Linux-Powered Auto-Parking Car 414
megmag writes "A really cool project using a Linux P4 machine for automatic parking of a Volvo S60 was presented last week. Take a look at the video. That's how your parking problem should be solved. It is a final-year student project within the mechanical engineering department at Linköping University, Sweden."
Notice how much space is available to park the car (Score:3, Interesting)
white cars only? (Score:2, Interesting)
Questions... (Score:5, Interesting)
2) How does the system deal with engine/linkage issues. Cars don't provide smooth power/steering at all times. If the engine is out of tune or has a catchy throttle, can the system deal with that as well as/better than a human?
3) How is it told where to park? It would have been nice if it was clear in the video what the driver did to tell it that. The article alludes to some sort of analysis system for this, but I like pretty pictures.
Pretty nifty anyway!
This reminds me of... (Score:3, Interesting)
I don't believe they recommended that you got out of the car before the parking manuever was completed.
Re:This is cool (Score:5, Interesting)
To be fair, in the same show they did an experiment where men and women were given a list of things to do in a time limit. Things like answering the door, the phone, cooking, cleaning, writing out a grocery list, and other basic household stuff. All but one of the men failed to finish in the time period, and experienced significantly more stress in accomplishing these tasks. They also tended to do one thing at a time and got flustered when their concentration was broken. All of the women had no trouble finishing the same tasks from the same list in the same environment.
Re:that space would almost fit two cars (Score:5, Interesting)
Hell, I can carry a disk between computers faster than that 300 baud modem. If it can't transmit data any faster than that, it doesn't have much practical purpose. What's that? It'll get faster once people start using it and the technology improves? Huh. Who'd a thunk it?
Sheesh, people, lighten up. Proof of concept.
Re:This is cool (Score:5, Interesting)
Men and women were asked to draw a bike from memory. Women always had all the right parts, but rarely in the right places. Men rarely had all the parts, but always in the right places. There is a difference in how men and women's minds work. It's not that one is better or worse than the other, it's that they're different.
Re:This is cool (Score:3, Interesting)
Maybe you could do an experiment to show that men "have a higher capacity for algebra", or "have a higher capacity for mechanical engineering," by doing some basic tests. However, in our society, women are encouraged NOT to understand algebra and engineering. Why? Because a woman who does is unattractive!! Think about the stereotypes and the status quo here - imagine what the general student opinion would be of a girl in high school who took and aced advanced calculus? Do you think the guys would be "all over her"? What would her friends think? In the greater scheme of the social environment, we enforce gender difference in a very complex interlocking web of pressures.
Re:Questions... (Score:3, Interesting)
I appreciate the detailed article. I will confess up front that I am going to reply without reading the entire thing, but from what I read (could grasp quickly without resorting to formulae) it appears to describe a normal feedback system augmented by visual sensors.
The thing is, AFAIK, that people usually don't handle controls the same way a machine does. People (although my statistical sample is the smallish one -- myself) will remember how a machine reacts to their inputs and (attempt to) compensate when dealing with those controls. In addition to sight I also rely on hearing and vibration to tell me what the machine is doing. Machines, even ones sophisticated enough to employ a visual augmentation, don't have those (occasionally critical) advantages.
As an example, my minivan has a sticky accelerator first thing in the morning. I know this, so to make sure it doesn't jump when I try to back out I rev the engine a bit while still in park to clear the stickiness. It doesn't happen after the first start of the day but it has become ritual. A computer, unless employing a learning system or very well programmed, won't act the same way. A feedback system would normally just keep applying force until it got an expected reaction (motion). In my sticky-peddal problem the computer may very well have just rammed the car ahead/behind before realizing it got more force than expected. If I find excessive resistance in the pedal I don't go any further unless I'm SURE I'm in Park.
Maybe at the millisecond resolution computers can handle the feedback and the my sticky pedal issue isn't a problem (not being in that particular field of computing I wouldn't know). But I suspect those sorts of tactile-memory skills wouldn't transfer easily or well to a machine.
OTOH, I could be completely off base. It wouldn't be the first time!
obligatory (Score:2, Interesting)
Re:This is cool (Score:2, Interesting)
This girl was one of the most physically attractive, popular, and intelligent people in our school, while also being a math and band nerd.
Already been streamlined (Score:2, Interesting)
Re:This is cool (Score:2, Interesting)