Columbia University

Technology Ventures

High resolution optical touch sensor

Technology #cu17105

Questions about this technology? Ask a Technology Manager

Download Printable PDF

Image Gallery
Ioannis John Kymissis
Managed By
Greg Maskel

This technology is an electronic pressure detector that can mimic tactile sensation, with primary applications in next-generation robotics and prostheses.

Unmet Need: Electronic systems that can reliably replicate the sensation of human touch

Current methods of tactile sensing employ wired sensors whose cost and complexity exponentially increase with improvements in size and resolution by simply adding more sensors per unit of area. Moreover, the typical solution of a matrix of sensing wires is difficult to adapt to curved surfaces and can be difficult to manufacture. Thus, current approaches are ill-suited for implementation in robotic and prosthetic applications that call for improved tactile accuracy without letting the computational complexity become too burdensome to performance.

The Technology: Simple and high resolution tactile sensor

This technology combines light-emitting diodes (LEDs) with light-receiving transducers to construct a tactical sensor that generates electrical signals in response to mechanical deformation (a “touch”). Deformation of the material correspondingly alters the light absorbed by the transducers and produces a transient electrical signal change that correlates to the depth and location of the deformation. The sensor uses data-driven machine learning methods to map between a rich extracted signal set and the variables associated with touch sensing. The platform offers greater high-resolution scalability than competing approaches.

This technology has been validated to detect touch location and indentation depth with sub-millimeter accuracy across a 400 square-mm sensing area.


  • Precision robotics
  • Robotic surgery
  • Fine manufacturing
  • Human-machine interfaces
  • Robotic limbs and prosthetics


  • High resolution tactile sensing without increased manufacturing complexity
  • Exponential scaling of sensitivity without equal increase in electrode number
  • Can be easily integrated with precision robotic systems
  • Easy and inexpensive to manufacture
  • Parts are already commercially available
  • Produces rich data set from physical contact
  • Offers high signal-to-noise ratio
  • Requires minimal power

Lead Inventor:

Ioannis Kymissis, Ph.D.

Patent Information:

Patent Pending

Related Publications:

Tech Ventures Reference: