The "Internet of Things" concept envisions a world in which myriad devices -- from toasters to streetlights to heating thermostats -- exist in the cloud, and can identify and connect with one another seamlessly. Now, a Swedish-based startup is offering a mini-OS that can be built into device microcontrollers to ease the development of devices that are ready for this kind of connectivity.
“The idea behind Thingsquare Mist is to make it easy to connect low-power wireless devices to the Internet, so that they can be connected with smartphone apps or backend databases,” Thingsquare chairman and chief architect Adam Dunkels told Design News. “We see a large interest in adding new forms of intelligence to products in the connected home, in smart lighting, and for smart cities -- this is what we call the Internet of Things.”
Mist comes in the form of source code that’s compiled into a firmware image that can run on a number of battery-powered microcontroller and Systems on a Chip (SoC) platforms. When Mist is designed into a wireless device, it allows that device to connect directly to a smartphone backend. Mist is based on the Contiki OS, which Dunkels himself developed, and supports the IPv6 Internet protocol for connectivity.
A diagram showing how Thingsquare Mist, a mini-OS for embedded devices, allows devices to create a mesh network that can enable the future Internet of Things. (Source: Thingsquare)
Dunkels explained to us how Thingsquare Mist builds a mesh network of devices in this way:
Thingsquare Mist devices build a wireless self-healing mesh network, based on standard Internet protocols. One of the devices has an Ethernet or WiFi connection. This device, called the router, becomes the access point to the Internet. Unlike previous systems, the Thingsquare Mist is very lightweight, and can run on a small 16-bit microcontroller. Existing solutions require a much more complex and costly device, often running a microprocessor large enough to run Linux.
Thingsquare has released the source code for Mist, so engineers can download it and use it with a number of embedded hardware platforms, including microcontrollers from Texas Instruments and STMicroelectronics. “We currently support the TI MSP430, the STM32L, the Freescale MC13224v, and the TI CC2538,” Dunkels told us.
Thingsquare Mist is already being used in small-device applications, such as a smartphone-connected thermostat from Tado that uses geolocation mapping to control a person’s thermostat at home, turning it up and down according to where someone is. For instance, Tado can intelligently detect when the last person leaves a residence and adjust a home heating system accordingly, but then turn it up again before a person arrives home so it’s warm by the time he or she gets there.
Thingsquare also is working on a cloud backend for Mist devices, Thingsquare Haven, that it will reveal in a few months' time, Dunkels said. Haven will monitor the status, connectivity, and health of devices connected using Mist.
This is quite an interesting offering to enable this so-called "Internet of Things" and help it go beyond mere industry hype. By building this mini-OS directly to devices as well as eventually providing a cloud-computing back-end for the devices, Thingsquare is trying to provide a key enabling technology to make this vision more of a technology reality.
I remember predictions of this type of technology 15 years ago during the dot com boom. The go-go folks of the early Internet days saw a home where everything was connected. They saw a fridge where sensors could read expiration dates and place milk on the list of groceries needed through the web-based grocery service.
Yes, Rob, I was writing about technology back then as well and remember execs from Microsoft and the now-defunct Sun Microsystems (part of Oracle now) blathering on about this. It took some time but it finally does seem to becoming a reality. I always knew it was possible but as usual, it just takes technology some time to get there.
I agree, Elizabeth, the technology is now there for the internet of things. Ultimately, though, it will gain traction in as much as it solves problems or provides pleasure. During the dot com boom, these two considerations were not in the forefront.
Interesting perspective, Rob, but I think you're right. During that time (the dotcom boom), it seems like a lot of people were more fascinated by the technology itself, geeking out on mere innovation, rather than thinking about the marketing and practicle aspect of it. Things have changed a bit now, as you noted.
Yes, I think we're in a healthier environment now when it comes to Internet businesses. Execution -- rather than a great idea -- seems to be the critical factor now. Google search was not a new idea, nor was Facebook, nor most of Apple's products. It was execution that took these companies to the top.
Indeed, Rob, execution is key. Although I think it was Microsoft way back when that taught us that lesson, which I think went a bit pear shaped during the dotcom time. Apple came out with the PC and a great OS idea and Microsoft commoditized it and made it ubiquitous. (I'm talking about Windows, of course.) I think now there is an idea that innovative technologies must prove themselves first rather than just be a great idea.
Good points, Elizabeth. Going back even further, Apple got its technology from Xerox. There are arguments about whether Steve Jobs actually "stole" the idea of a mouse and a graphic-user interface. But at any rate, Apple was the executer, not the innovator.
Are they robots or androids? We're not exactly sure. Each talking, gesturing Geminoid looks exactly like a real individual, starting with their creator, professor Hiroshi Ishiguro of Osaka University in Japan.
Truchard will be presented the award at the 2014 Golden Mousetrap Awards ceremony during the co-located events Pacific Design & Manufacturing, MD&M West, WestPack, PLASTEC West, Electronics West, ATX West, and AeroCon.
For industrial control applications, or even a simple assembly line, that machine can go almost 24/7 without a break. But what happens when the task is a little more complex? That’s where the “smart” machine would come in. The smart machine is one that has some simple (or complex in some cases) processing capability to be able to adapt to changing conditions. Such machines are suited for a host of applications, including automotive, aerospace, defense, medical, computers and electronics, telecommunications, consumer goods, and so on. This discussion will examine what’s possible with smart machines, and what tradeoffs need to be made to implement such a solution.