Tag Archives: Driverless

Dancing.

Teaching smart cars how humans move could help make them safer and better

Computers today can’t make heads and tails of how our bodies usually move, so one team of scientists is trying to teach them using synthetic images of people in motion.

Google driverless car.

Image credits Becky Stern / Flickr.

AIs and computers can be hard to wrap your head around. But it’s easy to forget that holds true from their perspective as well. This can become a problem because we ask them to perform a lot of tasks which would go over a lot smoother if they actually did understand us a tad better.

This is how we roll

Case in point: driverless cars. The software navigating these vehicles can see us going all around them through various sensors and can pick out the motion easily enough, but it doesn’t understand it. So it can’t predict how that motion will continue, even for something as simple as walking in a straight line. To address that issue, a team of researchers has taken to teaching computers how human behavior looks like.

When you think about it, you’ve literally had a lifetime to acquaint yourself to how people and other stuff behaves. Based on that experience, your brain can tell if someone’s going to take a step or fall over or where he or she will land after a jump. But computers don’t have that store of information in the form of experience. The team’s idea was to use images and videos of computer-generated bodies walking, dancing, or going through a myriad of other motions to help computers learn what cues it can use to successfully predict how we act.

Dancing.

Hard to predict these wicked moves, though.

“Recognising what’s going on in images is natural for humans. Getting computers to do the same requires a lot more effort,” says Javier Romero at the Max Planck Institute for Intelligent Systems in Tübingen, Germany.

The best algorithms today are tutored using up to thousands of pre-labeled images to highlight important characteristics. It allows them to tell an eye apart from an arm, or a hammer from a chair, with consistent accuracy — but there’s a limit to how much data can realistically be labeled that way. To do this for a video of a single type of motion would take millions of labels which is “just not possible,” the team adds.

Training videos

So they armed themselves with human figure templates and real-life motion data then took to 3D rendering software Blender to create synthetic humans in motion. The animations were generated using random body shapes and clothing, as well as random poses. Background, lighting, and viewpoints were also randomly selected. In total, the team created more than 65,000 clips and 6.5 million frames of data for the computers to analyze.

“With synthetic images you can create more unusual body shapes and actions, and you don’t have to label the data, so it’s very appealing,” says Mykhaylo Andriluka at Max Planck Institute for Informatics in Saarbrücken, Germany.

Starting from this material, computer systems can learn to recognize how the patterns of pixels changing from frame to frame relate to motion in a human. This could help a driverless car tell if a person is walking close by or about to step into the road, for example. And, as the animations are all in 3D, the material can also be used to teach systems how to recognize depth — which is obviously desirable in a smart car but would also prove useful in pretty much any robotic application. .

These results will be presented at the Conference on Computer Vision and Pattern Recognition in July. The papers “Learning from Synthetic Humans” has been published in the Computer Vision and Pattern Recognition.

Finland capital Helsinki starts driverless bus pilot

If you thought driverless cars are part of a distant past, think again. In Helsinki’s Hernesaari district, self-driving buses are already taking passengers to their destinations.

Image by Digital Trends.

Finnish laws don’t require cars to have a driver, which is why this kind of project can be initiated with relative ease, at least from a legislative standpoint.

“This is actually a really big deal right now. There’s no more than a handful of these kinds of street traffic trials taking place, if that,” said test project lead and Metropolia University of Applied Sciences project manager Harri Santamala.

It’s not the first time something like this was rolled out in Finland, as neighboring city Vantaa rolled out similar vehicles during its housing fair, though the buses were only allowed to circulate in areas shut off to other traffic. This time, it’s the real deal. Buses are tested in a challenging environment, with real drivers who can sometimes get angry and push or break circulation laws.

Authorities say that the buses, which are electric, carry up to 10 passengers and run at an average speed of 10 km/h, are not meant to replace conventional transportation, but rather to supplement it.

