Tesla's Elon Musk Talks With Google About Self-Driving Cars 199
Tesla Motors CEO Elon Musk has been thinking about bringing autonomous driving technology to Tesla's electric cars. Quoting Bloomberg:
"Musk, 41, said technologies that can take over for drivers are a logical step in the evolution of cars. He has talked with Google about the self-driving technology it’s been developing, though he prefers to think of applications that are more like an airplane’s autopilot system. 'I like the word autopilot more than I like the word self- driving,' Musk said in an interview. 'Self-driving sounds like it’s going to do something you don’t want it to do. Autopilot is a good thing to have in planes, and we should have it in cars.' ... Google’s approach builds on a push for the driverless-car technology long pursued by the U.S. military’s Defense Advanced Research Projects Agency, which held vehicle competitions for carmakers and research labs. Anthony Levandowski, product manager for Google’s self-driving car project, has said the company expects to release the technology within five years. 'The problem with Google’s current approach is that the sensor system is too expensive,' Musk said. 'It’s better to have an optical system, basically cameras with software that is able to figure out what’s going on just by looking at things.' ... 'I think Tesla will most likely develop its own autopilot system for the car, as I think it should be camera-based, not Lidar-based,' Musk said yesterday in an e-mail. 'However, it is also possible that we do something jointly with Google.'"
Musk later warned not to take this as an actual announcement.
Major problem here (Score:5, Interesting)
Re:Who wants a driverless tesla roadster? (Score:3, Interesting)
Tedious and dangerous. A combination practically designed to induce stress.
LIDARm but not Google's LIDAR (Score:5, Interesting)
As someone who's actually done this stuff, LIDAR gives solid data, but is range-limited. Cameras have more ambiguous results. Cameras are most useful when things are going well, as on a highway under good conditions. That was Stanford's approach in the Grand Challenge. All their vision system really did was answer the question "is the near section of road (within LIDAR range) like the far section of road"? If the LIDAR said the near section was OK to drive on and the vision system said the far section was like the near section, then the vehicle could speed up and out-drive the LIDAR range. That sped up travel on good sections of road.
Google is using Velodyne LIDAR units, which are effective but an expensive mechanical kludge. A better approach is from Advanced Scientific Concepts [advancedsc...ncepts.com], which has an eye-safe flash LIDAR. No moving parts.
ASC's units cost about $100K each, but that's because they're hand-made for DoD. The technology isn't inherently expensive if made in volume. It uses custom imaging ICs, and because they're made by tens, not millions, they cost far too much. If the cost can be brought down, the vehicle can have multiple LIDAR units around the car to get full coverage, rather than one big spinning thing up on the roof.
Millimeter radar is also useful. It's good to have a Dopper anticollision radar as a backup system. It provides an unambiguous "rapidly approaching big solid object" signal. We had one of those on our DARPA Grand Challenge vehicle as a backup to the fancier LIDAR system.
but think of the benefits! (Score:4, Interesting)
When our roads are autonomous, you will get to point B faster, safer, with less fuel, less wear on the car, and better rested. You can also stop worrying about parking, fueling, and maintaining you car as it can go and do all these things automatically while you go about your business.
Insurance (Score:5, Interesting)
At the same time I can see the insurance companies realizing that a huge huge HUGE market will simply go away when car accidents become unlikely enough for car companies to be able to cover it. Think about it. Every car that you see is paying in around $1,000+ for insurance. The only insurance people will want after robotic cars will be theft (hard to do with a hi-tech upgradable car), vandalism, trees falling on them kinds of insurance. Plus nearly every jurisdiction says you must have something like 2 million in liability; that need will vanish or at least be covered by the manufacturers.
So my robotic car prediction is that car companies will be trying to terrorize us into hating robotic cars. They will show videos of families being driven off cliffs, or saying it is our god given right to have control of our cars. And of course they will spend ungodly amounts of money lobbying everyone from the president down to your school board to stop this.
But the simple reality is that 35,000 people are killed every year in the US and robotic cars might take this down to a few hundred. (mechanical failure, trees falling on them, sinkholes, etc)
Re:Who wants a driverless tesla roadster? (Score:5, Interesting)
Electric motors need a transmission too.
If by transmission you mean a fixed ratio reduction gear. Electric motors have actually been used for years to eliminate the need for variable ratio transmissions, which often don't work well w/ high torques or other situations that electric motors handle gracefully. That's what the electric part of a diesel-electric locomotive is - an electric motor used in place of a transmission. They're built that way because mechanical transmissions can't cut it.
Teslas engineers were just too incompetent to build one correctly.
Tesla subcontracted the transmission design, and three companies, all of which have extensive experience, couldn't produce something that worked right. Tesla's solution was to improve the electric motor and drive electronics, which gave them equal or better performance than was originally anticipated with a transmission, but without the weight or unreliability of a mechanical transmission. Tesla's "incompetence" led to a better car.