Marrying movement with sensor technology_

A global technology company had built a sensor that aimed to make our relationship with technology more human and intuitive. By detecting sub-millimetre motion at high speeds, it could track movement with exceptional accuracy.

This would allow people to control devices with a simple, universal set of gestures. Devices needed to be able to detect the location of people in their sensing area and recognise body cues that typically start and end interactions, including hand and body gestures.

To bring this product to market, the company needed the support of a sophisticated team with skills that could support on machine learning, data collection and analytics, and quality assurance.

The technology company selected Mobica to collaborate on the project.

This included the development and testing of a miniature 60MHz radar that could detect numerous human motions - from the tap of a finger to more exaggerated body movements.

We also helped to build the initial machine learning algorithms – processing motion resolution by extracting subtle changes in received signals over time. These temporal signal variations helped to distinguish complex finger movements and hand shapes.

We also supported the development of a machine learning algorithm that could be applied to football. This recognised physical activities like kicks, shot power, distance and speed – and then replicated them in a digital simulation.  

Fifteen full time Mobica engineers are currently working with the customer both remotely and onsite at the company’s headquarters. 

This team has helped to develop commercial applications by working in partnership with the company’s own clients, which include a major sports retail brand and a digital technology company.