42 ms·

The Lowdown on Lidar

667 points by timdierks 12y ago

thrownaway2424 12y ago
Alternate title: some guy says he was doing the limit in a Porsche.
- timdierks 12y ago
  My clickbait technique is impeccable.
zinxq 12y ago
Josh Bloch is awesome. For the non-java inclined, he wrote (among other things) java.util.HashMap. Most all Java programs run (lots) of his code.
- shangxiao 12y ago
  I know him as the author of the awesome "Effective Java"
  - thescrewdriver 12y ago
    I know him as the guy who advocated the even wider use of factories...
taeric 12y ago
I confess I was expecting to not care for this. I'm not sure how any of it should be surprising, as it seemed fairly straight forward analysis of the tools. That said, this was a ridiculously fun read. More detailed than expected for some parts, and always informative. I'm glad it did not come across as anti speed detectors. Really more of a "know the limitations."
- Zelphyr 12y ago
  Me too. Though I would've been over the moon if he'd said he wore his "Back off man. I'm a scientist." shirt to court.
- mortov 12y ago
  Often (but not always) in the UK, the laser is accompanied by video footage which shows where the beam has been trained and it is a combination of both the video and laser reading (shows on the video) which is the evidence. (all tripod mounted - so I suspect many of the issues detailed are well known). Sounds like that would give good supporting evidence for a challenge. However, traffic violations is most places are about revenue raising so anything which reduces the opportunities - like accuracy and fairness - are unlikely to get much of a look in. edit: for clarity, it looks like this http://www.mirror.co.uk/news/uk-news/pictured-speeding-motorist-who-claimed-3100917 http://www.mirror.co.uk/news/uk-news/pictured-speeding-motor... also referenced in my reply below.
  - Alupis 12y ago
    how can a video possibly show where a laser is pointed at from > 1,000ft away? that seems unreasonable.
    - mortov 12y ago
      The video is required to zoom to show the driver of the vehicle as well for identification purposes. It prevents people claiming they were not the driver. (e.g. it looks like this http://www.mirror.co.uk/news/uk-news/pictured-speeding-motorist-who-claimed-3100917 http://www.mirror.co.uk/news/uk-news/pictured-speeding-motor...)
      - Alupis 12y ago
        That image may reasonably prove it was not a woman driving the vehicle, but no way can that prove it was the actual driver (or anyone's identity). It's way too blurry and pixelated. Also, a LIDAR beam shouldn't be pointed at a sloped window, so I doubt it actually proves where the beam was pointed at either.
ARothfusz 12y ago
Well, he fought with half science: he came up with a set of testable hypotheses, but didn't actually run any of the tests. I guess that's as scientific as cosmology :-)
- thrownaway2424 12y ago
  It's not even half science, it's just a big slide deck of FUD padded with a healthy amount of entitlement.
  - chiph 12y ago
    Be honest with yourself: If you got a ticket where the indicated speed was massively different than the speed you were actually travelling, you'd fight it too. And you might do some research into the validity of the reading, as well. You'd especially do it in a state where the law says "x" mph over the limit and your license may get suspended (meaning you walk for 60 days). It'd be worth fighting for.
    - gknoy 12y ago
      Especially if it were in a residential area where you had been deliberately careful. If I got a ticket saying that I'd been driving 40 in a 25 zone, when I had been carefully watching my spedometer at ~20s, I'd feel pretty wronged. I'm not sure I'd have the resources to fight it, but I would certainly want to.
  - Estragon 12y ago
    Can you cite some parts which reflect entitlement?
  - smackfu 12y ago
    FUD, eh? So that's Fear, Uncertainty, and Doubt. I don't really see any fear, and I hope that any legal defense is going to push pretty heavily on uncertainty and doubt. Don't quite see the problem.
- joncrocks 12y ago
PeterWhittaker 12y ago
Excellent presentation with considerable technical information, e.g., weaknesses in the LIDAR approach, operational realities that reduce accuracy, etc., and some legal background, e.g., limiting LIDAR to 1000' or less in some jurisdictions - but contrast with overall lack of case law. Could be of value to anyone wishing to fight a ticket.
privong 12y ago
