TouchTronix Robotics 1,797 followers 3w Report this post We’re excited to share that TouchTronix Robotics has been accepted into the NVIDIA Inception Program! 🎉🎉🎉 At TouchTronix, our mission is to build the multimodal data infrastructure for Physical AI — capturing and transforming human interaction and tactile experience into scalable vision-tactile data that helps robots better perceive, learn, and interact with the physical world. Joining NVIDIA Inception marks another important step in our journey to bridge human physical intelligence and robotic intelligence. We look forward to leveraging NVIDIA’s ecosystem and technology as we continue scaling our data collection systems, datasets, and AI infrastructure for the next generation of intelligent robots. From human touch to robotic intelligence. #NVIDIAInception #PhysicalAI #Robotics #EmbodiedAI #TactileSensing #MultimodalAI #TouchTronix 57 3 Comments Like Comment ShareTouchTronix Robotics has been accepted into the NVIDIA Inception Program! 🎉🎉🎉 At TouchTronix, our mission is to build the multimodal data infrastructure for Physical AI — capturing and transforming human interaction and tactile experience into scalable vision-tactile data that helps robots better perceive, learn, and interact with the physical world. Joining NVIDIA Inception marks another important step in our journey to bridge human physical intelligence and robotic intelligence. We look forward to leveraging NVIDIA’s ecosystem and technology as we continue scaling our data collection systems, datasets, and AI infrastructure for the next generation of intelligent robots. From human touch to robotic intelligence. #NVIDIAInception #PhysicalAI #Robotics #EmbodiedAI #TactileSensing #MultimodalAI #TouchTronixTouchTronix Robotics 1,797 followers 3w Edited Report this post Great to see so much interest in our piezoresistive thin-film tactile array sensor at #IROS! One question we heard a lot: “Where can we get them?” The good news: our standard tactile sensor arrays are readily available through DigiKey, making it easy for researchers and robotics teams to get started: https://lnkd.in/gabRpbGY Need something beyond the standard form factor? We also provide custom sensor designs and integration support tailored to your robotic grippers, dexterous hands, and other manipulation platforms. From research prototypes to scalable robotic systems, we’re excited to make high-performance tactile sensing more accessible to the robotics community. #IROS #Robotics #TactileSensing #DexterousManipulation #PhysicalAI #TouchTronix TouchTronix Robotics 1,797 followers 8mo After months' of efforts, we’re excited to share that TouchTronix Robotics SensX tactile array sensors are now available on DigiKey! 🎉🎉🎉 This big step makes our product accessible to a broader R&D community (robotics, healthcare, manufacturing, etc.) with faster, easier procurement🚀🚀🚀: SensX25 – 5×5 array sensor with sensor-on-PCB build: https://lnkd.in/eENB3rQf SensX192 – 12×16 array thin film sensor: https://lnkd.in/eBXnyV_H SensX160 – 20×8 array thin film sensor: https://lnkd.in/es7Sk_WE SensX187 – 11×17 large-area thin film sensor: https://lnkd.in/eSBE2SKy Moreover, we offer compatible Data Acquisition Modules with miniaturized form factor but complete capabilities for "Plug-an-Play" for all of our thin film sensors: SensX192: https://lnkd.in/eEiMMCeH SensX160: https://lnkd.in/eaUU48Pp SensX187: https://lnkd.in/e3MVzkrS 👏Explore the SensX portfolio on DigiKey and start building today🦾! Also, we offer customization for your own design (contact: infor@touchtronix.io)! #DigiKey #Robotics #Sensors #TactileSensing #Innovation #Engineering #HumanoidRobotics #DeepTech #HardwareStartup #SpatialIntelligence #RoboticManipulation #AI #Sim2Real #Simulation #Gripper #Hand #DexterousHand #Safety #PhysicalAI 7 Like Comment ShareTouchTronix Robotics 1,797 followers 3w Report this post We’re excited to announce a strategic partnership between TouchTronix Robotics and TransFinite AI, bringing together multimodal data infrastructure and training-ready data capabilities for Physical AI. 🤝 Together, we aim to enabling robotics teams to build, validate, and iterate intelligent models faster and with greater confidence. 🚀 35 1 Comment Like Comment ShareTouchTronix Robotics 1,797 followers 1mo Report this post Welcome to visit us at booth #1110 during #IROS 2026 here in Pittsburgh! We will be happy to showcase our solutions for tactile sensors, FusionX Vision-Tactile data collection system, and datasets services. 36 2 Comments Like Comment ShareDear friends, you are welcome to visit us at #IROS 2026, Booth 1110 from tomorrow to 9.30😊. We will be happy to learn more about your needs and introduce our solutions on Tactile sensing, Vision-Tactile data collection, and Multimodal datasets for your robotic research.