DAIMON Robotics Wants to Give Robot Hands a Sense of Touch

2026-05-04 · Source: IEEE Spectrum · Field: Technology & Digital — Robotics & Autonomous Systems, Artificial Intelligence & Machine Learning · Depth: Intermediate, long

Summary

DAIMON Robotics, a two-and-a-half-year-old company, released Daimon-Infinity in April, described as the largest omni-modal robotic dataset for physical AI. This dataset features high-resolution tactile sensing and covers tasks from laundry folding to factory assembly, supported by partners including Google DeepMind and Northwestern University. DAIMON, known for its monochromatic vision-based tactile sensor packing over 110,000 effective sensing units, leverages this technology and a distributed out-of-lab collection network to generate millions of hours of data annually. The company open-sourced 10,000 hours of this data to accelerate embodied AI deployment. Prof. Michael Yu Wang, DAIMON's chief scientist, pioneered the Vision-Tactile-Language-Action (VTLA) architecture, elevating tactile feedback to a modality on par with vision, addressing the "insensitivity" of current Vision-Language-Action (VLA) models.

Key takeaway

For Robotics Engineers or AI Scientists developing advanced manipulation systems, integrating high-resolution tactile sensing is critical. The Vision-Tactile-Language-Action (VTLA) model, supported by datasets like Daimon-Infinity, addresses limitations of Vision-Language-Action (VLA) by enabling precise force control, slip detection, and object localization in challenging environments. You should consider adopting vision-based tactile sensors to enhance robot dexterity and reliability, especially for tasks involving fragile items or confined spaces, accelerating real-world embodied AI deployment.

Key insights

Tactile sensing is crucial for advanced robot manipulation, enabling the VTLA model to overcome VLA limitations.

Principles

• Data scarcity bottlenecks robot learning.
• Tactile data is essential for dexterous manipulation.
• Vision-based tactile sensors offer high resolution.

Method

DAIMON's method involves high-resolution vision-based tactile sensors, a robust multimodal data processing pipeline, and a distributed out-of-lab collection network for large-scale data generation.

In practice

• Integrate tactile feedback for fragile object handling.
• Use VTLA for dark environment object location.
• Deploy robots in tight spaces like convenience stores.

Topics

• Robotic Manipulation
• Tactile Sensing
• Embodied AI
• Robot Datasets
• VTLA Architecture
• Vision-based Sensors

Best for: AI Engineer, Machine Learning Engineer, Research Scientist, Robotics Engineer, AI Scientist, Director of AI/ML

Related on AIssential

• From Abstraction to Instantiation: Learning Behavioral Representation for Vision-Language-Action Model — BehaviorVLA improves VLA model robustness by learning temporally coherent behavioral representations for generalized manipulation across environments.
• Video Friday: Figure, 1X Ramp Up Humanoid Robot Production — Robotics innovation is accelerating, with significant advancements in humanoid production, dexterous manipulation, and autonomous decision-making.
• SECOND-Grasp: Semantic Contact-guided Dexterous Grasping — Integrating semantic reasoning with physical stability enhances robotic dexterous grasping performance and adaptability.
• PaCo-VLA: Passivity-Shielded Compliance Prior for Contact-Rich Vision-Language-Action Manipulation — PaCo-VLA safely integrates high-level VLA semantic reasoning with low-level contact dynamics using a passivity shield.
• Why Do Humanoid Robots Still Struggle With the Small Stuff? — Despite AI advancements, humanoid robots still struggle with fundamental physics like force control for reliable real-world interaction.
• UMI-3D: Extending Universal Manipulation Interface from Vision-Limited to 3D Spatial Perception — Integrating LiDAR into a wrist-mounted manipulation interface significantly enhances data quality and task reliability for embodied robot learning.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by IEEE Spectrum.