Tag Archives: taxi

Sharing taxi rides could significantly reduce cost and emissions. Study suggests it can work in virtually any big city

By using ride-sharing algorithms, scientists have found there’s an immense untapped potential for reducing cost and emissions. Their findings suggest there’s a universal mathematical framework that can be applied to almost any city to match taxi rides. Two or three people could use the same cab instead each traveling solo with minimal delays to their destination, the researchers found.

credit: Pixabay

Almost every sizable city in the world is facing problems due to urban transportation. For one, there’s traffic which is the bane of any commuter, and then there’s the issue of air pollution. Municipalities have always found it challenging to address these problems in their urban planning partly because it’s genuinely difficult to convince people not to use their own cars. But almost everyone uses taxis or, as of more recently, ride-sharing apps  — the problem is most of the time the rides are taken solo.

Established ride-sharing companies such as Uber or Lyft already offer options to customers to share their journey but these are heavily outnumbered throughout the world by traditional taxis. Researchers led by Carlo Ratti, an architect and engineer at MIT, wanted to see what kind of impact ride sharing among taxis could have. They chose to work with GPS data sourced from taxis working in San Francisco, Vienna, Singapore, and Manhattan. All four urban areas have different layouts which proved useful in teasing out universal patterns.

The researchers designed an abstract network of trips which were connected if these could be shareable (no delay to either party of more than 5 minutes). The researchers then forecasted what would happen to the number of shareable rides if the number of taxis was to drop in any of the four cities. They found that despite lower average demand for mobility or fewer available cars, shareability followed a universal law with the same feasibility-demand curve exhibited in all four cities.

This trend suggests shareability remains high even if demands drops and mass adoption of ride sharing could theoretically be possible in any sizeable city, not jus the four included in the study published in Scientific Reports.


“Although these four cities superficially look different, their shareability curves look the same,” says Steven Strogatz, a mathematician at Cornell University in Ithaca, New York, and co-author on the latest study. “It’s amazing to me that it works as well as it does.”

Previously, Ratti and colleagues found most yellow cab rides which start and end in Manhattan could pick up more than one customer in the 5-minute shareable ride threshold. After they crunched some numbers based on GPS data the researchers found ride sharing could slash the total number of miles driven by yellow cabs by 40% and emissions by just as much.

“If people share more rides, traffic will be reduced because vehicle occupancy will increase. With less traffic, travel time will decrease, increasing the potential for further ride sharing,” Ratti said.

This model, however, does not take into consideration a wide range of social and economic factors. If people can share a taxi ride with a stranger, it doesn’t necessarily mean they would do it — not without an incentive. This incentive could be a much lower cost for the taxi ride or very tight safety regulations to ensure that people use ride sharing services with confidence. The study, however, only looked at abstract network connections so maybe the findings don’t actually describe a ‘universal’ law after all.

However, the results do suggest there’s a huge potential in ride sharing. The challenge is to develop an urban transport framework that can involve all stakeholders and distribute value. If taxi companies lose money, for instance, they will likely become against mass ride sharing. Likewise, if many taxi drivers lose their jobs, that’s also an undesirable effect.

The first self-driving taxis are here — and by “here” I mean Singapore

This Thursday, Singapore will become the first city in the world to use self-driving taxis.

The first ever autonomous taxis will be available in Singapore. Image credits nuTonomy.

The first ever autonomous taxis will be available in Singapore.
Image credits nuTonomy.

With huge players such as Google or Volvo toying around with self-driving cars on public roads, most of us probably expected that the future fully automated vehicles we’d zip around in would be developed by one such huge company. But, for a handful of Singapore residents, the future is now — and it’s brought by the small, autonomous vehicle software startup nuTonomy. The company will become the first to ever offer the public a chance to ride in autonomous vehicles, beating ride-hailing service Uber — which plans to offer a similar service in Pittsburgh — by a few weeks.

