Autonomous taxis get closer to reality as tests set to begin

MIT spin-out hopes to bring driverless cabs to Singapore to reduce traffic

NuTonomy: to begin testing self-driving cab service this year

NuTonomy: to begin testing self-driving cab service this year


The passenger takes a smartphone, calls the cab that picks them up.

You sit back and it drives you there. It’s a simple notion but one that could mark a fundamental change in urban mobility. Did I mention there is no driver?

The future vision of the autonomous taxi service is getting closer to reality in the city-state of Singapore where a spin-out of the Massachusetts Institute of Technology (MIT) is hoping to test the world’s first self-driving cab service this year.

“Our goal is for this pilot to grow into a commercial autonomous taxi service available across Singapore. This will require both technology and regulation to develop in parallel,” Doug Parker, chief operating officer for the company, nuTonomy, said by email.

Seed funding

United StatesEuropeGoogleUber

NuTonomy secured $3.6 million (€3.19 million) in seed funding in January and it hopes to provide a more convenient form of public transit while helping reduce greenhouse gas emissions in the densely populated city-state by using electric cars.

Co-founders Karl Iagnemma and Emilio Frazzoli, nuTonomy’s CEO and CTO, were doing mobility focused robotics research at MIT.

“This could make car-sharing something that is almost as convenient as having your own private car, but with the accessibility and cost of public transit,” said Frazzoli, a professor of aeronautical and astronautical engineering, who was recently part of an MIT experiment in Singapore which set up autonomous golf carts to ferry tourists around a park for a week.

‘Formal logic’

The self-driving taxis have algorithms for a “formal logic” that tells taxis when low-priority “rules of the road” can be broken in everyday situations.

For example, if the car drives around a double-parked car, with no cars oncoming, the taxi will recognise that it’s not violating the most important rule, of not hitting another object – and pass.

It also uses LIDAR data for better localisation, by detecting not only objects on the road but also stationary objects all around the car.

The scientists believe that 300,000 driverless taxis could do the work of the 780,000 privately-owned cars currently operating today in Singapore, while keeping waiting times below 15 minutes.

“That’s a 60 per cent reduction in the number of vehicles operating in Singapore,” Frazzoli told MIT News. “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.”

Among the innovations in the system is advanced fleet management, which derived from Frazzoli’s previous work: writing algorithms to co-ordinate swarms of drones for the US military.

Using similar concepts, the nuTonomy engineers designed algorithms to allow the minimal number of cars to cart people around a city, alleviating traffic congestion and reducing emissions.

The autonomous taxi successfully navigated a custom obstacle course, and is now getting approval for a small-scale pilot study of a fully autonomous on-demand mobility system in One North, a business park in Singapore near the national university.