Anecdotally, I heard of an astronomer in a similar situation, who attempted to prove the radar gun used to "catch" him speeding was incapable of measuring the speed accurately enough (presenting info on the radar gun's sampling properties and on sampling theory; I forget the specifics). Supposedly the judge maintained the speeding violation but reduced the man's fine for "entertaining the court".
justinsb 12y ago
Can anyone tell me why speeding tickets aren't enforced with video cameras (like red-light tickets are)? It would seem to be much more accurate, much easier to enforce, and much harder to dispute. Irrespective of the rights or wrongs of speed limit laws, I do believe that consistent and efficient enforcement can only be a good thing.
- toomuchtodo 12y ago
  One word: Discretion "Confronting Automated Law Enforcement" http://robots.law.miami.edu/wp-content/uploads/2012/01/Shay-EtAl-ConfrontingAutomatedLawEnf.pdf http://robots.law.miami.edu/wp-content/uploads/2012/01/Shay-...
  - Gracana 12y ago
    That's a pretty short summary of a 44 page document. Can you elaborate a little more?
    - toomuchtodo 12y ago
      Sure. Do you want to live in a world where every time you violate the letter of the law you're penalized? Before you answer, consider that you're going to have to keep 100-200 years of law hot in memory to ensure your daily compliance.
      - Gracana 12y ago
        Why bother asking a question when you word it in such a biased way? Obviously my answer is yes, that is the world I want to live in.
        toomuchtodo 12y ago
        Because that's a terrifying world to live in. I am biased. To accept the idea of compliance with the law to the letter constantly is unreasonable. Roll 1-3 feet over the white stop line? Ticket. 1-2mph over the speed limit? Ticket.
ihaveajob 12y ago
Speed radars are good for speed traps, but not so great to prevent speed related incidents. They share the same fatal flaw with all other single-point speed meters: You can just slow down for a minute, while you pass by the speed trap, and then speed up for the rest of your trip. Instead, a 2-point speed meter (i.e. license plate readers every few miles, measuring time between matching reads) system is superior because a) their margin of error is negligible and b) they measure sustained speed over a long distance rather than at a single hot spot, thus making roads safer. I can't believe it's not implanted at least in all interstate freeways.
- legohead 12y ago
  If you've traveled around the country, you've invariably come across those "speed is monitored by aircraft" signs, and what you described is how they do it (tracking how long it takes cars to go from point A to B)
  - onan_barbarian 12y ago
    I've always been a bit nervous about those "speed is enforced by aircraft" signs (as opposed to "monitored"). I always get a mental picture of a aircraft strafing near my car until I slow down...
    - dTal 12y ago
      You may enjoy this image from XKCD then: http://what-if.xkcd.com/imgs/a/87/radar_aircraft.png http://what-if.xkcd.com/imgs/a/87/radar_aircraft.png
  - cardiffspaceman 12y ago
    In California the CHP (California Hwy Patrol) plane has to do the same thing the CHP car would have to do: match your speed and then note its own speed. Measuring time between points is forbidden no matter what the observer's vehicle is. Radar is also used from the ground but was not available to CHP when they started using aircraft. The planes are on the inexpensive side so sometimes they have to dive to achieve the speed. Also recently CHP got to use radar statewide so they don't use aircraft so much.
chillingeffect 12y ago
The highlights: 1. Lidar gun manufacturers test them on limited and ideal circumstances 2. Beam divergence and shakiness make them less useful past 1000 feet, which is generally acknowledged.
awjr 12y ago
The reading was taken at a distance of 1300 feet with a handheld device that is not meant to be used handheld beyond 1000 feet. That's a significant distance. I'm assuming the cop was hiding or trying to stay hidden.
darksim905 12y ago
A lot of this is common knowledge now. You can check out guys of LIDAR who research these types of guns & how accurate their findings are. This is also why cops should be taught to do several speed readings & to point at the vehicle, not the headlights. Also, depending on the paint, it can also cloak the vehicle's speed or make it take longer to get back to the gun.
leephillips 12y ago
I hope this information, or the equivalent, filters into the awareness of judges. I was in traffic court in Virginia last year and, as I was waiting for my case, heard the judge explain to several defendants who were there to challenge their lidar tickets that the lidar was essentially infallible and could not be confused by other cars on the road. And this was an intelligent judge who seemed unusually sympathetic to the accused.
