Internet of things is an interconnected group of physical objects collecting and sharing data to make lives on this planet easier. This is done by enabling the need for minimal human intervention to operate. Thanks to the advancement in technology, anything ranging from a pen to an airplane can be connected to IoT networks.
Voice assistants like Alexa, Siri, and Cortana have kick-started the Smart Home revolution and are gradually getting people excited about smart speakers, thermostats, etc. It is definitely becoming a lot easier to adopt smart technologies, and smart speakers, coffee machines, lights, and air purifiers are doing their part of the job to make lives easier. But, that’s not exactly how a smart home is envisioned, right? A smart home is supposed to be intelligent enough to understand its job without being asked to do so.
Your coffee maker still needs your input to start brewing coffee. Still not much different from you having to get up from your bed, go to your kitchen and start preparing coffee for yourself. But what if it was smart enough to be able to start brewing your coffee by considering the time when you put a stop to your alarm? That’s just the tip of the iceberg, meaning just one of many things the incorporation of Machine Learning can do to your IoT-enabled smart home.
Your Smart home today might have a number of smart devices all working individually in perfection, requiring some control from your end. For instance, having an Amazon Echo would reduce your effort in having to switch songs, control certain devices with your voice, for instance, reducing the brightness of a bulb, etc. But what if it knew when to increase the brightness and when to dim it by learning your preferences?
Living Made Easier…
Let’s try to visualize some simple everyday scenarios, assuming you are utilizing the full potential of the current IoT ecosystem –
1. After office scene - It’s 6:30 in the evening and you start your travel home from the office. Based on the time you would need while travelling, you set your thermostat to get your desired room temperature. That’s how you automate your home devices to give you a desired welcome the moment you reach your home, with the right temperature and a cup of coffee. Simultaneously, you can set your shower to warm up the moment you reach home so you can have a warm shower.
2. Morning ritual - Let’s consider an early morning scenario, you have an alarm to wake you up which you will probably snooze once or twice before finally ending it, say at 7:30 AM. You have probably set your coffee machine and toaster to start preparing your breakfast at 8:00 AM and your shower to warm up at 8:30 AM. Next, you probably want your smart speaker to play some workout music to motivate you to flex your muscles, followed by giving you some daily news.
3. Getting a good night’s sleep – In order to get a good 8 hours of unperturbed sleep, you would have to sleep by say 11:00 PM. Accordingly, you might want to set up some sleep-inducing music to play in the background for a pre-defined time. Set up your room temperature to something optimal for peaceful sleep and probably want to turn your phone to sleep mode. All done individually.
4. Household choirs – Fortunately, there’s plenty of devices to support you in doing your household choirs efficiently. From a robotic vacuum cleaner to a robotic mop, you can have them all. Suppose you decide to clean your house and clothes during weekends. Thus, you set up your washing machine and vacuum cleaner to a convenient time for its operation to start. Further, your robotic mop can decide by itself whether your floor needs a dry mopping, wet mopping or damp mopping.
5. Monitoring your home – With the innovatory technologies in the field of home security solutions, it has become rather simpler for you to go for long vacations, leaving your house in the safe hands of smart security cameras, object detection, and tracking devices which would allow you to monitor your house from far away and get notified in case of any odd activity. Remote Elderly, Baby care too has been made possible with the monitoring solutions based on RADAR.
Similar to the above scenarios, there are millions of ways IoT-enabled devices can make our lives simpler. But things can get haywire and messy the moment you have changes in your schedules. For instance, the moment you decide to come home an hour late or get stuck in traffic, you would be greeted with a cold coffee, cold shower and a thermostat running for an hour extra for no reason. The good news is… all the above-mentioned scenarios can unfold without needing any direct input from you. Here’s how –
Living Made Smarter…
With the help of an intelligent and interconnected system, the need for human intervention further decreases, thus, reducing the scope of human errors. When devices are interconnected and share information about the status of each of their activities, you no longer have to individually set up every device to start operating at a specific time. For instance, you no longer have to worry about being greeted by a cold coffee, as your phone is already capturing real-time data about route traffic and simultaneously alerting your coffee machine of the same. Thus, however late you will be, your warm coffee will be there to greet you.
For each device to operate efficiently, they are required to be programmed intelligently. Here are the steps in the role of Machine Learning/Intelligence…
Machine Learning, the field of computer science which focuses on developing algorithms to enable systems to independently learn and continuously adapt with the help of data without being explicitly programmed for the particular data. Primarily, the inclusion of machine learning in an IoT system will enhance the following functionalities:
- 1. Optimization and Automation process
- 2. Extraction and Classification of data
- 3. Detection, Analysis and Prediction of patterns/trends
- 4. Enhancing interaction with humans and ambience
Let’s take another scenario to have a better understanding of the impact of ML in IoT.
In the instance of Remote Home Monitoring, suppose you have installed a smart CCTV camera outside your house to monitor who is entering the house and a Radar based Elderly Monitoring solution to keep an eye on your family member. The intelligence of your CCTV camera can be enhanced if it’s taught to identify bizarre activities and accordingly turn on alarm to alert the home owners. An elderly monitoring solution can be made smarter with the help of machine learning which can help differentiate between a fall and purposeful lying down, also between resting and being motionless due to trauma/shock.
Well, the above described smart home scenarios also applies to smart restaurants, offices, hospitals etc.
It’s imperative to mention the risks involved in an intelligent, connected world. Data Privacy, Physical Attacks, Grid Collapse are some of the threats keeping the IoT Security experts busy. So, the path to an Intelligent, IoT-enabled perfect world isn’t a bed of roses and involves a number of planning and preparations. Every new technology has its highs and lows. As for IoT + ML world, we can only hope that the pros exceed the cons.
- 4 ways Artificial Intelligence (and Deep Learning) is transforming automotive industry
- Industrial IoT, transforming the way modern warehouses operate
- How to Integrate Industrial Robots and Machine Vision in Manufacturing: Beyond Industry 4.0
- Role of Machine Vision technology in Packaging Industry
- Developing Deep Learning Algorithms on an Embedded Platform
- Build next-gen product with computer vision