An American start-up is about to start testing its electric driverless taxis in Singapore to help the city-state reduce congestion and fight pollution.
The company, nuTonomy, is a spin-out from the Massachusetts Institute of Technology, which has previously tested driverless golf-carts to ferry tourists in one of Singapore’s public gardens.
Unlike its high-profile rivals Google, Apple or Tesla, nuTonomy is more open about the technology behind its vehicles. Its major advantage, the firm believes, is sophisticated fleet management software based on algorithms developed previously for the US military to coordinate large swarms of drones.
The algorithms, which mimic human decision-making, would allow the fleet to work so efficiently that only 300,000 taxis would be needed in Singapore instead of the currently operated 780,000.
"That's a 60 per cent reduction in the number of vehicles operating in Singapore," said Emilio Frazzoli, an MIT professor of aeronautical and astronautical engineering and nuTonomy’s co-founder and chief technology officer, who led the research from the start.
"This was a big sign of impact for the Singaporean government. At first we were asking them to let us test cars there – then they were asking us to come test."
According to calculations that Frazzoli’s team carried out, this halving of the number of taxis wouldn’t have an adverse effect on waiting times – every passenger would be able to get his or her car within 15 minutes. However, it would considerably alleviate the city’s traffic congestion and reduce pollution, especially as the cars are electric.
The algorithms driving the fleet management system use the so-called formal logic, which enables them to decide in every situation in the most flexible way.
For example, when driving around a double-parked car, with no oncoming cars, the taxi will see it's not violating the most important rule – not hitting another object – and pass.
"These are situations we encounter everyday, and we use our judgment to understand the rules we can violate," Frazzoli said. "We have these same judgments embedded in our algorithms. That allows us to exhibit complex behaviour and respond in correct way in very many and very complicated scenarios."
The cars are also equipped with lidar to determine more accurately their exact position in the environment. nuTonomy has designed the lidar not only to detect objects on the road but also everything around.
"Even though stuff at road level can change all the time – you can have a car parked here or not, for example – a building is going to stay put," Frazzoli said.
It’s not just a quirk. Having a wider view can make the difference between being able to continue the autonomous journey or not in difficult conditions. When the researchers tested the vehicles in Michigan last year, the all-observing lidar enabled them to continue in the middle of a powerful snowstorm that hid everything on the road.
"There was nothing visible on the ground, but who cares: the car was looking at the buildings," said Frazzoli.
Last week, nuTonomy passed first driving tests in Singapore, navigating its driverless taxi through a custom obstacle course without incident. The company now plans to start trials in the city’s One North business district, which has been earmarked as the testbed for Singapore’s driverless car revolution.