- Guvante 12y ago
  Unfortunately judges are people and if you have tons of people coming in with just "LIDAR is not accurate enough to prove reasonable doubt" compared to cops bringing in studies proving their accuracy it is difficult to listen to it.
DEinspanjer 12y ago
All the talk in this thread about various forms of speed management and fines remind me of an article I read a while back about a Swedish test program that used fines from speeders as the funds for a lottery that people driving at or under the limit were entered into automatically: http://www.wired.com/2010/12/swedish-speed-camera-pays-drivers-to-slow-down/ http://www.wired.com/2010/12/swedish-speed-camera-pays-drive...
voidlogic 12y ago
I've wondered sometimes if it would be possible to fight erroneous speeding tickets using the repeated GPS pings your phone or nav unit makes while using them for directions.
MattGrommes 12y ago
As a college intern at a national lab my boss' boss got a speeding ticket on the military base we worked on. He spent some time doing research and when he presented his evidence in court the judge apparently told him he wasn't going to have to pay the fine but the judge wasn't going to find him not guilty because he wasn't going to risk invalidating all speed gun findings. I don't recall if it was lidar but it's funny the stories are so similar. It's hard to use technology to mess with engineers. :)
- aroch 12y ago
  Lower courts (even in the Military) can't really establish any meaningful precedence
- reustle 12y ago
  If you were really gung-ho about getting it marked as not guilty, could you appeal it and move up the chain?
ylhert 12y ago
LIDAR sucks and there is no judicial notice on it in California. If you ever get a LIDAR ticket, do a trial by written declaration (or TBD). Just state that you dispute the validity and accuracy of the LIDAR reading and they'll usually throw it out. Remember to ask the officer(after he has given you your ticket) what model laser/radar detector he was using, and at what distance he measured you at. Anything over 800/1000 ft(this is really close, and rarely do they get you this close) will usually automatically be struck down if you question LIDAR's validity at that distance. That is a case you will win every time, if you do it right. The court cannot prove it's accuracy at these ranges and they know it. Also, rule of thumb, all CHP officers have laser only on their bikes now. source: I used to have a Lancer Evolution
- gknoy 12y ago
  >> Remember to ask the officer(after he has given you your ticket) what model laser/radar detector he was using, and at what distance he measured you at. That's an interesting point. Would that be admissible, though? Wouldn't you saying "He said it was at X feet" would be considered hearsay in court, versus a cop saying "I took a reading at X/2 feet"?
  - ylhert 12y ago
    He/she should normally write it on the ticket(pretty sure they have to, it's always been there on tickets I've seen), but if you can get him/her to say it out loud, it's evidence that you can probably use against him/her. Also you should record your interactions with all officers. My final not a lawyer advice on beating speeding tickets: shut your mouth and don't answer the officers questions. If the officer asks how fast you were going, nothing you can say helps you, especially "I don't know". Just do your best to be polite, but firm, and don't give him anything more than a nice attitude
    - lawnchair_larry 12y ago
      This isn't always the best strategy. If you're unlikely to make a court date, being remorseful and apologetic has a decent chance of getting you a reduction from the officer. If you play hardball, so will he, then you have no choice on going to court if you want a reduction. If you're in serious trouble, STFU is the correct choice, but it's not necessarily the most economical for a moving violation.
ape4 12y ago
Don't Google's self driving cars use LIDAR.
- diminoten 12y ago
  They probably take all of these things into account. If they didn't, I suspect Google's self driving cars wouldn't use LIDAR.
kudu 12y ago
Why do HN mods keep changing the title from meaningful ones, which make the reader interested in the content, to generic, vague ones? You'll see that there's definitely a correlation between the moment that HN mods change the title and the post dropping off the top frontpage spot.
- Karunamon 12y ago
  Agreed - the original title, about a guy beating a LIDAR ticket, was much more descriptive than this headline, which effectively says "This article is about LIDAR".
