December 6, 2016 | Written by: Kevin Larsen
Share this post:
Smart products that contain customized sensors and sophisticated software take time and money to plan, design and build, pushing the opportunity to evaluate them into the later stages of the product development process.
An instant impact
To be able to walk into a room of decision makers with a prototype in hand is worth a thousand person-hours of design and planning meetings. Whether you are simulating heavy machinery home appliances and security systems, or different types of wearables, a prototype can easily be expanded or extrapolated into home, enterprise or industrial settings.
This is significant because as soon as a real device is used (as opposed to a simulated one), a development team will quickly encounter the same problems that customers or users are going to have in the real world: How do you handle passwords and user IDs across multiple devices? How do you authenticate and make sure a device is right? What happens when the device restarts or you lose your network? Does it come back? Does it recover nicely? These are all real world implications that can be seen the moment you start using a real device.
Simulating a solution offers developers a glimpse into the real world
Developers who are thinking about IoT want to think about the physical thing too, not just the application. Having boards, sensors and devices which are inexpensive to break gives a developer or engineer the freedom to prototype and try things – at very low risk. If something gets messed up, it’s easy to hit the factory reset and start over, creating the perfect environment for simulation with a low-risk platform to not only try and experiment, but to actually do things.
When developers use a device that historically has been a dumb device, and turn it into a smart, connected device, they need to build that on something first. Having a risk-free prototyping environment lets developers and engineers explore an idea before investing in designing new chipsets and boards, or engaging thousands of dollars’ worth of engineering.
For example, if someone wants to turn a washing machine into a smart washing machine, or test out how to interact with an automobile to mock up and create what the end user experience might be, it can be done easily without breaking the bank. Then, as the developer learns what the possibilities are, and what else can be done with that new smart device, they can spec out and build whatever custom dedicated hardware might be required to support the smart product.
Test and implement new ideas using a low-risk prototype
IoT makes regular objects an active part of the business process. Every light, every sensor, every thermostat, every printer, is all part of someone’s business. IoT allows us to see what’s happening with them; gives us the ability to interact with them, to collect data from them. They can tell us about the environment they are in or the people that use them. They can tell us when things are breaking. For instance, rather than having to go and check whether the printer is out of toner, if the coffee machine’s working, or the light’s on, we can find out from the things themselves.
A lot of smart products start with devices, boards, and a range of sensors connected up to cloud platforms such as IBM Watson IoT Platform, using the popular Node-RED IoT wiring tool. Collected data can then be analysed by applications hosted on IBM Bluemix. Proof of concept and prototypes can be created using kits and devices from ecosystem partners and vendors like Avnet, Premier Farnell, Texas Instruments, and many others. The range of available devices and boards offer a low-cost way to visualize and test various configurations and features to see how a ‘thing’ or solution might work.
Prototyping IoT use cases
Take for example the Raspberry Pi which integrates nicely with the IBM Watson IoT Platform. Developers can connect directly into the Watson IoT platform, get data from it, interact with it, do device management on it, and treat is as any other smart device that you might use in IoT. There are hundreds of recipes and examples where developers and makers are exploring IoT concepts using Raspberry Pis. From there, it’s easy to start prototyping and simulating ideas in the Raspberry Pi environment. Relatively inexpensive, the Raspberry Pi can be connected to other things, easily programmed to do something from almost any operating system (including Linux, Windows, and Mac.)
The DragonBoard™ 410c from Arrow Electronics is another example of an intelligent device you can programme and place at the network edge for collecting data and handing it off to the cloud. In parallel, you can use REST and real-time APIs to connect your applications and analytics to your data. Here’s a three-step recipe for connecting the DragonBoard 410c to the IBM Watson IoT Platform using a Message Queuing Telemetry Transport (MQTT) client app; or, try this recipe which shows how to connect a DragonBoard 410c to IBM Watson IoT Quickstart service using Arrow’s IBM IoT Starter kit.
Visualizing data using sensors
Visualization is an important part of any IoT deployment, especially when you are using sensors to detect temperature, humidity, light, and accelerometers. The TI SimpleLink™ Bluetooth® low energy CC2650 SensorTag kit is an IoT demo platform that allows users to stream data from 10 sensors to the cloud in minutes via a mobile phone application. It features an ultra-low-lower MCU and radio that can function for years on a simple coin-cell battery. Other versions of the SensorTag offer 802.15.4, ZigBee, and Sub-1 GHz connectivity. A C SDK supports application development.
The CC2650 is a good choice for ultra-low power endpoints that can connect to the cloud via a gateway, and there is already a lot of integration between TI and WIoTP. For example, there is a free SensorTag app available in the app store which comes preconfigured to connect to Watson IoT Platform. Additionally, the environment is built into Watson IoT Platform and can be customized to consume data from TI devices by adding TI Nodes that are available from the Node-RED library.
It’s easy to start developing IoT apps using a combination of the Watson IoT Platform with Node-RED. Try visualizing IBM Watson IoT Platform data using this tutorial featuring a TI SensorTag kit. To learn more about IoT use cases, consider also attending a webinar available from Friday, December 16th at 9am ET. This session includes a demonstration of TI SensorTag and Watson IoT Platform.
Open innovation and collaboration creates endless possibilities
Adherence to open standards and protocols mean these devices are easy to code, easy to access, cheap enough, and can often be connected directly to legacy devices, giving developers the ability to mock up how their idea or concept might work in the real world later.
From there it’s only natural for developers to get plugged into the different IoT communities and resources available to encourage further collaboration and exploration. These communities offer visibility into what other people are doing, how they are using IoT devices, what boards they are using, what sensors they’ve added, what add-ons they’ve put on; what code snippets and examples are available to try.
In any of these examples, the ability to get real hardware that combines power, openness, a community around it, to work with real things instead of simulation, gives developers the freedom to take dumb products and turn them into smart, connected products, while building entire solutions around them.
When compute and design capabilities are democratized through flexibility, access, power and ease of use – virtually available to anyone – suddenly we become part of an environment with limitless possibilities. When creativity and technology come together in an environment that is so easy to prototype and simulate, there’s no telling what amazing ideas and innovations will be discovered.
Register for a no-charge 30-day trial of Watson IoT Platform and explore the depth of developer resources available through IBM developerWorks and Bluemix.