As a result of the ongoing digitalization process in the industry 4.0, the adoption of touch screens has increased but the technology still presents a few drawbacks:
- Low precision operating in harsh ambient conditions (e.g., dirt, dust, liquids…).
- Personal Protection Equipment, in particular gloves, can interfere with touch detection.
An operator using a touch screen
SC Robotics and Edge Impulse are leading a project to create a new device that can not only overcome these issues, but also bring new types of user interactions. To do so, SC Robotics provides its previous experience designing contactless keyboards to fight COVID-19 whereas Edge Impulse contributes with their platform and know-how to build ML algorithms that can run on the device itself.
Even though the end goal is to create a low-cost device, during this first phase, SC Robotics’s development team has been using a Raspberry Pi to collect the data generated by the sensors and upload it to the cloud. Testing all available sensor technologies is also of key importance in this initial stage. For this device, Time-of-Flight (ToF) technology has been chosen, however, within the same type of devices there are many alternatives depending on the performance and/or cost.
Raspberry Pi connected to Ntouch V1
Once the data has been collected, the training phase starts. This used to be a tedious task, but thanks to the powerful Edge Impulse platform and their expertise, this is not the case anymore. Obviously, it’s necessary to gather as much data as possible and that’s a manual repetitive job, but you can already start to see results that validate the idea after only a few iterations.
Gestures-recognition training on Edge Impulse
Thanks to this partnership, SC Robotics and Edge Impulse will design and test a low-cost device that can recognize contactless gestures in all kind of conditions, bringing enhanced safety to operators and new kind of interactions.
About the consortium
SC Robotics (SCR) is a young Spanish company specialized in the design, development and manufacture of customized integrated electronic devices. The core of its know-how is based around intelligent, low power and remotely connected sensors in the logistics and agrotech industry, offering a broad portfolio of devices that can cover many different user needs. A solar powered datalogger for crops and farms with all kind of needs and sizes, or a smart bung for wine and liquor are recent examples. Since it was created, SCR has participated in many R&D projects to improve their own products or explore new ideas and markets.
Edge Impulse (EI) is the leading development platform for embedded machine learning, used by over 1,000 enterprises. EI is on a mission to enable the ultimate development experience for machine learning (ML) on embedded devices for sensors, audio, and computer vision, at scale. EI enables highly optimized ML deployable to a wide range of hardware from MCUs to CPUs. With Edge Impulse, developers, engineers, and domain experts solve real problems using ML in embedded solutions, speeding up development time from years to weeks.