CAPE TOWN – The Fourth Industrial Revolution (4IR) is lead by two particular streams of technology that will considerably transform our world: artificial intelligence (AI) and fifth generation (5G) telecom technology. The impact of these two technologies will be felt on almost every aspect of our society.
One of the promised advantages of 5G is that it will seamlessly work with wireless Internet of Things (IoT) sensors, vehicle-to-vehicle communication technologies, smart traffic lights, smart energy grids, augmented reality, mobile wearables, smart home devices and other cutting-edge technologies.
Since IoT applications typically need to gather huge amounts of data from a large number of devices and sensors, the technology necessitates an efficient and fast network for data collection, processing, transmission, control and real-time data analytics. 5G's high speed, lower cost, flexibility and eventual ubiquity, make it a strong choice for IoT networking.
But when 5G technology with mobile internet speeds up to 20 times faster than today’s speeds, is combined with the fast advancing technology of AI, it leads to a new potentially game-changing era of technological development or “Fifth Generation Intelligence”.
For instance, AI together with 5G, has the potential to make smart cities a reality. 5G's high speed and ultra-low latency is vital for supporting smart city applications like public transit scheduling, smart streetlights, real-time crime detection and reporting, and IoT sensors that monitor things such as parking spaces, traffic flow, air quality, water use, sewers and trash collection.
Yinchuan, an isolated Chinese city on the outskirts of the Ordos Desert, is China’s example of their ambitions to lead the world in 5G technology. Since 2017 Yinchuan has implemented high-definition CCTV cameras capable of facial identification for public-security monitoring. With the benefit of AI, the city also deployed smart street lighting with integrated electric-vehicle charging and advertising, and smart traffic lane management on the city’s highways.
But it is Yinchuan’s facial-recognition-enabled payments for public transport and their smart trashcans that really catch one’s imagination. Through the use of AI and 5G, facial recognition makes the convenience of ticket-less travelling possible and prohibits illegal travelling on public transport, while constantly monitoring the security of the commuters.
The smart trashcans are just as innovative. As many other cities in China, Yinchuan had a problem that deserted buildings were often used as trash stations where residents dumped their garbage. These unofficial trash stations became a serious health hazard since they were mostly overflowing and not serviced. This is why the city developed smart trashcans. The moment the trashcans are full, they send a signal to relevant department of the municipal authorities, and municipal workers promptly turn up in a truck to clear the waste.
At the hart of Yinchuan’s vast efforts is a municipal government command centre where continuous data streams from numerous sensors and CCTV cameras spread across the city. This data is filtered, sorted, analysed, prioritised and then acted upon. It emphasises the importance of big data analytics in the era of the 4IR.
The success of the smart city of Yinchuan is significant and is probably due to the joint ventures between the government, city and private telecom firms. China is well known for the close ties between authorities and technology companies since technology is central to China’s future strategy. But the extremely fast implementation of 5G intelligence could also be ascribed to low levels of land-use regulation, easy access to capital, and a strong manufacturing industry.
Due to the ever growing population and traffic in cities, AI and 5G are expected to particularly make a difference in the integration of a smart transportation infrastructure, a network built on the ability of multiple devices and sensors to communicate with each other quickly and efficiently in real time. Ubiquitous connectivity, fast transmission speeds, and low latency of 5G technology, as well as AI automation technology, are crucial for the implementation of a smart transportation system that allows faster and smarter commutes.
5G would allow vehicles to communicate with other vehicles and even traffic signals in real time, facilitating driverless cars, pedestrian and cyclist warnings, and real-time rerouting to avoid accidents or traffic congestion. It is generally expected that such an intelligent system will significantly reduce road accidents, as well as save commuters time, fuel and money. The lower probability of road accidents will in turn lead to much lower car insurance rates.
However, the most important contribution of a transportation system using AI and 5G will be to help employees get to work faster. Traffic flow is presently mostly managed by traffic lights that are programmed according to time schedules and feedback from inductive loop technology on the road that can detect metal in vehicles. Light signal time schedules are devised to coordinate lights from one stop to the next so that cars will experience a synchronised set of green lights and a smooth flow of traffic. Unfortunately, in reality, many transportation systems in cities rely on out-dated schedules (often due to unavailability of reliable data) and old technology, resulting in poorly timed lights, which causes a longer commuting time, higher costs, a loss of productivity, and frustration.
AI and 5G technology could really make a difference to the flow of traffic by providing real-time information to modern smart traffic lights about current traffic patterns as detected in real-time by CCTV cameras and IoT sensors dispersed throughout smart cities. It is even possible to use drones in rush-hour traffic to communicate real-time traffic data. Traffic lights could thus be more reactive to the real traffic situation, keeping traffic flowing and reducing unnecessary stops at red lights. A study on smart traffic light systems by Carnegie Mellon University in Pittsburgh yielded a 40% decrease in traffic stops, a 21% drop in emissions and a 26% reduction in travel time.
A further reduction in commuting time could come from smart parking systems, where the transportation system uses sensors to determine which parking bays are available and direct the traffic accordingly. Los Angeles has gone one step further and uses dynamic pricing for its parking spaces on public streets and city-operated parking garages in downtown Los Angeles. The demand responsive pricing model means that prices vary according to the geographical area, time-band and whether it is a weekday or weekend. The parking place on a particularly busy and popular downtown street could be much more expensive than the parking place a few blocks away in a not so busy area.
What is nifty about the system is that it is based on sensor data and directs the flow of traffic. It prevents drivers from circling and clogging the roads, thus wasting time and fuel. As soon as parking place occupancy reaches 80% in a given area, the price will increase and if it drops below 60% the price decreases. Ultimately the goal of demand-based rates is to get drivers off the streets and into a parking space as quickly as possible. Drivers can plan their commute by using the real-time map on the LA Express Park website that shows the prices for a given parking space at a given time. Parking meters and pay kiosks also display the rate for the specific time and place.
The city of San Francisco also uses a demand-driven parking system very similar to the one in Los Angeles and made the wonderful discovery that when parking becomes easier with better traffic flow and less time wasted, the economy benefited tremendously. Sales tax collections increased with 35% comparing to the 20% in other parts of the city.
But alas, since the South African Department of Telecommunications and Postal Services is dragging its feet with the digital rollout of TV and radio and reassigning the frequencies to 5G, we will have to enjoy the comfort of our dream cars while stuck in the traffic for many more years to come.
Professor Louis Fourie is the deputy vice-chancellor: Knowledge & Information Technology – Cape Peninsula University of Technology. The views expressed here are his own.