“Their purpose is to supplement but not to replace it. For example the goal could be to use them as a feeder service for high-volume bus or metro traffic, like Kutsuplus. In other words the mini-bus would know when the connecting service is coming and it would get you there on time,” Santamala explained.

What do you think, would you hop along a driverless bus?

Expert warns smart-cars will promote sex behind the wheel and distracted driving

Will widespread use of smart cars make roads safer or actually more dangerous? One Canadian expert is raising concerns that as automated systems take up the bulk of navigating tasks, drivers will keep their hands less on the driving wheel…and more on the person (persons?) next to them.

Image via scmp.com

Drop whatever you were doing and rejoice because science has delivered.

“I am predicting that, once computers are doing the driving, there will be a lot more sex in cars,” said Barrie Kirk of the Canadian Automated Vehicles Centre of Excellence.

It truly is a wonderful time to be alive. But, before we go about congratulating and patting each other on the back in satisfaction, is this a good thing? I mean beyond the obvious fact that we all like to get it on.

There is legitimate concern around this question, not because of the cars themselves but because of the drivers. I see people texting or talking on the phone at the wheel — and these aren’t particularly enjoyable activities — every day, driving regular vehicles without any computers to watch the road for them. But if people trust their cars enough to handle themselves in traffic, they’ll throw their phones along with their pants on the back seat before you can say “responsible driving practices.”

“That’s one of several things people will do which will inhibit their ability to respond quickly when the computer says to the human, ‘Take over.'”

Canadian Press obtained several federal emails discussing Tesla’s self-driving cars under the Access to Information Act. In them, officials tasked with constructing the legislative framework for autonomous cars took up the issue in the briefing notes compiled for Transport Minister Marc Garneau after his appointment last fall.

“The issue of the attentive driver is … problematic,” one such email reads. “Drivers tend to overestimate the performance of automation and will naturally turn their focus away from the road when they turn on their auto-pilot.”

The emails cite several pieces of footage showing Tesla drivers doing anything else than paying attention to the road, such as reading a newspaper for example. Other videos show Tesla owners recording flaws in how the car’s autopilot system reacts to changes in road markings.

Therein lies the problem: Tesla itself made it clear that the autopilot system only has limited autonomy and functionality. It’s designed to work in tandem with a human, not to replace him. And people still behave like it’s their personal chauffeur. Transport Canada tested several semi-autonomous vehicles, such as Mercedes’ C-Class or the Infiniti Q50 (but not the Tesla so far,) the documents go on to detail. While they found the systems efficient at what they do, the technology is still in its infancy.

“It really needs to be emphasized that these vehicles are not truly self-driving,” officials said. They predicted that fully-autonomous cars and trucks are “still a few years away.”

Current vehicle safety standards don’t prohibit driverless cars from zooming on Canada’s roadways, and the country is now considering how to regulate such vehicles.

“But last month’s federal budget included money for Transport Canada to develop regulations around automated vehicle design. Those regulations, at least initially, would require that the vehicles are equipped with a ‘failsafe mechanism that can respond to situations when the driver is not available,'” CBC writes. “Ontario also set out some regulations, including a requirement that an expert in autonomous vehicles be in the driver’s seat and able to assume full control at a moment’s notice.”

The “failsafe mechanism” basically means that the car should be able to safely get out of traffic until a human assumes control — and that should be at the center of how we handle this I think. Because that “expert in autonomous vehicles,ready at a moment’s notice” part? I think that’s wishful thinking.

The whole point of having autonomous cars is that no driver is required, and people won’t be willing to wait, clutching the wheel, on the off chance they’re needed. It’s got to go all the way, or at least allow for a window of time in which the driver can analyze the situation, plan his movements and assume control. Assuming that a driver who may not have been paying attention to his or her surroundings can control a vehicle right off the bat is a tall order however, Kirk believes.

“People will not be able to respond in time.”

It’s a good thing that we come face to face with these issues now, before autonomous vehicles truly hit the roads. But they just aren’t here yet, so you’ll have to keep your eyes on the road until they do. And yes, your hands on the wheel, too.