More than just a 3-D rendering, it will be an intelligent representation. Want to see just the roof surfaces on buildings 10 stories or taller? That will happen. “It may look like just another video image, but the platform actually knows the three-dimensional position of the point in space,” said Ronnie Lee, deputy director of Singapore’s Geospatial Specialist Office. “It can give you a 360-degree point of view from any position, so it becomes a kind of virtual drone.”
The $73 million project launched at the end of 2014. Initial rollout is slated for 2017 with completion expected in 2020. The effort is supported by Singapore’s National Research Foundation in collaboration with French multinational software company Dassault Systèmes with support from various government agencies.
To build the base imagery, the Singapore Land Authority used aerial light detection and ranging (lidar) to image the city, while vehicle-mounted lidar recorded details at ground level. The geospatial team uses the GML standard to define the elements of the city and employs FME software for data conversion.
Possible uses
As a research project, Virtual Singapore is somewhat open-ended in terms of specific future implementations. Planners are working with diverse government agencies to see how they might best make use of the simulated city.One area of interest involves the rise of the Internet of Things and smart cities communications. In a densely packed urban environment, where it may be difficult to make the needed connections between sensors and other devices, a good 3-D visualization could help urban architects deploy those resources effectively.
“We could perhaps create a map of where the urban canyons are, those places where your GPS may not be able to receive proper satellite signals,” Lee said. “If we are going to have autonomous vehicles, we are going to need communications devices on lampposts, but those will be affected by the buildings around them. With this map we can study mitigation measures.”
A virtual city might also help transportation authorities to do planning in richer, more complex ways. “We are doing a traffic simulation and instead of looking at two-dimensional dots, we are looking at virtual cars or trucks or taxis. Then we visualize the traffic from the sky and from the pedestrian view, or we can add in pedestrians or bicycles and scooters,” Lee said.
“Putting them all into the same model allows us to assess their behaviors and think about how we can govern the different uses,” he said. “We can develop tools that allow planners to select an area and to put together things that are usually looked at separately. So you get a more complex analysis.”
The research team already is implementing one practical application for the virtual city, using it to explore the impact of sustainability efforts in Yuhua, an older housing estate that the city has identified for green-style upgrades.
“We went to the site and looked at the solar panels and modeled them in Virtual Singapore. Then we worked with local authorities to get data from energy bills," Lee said. "When we put that information together, we can show how much energy is captured by the solar panels for the day or the week or the month. We can show how the energy powers the common areas and visually demonstrate the positive advantages of deploying solar panels.”
This helps to drive community buy-in for the sustainability push, while also validating the city’s investments. Planners also are using the tool to draw hypothetical deployments of bike racks around the community, to solicit community input as to where these might be mounted most effectively.
The biggest challenge thus far has been in determining the level of detail that ought to go into the virtual rendering.
“We want this to deliver on an Internet Web browser so that many people can have access. But what you can deliver through the Web is sometimes limited,” Lee said. It’s possible to model every door, window, lamppost and tree, “but the more you put in the more costly it is and the harder it is to maintain. So you have to make choices about which details to include.”
Crowdsourced buses
Even as Singapore officials wait for the virtual map to be completed, they are pursuing other smart city initiatives meant to improve the quality of life on this crowded island.Take for instance Beeline, a smart mobility micro-transit platform developed by the city’s GovTech office in collaboration with Land Transport Authority and launched in late 2015. Bus riders can use the app to pre-book travel and even to propose new bus routes.
If 10 riders each buy five days' worth of tickets on a proposed route, the city will work with a private bus company to launch that new route.
The idea is to give the city a new ability to shift on the fly, based on actual consumer demand.
“Transportation is usually quite top down. I know where the metro lines need to go; I need to know where my bus lines need to go. You do broad brush strokes and you have bus lines that never change for years,” said Feng-Yuan Liu, director of the Data Science Division in Singapore’s GovTech division.
“With metro lines, you can’t rip it out and change it, but with bus lines we can be a lot more flexible,” he said. “It’s not going to change day by day, but over months and weeks, you want your bus lines to evolve and to adapt, almost like an evolutionary, biological approach.”
The app user base includes 5,000 active riders and 100 bus drivers. GovTech reports it has sold 37,000 tickets through Beeline.
In addition to communicating with riders and bus lines, the Beeline team also has been working with local businesses. For cities looking to get smarter about public transportation, input from big employers can be a significant guidepost, Liu said.
“We work with them to better understand where their employees live and what kind of transportation needs they may have,” he said. “This way the bus service can be more closely aligned with that actual need.”