🤝🏻🤝🏻🤝🏻TouchTronix Robotics 1,797 followers 1mo Edited 🚀 Introducing the Next Generation of TouchTronix Robotics FusionX— and our new Vision-Tactile Datasets We’re excited to introduce the latest generation of FusionX, our multimodal data collection system built for robotic learning and Physical AI. This release brings major upgrades across sensing, perception, synchronization, and data annotation: 🔹 State-of-the-art tactile sensors A newly developed thin-film tactile sensor architecture and hot-pressing manufacturing technique delivers significantly improved sensitivity, stability (especially addressing “grasp-free” baseline signal drift), and pixel-to- pixel consistency. 🔹 Three-view visual perception FusionX now combines stereo vision with dual wrist-mounted cameras, capturing manipulation from three complementary perspectives. 🔹 Highly aligned vision-tactile data Tactile signals and visual observations are tightly synchronized and aligned (<5 ms lantency), enabling models to better connect what happens visually with the underlying physical contact. 🔹 In-house hand pose estimation algorithm Our proprietary hand-pose pipeline provides structured hand kinematics alongside tactile and visual observations. 🔹 High-precision natural-language annotations Each dataset can include detailed semantic annotations describing actions, objects, interactions, and manipulation processes — making the data more directly usable for robot learning and multimodal model training. But the most important part is the data. We have already collected thousands of hours of real-world vision-tactile manipulation data, and are scaling toward 10,000+ hours in the coming weeks. Both the complete FusionX system and our vision-tactile datasets are now commercially available. (Product/Datasets contact: info@touchtronix.io) We are building the data infrastructure that connects vision, touch, hand motion, gaze, semantics and more— so robots can learn not only what the physical world looks like, but how it feels and how to interact with it. Vision + Touch → Robotic Intelligence. #PhysicalAI #Robotics #EmbodiedAI #RobotLearning #TactileSensing #MultimodalAI #DexterousManipulation #RoboticsData #TouchTronix #FusionXTouchTronix Robotics FusionX— and our new Vision-Tactile Datasets We’re excited to introduce the latest generation of FusionX, our multimodal data collection system built for robotic learning and Physical AI. This release brings major upgrades across sensing, perception, synchronization, and data annotation: 🔹 State-of-the-art tactile sensors A newly developed thin-film tactile sensor architecture and hot-pressing manufacturing technique delivers significantly improved sensitivity, stability (especially addressing “grasp-free” baseline signal drift), and pixel-to- pixel consistency. 🔹 Three-view visual perception FusionX now combines stereo vision with dual wrist-mounted cameras, capturing manipulation from three complementary perspectives. 🔹 Highly aligned vision-tactile data Tactile signals and visual observations are tightly synchronized and aligned (<5 ms lantency), enabling models to better connect what happens visually with the underlying physical contact. 🔹 In-house hand pose estimation algorithm Our proprietary hand-pose pipeline provides structured hand kinematics alongside tactile and visual observations. 🔹 High-precision natural-language annotations Each dataset can include detailed semantic annotations describing actions, objects, interactions, and manipulation processes — making the data more directly usable for robot learning and multimodal model training. But the most important part is the data. We have already collected thousands of hours of real-world vision-tactile manipulation data, and are scaling toward 10,000+ hours in the coming weeks. Both the complete FusionX system and our vision-tactile datasets are now commercially available. (Product/Datasets contact: info@touchtronix.io) We are building the data infrastructure that connects vision, touch, hand motion, gaze, semantics and more— so robots can learn not only what the physical world looks like, but how it feels and how to interact with it. Vision + Touch → Robotic Intelligence. #PhysicalAI #Robotics #EmbodiedAI #RobotLearning #TactileSensing #MultimodalAI #DexterousManipulation #RoboticsData #TouchTronix #FusionXTouchTronix Robotics 1,797 followers 1mo Edited Report this post 🚀 Introducing the Next Generation of TouchTronix Robotics FusionX— and our new Vision-Tactile Datasets We’re excited to introduce the latest generation of FusionX, our multimodal data collection system built for robotic learning and Physical AI. This release brings major upgrades across sensing, perception, synchronization, and data annotation: 🔹 State-of-the-art tactile sensors A newly developed thin-film tactile sensor architecture and hot-pressing manufacturing technique delivers significantly improved sensitivity, stability (especially addressing “grasp-free” baseline signal drift), and pixel-to- pixel consistency. 🔹 Three-view visual perception FusionX now combines stereo vision with dual wrist-mounted cameras, capturing manipulation from three complementary perspectives. 🔹 Highly aligned vision-tactile data Tactile signals and visual observations are tightly synchronized and aligned (<5 ms lantency), enabling models to better connect what happens visually with the underlying physical contact. 🔹 In-house hand pose estimation algorithm Our proprietary hand-pose pipeline provides structured hand kinematics alongside tactile and visual observations. 