Technology plays a crucial role in empowering smart citizens, strengthening smart societies, building smart cities and, ultimately, becoming smart nations.
What does the word smart mean, particularly when it comes to a smart home and a smart city? It is really confusing because smart could mean different things to different people. There are smart devices, smart light bulbs, smart meters, smart homes, smart citizens, smart society, smart cities, smart nations and so on. For example, some people think that a smart device should have the ability to communicate with other devices through a network like the Internet, while others think that it should have the capability to learn.
We often come across the word smart while talking to our friends, colleagues, watching TV or reading. It seems almost everything has become smart. Does this mean the world is becoming a smarter place? Yes, application of smart solutions, and use of smart devices and digital technologies have made the world much smarter than before!
Let us take the example of lights in a hall. You manually set the brightness level of the light at various times of the day to reduce energy consumption. Do you think controlling the light with a remote is smart enough? What about making the device learn, so it can figure out what brightness level you want and when? A smart light can adjust the brightness automatically any time of the day, depending on the brightness level of the Sun and surrounding areas, to save energy.
If you want to change the brightness to zero per cent at, say, 10PM, the device figures it out by itself and does it for you. That is a smart device. Experts say that everything that is intelligent is smart, but not vice-versa. This article covers the use of artificial intelligence (AI) and smart devices in smart homes and smart cities.
Smart home technology is also referred to as home automation that provides convenience, comfort, security and energy-efficiency through the use of smart devices including smartphones, sensors or other networked devices. The first general-purpose home automation network technology, X10, was developed in 1975. It is a communication protocol for electronic devices that uses electric-powered transmission wiring for signalling and control. A transmitter sends a signal along the house’s electric wiring, telling the device to turn on at a specific time.
Household electrical wiring is used to send digital data between X10 devices to control lights and other electrical appliances. Digital data consists of an address and a command sent from a transmitter to a device. This data is encoded onto a 120kHz carrier, which is transmitted as bursts during the relatively-quiet zero crossings of 50Hz or 60Hz AC waveform. One bit is transmitted at each zero crossing. X10 controls in one house do not interfere with X10 controls in a neighbouring house. Inductive filters are used to attenuate X10 signals coming into or going out of the local area.
AI in smart homes
X10 was initially a one-way technology, so while smart devices can take commands, these cannot send data back to the central network. But later, two-way controller devices were introduced that allow for a more robust network. As technology advanced, RF protocols were defined to allow for wireless keypads, remote switches, motion sensors and so on. The wireless protocol operates at a frequency of 310MHz in the US and 433.92MHz in European systems.
While X10 devices are still around, other technologies have emerged to compete for home networking. Instead of going through power lines, many new systems use radio waves to communicate. Now, we are witnessing technologies like the Internet of Things (IoT), Bluetooth, Wi-Fi and cellphones in home automation.
A wireless network provides more flexibility, convenience and comfort at finger tips. With such emerging technologies as AI, home automation has become smarter and more advanced. Integration of AI has redefined the role of home automation and added new features including home security and energy-efficiency.
How AI-based smart home systems work
All devices in an AI-powered smart home system should be able to communicate with each other and continuously acquire new data for further learning. A truly intelligent smart home device employs the learning system—a multi-layer system capable of making decisions based on historical and real-time data, which requires little to no user control.
For instance, when you leave a room either to go to the bathroom or the balcony, the system should be able to identify your actions and movements, and issue appropriate commands to other devices within the network, either to turn off the lights in that room and turn on the lights in another. The goal of smart home devices is to bring down manual settings to zero. Using AI and machine learning, you can significantly minimise the inconvenience for connected home-owners. Let us take a look at some applications of AI in smart home automation.
Location-based smart devices
Smart home devices can be remotely controlled through smartphone apps and preferences can be auto-managed. For instance, when you are approaching your home and are within a given radius, the system can set up your room with your own preference of lights, curtains and AC temperature. These features add to convenience and comfort.
