In other news today...
-
@Rhywden said in In other news today...:
@Carnage said in In other news today...:
@Rhywden said in In other news today...:
@acrow said in In other news today...:
@Bulb said in In other news today...:
@acrow said in In other news today...:
proper co-ordinate accurate maps
Nobody's ever going to rely on that. Any self-driving or self-flying aircraft must have sensors to survey the surroundings in real time. No map can ever do that.
Teslas have shown that machine vision is insufficient. And a LIDAR gives up if there's snow falling. So the only practical sensing options left for staying on the road (that I know of) are GPS, and guide signal wires over/under the road. Also maybe this: https://hackaday.com/2020/04/06/navigating-self-driving-cars-by-looking-at-whats-underneath-the-road/
Erm, you do realize that you can combine several types of sensors? You also forgot about radar.
And SONAR, which is more relevant for cars.
Erm, radar is already in widespread use. Most adaptive cruise control systems rely on that.
Narrow-beam radar. I forgot about that. A very narrow beam, to keep it from picking up potholes in the road.
-
@acrow said in In other news today...:
@Rhywden said in In other news today...:
@Carnage said in In other news today...:
@Rhywden said in In other news today...:
@acrow said in In other news today...:
@Bulb said in In other news today...:
@acrow said in In other news today...:
proper co-ordinate accurate maps
Nobody's ever going to rely on that. Any self-driving or self-flying aircraft must have sensors to survey the surroundings in real time. No map can ever do that.
Teslas have shown that machine vision is insufficient. And a LIDAR gives up if there's snow falling. So the only practical sensing options left for staying on the road (that I know of) are GPS, and guide signal wires over/under the road. Also maybe this: https://hackaday.com/2020/04/06/navigating-self-driving-cars-by-looking-at-whats-underneath-the-road/
Erm, you do realize that you can combine several types of sensors? You also forgot about radar.
And SONAR, which is more relevant for cars.
Erm, radar is already in widespread use. Most adaptive cruise control systems rely on that.
Narrow-beam radar. I forgot about that. A very narrow beam, to keep it from picking up potholes in the road.
You do realize that phase-shift arrays are a thing? Besides, all radar systems are "narrow-beam".
-
@Rhywden said in In other news today...:
@acrow said in In other news today...:
@Rhywden said in In other news today...:
@Carnage said in In other news today...:
@Rhywden said in In other news today...:
@acrow said in In other news today...:
@Bulb said in In other news today...:
@acrow said in In other news today...:
proper co-ordinate accurate maps
Nobody's ever going to rely on that. Any self-driving or self-flying aircraft must have sensors to survey the surroundings in real time. No map can ever do that.
Teslas have shown that machine vision is insufficient. And a LIDAR gives up if there's snow falling. So the only practical sensing options left for staying on the road (that I know of) are GPS, and guide signal wires over/under the road. Also maybe this: https://hackaday.com/2020/04/06/navigating-self-driving-cars-by-looking-at-whats-underneath-the-road/
Erm, you do realize that you can combine several types of sensors? You also forgot about radar.
And SONAR, which is more relevant for cars.
Erm, radar is already in widespread use. Most adaptive cruise control systems rely on that.
Narrow-beam radar. I forgot about that. A very narrow beam, to keep it from picking up potholes in the road.
You do realize that phase-shift arrays are a thing?
I do. They're used to steer the radar. They also take up space, making them unsuited for a car's nose, AIUI.
And I don't deny that radar is handy for finding just how far the next car is. But that's an easy case, since it can ignore returns that have a difference of more than 10-20km/h to the car's own speed. I.e. potholes and road seams.
All radar systems are "narrow-beam".
Not quite. If you want to spot e.g. incoming missiles, you'd do a wide beam and filter on speed. Also, to find precise angle to target, you'd need to "sweep" the view. As opposed to a spot-beam on the cruise control.
Also, the cruise control doesn't need to actively steer the beam, I think. Does it? If not, then a formed antenna suffices. No need for active beam control.
-
@acrow The size is most likely due to the power requirements. With a reduction in power you can probably reduce the size quite a bit.
-
@acrow said in In other news today...:
Also, the cruise control doesn't need to actively steer the beam, I think. Does it? If not, then a formed antenna suffices. No need for active beam control.
It might need to adjust in corners? Probably by responding to the steering angle.
-
Speaking of saturating the medium with signals so that each individual radar, lidar and sonar can't tell which signal is it's own, can't this be worked around by having the ping be a signal with identity encoded in it and filtering the noise to find your own signal?