🔹 High-precision natural-language annotations Each dataset can include detailed semantic annotations describing actions, objects, interactions, and manipulation processes — making the data more directly usable for robot learning and multimodal model training. But the most important part is the data. We have already collected thousands of hours of real-world vision-tactile manipulation data, and are scaling toward 10,000+ hours in the coming weeks. Both the complete FusionX system and our vision-tactile datasets are now commercially available. (Product/Datasets contact: info@touchtronix.io) We are building the data infrastructure that connects vision, touch, hand motion, gaze, semantics and more— so robots can learn not only what the physical world looks like, but how it feels and how to interact with it. Vision + Touch → Robotic Intelligence. #PhysicalAI #Robotics #EmbodiedAI #RobotLearning #TactileSensing #MultimodalAI #DexterousManipulation #RoboticsData #TouchTronix #FusionX 25 Like Comment ShareTouchTronix Robotics reposted this Jun Liu 2mo Report this post Today marks my last day as a professor. Since the launch of TouchTronix Robotics in February 2025, the company has grown at an incredible pace 🚀. That journey has led me to make one of the biggest decisions of my career: transitioning from academia to full-time entrepreneurship. I want to express my deepest gratitude to all of my colleagues, students, collaborators, and friends in academia. Your support, mentorship, collaboration, and friendship over the past 10+ years have shaped who I am today. I couldn't have reached this point without you. Going forward, I'll be wearing the hat of Founder & CEO of TouchTronix Robotics, where our mission is to tackle one of the biggest challenges in embodied AI: Enabling robots with rich vision-tactile perception through large-scale multimodal sensing data and foundation models for dexterous manipulation and human-robot interaction. Although I'm leaving academia, I'm certainly not leaving science. I'll continue publishing research (perhaps more as a hobby 😊) because my passion for scientific discovery will always remain. A new chapter begins. I'm excited for what's ahead. 🤞 379 81 Comments Like Comment ShareTouchTronix Robotics in February 2025, the company has grown at an incredible pace 🚀. That journey has led me to make one of the biggest decisions of my career: transitioning from academia to full-time entrepreneurship. I want to express my deepest gratitude to all of my colleagues, students, collaborators, and friends in academia. Your support, mentorship, collaboration, and friendship over the past 10+ years have shaped who I am today. I couldn't have reached this point without you. Going forward, I'll be wearing the hat of Founder & CEO of TouchTronix Robotics, where our mission is to tackle one of the biggest challenges in embodied AI: Enabling robots with rich vision-tactile perception through large-scale multimodal sensing data and foundation models for dexterous manipulation and human-robot interaction. Although I'm leaving academia, I'm certainly not leaving science. I'll continue publishing research (perhaps more as a hobby 😊) because my passion for scientific discovery will always remain. A new chapter begins. I'm excited for what's ahead. 🤞TouchTronix Robotics 1,797 followers 5mo Report this post We are grateful for the support from Business & Entrepreneur Partnerships | University at Buffalo. As we are growing fast 🚀🚀🚀, we will announce our new space (much bigger!) at Baird Research Park soon! TouchTronix Robotics 1,797 followers 5mo Edited Today, we are excited to announce the new feature of FusionX human-centric multimodal data collection system: Fused Hand Pose Tracking+Tactile Force Distribution in physical 3D!🚀🚀🚀 Robots can only interact intelligently with the physical world when they understand both motion and touch. Now our solution combines: ✅ Vision-based hand pose extraction via ego-centric camera ✅ High-performance tactile sensing via data gloves ✅ Synchronized vision-tactile data via proprietary data pipeline ✅ Fused hand pose tracking and tactile force distribution in physical 3D By synchronizing hand kinematics with tactile interaction signals, robots can learn not only how humans move, but also how humans physically interact with objects and even human body— including grasping dynamics, slip events, force adaptation, and fine manipulation behaviors. Our technology can support applications including: 👉 Dexterous robotic manipulation 👉 Human demonstration learning 👉 Imitation learning & policy training 👉 Embodied foundation models 👉 Humanoid robotics 👉 Smart teleoperation systems As the robotics industry moves toward more general-purpose embodied intelligence, multimodal interaction data will become increasingly critical. We’re excited to continue pushing forward tactile intelligence for robotics. #Robotics #EmbodiedAI #HumanoidRobots #ArtificialIntelligence #MachineLearning #TactileSensing #ComputerVision #RobotLearning #ImitationLearning #PhysicalAI #TouchTronix 2 1 Comment Like Comment Share