The Self-Driving Car’s Bicycle Problem

6210901187_dc5fc87af0_b

Human error plays a role in 94% of U.S. traffic fatalities. Image: William Murphy via Flickr

Robotic cars are great at monitoring other cars, and they’re getting better at noticing pedestrians, squirrels, and birds. The main challenge, though, is posed by the lightest, quietest, swerviest vehicles on the road.

“Bicycles are probably the most difficult detection problem that autonomous vehicle systems face,” says UC Berkeley research engineer Steven Shladover.

Nuno Vasconcelos, a visual computing expert at the University of California, San Diego, says bikes pose a complex detection problem because they are relatively small, fast and heterogenous. “A car is basically a big block of stuff. A bicycle has much less mass and also there can be more variation in appearance — there are more shapes and colors and people hang stuff on them.”

That’s why the detection rate for cars has outstripped that for bicycles in recent years. Most of the improvement has come from techniques whereby systems train themselves by studying thousands of images in which known objects are labeled. One reason for this is that most of the training has concentrated on images featuring cars, with far fewer bikes.

Consider the Deep3DBox algorithm presented recently by researchers at George Mason University and stealth-mode robotic taxi developer Zoox, based in Menlo Park, Calif. On an industry-recognized benchmark test, which challenges vision systems with 2D road images, Deep3DBox identifies 89 percent of cars. Sub-70-percent car-spotting scores prevailed just a few years ago.

Deep3DBox further excels at a tougher task: predicting which way vehicles are facing and inferring a 3D box around each object spotted on a 2D image. “Deep learning is typically used for just detecting pixel patterns. We figured out an effective way to use the same techniques to estimate geometrical quantities,” explains Deep3DBox contributor Jana Košecká, a computer scientist at George Mason University in Fairfax, Virginia.

However, when it comes to spotting and orienting bikes and bicyclists, performance drops significantly. Deep3DBox is among the best, yet it spots only 74 percent of bikes in the benchmarking test. And though it can orient over 88 percent of the cars in the test images, it scores just 59 percent for the bikes.

Košecká says commercial systems are delivering better results as developers gather massive proprietary datasets of road images with which to train their systems. And she says most demonstration vehicles augment their visual processing with laser-scanning (ie lidar) imagery and radar sensing, which help recognize bikes and their relative position even if they can’t help determine their orientation.

Further strides, meanwhile, are coming via high-definition maps such as Israel-based Mobileye’s Road Experience Management system. These maps offer computer vision algorithms a head start in identifying bikes, which stand out as anomalies from pre-recorded street views. Ford Motor says “highly detailed 3D maps” are at the core of the 70 self-driving test cars that it plans to have driving on roads this year.

Put all of these elements together, and one can observe some pretty impressive results, such as the bike spotting demonstrated last year by Google’s vehicles. Waymo, Google’s autonomous vehicle spinoff, unveiled proprietary sensor technology with further upgraded bike-recognition capabilities at this month’s Detroit Auto Show.

Vasconcelos doubts that today’s sensing and automation technology is good enough to replace human drivers, but he believes they can already help human drivers avoid accidents. Automated cyclist detection is seeing its first commercial applications in automated emergency braking systems (AEB) for conventional vehicles, which are expanding to respond to pedestrians and cyclists in addition to cars.

Volvo began offering the first cyclist-aware AEB in 2013, crunching camera and radar data to predict potential collisions; it is rolling out similar tech for European buses this year. More automakers are expected to follow suit as European auto safety regulators begin scoring AEB systems for cyclist detection next year.

That said, AEB systems still suffer from a severe limitation that points to the next grand challenge that AV developers are struggling with: predicting where moving objects will go. Squeezing more value from cyclist-AEB systems will be an especially tall order, says Olaf Op den Camp, a senior consultant at the Dutch Organization for Applied Scientific Research (TNO). Op den Camp, who led the design of Europe’s cyclist-AEB benchmarking test, says that it’s because cyclists movements are especially hard to predict.

Košecká agrees: “Bicycles are much less predictable than cars because it’s easier for them to make sudden turns or jump out of nowhere.”

That means it may be a while before cyclists escape the threat of human error, which contributes to 94 percent of traffic fatalities, according to U.S. regulators. “Everybody who bikes is excited about the promise of eliminating that,” says Brian Wiedenmeier, executive director of the San Francisco Bicycle Coalition. But he says it is right to wait for automation technology to mature.

In December, Wiedenmeier warned that self-driving taxis deployed by Uber Technologies were violating California driving rules designed to protect cyclists from cars and trucks crossing designated bike lanes. He applauded when California officials pulled the vehicles’ registrations, citing the ridesharing firm’s refusal to secure state permits for them. (Uber is still testing its self-driving cars in Arizona and Pittsburgh, and it recently got permission to put some back on San Francisco streets strictly as mapping machines, provided that human drivers are at the wheel.)

Wiedenmeier says Uber’s “rush to market” is the wrong way to go. As he puts it: “Like any new technology this needs to be tested very carefully.”