I mean, I'm pretty sure that this is a solved problem already, as all the technologies are fairly old and mature, and for using them to read the near area in cars, you'd not need extremely high performance so you could spend a bit of time adding extra bells and whistles to the system and adding a few ms of roundtrip for each ping.
-
@Rhywden Possibly. ...Depending on the frequency. The positioning and size of the antennas depends a bit on the wavelength, doesn't it?
For an example, size-wise, of a radar capable of sweeping and tracking, well, there's the tank point-defence systems. E.g.: https://en.wikipedia.org/wiki/Trophy_(countermeasure)
I think they made those as small as possible, though the tech is a few years old at this point.
-
@Carnage I think LIDAR only observes a very narrow slit view (where it shone the laser in the first place, plus a few degrees in one direction for beam shift from distance) for the return, so two of them would need to paint the same place at the same time for interference to happen. And they spin contnuously, so that doesn't happen often.
But it's continuously spinning, so trying to encode the transmit (with anything except a continuos wave maybe?) would lower the resolution to the span required for a complete ID round-trip.
Sonar... your ID transmit needs to be shorter than the round-trip time. Plus, it'll echo from more than one point, so trying to read the info contained again may not be possible even if it's alone in transmitting.
-
@acrow said in In other news today...:
@Rhywden Possibly. ...Depending on the frequency. The positioning and size of the antennas depends a bit on the wavelength, doesn't it?
Well, the default answer would be "length is a integer multiple of the half wavelength".
Though, of course, there's weird stuff like fractal antennas.
-
So this is kinda huge:
-
@acrow said in In other news today...:
Also, the cruise control doesn't need to actively steer the beam, I think. Does it? If not, then a formed antenna suffices. No need for active beam control.
It really doesn't take that much power or space to make a phase array, and by using a modulated beam you can spot it coming back in the reflection even among a lot of other chatter and noise. An upgraded version of that is used in 5G telecoms, and the transceivers for those are able to run in the power and size budget of a mobile phone. On a car? Trivial!
Signal processing in hardware is a hell of a thing.
-
-
The economic implications will hav the same kind of lasting effects that world war 2 did on Japan and Germany.
Only 2 countries you can count on to not have economic impact are India and China.
-
@acrow said in In other news today...:
, for one thing; thay're all
effectivelyliterally radio transmittersIt's in the name.
@acrow said in In other news today...:
and will hear each other if jammed to every car.
Yes, there's algorithms to ensure they know which signal is theirs. We recommend finding the article on how a cave of bats manages to not have mid-air collisions despite every participant literally screaming at each other.
-
@acrow said in In other news today...:
would lower the resolution to the span required for a complete ID round-trip.
Just how slow do you think light travels?!?
-
@Nagesh said in In other news today...:
The economic implications will hav the same kind of lasting effects that world war 2 did on Japan and Germany.
Only 2 countries you can count on to not have economic impact are India and China.
I think you're lost. The cake news is
-
@Tsaukpaetra said in In other news today...:
I think you're lost. The cake news is
The cupcake thread is too, but different arrows.
-
@dkf said in In other news today...:
@acrow said in In other news today...:
Also, the cruise control doesn't need to actively steer the beam, I think. Does it? If not, then a formed antenna suffices. No need for active beam control.
It really doesn't take that much power or space to make a phase array, and by using a modulated beam you can spot it coming back in the reflection even among a lot of other chatter and noise. An upgraded version of that is used in 5G telecoms, and the transceivers for those are able to run in the power and size budget of a mobile phone. On a car? Trivial!
Signal processing in hardware is a hell of a thing.
But only if you have a type R phase discriminating amplifier. And don't forget to remodulate the reflector dishes and isolate the main core with a fractal code.
-
@topspin Don't forget the turboencabulator.
-
-
@Nagesh said in In other news today...:
The economic implications will hav the same kind of lasting effects that world war 2 did on Japan and Germany.
Only 2 countries you can count on to not have economic impact are India and China.
China already has an economic impact, so I doubt that.
-
@Tsaukpaetra said in In other news today...:
@acrow said in In other news today...:
would lower the resolution to the span required for a complete ID round-trip.
Just how slow do you think light travels?!?
Round-trip meaning here the time required to transmit and receive your ID. Light traveling fast is moot when your baudrate is limited by the return filtering and decoding.