  - shangxiao 12y ago
    It was the fact that the guy was Joshua Bloch which got me to read the presentation - and I'm totally glad I did! I would never have read it if it was just named "The Lowdown on Lidar"
    - puredemo 12y ago
      So appeals to authority work really well on you..
      - gwern 12y ago
        BTW, your downvotes are an opportunity to reflect on why your comment was obviously wrong and counterproductive.
  - lnanek2 12y ago
    Yeah, I had to read a lot of slides wondering why it was repeating the same thing over and over before I realized he was trying to beat a ticket. It definitely isn't a general purpose tech primer like the new title makes it sound, it's just focused on covering an error situation that gives you a legal out.
    - DanBC 12y ago
jstalin 12y ago
In Michigan, a speeding ticket is a civil infraction, not a criminal offense, so the legal standard is preponderance of the evidence, not beyond a reasonable doubt, as the slide deck says. I suspect it's the same in California.
- dtrizzle 12y ago
  In CA, infractions are crimes so the reasonable doubt standard applies. See Cal. Pen. Code section 16. http://law.onecle.com/california/penal/16.html http://law.onecle.com/california/penal/16.html
  - jervisfm 12y ago
    Wait a minute, if that's true won't it imply that traffic violations/infractions convictions get you a criminal record ?
    - dtrizzle 12y ago
      Yes. Infractions are crimes so yes. For the most part, they are not treated as crimes. You cannot be jailed based on an infraction. Most job applications I've seen typically explicitly exclude requiring information about past infraction violations.
mrfusion 12y ago
When I was a teenager I tried to fight a speeding ticket, and prepared a presentation like this with several good points (I think). Come court day, I got all dressed up, stood up when called and started my arguments ... and the judge simply cut me off after two sentences and said "guilty". I'm surprised this guy was allowed to present his whole argument and it actually convinced a judge.
- jc4p 12y ago
  From the beginning, it sounds like he's presenting this to a group of people invited by his lawyer, and the lawyer took care of the actual ticket resolution.
- timdierks 12y ago
  This presentation was not to a court; see slide 51 for the resolution: his attorney bargained it down to a zero-points violation before trial.
- scotty79 12y ago
  I'm sure it was valuable lesson for you that law is not concerned with the truth as much as it is with order.
- driverdan 12y ago
  Did you appeal? You can always appeal your first court hearing. The key is to talk to the prosecutor beforehand. They will almost always plea it down.
  - mrfusion 12y ago
    This was in Maryland. It seemed you went right to the courtroom and talked to a judge along with 40 other people. There was no prosecutor present as far as I could tell. I've fought them in Connecticut later and there you only meet with a prosecutor and they warn you all over the place that if you don't take the deal and try to go to court the judge can significantly increase the fine :-(
supernova87a 12y ago
So regardless of the science behind it, what is the successful legal forum / mechanism to raise doubt about the accuracy of the technology being used by police?
darkmighty 12y ago
I think he missed a couple important points. The sweep effect, shaking and aiming are more amenable to signal processing than it seems, depending on the resolution and noise in the system. I tried to illustrate in pictures: 1) Just to demonstrate that a nonuniform sweep is amenable to detection or processing, consider this picture, with the correct estimate in green: http://imgur.com/FsPAwCn http://imgur.com/FsPAwCn 2) Now imagine the device is more sophisticated. Consider it has a very large resolution around the incoming pulse: http://imgur.com/z8UrhUw http://imgur.com/z8UrhUw In this case the beam width is actually beneficial, and so are large distances. The sweep effect would cause a shift in the peak intensity shown in the graph, but the object depth frustrum should be clear across samples (and dislocating at t'=t-v*dt/c). 3) Officers can be trained to recognize situations unforable, regardless of distance. One fast car passing a slow one would make it easy to confuse the equipment. A clear desert however doesn't present many threats for error.
- Trombone5 12y ago