This post was created for Cars That Think, IEEE Spectrum’s blog about the sensors, software, and systems that are making cars smarter, more entertaining, and ultimately, autonomous.

Trump’s Impact on Clean-Energy Businesses

Published today at MIT Technology Review:

President-elect Donald Trump is a self-declared climate-change denier who, on the campaign trail, criticized solar power as “very, very expensive” and said wind power was bad for the environment because it was “killing all the eagles.” He also vowed to eliminate all federal action on climate change, including the Clean Power Plan, President Obama’s emissions reduction program for the power sector.

So how will renewable-energy businesses fare under the new regime?

Trump’s rhetoric has had renewable-energy stocks gyrating since the election. But the impact could be far less drastic than many worst-case scenarios. “At the end of the day what Trump says and what is actually implemented are two completely different things,” says Yuan-Sheng Yu, an energy analyst with Lux Research …

For the whole story see “Trump’s Impact on Clean-Energy Businesses

Micro-Satellite Tracks Carbon Polluters From Space

CLAIRE.PetitSaut.MethaneConcentration

Simulated satellite image of methane plume from French Guyana’s Petit Saut hydroelectric power plant. Image: GHGSat

Attention greenhouse gas emitters: There’s a new eye in the sky that will soon be photographing your carbon footprint and selling the images to any and all. It’s a micro-satellite dubbed “Claire” (clear, bright, and clean in French) by its Montreal-based developer, GHGSat.

This microwave-oven-sized pollution paparazzo rocketed to a 512-kilometer-high orbit in mid-June care of the Indian Space Agency, with a mission to remotely measure the plumes of carbon dioxide and methane wafting up from myriad sources on Earth’s surface. Claire’s targets include power plants, natural gas fracking fields, rice paddies, and much more—just about any emissions source that someone with a checkbook (corporations, regulators, activists) wants tracked, according to GHGSat president Stéphane Germain. Continue reading

Does Electrification Really Cause Economic Growth?

Mjc2OTQ4NQ

Villages brightened from 2001 (L) to 2011 (R). Images: Burlig & Preonas / NOAA

Electrification is associated with a seemingly endless list of social and economic goods. Nations that use more power tend to have increased income levels and educational attainment and lower risk of infant mortality, to name but a few. So I was baffled to stumble across a pair of economic analyses on electrification in India and Kenya, posted last month, that cast serious doubt on what has long assumed to be a causal link between the glow of electricity and rural development.

“It is difficult to find evidence in the data that electrification is dramatically transforming rural India,” concludes Fiona Burlig, a fourth-year UC Berkeley doctoral student in agricultural and resource economics who coauthored the India study. “In the medium term, rural electrification just doesn’t appear to be a silver bullet for development.” Continue reading

Wind Could Provide Over 26% of Chinese Electricity by 2030

Last month I argued that the primary reason Chinese wind farms underperform versus their U.S.-based counterparts is that China’s grid operators deliberately favor operation of coal-fired power plants. Such curtailment of wind power has both economic and technical roots, and it has raised serious questions about whether China can rely on an expanding role for wind energy. New research published today appears to put those concerns to rest, arguing that wind power in China should still grow dramatically.

The report today in the journal Nature Energy projects that wind energy could affordably meet over one-quarter of China’s projected 2030 electricity demand—up from just 3.3 percent of demand last year.

In fact the researchers, from MIT and Tsinghua University, project that modest improvements to the flexibility of China’s grid would enable wind power to grow a further 17 percent. That, they argue, means that China’s non-fossil resources could grow well beyond the 20 percent level that China pledged to achieve under the Paris Climate Agreement. Continue reading

The Natural Gas Accounting Gap

Last month the U.S. EPA admitted it was way off in its estimate of how much methane producers leak into the atmosphere in the process of wresting natural gas from the ground and piping it across the continent. It’s a big deal since methane is a far more potent greenhouse gas than carbon dioxide and likely responsible for a substantial fraction of the climate change we’re already experiencing. And it’s been a long time coming. For many years now methane measurements by airplanes and satellites have strongly suggested that methane emissions from the oil and gas patch could be double what EPA figures captured.

Today the online earth observation pub Earthzine has my take on an unusual research project that helped convince EPA — and the industry — to change their tune on methane emissions. Take me to the article…

Beetles, Cacti, and Killer Plants Inspire Energy Efficiency

What do you get when you mix a desert beetle, a pitcher plant, and a cactus? Pick the right parts and you get an extremely slippery surface with an uncanny capacity to condense and collect water, according to research reported today in the journal Nature.

The advance could be a big deal for the energy world because, when it comes to energy efficiency, condensation lies somewhere between a necessary evil and a major drag. Nuclear, coal, and thermal solar power plants, for example, require large heat exchangers to condense the steam exiting from their turbines so that they can raise a new round of hotter steam. For other devices, such as wind turbines and refrigerator coils, condensation is the first step towards energy-sapping ice formation. Continue reading