-
@dkf On the other hand,
- It still takes extra money make an active array. Car manufacturers like profit.
- You need the frequency band legally reserved or allotted, which may or may not be near the 5G band.
- Radars are not yet as common as cellphones, so manufacturing scale didn't bring the price down (barring commonality with 5G HW).
-
@acrow said in In other news today...:
your baudrate is limited by the return filtering and decoding.
Why would it be though?
Put in another way, why would a phase-locked-loop signal be limited by the data it's transmitting?
It would be like saying you can't verify a set of UDP packets until the whole set has been fully transmitted....
But anyways, recommend taking this out of the news thread if you're so interested.
-
@Tsaukpaetra said in In other news today...:
@acrow said in In other news today...:
your baudrate is limited by the return filtering and decoding.
Why would it be though?
Because you can't data faster than yor hardware can...?
Put in another way, why would a phase-locked-loop signal be limited by the data it's transmitting?
A point, you have there.
It would be like saying you can't verify a set of UDP packets until the whole set has been fully transmitted....
No. It's like saying you can't verify a packet until you receive all of it.
But anyways, recommend taking this out of the news thread if you're so interested.
Not that interested. And this is like the last topic to stay on topic. ...And this time started
on-topic.with tamper-proof pizza boxesEdit: FTFM
-
@acrow said in In other news today...:
@Tsaukpaetra said in In other news today...:
@acrow said in In other news today...:
your baudrate is limited by the return filtering and decoding.
Why would it be though?
Because you can't data faster than yor hardware can...?
if software-defined radio is something a hobbyist can do for a few lincolns, the problem might not be in the hardware....
Put in another way, why would a phase-locked-loop signal be limited by the data it's transmitting?
A point, you have there.
It would be like saying you can't verify a set of UDP packets until the whole set has been fully transmitted....
No. It's like saying you can't verify a packet until you receive all of it.
So you drop this nanosecond's packet and try again in the next. The loss in detail should be negligible for the distances being spoken of.
-
@Tsaukpaetra said in In other news today...:
@acrow said in In other news today...:
@Tsaukpaetra said in In other news today...:
@acrow said in In other news today...:
your baudrate is limited by the return filtering and decoding.
Why would it be though?
Because you can't data faster than yor hardware can...?
if software-defined radio is something a hobbyist can do for a few lincolns, the problem might not be in the hardware....
We are talking LIDARs here, right? You need to receive the signal back and process with hardware before you can actually get into the digital side. Money gets you more speed but less sales. And, btw, you're sharing the rx hardware with the distance measurement process that's the actual job of the LIDAR.
Also also, forget software-defined radio. This kind of pattern matching is a job for FPGAs.
Put in another way, why would a phase-locked-loop signal be limited by the data it's transmitting?
A point, you have there.
It would be like saying you can't verify a set of UDP packets until the whole set has been fully transmitted....
No. It's like saying you can't verify a packet until you receive all of it.
So you drop this nanosecond's packet and try again in the next. The loss in detail should be negligible for the distances being spoken of.
Yes. Exactly. And the amount of lost data depends on the ratio of ID baudrate to scan speed. After all, it's one ID packet's worth of angle that's now "at doubt".
So are we in total agreement yet?
-
@acrow said in In other news today...:
We are talking LIDARs here, right?
I didn't think so, but now I guess we are...
@acrow said in In other news today...:
So are we in total agreement yet?
Were we ever in disagreement?
You sure are passionate for being
@acrow said in In other news today...:
Not that interested.
-
@Tsaukpaetra said in In other news today...:
You sure are passionate
I'm a compulsive replier.
Also .
As you well know.
-
-
@Luhmann said in In other news today...:
@acrow said in In other news today...:
I'm a compulsive replier.
So you answer every question?
Nearly.
And damn you for abusing it.
-
@acrow said in In other news today...:
And damn you for abusing it.
You think I'm abusing this? Why?
-
@Luhmann said in In other news today...:
@acrow said in In other news today...:
And damn you for abusing it.
You think I'm abusing this? Why?
Your daily abuse allotment has been exceeded. Fuck you, pay him.
-
@Luhmann said in In other news today...:
@acrow said in In other news today...:
And damn you for abusing it.
You think I'm abusing this? Why?
I'm also interested in the answer to this.
@acrow Can you elucidate?
-
@Rhywden said in In other news today...:
elucidate
-
@Applied-Mediocrity said in In other news today...:
@Rhywden said in In other news today...:
elucidate
Forsooth! Woe is me!