  I think its funny how he claims location sampling at different times is not a measurement of speed. It's strange how he mentions the sampling speed, but then pretends it's 1/3 second. Basically what he is attacking is the worst design of a speed gun that he could imagine from the public information. A more fair challenge would be to find sufficient error from a competent design of a speed gun, but then it wouldn't be as self serving.
dzhiurgis 12y ago
Perhaps anyone can explain me why Lidar is not dangerous to sight? Whenever you see anyone working with lasers they'd wear protective glasses. Now I realise Lidar uses infrared wavelength which is not visible to us, but that does not mean it's not dangerous, or does it?
- bicubic 12y ago
  Non-visible lasers pose a much bigger hazard than visible. Intense visible light triggers an involuntary defence called the blink reflex whereby we blink before the retina absorbs enough energy to get damaged. Lasers outside the visible range don't trigger the blink reflex and can destroy the retina before you know what happened. Lidar operates at relatively low power and relies on very sensitive detection in a very narrow wavelength range to identify reflections.
xchaotic 12y ago
Exec summary: a geek/nerd gets a speeding ticket driving a porsche. 5 years later he's still doing powerpoints on it rather than getting a life
trhway 12y ago
now if only somebody published a similarly powerful defense on "traffic control device violation" :)
Shivetya 12y ago
How does the officer confirm which vehicle the LIDAR selected? I have never been clear on this point.
- njharman 12y ago
  I believe they have "sights" and are aimed like a gun.
  - icelancer 12y ago
    Cops routinely shoot innocent bystanders.
- john_b 12y ago
  Having fought a LIDAR ticket before, the officer simply claims that he or she targeted your car and the judge admits the claim as a fact without question. Any attempts on your part to question this will be met with a reminder that the officer is Certified and Trained and you are just a random person with no qualifications to question the officer's abilities. You can ask the officer if there was a wind that day, and if so how heavy. You can ask if the officer used a tripod or held the gun freehanded. Even if the officer admits that there was a heavy wind, the gun was used freehanded, and he or she pinged you in traffic at 950 ft, the judge will still accept the officer's word that your car, and only your car, was targeted. Why? Because it's easier for the court, because they know you can't prove without a doubt that the officer targeted a different car, and because it is an important source of income for the city or county which has summoned you. It's a scam, but it's one which Americans in general are comfortable with because it gives them the illusion of safety.
Marcus10110 12y ago
Next time I get a parking ticket, I'll be sure to investigate the temperature stability of the parking meter's clock.
Relaxx 12y ago
I'm surprised the angle of measurement never came up. Isn't it the case that a reading has to be head-on for it to be accurate? It seems pretty obvious that at higher angles the observed distance delta will be smaller.
disillusioned 12y ago
I have lidar jammers on my car, but Arizona doesn't use a lot of lidar. (Motors and sometimes bored DPS will, but typically it's Ka band radar via Stalker Duals.) Not really knowing how lidar worked, except that a "laser" pulse hits the car and reads back a speed, this article was great because it explains that the measurement process is far slower than you'd think (0.3 seconds!) which is easily enough time for my jammer to detect the inbound infrared and start spamming light on that frequency, which is what they do. I don't think I've ever successfully jammed one, since again, AZ is mostly Ka, but it's pretty neat to know that they can work as advertised (since they're built in to the front of the car most likely to be targeted by lidar).
- tempestn 12y ago
  I've read that in some jurisdictions police have started confiscating jammers and even charging users with obstruction of justice, on the theory that while owning them is explicitly legal, using them prevents police from carrying out their duties. Do you know what the history in AZ is like?
hellbreakslose 12y ago
Great thread loved it... But thats why we GREEKS invented another patent that works. You won't get a fine if you are 15-20mph over the limit, EVER!
Jimny22 12y ago
Well the author may be a clever computer scientist but he knows nothing about physics and LIDAR equipment. Pathetic I'm afraid.
- BHSPitMonkey 12y ago
  Given that (1) you're an expert in these areas and (2) you found no points in the author's work worth refuting, I guess it follows that the author is actually an expert as well.

