Adexis is a frontier robotics lab developing the "somatosensory cortex" for Physical AI, enabling machines to achieve dexterity through touch. The company builds tactile sensing and processing systems that let robots understand and manipulate physical objects with human-like precision. Their technology focuses on bridging the gap between machine perception and physical interaction, targeting applications requiring fine motor control.
Funding
Funding not disclosed
Founders
Product
Problem
Current robotic systems rely primarily on vision and proprioception, leaving them "numb" to the physical properties of objects they handle. This sensory gap prevents machines from performing dexterous manipulation tasks that require nuanced touch feedback, such as grasping fragile items or assembling intricate components. Without tactile intelligence, robots cannot achieve the fine motor control needed for advanced physical tasks.
Solution
Adexis is building the somatosensory cortex for Physical AI, a specialized layer that gives machines the power of touch. The company develops tactile sensing technologies and processing algorithms that enable robots to interpret physical contact, pressure, texture, and force feedback in real time. By integrating this touch-based intelligence into robotic systems, Adexis aims to solve the dexterity problem that currently limits physical AI applications. Their approach combines advanced sensor hardware with machine learning models that translate tactile signals into actionable motor commands, bringing machines closer to physical general intelligence.
Target Audience
Adexis targets robotics manufacturers, industrial automation companies, and research institutions developing advanced physical AI systems that require dexterous manipulation capabilities.
Features
- Proprietary tactile sensor technology that captures high-resolution pressure and texture data
- Machine learning algorithms that process touch signals into real-time motor control commands
- Integration framework designed to work with existing robotic platforms and control systems
- Focus on dexterous manipulation capabilities beyond current vision-based approaches
- Research-driven development targeting physical general intelligence applications