Voice-enabled devices
Smart devices can be linked with such devices as Google Assistant and Amazon Alexa to provide voice-control options across all home automation systems. Voice technology can be used to control lights, make voice or video calls, read news, place orders, play music and so on. This feature can be used during a medical emergency in notifying family and friends for immediate attention.
Smart security systems
Security systems include facial recognition, motion detection, voice recognition and biometric access control.
Facial recognition
Smart home devices include Wi-Fi-enabled devices with video feed security features. You can remotely access these features on a smartphone from any location. Smart systems incorporate facial recognition technology into connected video cameras. AI-based neural networks can identify eyes, cheekbones, nose and chin of a visitor through the relevant camera and compare with the faces of the home’s residents and send notifications about suspicious activity to the home-owner’s smartphone.
Motion detection
Motion sensors can detect random movement through video surveillance like detecting intruders and warning home-owners.
Biometric access control
Biometric door locks such as August Smart Lock can be seamlessly integrated into Amazon and Google smart home systems. These usually take photos of the user’s fingertips via optical scanners and store them for matching, or use two-factor authentication using fingerprints and passwords.
Voice recognition
AI-based voice recognition employs Natural Language Processing (NLP). This technology isolates a person’s voice from background noises, converts the audio to a digital file, sends it to the cloud for NLP analysis and triggers an appropriate action. With machine learning, it is possible to teach a system to recognise a variety of different terms or messages.
The term voice recognition or speaker identification refers to identifying the speaker rather than meaning of the voice or interpretation. It can be trained on a specific person’s voice, or used to authenticate or verify the identity of the speaker as part of a security process.
Optimising energy consumption. Smart home automation products provide functionality beyond controlling devices or notifying warning signals—these also optimise energy performance. With integration of AI and machine learning, smart devices offer energy consumption patterns to help users limit the usage of devices and appliances when not required.
Energy saving and efficiency are hot topics due to climate changes and energy challenges globally. AI-powered smart devices can reduce the carbon footprints by controlling many home automation systems including smart thermostats, smart plugs and automated lighting sensors.
Challenges in smart home systems
While there are many benefits of smart home devices, there are also challenges to fully implement a truly smart home system. This is because, to set up a secure, scalable and highly functional connected home solution, you need to integrate smart systems created by different vendors. This brings interoperability issues. However, some manufacturers are trying to bring down the differences and interoperability issues. For example, Z-Wave allows connected home manufacturers to integrate their devices with multiple smart home solutions.
But there are still many products that are not compatible with each other in one way or the other. That is, a Wi-Fi-enabled device cannot simply brew a cup of coffee of your choice unless you combine a coffee maker with WeMo and Google Assistant or Amazon Echo.
An AI-driven home automation system certainly offers peace of mind, convenience, home security and energy-efficiency. However, higher initial investment cost of the new technology causes a lot of hesitation to many consumers.
Another important aspect is security issues. Smart systems need to have a good firewall to prevent hacking, and to build a highly secure mechanism to guard against privacy and loss of sensitive information. But note that no security system is fool-proof. For example, facial recognition may work for now but may not work after twenty years, because your face may not be recognisable by then. Besides, it is not very accurate.
What a smart city is
The concept of the smart city lies in the use of technology to facilitate new experiences for its inhabitants to make living more comfortable and secure, among others. The definition of a smart city varies from city to city and country to country, depending on the requirements and aspirations of the city residents.
According to forbes.com, the top ten smart cities in the world are London, New York, Amsterdam, Paris, Reykjavik, Tokyo, Singapore, Copenhagen, Berlin and Vienna. These cities have been analysed across nine key points, including human capital, social cohesion, economy, environment, governance, urban planning, international outreach, technology, and mobility and transportation.
Smart city features include smart urban planning, smart housings, creating walkable localities and reducing traffic congestion, reducing air pollution, preserving and developing open spaces (parks, playgrounds and recreational spaces), promoting a variety of transport options, making citizen-friendly location, providing adequate infrastructure and more.
AI in a smart city
AI and machine learning combined with the IoT have the potential to address many key challenges faced by the urban population, including water management, energy crisis, traffic management, healthcare and businesses.
