Our expert predicts upcoming trends in connected device development
There is a huge need for custom hardware and IoT implementations to take advantage of connected technology for industry.
BY JEREMY LOSAW
Although it has only been about 10 years since smart home and IoT products exploded into the marketplace, there could be up to 50 billion connected devices in service by the end of next year. That is approximately 6 1/2 devices for everyone on the planet.
The Internet of Things has seemingly penetrated every corner of our lives. Even graffiti artist Banksy used wireless technology to shred his “Girl With Balloon” painting during a Sotheby’s auction last year.
Despite the saturation of IoT devices, there is still plenty of room in the market for improvement and change. Consider these IoT trends that I think will drive connected device development in coming years:
When IoT devices were first conceived, they were “add-on” products. The Nike+, one of the first connected fitness trackers, clipped to your sneaker. It was clearly an accessory.
As connected technology and the need for it has evolved, we are starting to see products that are designed holistically with IoT embedded in them—such as Digitsole, an IoT-enabled shoe insert capable of advanced fitness monitoring. You would never know the difference between a sneaker fitted with one and one without.
Sensors and actuators will continue to shrink in size, so they will be easier to be embedded in more different types of devices—including irregular products such as clothing.
The Fitbit—the first superstar of IoT wellness—merely opened the floodgates. People have always had enthusiasm for fitness and well-being, and connected devices provide hard data to help us make meaningful changes. Existing devices can help you improve posture, track your fertility, or monitor your workouts.
However, there are still opportunities in the market. Sleep tech is an emerging trend. The Somnox sleep pacing pillow and the Kryo heating and cooling mattress topper are just two connected sleep devices that have scratched the surface of the capability of IoT to help us rest and recover.
There is also an emerging promise that IoT devices can help persons living with disabilities. The emergence of voice control and advanced laser vision technology opens the door to bring sense and control back to people with disabilities.
IoT has been dubbed the fourth Industrial Revolution, and with good reason. Data from manufacturing and other industries are helping to optimize the time and energy it takes to make, transport and warehouse goods. Although devices in this category do not get as much hype as the latest smart home gadgets, they can be far more impactful to the global community.
Smart agriculture devices monitor food crops and can tell farmers when to water fields, as well as indicating which areas are having pest issues. Sensored assembly lines allow manufacturers to assemble faster and with fewer mistakes. Inventory trackers are monitoring shipments and making sure that condition-sensitive deliveries are not exposed to adverse conditions.
The possibilities are endless. There is a huge need for custom hardware and IoT implementations to take advantage of connected technology for industry.
Collecting masses of data is futile if it is not analyzed. Machine learning leverages the massive computing power of the cloud to find trends in IoT data and help create algorithms that are fed back to our devices to make them smarter.
For example, the Nest thermostat collects data that are then analyzed to find trends and adapt its heating and cooling cycles to match your habits. Machine learning software and techniques are becoming more mature and easier to deploy; these will continue to be an integral part of IoT products.
An interesting side effect of the boom in machine learning is an emerging trend toward the commodification of hardware. Some companies are choosing to spend less development time and dollars developing custom circuitry, instead using off-the-shelf hardware. This allows them to devote more resources to develop learning algorithms for their devices to make them smarter and more useful.
This process leverages the processing power on the device to perform calculations, instead of sending it to the cloud to be analyzed. The need for fast processing speed is one driver of edge computing technology, and an obvious application is driverless/driver-assisted vehicles.
In this environment, vehicles are collecting lots of data and moving at a rapid rate of ground speed. The need to make split-second decisions to modulate their controls means they simply do not have time for data to be pushed to and from the cloud before taking action, such as applying brake pressure to avoid a collision.
The emergence of long-range wireless networks such as LoRaWAN is also driving the need for edge computing. LoRaWAN offers a transmission range of miles but can only transmit small packets of data.
In these types of devices, it makes more sense to process data and make decisions on the device and then only transmit the result of the data to the cloud. Fortunately, processing power continues to drop in cost, allowing very capable edge computing to be deployed economically.
Arnold Schwarzenegger’s character in the 1991 movie “Terminator 2: Judgment Day” originally predicted that smart devices would take over the world. (The Terminator was a tad optimistic on the day when computers would take over: Skynet was supposed to be fully autonomous on Aug. 29, 1997.)
Smart devices are being deployed to help us live better lives—and fortunately, there is plenty of room for inventors to advance the technology.