Last week, police in Tempe, Arizona, released dash cam footage showing the final seconds before an Uber self-driving vehicle crashed into 49-year-old pedestrian Elaine Herzberg, who died at the hospital shortly afterward.

While Herzberg is visible for less than two seconds in the camera footage, she might have become visible earlier to a human driver, since human eyes are better at picking out details in low-light situations. The police also released internal dash cam footage showing the car’s driver, Rafaela Vasquez, in the seconds before the crash. Vasquez can be seen looking down toward her lap for almost five seconds before glancing up again and realizing what is about to happen.

This is not the first collision involving a self-driving vehicle. Other crashes have involved Ubers in Arizona, a Tesla in ‘autopilot’ mode and many others in California. While it is still not clear as to who is to be blamed for the latest incident, most of the other crashes were due to human error.

Image source: ReadWrite

Why do autonomous vehicles crash? The first cause is when the sensors don’t detect what is happening around the vehicle. In the case of Herzberg, she was walking with a bicycle across a poorly lit roadway. About 1.4 seconds elapse between the time when Herzberg starts to become visible and the video’s final frame. Each car and each sensor works differently. In some, GPS may work only with a clear view of the sky or the camera may function properly only with ample light. Researchers aren’t clear about the ‘ideal’ set of sensors in an autonomous vehicle, and can’t keep adding more and more to the fleet because of cost limitations.

The second cause is when the vehicle encounters a situation that the software didn’t plan for. Much like human drivers, self-driving cars also make many decisions at any given second. So, when a self-driving car sees something it’s not programmed to handle, it becomes difficult to assess what is going to happen next. So, the challenge for programmers is to combine information from all sensors to create an accurate representation of the surroundings. If that doesn’t happen, the vehicle can’t take a call.

Before we can release autonomous vehicles on the road completely, they need to be programmed with a whole set of instructions about dealing with the out-of-the-ordinary, and plan for extreme adverse situations. A human driver, for instance, might even break a few traffic rules to save lives. The autonomous car needs to be taught how to do the same instead of waiting or the situation to pass.

Additionally, self-driving cars must be taught to understand their surroundings and the context. In the crash involving Herzberg, the car should have been tested and rated based on how well it performed difficult tasks, including assessing danger in fraction of a second. This is a lot like a human driving test, but that is how it should be if the two have to co-exist safely on the road.

Another approach to this could also be explicitly stating rules in formal logic, like how nuTonomy, which is running an autonomous taxi pilot in cooperation with authorities in Singapore, is doing. NuTonomy’s approach to controlling autonomous vehicles is based on a rules hierarchy. Top priority goes to rules such as “don’t hit pedestrians”, followed by “don’t hit other vehicles” and “don’t hit objects”. Rules such as “maintain speed when safe” and “don’t cross the centreline” get a lower priority.

Image source: The Verge

Learning how to drive is an ongoing process for both humans and their autonomous cars, as both adapt to new situations, road rules and technology. Time will tell the better approach, but for now, the research still has a long way to go.

The post Driverless cars still have a long way to go appeared first on TravHQ.

NuTonomy, the autonomous car startup spun out of MIT will soon begin testing its autonomous vehicles in Boston. The company beat Uber to public streets in the race for autonomous cars by rolling out driverless taxis in Singapore. It announced that it has signed an agreement with officials in Boston to roll out driverless cars to the city streets and the testing will begin in an industrial park in South Boston later this year.

Similar to Singapore, their autonomous Renault Zoe electric cars will be operated in public area but somewhat removed from the whole urban chaos an average driver deals with. Initially only company engineers will be riding the cars in Boston however, there could be plans to open it up for public like the Singapore program.

“These tests in the city of Boston will enable our engineers to adapt our autonomous vehicle software to the weather and traffic challenges of this unique driving environment,” company CEO and co-founder Karl Iagnemma said in a statement. “Testing our self-driving cars so near to nuTonomy’s home is the next step towards our ultimate goal: deployment of a safe, efficient, fully autonomous mobility-on-demand transportation service.”

