sensing can now transport unsecured objects — no mounts, no straps. The tactile policy transfers zero-shot from sim to real. Core Task-Agnostic Features: 1. High-fidelity contact simulation for distributed tactile sensors 2. Adaptive gait reward for symmetric, frequency-adaptive locomotion Website: Paper:
Changyi Lin38,782 Aufrufe • vor 1 Jahr

Open-sourcing three repos from our CoRL paper LocoTouch ( 1. Perceptive Learning in IsaacLab ( 2. Policy Deployment (MoCap or onboard sensors) ( 3. Mac Mini for Onboard Robotics ( Yuxiang Yang and I have used the compact Mac Mini and stable deployment pipeline across several legged robot projects (
Changyi Lin20,325 Aufrufe • vor 1 Jahr
Keine weiteren Inhalte verfügbar