Internet of Things (IoT)

IoT meets Big Data and Artificial Intelligence

data streamIoT is the inter-networking of any physical device, while Big Data refers to storage and analysis of large amounts of data.

So, every time an IoT device produces a lot of data, these two technologies meet. Artificial Intelligence algorithms can use that quantity of data to train algorithms and achieve an high efficiency.

An infrastructure may contain many IoT devices and the quantity of data that each of them produce may be huge.

To store and retrieve such amount of data, often in a constrained time frame, non-relational database are used. Those systems basically drop functionalities in favor of speed and flexibility.

Scenarios of convergence and sectors where those technologies can be used are countless: economy, finance, medicine, agriculture, robotics, traffic shaping in data networks and in transports, social networks, marketing, and so on.

For example, consider many meteorological buoys scattered on a wide area, each of them sending data at relatively high frequency. The amount of data collected over time will be huge and Big Data analysis may be used to improve weather forecasts and climate models.

Another example of IoT objects could be autonomous cars or robots. Many sensors are used to send data at a very fast pace to react to environment changes and to assess vehicle or robot position and dynamic status. Those data need to be analyzed in real-time. Big Data techniques are necessary to process an enormous quantity of data in fractions of a second and AI algorithms decide the next actions of the car or the robot.

Wearable sensors may be used to detect when you are walking, running, sitting or sleeping, measure your hearth rate or values of blood indicators. Those data can be analyzed to understand if and when your do physical exercise and what you eat and drink. So AI algorithms can be trained to suggest you to do some work out in your spare time or to improve your diet.

There’s a lot of talk and work about those subjects. Those technologies and algorithms will have a big impact in many aspects of our lives in the near future.

A new era for road transport is on the way

autonomous drivingThe world is approaching a revolution in road transport. There will be tremendous changes in many aspects:

  • autonomous driving
  • propulsion using clean energy sources
  • smart infrastructures and communication systems

Autonomous driving

Soon vehicles will be able to drive without human intervention. A increasing number of companies is investing in self-driving and in the related Artificial Intelligence (AI) algorithms, mostly automakers and technology providers.

But apart from technical difficulties, there are also legal and ethical problems. Lawmakers seem to gain awareness of this big change. For example, who is responsible in case of an accident? The owner, the human “emergency” driver (if any) or the producer? And also, if a vehicle cannot safely stop before hitting a pedestrian, should it choose to hit him anyway or steer and crash its passengers against a wall?

Especially in cities, cars will be more a public service than a private good because they could be requested on call instead of needing one constantly out of the door, waiting to be used. That will change people daily habits and family budget plannings.

Propulsion using clean energy sources

Electric cars are not a news but there are not so many on the roads. Now electricity storage is improving and that involves lower costs and shorter and less frequent recharges, so that electric cars will be more attractive.
Due to those improvements, also trucks are starting to use electricity for their propulsion.
Energy can be obtained from several sources but it is essential to get it in a clean way to fight pollution and global warming.

Smart infrastructures and communication systems

Self-driving associated with smart infrastructures and communication system will allow an overall control of traffic and emergency signalling. We will achieve rerouting because of a closed road or traffic congestion, optimization of traffic lights, statistics collection for future planning of public transport or new roads.

Let’s consider an unfortunate scenario. A car detects it just had an accident by reading peak values from its accelerometers. Immediately it sends an alert to the vehicle behind for an emergency brake. Then it scans the vital signs of the occupants and tries to verbally communicate with them. After this investigation phase, it may decide to call for help to a central system. That one in turn sends an ambulance and reroutes the traffic to avoid the accident area. The ambulance transmits its position along the way and the central system diverts again the traffic to free the path of the ambulance.

In the initial phase, manned and autonomous vehicles will coexist but when all vehicles will be autonomous, road infrastructures like streetlights and road signs will not be needed anymore, thus saving energy, raw materials and maintenance costs.

No more time and energy wasted searching for a parking place, because the car will know where one is available.

Without a driver, cars will be like small offices or rest rooms. Time spent in a car will be more useful than today. There will be advanced interfaces and entertainment systems.

The saying “It’s not about the destination;¬†it’s about the journey.” is going to be even more true.

Old friends and new technological gardening

smart gardeningI have a friend who spends whole days working at his garden. He often changes the position of his bushes and stones, sometimes he adds a tree and I don’t know what else. As a gardening-addicted he’s considering to buy some smart soil probes and put them in some strategic spots. He’d love to check them when he is far from his beloved plants.

Geographical maps are not so detailed for his garden and of course they don’t update at every relocation of his plants. So he needs an approximate drawing of his garden and virtually set the position of his probes on it.

I suggested him to have a look at Sensorbis. He could do all that by uploading a plan of his garden and adding his probes as a federation of devices. Sending data in a simple format, humidity and pH rates can be displayed on the plan in real-time. He could also see a real-time chart or a historical chart of his probes’ data, calculate, for example, an average and also save those data for some nostalgic days.

That’s just a little example of the opportunities offered by IoT systems in gardening or in agriculture.
There are IoT sensors to measure temperature, humidity, wind, rain, sunlight, soil nutrients, plant growth, fruit ripening. Using data collected by those sensors, IoT devices can control automatic irrigation, nutrients feeding, shielding from hailstorms and so on.

Those new technologies allow, for example, to grow plants where water is very scarce, calculating the exact amount needed to save it as much as possible. That is of vital importance for people living in such disadvantaged areas.

And after harvesting, IoT systems are involved in the subsequent phases of processing, preservation and delivery.

IoT systems automate and fine tune many aspect of production, reducing effort, resources usage and energy consumption while increasing yield and quality of the crops.

How to visualize your IoT infrastructure?

telemetry tracking on mapsAn IoT infrastructure is composed of a set devices that could be located in a room, like socket consumption meters, or spread through the planet, like oceanographic buoys. Such devices could also change their location over time, like vehicles.

Displaying your devices at their location can greatly help to have an overall view of the situation and immediately identify a device much faster than an identification number or a textual description. Depending on the purpose and on the physical area covered by your set of devices, you could display them on a geographical map (e.g. marine buoys or vehicles) or over a plan (e.g. domestic power consumption meters or machinery in a factory).

Sensorbis allows using a geographical map or a user-provided plan or image to visualize real-time position and telemetry data of your devices. If a device is static, you can set its coordinates in the device properties, while, if it can move, you can use coordinates provided by the device itself. Users can choose to view a track of the last positions of a device.

Coordinates are typically acquired using GPS, but any sort of custom positioning is fine, e.g. those calculated by indoor smart cleaning robots. All you need to do is specifying the coordinate channels of your device. If you use a custom plan, you can set its physical height and width to match the device measurements.

A deeper insight of the telemetry data is provided through charts with selectable period range, data coarseness and statistical indicators.

Let’s consider the case of robots that move on rails inside a factory. Those robots could be programmed to calculate their location on the track and thus inside the building. They could wirelessly send the position at regular intervals. By uploading a plan of the factory in the system, it is possible to follow the movement of the robots in real-time. And, by clicking on one of them, it would possible to view its battery level, the weight of its load, and so on. Those information can be expanded on a dedicated chart for a more detailed view over time.