According to company statement, there are plans to have a dozen cars on the roads in Singapore as this year approaches a close. In Singapore, the cars are picking and dropping passengers through a partnership with Grab. nuTonomy will have to put together a similar partnership in Boston which is highly unlikely considering the fact that it would cause conflict of interest as both Uber and Lyft are working on their own versions of self-driving cars. Uber is already running tests in Pittsburgh which kicked off a short while after nuTonomy launched the program in Singapore.

The race for autonomous vehicles is getting competitive as almost all the players in the space are registering good progress. Alphabet (Google) seems to be much further ahead of the others but it is still a long way to go before any one of these players could actually start affecting the ground transportation.

Source: WSJ

The post NuTonomy heads closer to home after Singapore to test autonomous cars appeared first on TravHQ.

Yes you can be excited about driverless cars now, especially if you live in Singapore. While Uber recently announced that it will start testing driverless cars in Pittsburgh starting this month, NuTonomy has officially launched driverless taxis in Singapore, probably beating Uber to it by few days. This will be the first ever public trial for autonomous vehicles and it will be held in One-North business district.

NuTonomy, a US based driverless vehicle R&D firm founded by MIT PhD graduates in 2013, has partnered with EDBI, the corporate investment arm of Singapore Economic Development Board. NuTonomy will invite residents of Singapore to try out the service for free by booking a ride through its app. The driverless cars being used for testing are built upon Renault Zoe and Mitsubishi i-MiEV as base.

CEO and co-founder of nuTonomy, Karl Iagnemma, said, “nuTonomy’s first-in-the-world public trial is a direct reflection of the level of maturity that we have achieved with our AV software system. The trial represents an extraordinary opportunity to collect feedback from riders in a real-world setting, and this feedback will give nuTonomy a unique advantage as we work toward deployment of a self-driving vehicle fleet in 2018.”

In May of this year, nuTonomy completed a $16M Series A funding led by Highland Capital Partners that included participation from Fontinalis Partners, Signal Ventures, Samsung Ventures, and EDBI, the dedicated corporate investment arm of the Singapore Economic Development Board.

Uber is working with Volvo to bring driverless cabs and similar to their plan, these public rides will have an engineer from the company to observe the vehicle behaviour and take control in case things don’t proceed how they are supposed to. The data collected through the testing program will help them improve the software for the future.

While NuTonomy is the first company to start public testing of driverless cabs, Uber isn’t far behind. There are many other startups working in different ways to achieve autonomy in vehicles and will be seen in public sooner or later.

The post Uber has just been beaten by NuTonomy in the race to launch self-driving cabs appeared first on TravHQ.

Self-driving taxis are taking the commuting sector by storm. Otto offered autonomous technology for trucks lately and following it was nuTonomy, an MIT spinoff which recently got funded $16million for the same goal to offer self-driving cabs within two years.

Karl Iagnemma (CEO) and Emilio Frazzoli (CTO), MIT PhDs graduates founded nuTonomy, a Cambridge MA-based self-driving technology for cars, the kind of technology that Uber, BMW, Google are currently developing. nuTonomy refits existing cars with technology to make them driverless.

Singapore has been an experiment ground for nuTonomy. Apart from the investments that pour in, it has an office there and has tested self-driving vehicles. It had started testing out self-driving cart style buggies summoned by smartphones in 2014 but is now believed to test a fleet of self-driving fully functional taxis which it claims can hit the roads by 2018.

The start-up s working closely with SMART (Singapore-MIT Alliance for Research and Technology) as a part of organisation’s expansive decree to enhance the lifestyle of the residents. Being one of the many on agenda, Singapore’s smart city initiative also covers areas like cashless wallets, open government APIs, a push to promote entrepreneurialism, and increased security surveillance.

It is the best time for nuTonomy to hit the Singapore roads as the technology has made its debut in the U.S. and it’s already trying to create its own space here. The country only introduced its on-demand taxi services in April where driverless cars are now going to become a reality.

The nuTonomy cars being tested in Michigan and the UK where they’ve partnered with Jaguar and Land Rover.

Source: TechCrunch

The post nuTonomy to bring self driving cars to Singapore by 2018 appeared first on TravHQ.