RE: Robot car closes on driver track times
Discussion
McSam said:
405dogvan said:
McSam said:
Almost no chance whatsoever.
A car and a circuit are far, far more variable than any computer can be programmed for, and no matter how thorough a job is done, it isn't going to come close to a proper, experienced race driver in any of our lifetimes.
To say a computer could ever drive faster than the fastest human misunderstands how much of an art driving a race car is, and the limitations of programming!
'I think there is a world market for about five computers.' T Watson - Founder of IBMA car and a circuit are far, far more variable than any computer can be programmed for, and no matter how thorough a job is done, it isn't going to come close to a proper, experienced race driver in any of our lifetimes.
To say a computer could ever drive faster than the fastest human misunderstands how much of an art driving a race car is, and the limitations of programming!
You're THAT wrong...
All you need is money - the technology already exists but you're going to trash a few cars along the way and it will take time for the computers to 'learn' the fastest way around the track (and their mistakes will be expensive and noisy).
Thing is - once you've done this, you'll have a program which can drive the track in any conditions - could learn changes to the track - could analyse the surface and allsorts - all WAY beyond the abilities of a human being.
The only reason it's not happening is
a - it costs a lot
b - I'm not sure what you gain by it - nothing new need be invented.
Edited by 405dogvan on Monday 5th November 11:35
But that completely disregards that the conditions change the fastest line around the circuit, and that's where the art lies. The prediction, not only by looking at the condition of the surface at an exact point but also by the feeling through his arse, that a race driver can make is far beyond the abilities of computing, and it means that the driver will be able to find the fastest way far more easily and more intuitively than the computer. Iteration methods, even really good ones, are such that a large change in conditions could take thousands of iterations for the computer to work its way back to the very fastest line. An experienced driver will have it in five laps.
And even assuming the programming is possible, all this could be done with only one car, and would need complete reprogramming to use anything else, unlike a real driver.
There's almost no question of ever making a "better" race driver than a person.
ETA - Clarifying I'm talking only about race drivers.
Edited by McSam on Monday 5th November 11:52
For example the senses a driver relies upon:
"condition of the surface at an exact point" - a robotic vehicle could benefit from a spectral range far greater than a humans and with a greater resolution. The driver thinks that patch of tarmac might be wetter and colder so alters their line. A robot knows thats it wetter and colder and knows that given its own current tyre temps the deviation to a different line would cost more time or speed than it'll save.
"the feeling through his arse" - Accelerometers and microphones can again monitor a wider range of frequencies than a driver and know the cause of the symptom that is the vibration. It could perceive tread shuffle or a failing damper long before a driver.
As was said, your only limits are how much you want to pay and the time invested in developing a system that monitors everything it needs to and and combines new information into existing known successful actions.
And once you've paid even more to rationalize this system you could put it in any car.
On a controlled circuit I really could see manufacturers using robo cars, certainly for non-stop durability testing.
And if you wanted to get the best laptime from your CAD designed car, thats been through CAE testing, built to precise tolerances on a robotic production line why would you not want to know it was being driven as fast and repeatedly as it could to attain the very best laptime where every millimetre of its route had been calculated and refined with the precision the car itself was made and designed?
McSam said:
But that completely disregards that the conditions change the fastest line around the circuit, and that's where the art lies. The prediction, not only by looking at the condition of the surface at an exact point but also by the feeling through his arse, that a race driver can make is far beyond the abilities of computing, and it means that the driver will be able to find the fastest way far more easily and more intuitively than the computer. Iteration methods, even really good ones, are such that a large change in conditions could take thousands of iterations for the computer to work its way back to the very fastest line. An experienced driver will have it in five laps.
And even assuming the programming is possible, all this could be done with only one car, and would need complete reprogramming to use anything else, unlike a real driver.
There's almost no question of ever making a "better" race driver than a person.
ETA - Clarifying I'm talking only about race drivers.
I think you're completely underestimating the increasing power of computers and the rise of self-learning, heuristic software.And even assuming the programming is possible, all this could be done with only one car, and would need complete reprogramming to use anything else, unlike a real driver.
There's almost no question of ever making a "better" race driver than a person.
