1982 Cadillac Eldorado Ht 4100 Digital Fuel Injection 4.1l V8 on 2040-cars

Auto blog

Teaching autonomous vehicles to drive like (some) humans

While I love driving, I can't wait for fully autonomous vehicles. I have no doubt they'll reduce car accidents, 94 percent of which are caused by human error, leading to more than 37,000 road deaths in the U.S. last year. And if it means I can fly home at night in winter and get safely shuttled to my house an hour-plus away — and not have to endure a typical white-knuckle drive in the dark with torrential rain and blinding spray from 18-wheelers on Interstate 84 — sign me up. Autonomous technology will also take some of the stress, tedium and fatigue out of long highway drives, as I recently discovered while testing Cadillac Super Cruise. AVs are also supposed to eventually help increase traffic flow and reduce gridlock. But according to a recent Automotive News article, as the first wave of AVs are being tested on public roads, they're having the opposite effect. Part of the problem is they drive too cautiously and are programmed to strictly follow the written rules of the road rather than going with the flow of traffic. "Humans violate the rules in a safe and principled way, and the reality is that autonomous vehicles in the future may have to do the same thing if they don't want to be the source of bottlenecks," Karl Iagnemma, CEO of self-driving technology developer NuTonomy, told Automotive News. "You put a car on the road which may be driving by the letter of the law, but compared to the surrounding road users, it's acting very conservatively." I get it that, like teen drivers, AVs need a ramp up period to learn the unwritten rules of the road and that a skeptical public has to be convinced of the technology's safety. But this is where I become less of a champion on AVs, since where I live in the Pacific Northwest we already have more than our share of overly cautious human drivers. Since moving here 12 years ago, I've found it's an interesting paradox that a region famous for its strong coffee, where you'd think most drivers would be jacked up on caffeine, is also the home to annoyingly measured motorists. As an auto-journo colleague living in Seattle so aptly put it: "People in the Pacific Northwest drive as if they have nowhere to go." If you drive like me and always have somewhere to go — and usually are in a hurry to get there — it's absolutely maddening.

Our favorite mid-engine Corvette theories and rumors

The mid-engine Corvette! We just saw blurry photos through the foliage, and are hearts are all aflutter because it means that this car is real. And it's the most " never before" Corvette since 1984. Which leads us to all kinds of wacky theories and speculation. Everybody knows something or heard from a guy. None of those guys (or gals) from GM are talking to Autoblog, to be clear. But that won't stop us from stirring the pot. Let's go over a few point-by-point. The pushrod engine will be replaced by an overhead-cam V8 Car and Driver has repeatedly reported that an overhead-cam engine will come after the mid-engine Corvette's debut with an old-school pushrod. We've heard the same thing firsthand, albeit from a source with no connection to the development of the Corvette or powertrain. "At some point two valves can't pass future emissions regulations," said the guy we know. While the pushrod engine is compact, the design makes it difficult - if not impossible - for sophisticated variable valve timing (and lift) systems that control combustion with more precision. Hasty conclusion: The pushrod engine has a finite life in front of it. This one's a lock, it's a just a matter of when. Bowling Green's new paint shop is really the mid-engine assembly line Credit to Reddit for this one. Like we said, everybody knows a guy. But let's look deeper. The new paint facility costs $439 million dollars and adds 450,000 square feet, almost half the size of the existing plant. Plus GM announced another $290 million in upgrades at Bowling Green Assembly. That sure sounds like a lot of money, but you can actually spend that much on a paint booth. Porsche spent 500 million Euros (about $561 million dollars) to add the Macan assembly to its Leipzig, Germany plant in 2014. Chrysler shelled out $850 million for a paint shop in Sterling Heights, Michigan. GM spent $600 million for a new paint shop at the Fairfax (Kansas City) plant. Honda, on the other hand, is working on a thrifty $210 million project in Marysville, Ohio that includes a 300,000 square foot expansion, and the low-volume Acura NSX facility only cost $70 million. Hasty conclusion: The square footage and cost could easily mean a new assembly line instead of (or in addition to) a paint line. And if the NSX plant was really that cheap, GM could have hidden a similar sum in its existing announcements.

MIT puts V2V technology on its 2015 Top Ten list

Of all the technologies swimming around the automotive world, it is vehicle-to-vehicle communication that the Massachusetts Institute of Technology has fished out as one of its Ten Breakthrough Technologies of 2015. It joined emerging tech like brain organoids, supercharged photosynthesis, and Project Loon on the list, and got the nod over autonomous driving because, as the MIT Technology Review wrote, V2V communication "is likely to have a far bigger and more immediate effect on road safety." How so? Because actual cars transmitting data like their location, speed, steering angle, and state of braking to one another at least ten times per second provides a greater degree of awareness than sensor readings and algorithms. The US Department of Transportation and the National Highway Traffic Safety Administration have been working for years on standards and a regulatory schedule for introducing V2V to the marketplace, and Cadillac plans to incorporate V2V into at least one of its vehicles by 2017. Since we've begun the year with a number of stories of cars being hacked into, that got us wondering about the security of V2V communications. In a recent piece by our own Pete Bigelow on what motorists should know about getting their cars hacked into, he wrote that although cyber break-ins are extremely difficult, expensive, and time-consuming to do remotely, V2V is "one more conceivable avenue a hacker could use to impact multiple cars at a given time." So we spoke to Wilmington, Massachusetts-based Security Innovation about it. The automotive consultancy company has been working with the DOT since 2003 on V2V technology and the issues around it - namely security and privacy - and its chief scientist, William Whyte, is the technical editor of the Institute of Electrical and Electronics Engineers (IEEE) 1609.2 standard outlining its security protocols. Those protocols are expected to be finalized by the DOT toward the end of this year and then come into effect in 2016, and the company's Aerolink product is the security solution Cadillac will use. Whyte said, "If you hack into a car, V2V is the hardest place to start," and Pete Samson, the general manager of Security Innovation's automotive team, said "There are ten or 12 alternate attack surfaces" around the car that would make much easier targets.