667 points by timdierks 12y ago

thrownaway2424 12y ago

Alternate title: some guy says he was doing the limit in a Porsche.

timdierks 12y ago
My clickbait technique is impeccable.

zinxq 12y ago

Josh Bloch is awesome. For the non-java inclined, he wrote (among other things) java.util.HashMap. Most all Java programs run (lots) of his code.

shangxiao 12y ago
I know him as the author of the awesome "Effective Java"
- thescrewdriver 12y ago
  I know him as the guy who advocated the even wider use of factories...

taeric 12y ago

I confess I was expecting to not care for this. I'm not sure how any of it should be surprising, as it seemed fairly straight forward analysis of the tools. That said, this was a ridiculously fun read. More detailed than expected for some parts, and always informative. I'm glad it did not come across as anti speed detectors. Really more of a "know the limitations."

Zelphyr 12y ago
Me too. Though I would've been over the moon if he'd said he wore his "Back off man. I'm a scientist." shirt to court.
mortov 12y ago
Often (but not always) in the UK, the laser is accompanied by video footage which shows where the beam has been trained and it is a combination of both the video and laser reading (shows on the video) which is the evidence. (all tripod mounted - so I suspect many of the issues detailed are well known). Sounds like that would give good supporting evidence for a challenge. However, traffic violations is most places are about revenue raising so anything which reduces the opportunities - like accuracy and fairness - are unlikely to get much of a look in. edit: for clarity, it looks like this http://www.mirror.co.uk/news/uk-news/pictured-speeding-motorist-who-claimed-3100917 http://www.mirror.co.uk/news/uk-news/pictured-speeding-motor... also referenced in my reply below.
- Alupis 12y ago
  how can a video possibly show where a laser is pointed at from > 1,000ft away? that seems unreasonable.
  - mortov 12y ago
    The video is required to zoom to show the driver of the vehicle as well for identification purposes. It prevents people claiming they were not the driver. (e.g. it looks like this http://www.mirror.co.uk/news/uk-news/pictured-speeding-motorist-who-claimed-3100917 http://www.mirror.co.uk/news/uk-news/pictured-speeding-motor...)
    - Alupis 12y ago
      That image may reasonably prove it was not a woman driving the vehicle, but no way can that prove it was the actual driver (or anyone's identity). It's way too blurry and pixelated. Also, a LIDAR beam shouldn't be pointed at a sloped window, so I doubt it actually proves where the beam was pointed at either.

ARothfusz 12y ago

Well, he fought with half science: he came up with a set of testable hypotheses, but didn't actually run any of the tests. I guess that's as scientific as cosmology :-)

thrownaway2424 12y ago
It's not even half science, it's just a big slide deck of FUD padded with a healthy amount of entitlement.
- chiph 12y ago
  Be honest with yourself: If you got a ticket where the indicated speed was massively different than the speed you were actually travelling, you'd fight it too. And you might do some research into the validity of the reading, as well. You'd especially do it in a state where the law says "x" mph over the limit and your license may get suspended (meaning you walk for 60 days). It'd be worth fighting for.
  - gknoy 12y ago
    Especially if it were in a residential area where you had been deliberately careful. If I got a ticket saying that I'd been driving 40 in a 25 zone, when I had been carefully watching my spedometer at ~20s, I'd feel pretty wronged. I'm not sure I'd have the resources to fight it, but I would certainly want to.
- Estragon 12y ago
  Can you cite some parts which reflect entitlement?
- smackfu 12y ago
  FUD, eh? So that's Fear, Uncertainty, and Doubt. I don't really see any fear, and I hope that any legal defense is going to push pretty heavily on uncertainty and doubt. Don't quite see the problem.
joncrocks 12y ago

PeterWhittaker 12y ago

Excellent presentation with considerable technical information, e.g., weaknesses in the LIDAR approach, operational realities that reduce accuracy, etc., and some legal background, e.g., limiting LIDAR to 1000' or less in some jurisdictions - but contrast with overall lack of case law. Could be of value to anyone wishing to fight a ticket.

privong 12y ago

Anecdotally, I heard of an astronomer in a similar situation, who attempted to prove the radar gun used to "catch" him speeding was incapable of measuring the speed accurately enough (presenting info on the radar gun's sampling properties and on sampling theory; I forget the specifics). Supposedly the judge maintained the speeding violation but reduced the man's fine for "entertaining the court".

justinsb 12y ago

Can anyone tell me why speeding tickets aren't enforced with video cameras (like red-light tickets are)? It would seem to be much more accurate, much easier to enforce, and much harder to dispute. Irrespective of the rights or wrongs of speed limit laws, I do believe that consistent and efficient enforcement can only be a good thing.

toomuchtodo 12y ago
One word: Discretion "Confronting Automated Law Enforcement" http://robots.law.miami.edu/wp-content/uploads/2012/01/Shay-EtAl-ConfrontingAutomatedLawEnf.pdf http://robots.law.miami.edu/wp-content/uploads/2012/01/Shay-...
- Gracana 12y ago
  That's a pretty short summary of a 44 page document. Can you elaborate a little more?
  - toomuchtodo 12y ago
    Sure. Do you want to live in a world where every time you violate the letter of the law you're penalized? Before you answer, consider that you're going to have to keep 100-200 years of law hot in memory to ensure your daily compliance.
    - Gracana 12y ago
      Why bother asking a question when you word it in such a biased way? Obviously my answer is yes, that is the world I want to live in.
      - toomuchtodo 12y ago
        Because that's a terrifying world to live in. I am biased. To accept the idea of compliance with the law to the letter constantly is unreasonable. Roll 1-3 feet over the white stop line? Ticket. 1-2mph over the speed limit? Ticket.