ETA - Clarifying I'm talking only about race drivers.
Edited by McSam on Monday 5th November 11:52
As has been stated earlier in the thread, if this was a mission-critical development project it would just need a larger budget to beat racing drivers within a couple of years. Within 5 years i doubt the best racing drivers would see which way the robot car went.
Mr Whippy said:
What's the point?
You may as well just run a simulation and get the result in seconds
Human drivers might naturally look at a shaded area in winter and think ice, slow down, will the AI driver do that? Leaves on the road reducing grip?
the point would be to extract the fastest possible laptime to help sell/promote the car, though I guess that'd be dependent on how much faster a robot could lap a circuit in comparison to a top driver?You may as well just run a simulation and get the result in seconds
Human drivers might naturally look at a shaded area in winter and think ice, slow down, will the AI driver do that? Leaves on the road reducing grip?
Edited by Mr Whippy on Monday 5th November 11:50
a simulation will never be 100% accurate :P
I don't see why a car couldn't have some kind of laser scanner(s) and analyze the road ahead for leaves and other possible hazards
otolith said:
Mr Whippy said:
Human drivers might naturally look at a shaded area in winter and think ice, slow down, will the AI driver do that? Leaves on the road reducing grip?
It will if you program it to - and it has as many extra senses as you can devise sensors for, including those belonging to the car.It's is these elements that are still very tricky in AI terms. A human can reason based on all sorts of past experiences to then interpret a new situation appropriately, and do it quickly too.
Just how much data do we want the computer to assess, and how much 'experience' do we expose it to to make the right choices even in abstract conditions?
Ie, make it know the difference between an old person and a young child, and then say run over the old person as they have lived their life plenty, if the only two choices for vehicle trajectory lead through people?
Or do we leave it up to the computer to pick purely rationally based on some other factors?
An extreme case but how far do we go?
Ie, might a program see a bus-stop full of people and ram you into a wall killing you rather than letting you skid into the 'soft' bus-stop?
All very worrying that computers will be determining such things rather than fundamentally compassionate and forgiveable humans!
I bet I could trick one to run some poor bugger over if I knew how the algorithms worked... at least until they are so computationally powerful and aware they match humans, but I don't believe AI will be that close for another 50yrs.
Hmmm
Dave
They'll need a lot more than 3 computers in the end to mimic the reactions, thoughts and experience of a human.
Computers may be able to super multi-task all day long every day but they don't have the deft and instinctive touch of a professional human driver, nor can they make an on the spot decision based upon the driving conditions, steering feel and sensation a pro driver uses to action and control inputs.
The real world is the analogue world and even in technology terms you still need some sort of A-D converter to interface between the Analogue and Digital world. The human is that interface and always will, period!!
The day a computer can do everything a human can as well as a human can is the day we should all be stting it!!
I'll be back!!
Computers may be able to super multi-task all day long every day but they don't have the deft and instinctive touch of a professional human driver, nor can they make an on the spot decision based upon the driving conditions, steering feel and sensation a pro driver uses to action and control inputs.
The real world is the analogue world and even in technology terms you still need some sort of A-D converter to interface between the Analogue and Digital world. The human is that interface and always will, period!!
The day a computer can do everything a human can as well as a human can is the day we should all be stting it!!
I'll be back!!
otolith said:
Johnspex said:
Blimey, I'm not the only one on here who doesn't think he's a driving god. Wow, 2 of us. Imagine that, amongst all these "expert" drivers on here.
I think he's commenting more on how fast he can run...I am, like all PHers except Johnspex, a driving god
otolith said:
Johnspex said:
Blimey, I'm not the only one on here who doesn't think he's a driving god. Wow, 2 of us. Imagine that, amongst all these "expert" drivers on here.
I think he's commenting more on how fast he can run...julian64 said:
Yes, but at least you missing the joke was funnier than the joke itself, so that cheered me up.
My witty remark was inspired by this, earlier;
4key said:
kambites said:
What's best to put on components for rust protection is a huge can of worms.
do you have to rub the worms into the metal? Do they come in some sort of paste form or are they still live?Gassing Station | General Gassing | Top of Page | What's New | My Stuff