Their fleet will start out small, six cars right now, but will double by the end of the year. Their end goal is to have Singapore’s taxi system fully automated by 2018, NuTonomy officials said. This will drastically reduce the numbers of cars clogging Singapore’s congested roads. But there’s nothing to stop them from expanding the model to other cities later, they added.

Right now, however, the taxis will only operate in a 2.5-square-mile office and residential district known as “one-north,” with specific pick-up and drop-off locations. The users will have to receive an invitation from nuTonomy to use the service, and the company says dozens have signed up for the launch, with several thousands more expected to join within the next few months.

NuTonomy vehicles — modified Renault Zoe and Mitsubishi i-MiEV electric models — will be fitted with six sets of Lidar to help it navigate its surroundings, including one that spins on the roof for a full 360-degree field of detection. Lidar is a system very similar to radar, only it uses laser beams instead of radio waves. Two cameras installed on the dashboard will scan for obstacles and detect changes in traffic lights. And, if all these fail, there’s a human driver in the front seat ready to take the wheel in an emergency, and a researcher in the back monitoring the car’s computers.


The testing time frame is open-ended, said nuTonomy CEO Karl Iagnemma. Users will eventually start paying for the service, as more pick-up and drop-off points are added. He also said that company plans to test similar services in other Asian, European, and U.S. cities, but didn’t give any dates or places.

“I don’t expect there to be a time where we say, ‘We’ve learned enough,'” Iagnemma told the Associated Press.

Doug Parker, nuTonomy’s chief operating officer, says that their autonomous taxis could reduce the number of cars on Singapore’s roads by two-thirds, from 900,000 to some 300,000 cars.

“When you are able to take that many cars off the road, it creates a lot of possibilities. You can create smaller roads, you can create much smaller car parks,” Parker added. “I think it will change how people interact with the city going forward.”

NuTonomy was formed in 2013 by Iagnemma and Emilio Frazzoli, two Massachusetts Institute of Technology researchers working on robotics and autonomous vehicles for the Defense Department, and currently has offices in Massachusetts and Singapore. They received Singapore’s approval to test self-driving cars in one-north earlier this year, and it announced a research partnership with Singapore’s Land Transport Authority earlier this month. Pretty impressive for a 50-man strong company.

The team behind the autonomous taxis. Image via nuTonomy.

The team behind the autonomous taxis.
Image via nuTonomy.

Iagnemma says that they chose Singapore for their testing area because it has good weather, quality infrastructure, and drivers who tend to obey traffic laws — an ideal spot to test autonomous vehicles. Due to limited available space, Singapore’s also been on the lookout for non-traditional ways to grow its economy, so it’s been supportive of autonomous vehicle research. Delphi Corp., which is also working on autonomous vehicle software, was recently selected to test autonomous vehicles on the island and plans to start next year.

“We face constraints in land and manpower. We want to take advantage of self-driving technology to overcome such constraints, and in particular to introduce new mobility concepts which could bring about transformational improvements to public transport in Singapore,” said Pang Kin Keong, Singapore’s Permanent Secretary for Transport and the chairman of its committee on autonomous driving.

Users are reporting that nuTonomy’s taxis work like a dream. Olivia Seow, one of the riders the company selected, took a one-mile ride on Monday. She said she was nervous at first, then surprised as the steering wheel started turning by itself.

“It felt like there was a ghost or something,” she said.

The ride was smooth and controlled, she said, and she was relieved to see that the car recognized even small obstacles like birds and motorcycles parked in the distance.

“I couldn’t see them with my human eye, but the car could, so I knew that I could trust the car,” she said.

An Associated Press reporter took a ride on Wednesday and reported that the safety driver had to step on the brakes once when a car was obstructing the test car’s lane and another vehicle, which appeared to be parked, suddenly began moving in the oncoming lane.

Iagnemma is confident that the software can be relied on to make good decisions. For the future, nuTonomy hopes to eventually partner up with automakers, tech and logistics companies and other industry leaders to further the development of autonomous vehicles.

“What we’re finding is the number of interested parties is really overwhelming,” he said.