Author: Jonathan M. Gitlin

A Google self-driving car. (credit: Google)

Google has spent several years working to teach robots how to drive perfectly. The Wall Street Journal reports that the company now has to teach the cars how to break speed limits and aim for corner apices instead. The company’s self-driving fleet of cars is pretty safe, with just 16 accidents since 2009.

But the vast majority of those were human drivers rear-ending the Google car, something that many in the field attribute to the very conservative nature of the cars’ programming. The self-driving cars will slow down or stop based on cues they detect from car passengers or other road users that humans would ignore.

A Google car was paralyzed by a cyclist balancing on his fixed-gear bike at a stop sign in Austin last month. The cyclist’s motion of rocking back and forth to stay upright fooled the car into thinking he was moving, so the car refused to move. And Google had to change the cars’ programming with regard to crossing double yellow lines after discovering the autonomous vehicles would instead just sit permanently behind other vehicles double-parked on the road.

Is “security researcher” a synonym for “this is why we can’t have nice things”? It might be. Jonathan Petit of security firm Security Innovation has decided that the lidar sensors used by self-driving cars need to be taken down a peg or two, and he’s presenting a paper on how to exploit them using a low-powered laser.

The lidar sensors used by self-driving cars work by sending out laser beams and then measuring the light that is reflected back. Do this with several sensors arrayed around a vehicle and you can create a 360˚ view of the physical objects around the car. Petit’s exploit involves shining a laser at the sensors to create a false return. “I can take echoes of a fake car and put them at any location I want,” he said to IEEE Spectrum, “and I can do the same with a pedestrian or a wall.”

Petit wants car makers to use more than one kind of sensor and to perform filtering to prevent this from being a problem in the future. Meanwhile, we should note that the optical sensors used by human-driven vehicles are also susceptible to security exploitation thanks to exploits like fitting a rear-facing million candlepower strobe to a car.

One of the more challenging jobs the auto industry has right now is explaining to consumers that the future isn’t going to be like the past. We desperately need to reduce vehicle carbon emissions in order to avoid turning the planet into a hellscape, and that means turning to cars with some kind of energy storage other than hydrocarbons we’ve dug up from the ground and then distilled. That’s where people get confused and the message stalls, a problem laid out in a recent report from the National Academy of Sciences.

Danev’s 2010 paper “Relay Attacks on Passive Keyless Entry and Start Systems in Modern Cars” demonstrated the vulnerability of keyless entry fobs to signal amplification, but doing so required a lab bench full of equipment, and an AC power supply. Tatarevic was unable use Danev’s approach to create a low-cost cordless signal amplifier and instead concludes that the burglaries were more likely the result of a brute force attack against the rolling codes that some manufacturers use for their unlocking signals.

Tatarevic bases this on the work of Silvio Cesare, another security researcher who demonstrated such an attack at last year’s Black Hat conference. That attack involved using a laptop and a software-defined radio (SDR) to send the car code after code until the right one unlocked the doors, something that could take up to two hours. That could fit with the facts; in each burglary, the cars had been parked for some time. This trick would also only unlock the car, unlike amplifying the signal of a keyless entry system, which would allow the car to be started, if only once.

Apple’s keynote this week produced a lot of hype-worthy news, as such Apple announcements often do. Laptops with no ports! Wearables with five-figure price tags! And who could forget the claims that the iPhone is now going to transform medical research? The keynote seemed to imply that iPhones, and ResearchKit, would transform our smartphones into medical tricorders, ready to diagnose Parkinson’s disease or atherosclerosis at the tap of a screen. Not so fast.

As much as Silicon Valley likes to think of itself as a force for good, disrupting this and pivoting that, it sometimes forgets that there’s a wider world out there. And when it comes to using devices in the practice of medicine, that world contains three very important letters: FDA. That’s right, the US Food and Drug Administration, which Congress has empowered to regulate the marketing and research uses of medical devices. 

Oddly, not once in any of the announcement of ResearchKit did we see mention of premarket approval, 510k submission, or even investigational device exemptions. Which is odd, because several of the uses touted in the announcement aren’t going to be possible without getting the FDA to say yes.

Last week, we brought you news that Apple appears to be working on electric car technology. Those rumors are edging toward !CONFIRMED! status thanks to news that Apple is being sued by battery maker A123 Systems for poaching staff in order to develop a “large scale battery division.”

The lawsuit, first reported by Law360, involves five engineers and scientists who A123 Systems alleges were poached by Apple over the course of the last few months, in breach of various NDAs and non-compete agreements. A123 also claims that it’s had to shut down several projects as a result.

In addition to the various contracts, A123 Systems’ suit also included e-mail left on their servers by one of the poached scientists (as well as a résumé) as proof that Apple had been raiding its talent pool.

Chevrolet announced yesterday at the Chicago Auto Show that the company has greenlit the production of the Bolt, its new electric vehicle (EV). The car was first shown to the world at last month’s North American International Auto Show in Detroit, MI as a concept, but according to Alan Batey, president of GM North America, the response from the public has convinced the company to build the car.

Although details are still scarce, the Bolt will sell for around $30,000, and GM promises a range of 200 miles on a charge. The car will be built at GM’s Orion Assembly, located near Detroit. Boosting the car’s green credentials are the fact that the plant is powered by gas from two nearby landfills and a 350 kW solar array.

GM

The Bolt looks like a cross between the Volt and BMW’s i3.

3 more images in gallery

.related-stories { display: none !important; }

The Bolt will be a direct rival to Tesla’s future affordable EV, the Model 3, and it may indeed beat the as-yet unseen Tesla to market. It’s a relatively conventional looking car, less outrageously styled than BMW’s clever (but expensive and range-limited) i3, but with an equally interesting interior—assuming that the production car hews closely to the concept GM is showing off. Bolts should start appearing in showrooms in 2017. Here’s hoping they drive well.

ars.AD.queue.push([“xrailTop”, {sz:”300×250″, kws:[], collapse: true}]);