-
@Rhywden I also learned that Baba Yetu is the Lord's Prayer in Swahili. When searching for scripture fragments in Swahili, to counter-troll with.
-
@acrow said in In other news today...:
@Rhywden I also learned that Baba Yetu is the Lord's Prayer in Swahili. When searching for scripture fragments in Swahili, to counter-troll with.
Oh, but if the trolled cannot understand what you're trolling with is a troll truly trolling the trolled?
-
-
Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook! Ook? Ook. Ook? Ook. Ook. Ook. Ook? Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook? Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook? Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook? Ook. Ook? Ook. Ook? Ook. Ook? Ook. Ook! Ook! Ook? Ook! Ook. Ook? Ook. Ook? Ook. Ook? Ook. Ook? Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook! Ook. Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook! Ook. Ook? Ook. Ook? Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook! Ook. Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook. Ook. Ook? Ook. Ook? Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook! Ook. Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook! Ook. Ook? Ook. Ook? Ook. Ook! Ook. Ook. Ook? Ook. Ook? Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook! Ook. Ook! Ook! Ook! Ook! Ook! Ook. Ook. Ook. Ook! Ook. Ook? Ook. Ook? Ook. Ook! Ook. Ook. Ook? Ook. Ook? Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook. Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook! Ook. Ook. Ook. Ook! Ook. Ook? Ook. Ook? Ook. Ook! Ook. Ook. Ook? Ook. Ook? Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook! Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook! Ook. Ook? Ook. Ook? Ook. Ook! Ook. Ook. Ook? Ook. Ook? Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook. Ook! Ook! Ook! Ook. Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook! Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook! Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook! Ook. Ook? Ook. Ook? Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook. Ook! Ook.
-
@JBert said in In other news today...:
Does the Honey Bee Sound System make this noise? Because methinks it would ward off anyone...
-
I didn't read all of that () and so I'm probably not doing it justice, but it looks like just in time for the demise of Conway, Stephen Wolfram is at it again trying to explain the whole universe with automata. Which, thinking of it, isn't that the same thing he has so boldly called "a new kind of science" 20 years ago?
Sure, all of these things are immensely interesting, but the realization that simple rules can create very complex emergent behavior isn't particularly new. Since cellular automate are Turing complete, yeah, I'm sure you can find one that produces the same emergent behavior we observe. But enumerating the possible automata and computing if they fit our universe doesn't seem to be any less infeasible than likewise enumerating the possible equation systems that could model it. Einstein or Maxwell probably didn't arrive at their equations by invoking the infinite monkey theorem.
-
@topspin said in In other news today...:
Einstein or Maxwell probably didn't arrive at their equations by invoking the infinite monkey theorem.
I am, however, becoming increasingly convinced that god does play dice with the universe.
-
@Applied-Mediocrity said in In other news today...:
@topspin said in In other news today...:
Einstein or Maxwell probably didn't arrive at their equations by invoking the infinite monkey theorem.
I am, however, becoming increasingly convinced that god does play dice with the universe.
Which one of them is winning?
-
@Applied-Mediocrity said in In other news today...:
@topspin said in In other news today...:
Einstein or Maxwell probably didn't arrive at their equations by invoking the infinite monkey theorem.
I am, however, becoming increasingly convinced that god does play dice with the universe.
The real question is whether God plays Poker or Craps.
-
-
Pentagon-watchdog-clears-microsoft-$10billion-jedi-project
Hey, the government said the government was right in picking Microsoft, color me shocked.
-
@dkf said in In other news today...:
@Applied-Mediocrity said in In other news today...:
@topspin said in In other news today...:
Einstein or Maxwell probably didn't arrive at their equations by invoking the infinite monkey theorem.
I am, however, becoming increasingly convinced that god does play dice with the universe.
The real question is whether God plays Poker or Craps.
God using dice to play Poker reminds me of this Roddenberry quote:
We must question the story logic of having an all-knowing all-powerful God, who creates faulty Humans, and then blames them for his own mistakes.
-
@topspin To be fair, I create faulty programs all the time, and blame the programs for their faults afterwards. (I might be able to be convinced that I'm not all-powerful or all-knowing, though.)
Buggy program? Off to program-purgatory with you.
-
@acrow said in In other news today...:
Baba Yetu is based on the Lord's Prayer in Swahili
It takes a lot of liberties, in the name of turning it into a song that's fun to listen to.