Before assessing the current state of the technology or making predictions about its future, I think it’s important to break down the problem of making a self-driving car into several fuzzily overlapping domains:
- Last 100 meters – getting in and out of parking spaces, driveways, etc. (endpoints)
- Low speed, unlimited access, stop signs, “negotiated” yields, etc. – getting between endpoints and main roads
- Traffic light-managed (main road)
- High speed, limited access (highway)
And overlay that with several features not widely found on roads now, such as:
- Computer vision registration signage – distinct shapes that a computer image processing program can easily distinguish from its surroundings to determine the position of a car on a road
- Imbedded guideways – anything from physical rails such as one finds in a children’s amusement park ride to magnetic spikes driven into pavement to buried and rail-mounted magnetic sensors and radio transceivers. Such a system, the AHS, was prototyped in the early 1990s on portions of CA USA highways.
- Traffic signal transponders -
It’s useful also to outline at a high level some basic, overlapping driverless car technologies:
- Specialized sensors/detectors – the vehicle components of guideways and signal transponders systems described above.
- Vehicle transponders – usually radio-based systems that allow the guidance systems of cars near one another to communicate.
- Radar.
- Video camera based computer vision
All but the last of these were used in the AHS
A (at seems to me) thorough overview of existing self-driving car technology can be found in the
wikipedia article “Driverless car”. A read of its or other sources history section will reveal something surprising to those previously unfamiliar with it: the most advanced research in the field occurred in the 1980s and 90s (eg: the
VaMP robot vehicle), peaking in the mid 1990s, with interest and investment dwindling since then. The most prominent 21st century driverless vehicle project is the 2004 and 2005 “
DARPA Grand Challenges”, which have involved closed, semi-offroad courses with no vehicle interaction. The 2007 challenge (there was no event in 2006) requires vehicles to “obey traffic regulations while negotiating with other traffic and obstacles and merging into traffic”, though the speed requirement – 60 miles (96 km) in under 6 hours – is far from practical human driving speed.
In short, then, I think the answer to the original question
Quote:
Originally Posted by Agen
So the question is when will this be available to the public (in a decent price range)
|
is, as Alexander surmises, “not soon”, and possibly “not ever”. Both basic technology and market/economic challenges appear to be preventing this technology from achieving practical viability.
----------------
Moderator: Computers and Technology; Medical Science; Science Projects and Homework; Philosophy of Science; Physics and Mathematics; Environmental Studies
