The Role of AI and IoT to Establish Smart Mobility Environment
What are AI and IoT?
The Internet of Things (IoT) and Artificial Intelligence (AI) are recent hot topics in the world of technology. But before we start talking about smart mobility, let’s find out what those technologies mean.
Oracle.com defines IoT as “The network of physical objects — “things” — that are embedded with sensors, software, and other technologies for the purpose of connecting and exchanging data with other devices and systems over the internet.”
Which roughly means devices are connected and talking to each other via the internet. The capability of the connected devices creates a new dimension and a new perspective on how to solve problems. From simple things such as turning on/off lights to more complex implementation like self-driving cars.
On the other hand, according to jmc.stanford.edu, Artificial Intelligence is “The science and engineering of making intelligent machines, especially intelligent computer programs. It is related to the similar task of using computers to understand human intelligence, but AI does not have to confine itself to methods that are biologically observable.”
Based on the explanation, we understand that devices can exchange their data via the Internet. But, what is the outcome of that process? How can it help humans’ problems? This is where the combination of AI and IoT comes to play. The billions of raw data collected from the connected devices and sensors can be processed further by AI, turning it into something more “useful” for humans to make decisions.
What is the problem?
The data generated from sensors can vary from speed, image, temperature, pressure, current, voltage, etc. People may think that sensors and other IoT devices are expensive. In fact, you can buy a microcontroller with Wi-Fi capability with only less than $5. While for the AI side, a huge number of researches and papers are available for free on the internet. Basically, the sky is the only limit for AI and IoT technology development.
Let’s get back to the topic. Talking about problems that can be solved by AI and IoT alone will take forever. We will start by narrowing it down to specific problems that happen every day in most of the big cities in the world, especially here in Jakarta, Indonesia. The problem that has been a pain point for every citizen around the world, traffic congestion.
Tomtom.com, a Dutch-based developer & creator of location technology and consumer electronics, revealed that people in Jakarta lost 174 hours or 7 days plus six hours per year on average during rush hour due to traffic congestion. It means that people in Jakarta spend one extra week only for driving! Also, people in Jakarta didn’t only lose their time, but also their money, a lot. Based on the World Bank’s data, traffic congestion in 28 areas in Indonesia caused US$4B loss, where US$2,6B of it was contributed by Jakarta.
What is the solution?
When it comes to solving traffic congestion, the most obvious and significant way is by seeing the core problem, vehicles. If we can reduce the number of cars, motorbikes, trucks, and other vehicles, traffic congestion will also be reduced. It is simple, isn’t it?
But… Is it feasible and easy to be implemented? I guess not. The simplest example can be seen from the car purchase in Indonesia. Detik.com wrote that although the number of car purchases is decreasing in recent years (1.026.921 cars in 2019 compared to 1.151.413 cars in 2018), it is still a huge number. More than 1 million cars are bought by the Indonesians every year. In conclusion, the number of cars is always increasing, making the car purchase limit is not feasible.
So what can we do? To solve this congestion problem without reducing the number of cars or vehicles, we need to see it from a new perspective. Another thing that causes traffic congestion is traffic lights. Sometimes, the duration of the traffic lights does not match with the real-time situation on the intersection. For example, the green light duration is longer for lanes with less traffic, while red light duration is longer for lanes with more traffic. It caused a long queue of vehicles on one side of the road, while there are fewer vehicles on the other side of the road.
Most of the time, it also causes havoc in most of the intersections in Jakarta. To overcome the problem, Qlue develops a smart solution called Dynamic Traffic Controller (DTC). The main principle of DTC is to convert a conventional traffic light into a smart traffic light using a smart module powered by AI and IoT technology
The working principle is as follows:
DTC utilizes the stream from on-site CCTV streams which pointed to the road. The stream is then transmitted to an AI engine. In this case, we are using a vehicle counting engine, which is an AI engine that will process lots of vehicle-image-data and use specific algorithms. The engine will learn to distinguish and classify vehicles that are on the user-defined frame. After that, we can count or determine the density of the vehicles in the frame. Finally, the DTC device will get the data and process the result by dynamically adjusting the traffic lights duration.
Conclusion
Traffic congestion is a problem that a lot of people face every day. It doesn’t only cost us money, but also our precious time that we can use for more important activities, rather than sitting in your car in the middle of traffic congestion. With DTC devices and AI engines, we can reduce one of the traffic congestion causes while increasing the effectiveness of traffic lights.
But DTC is only one of a million examples of how AI and IoT can be implemented together to solve mobility problems. And this “dynamic duo” is not only limited to solve mobility problems, but it can also solve other problems like health, air pollution, security, and many more.
