Trends to Track for Next Year, According to SiLabs

If you want to know the latest trends in our industry, a good place to start is with one of the industry leaders, in this case Silicon Labs. While nobody gets it right all the time, it behooves the industry leaders to back the technologies that are going to have an impact, both in the Ultra-low-power computing, connectivity, aka IoT, and machine learning are three of the areas that will dominate the tech news over the next year or two, along with security. That’s the response I got from Daniel Cooley, the newly appointed Chief Strategy Officer for Silicon Labs. Hear what else Daniel had to say in our brief discussion.

Embedded Computing

wwwww

An efficient deep-learning tool for detecting eye disease

d41586-018-02431-1_15480472
Model scans images to detect urgent signs of conditions leading to blindness.

A new artificial-intelligence tool deploys a highly efficient form of deep learning to diagnose eye disease from medical images.

Convolutional neural networks are deep-learning algorithms adept at processing images, but researchers typically need to train them on more than a million medical images before they can test how well the algorithms work. Kang Zhang at the University of California, San Diego, in La Jolla and his colleagues created a kind of convolutional neural network capable of learning with many fewer images.

The team trained the model on 108,000 images of retinas. All had been classified by experts as either healthy or showing signs of a leading cause of blindness: macular degeneration or diabetic macular edema, a build-up of fluid in the retina. The algorithm identified critical cases of these conditions as accurately as six experts in ophthalmology.

The model also identified pediatric pneumonia from chest X-rays, suggesting that the technique could be broadly applied across medicine.

Source: Nature.com []

wwwww

AI drones: today’s security guards

adw---nightingale-blackbird---1100x400

In today’s disruptive economy of unmanned aircraft systems, we are constantly on the lookout for inspiring stories and innovative applications. Last week we spoke to Arnd Schöter, co-founder of Nightingale Security, a Mountain View Silicon Valley based service provider that uses autonomous drones to enhance or replace security guards for patrolling, surveillance, and response. Together with GMP Defence they won the Drone Hero Europe 2018 awards with their fully autonomous ‘Blackbird’.

Drone Machine Learning

“Our fully-autonomous system flies patrols day and night (in rain and snow), responds to alarms, transmits live video, lands, recharges, communicates, collaborates and reports maintenance needs—all by itself” explains Arnd. “Nightingale develops software that notices problems before you do. The system learns your facility’s normal attributes, which enables it to detect abnormal deviations and bring them to your attention. Based on all data collected by our system (telemetry, diagnostics, video, weather data, 3Dmapping data and more) our system is able to learn and improve.”

Autonomous drones in a participation economy

Nightingale’s goals in the long run are to become the #1 RaaS (Robot as a Service) provider in the security sector and eventually expand to other verticals, which opens up an endless list of possibilities. “In the long term we envision building a Robotic Applications Platform (RAP), allowing developers to build software for Nightingale’s already on-site, autonomous drones”, says Arnd. “For example, we are flying security at a site of solar fields and our solution works perfect in terms of security. However, the local company could identify an additional use case for the drone, for example putting it to use for detecting abnormalities in their solar fields on sub-cell level. The companies developers could then develop this feature and integrate it into our solution, potentially making it available for other parties to use and improve. A participation model if you will.”

About

Arnd: Arnd Schöter, based in Brussels, co-founded several consulting companies in Germany, Spain and Belgium, developing digital strategies for German blue chip companies. He has more than 10 years of experience in running global projects in the B2C, B2B area in the logistics, automobile and FMCG industries. He received his MBA from Sheffield Hallam University in England. Being able to live and work abroad for almost 20 years he acquired deep cultural understanding, speaks 5 languages and has built up an extended European business network.

Nightingale’s Blackbird:
‘Blackbird’ is the only UA system available on the market with fully autonomous takeoff, operation, landing and recharching. Together with its own base station and integrated weather system it is capable to perform patrol, inspection, escort and alarm functions, as well as to work even during difficult weather conditions. It can also monitor, recognize and follow objects, as well as avoid obstacles on its way using set of cameras and sensors.

Curious to see where Nigthingale’s headed in the next couple of years? Follow them here.

Source: Amsterdam Drone Week’s Blog []

wwwww