Smart water management
When we shift from rural to urban cities, some of the important requirements are water, food, shelter and lights. Water and food are basic amenities for living beings. Technology helps us get our basic amenities and minimise the cost of living in cities. With AI, it is possible to streamline the use of water and power in a city. Smart grid technology can be employed to better manage the use of power and water metering to locate leaks and prevent excess use of water.
Smart lighting
Everybody wants streets to be properly lit. While streetlights are necessary, these consume a lot of energy. Use of smart lighting can certainly reduce consumption of energy by allowing lighting only those areas when needed.
Lamps can be used to adjust the brightness based on the presence of pedestrians, cyclists or cars. It is also possible to employ a real-time mesh network to trigger neighbouring lights and create a safe circle of lights around a human occupant.
Lamp posts with additional sensors can be employed for security applications or serve as Wi-Fi network hotspots. Thus, apart from smart grid metering, there are other smart ways to fully utilise energy in an optimised way.
Smart traffic management
Modern urban cities are crowded with vehicles. AI-based smart traffic solutions can be employed to ensure people moving from one point to another as safely and efficiently as possible. Some congested cities in the world have adopted smart traffic solutions to control the flow of traffic. Road-surface sensors and CCTV cameras are installed to send real-time updates about traffic flow to a central traffic management system. Data feed from cameras are analysed and users are notified regarding congestion and traffic signal malfunctions.
Smart parking
Solving the traffic congestion issue is not enough; finding a parking slot especially during holiday time is a real struggle. Smart parking solutions can determine whether parking slots are free or occupied. Smart parking creates a real-time parking map by using road surface sensors embedded in the ground on parking area. This can reduce the time to wait, to find an empty slot for parking, which, in turn, helps in reducing congestion and pollution.
Smart waste management
Waste cannot be avoided but it can be properly managed. Waste management is an essential city service. With the ever-increasing urban population, there is a need to adopt smart methods for waste management. Use of AI technology not only provides smart waste management but also provides a sustainable waste management system. For instance, smart sensors fitted on waste bins send notifications to authorities to dispatch waste collection trucks as soon as they are about to be full.
Separate bins for paper, plastic, glass and waste food items should be maintained. These waste materials can be recycled and used later for other useful purposes.
Smart policing
Nobody wants to live in a crime-ridden city. Smart policing ensures law enforcement agencies employ evidence-based and data-driven strategies to effectively and economically control in an efficient way. For example, a network of cameras and sensors can be installed in every corner to help identify people who commit crime, smoke in prohibited zones or seem suspicious. Cameras can be used to monitor crowd density, vehicle movement and cleanliness in a particular area, among others.
Smart governance
Smart city infrastructure is not complete without smart governance. Smart governance implies the use of information and communication technology intelligently to reduce corruption in governments and public administration organisations. The main goal is to improve decision-making through better collaboration among different stakeholders, including government and citizens. Smart governance uses data, evidence and other resources to improve decision-making and compliance for citizens.
Smart nation
Smart Nation concept was launched by the government of Singapore in 2014. Its main goal is to harness information and communication technologies, networks and Big Data tech to create tech-enabled solutions. AI plays an important role in the journey towards creating a smart nation. Currently, Singapore has identified five national AI projects—transport and logistics, smart cities and estates, healthcare, education and safety and security. These projects can bring about many changes through the creation of new solutions to almost all the challenges faced today.
Digital technologies and AI are going to revolutionise the modern world in many ways including creating a smart society and smart city. Smart society, in modern context, is where people in a particular geographical area get all benefits and solutions through the use of technology, including easy transportation, protection, security, basic amenities and social well-being. A smart nation is ideally made up of smart societies and smart cities.
Conclusion
This article covered the importance of AI technology in smart homes and smart cities. Smart home refers to a modern home with all electrical appliances and devices controlled from a remote distance often via mobile apps and other smart devices that can operate interactively and autonomously. Smart cities employ technology in a way that positively impacts their inhabitants. Technology plays a key role in empowering smart citizens, strengthening smart societies, building smart cities and becoming smart nations.