ihaveajob 12y ago

Speed radars are good for speed traps, but not so great to prevent speed related incidents. They share the same fatal flaw with all other single-point speed meters: You can just slow down for a minute, while you pass by the speed trap, and then speed up for the rest of your trip. Instead, a 2-point speed meter (i.e. license plate readers every few miles, measuring time between matching reads) system is superior because a) their margin of error is negligible and b) they measure sustained speed over a long distance rather than at a single hot spot, thus making roads safer. I can't believe it's not implanted at least in all interstate freeways.

legohead 12y ago
If you've traveled around the country, you've invariably come across those "speed is monitored by aircraft" signs, and what you described is how they do it (tracking how long it takes cars to go from point A to B)
- onan_barbarian 12y ago
  I've always been a bit nervous about those "speed is enforced by aircraft" signs (as opposed to "monitored"). I always get a mental picture of a aircraft strafing near my car until I slow down...
  - dTal 12y ago
    You may enjoy this image from XKCD then: http://what-if.xkcd.com/imgs/a/87/radar_aircraft.png http://what-if.xkcd.com/imgs/a/87/radar_aircraft.png
- cardiffspaceman 12y ago
  In California the CHP (California Hwy Patrol) plane has to do the same thing the CHP car would have to do: match your speed and then note its own speed. Measuring time between points is forbidden no matter what the observer's vehicle is. Radar is also used from the ground but was not available to CHP when they started using aircraft. The planes are on the inexpensive side so sometimes they have to dive to achieve the speed. Also recently CHP got to use radar statewide so they don't use aircraft so much.

chillingeffect 12y ago

The highlights: 1. Lidar gun manufacturers test them on limited and ideal circumstances 2. Beam divergence and shakiness make them less useful past 1000 feet, which is generally acknowledged.

awjr 12y ago

The reading was taken at a distance of 1300 feet with a handheld device that is not meant to be used handheld beyond 1000 feet. That's a significant distance. I'm assuming the cop was hiding or trying to stay hidden.

darksim905 12y ago

A lot of this is common knowledge now. You can check out guys of LIDAR who research these types of guns & how accurate their findings are. This is also why cops should be taught to do several speed readings & to point at the vehicle, not the headlights. Also, depending on the paint, it can also cloak the vehicle's speed or make it take longer to get back to the gun.

leephillips 12y ago

I hope this information, or the equivalent, filters into the awareness of judges. I was in traffic court in Virginia last year and, as I was waiting for my case, heard the judge explain to several defendants who were there to challenge their lidar tickets that the lidar was essentially infallible and could not be confused by other cars on the road. And this was an intelligent judge who seemed unusually sympathetic to the accused.

Guvante 12y ago
Unfortunately judges are people and if you have tons of people coming in with just "LIDAR is not accurate enough to prove reasonable doubt" compared to cops bringing in studies proving their accuracy it is difficult to listen to it.

DEinspanjer 12y ago

All the talk in this thread about various forms of speed management and fines remind me of an article I read a while back about a Swedish test program that used fines from speeders as the funds for a lottery that people driving at or under the limit were entered into automatically: http://www.wired.com/2010/12/swedish-speed-camera-pays-drivers-to-slow-down/ http://www.wired.com/2010/12/swedish-speed-camera-pays-drive...

voidlogic 12y ago

I've wondered sometimes if it would be possible to fight erroneous speeding tickets using the repeated GPS pings your phone or nav unit makes while using them for directions.

MattGrommes 12y ago

As a college intern at a national lab my boss' boss got a speeding ticket on the military base we worked on. He spent some time doing research and when he presented his evidence in court the judge apparently told him he wasn't going to have to pay the fine but the judge wasn't going to find him not guilty because he wasn't going to risk invalidating all speed gun findings. I don't recall if it was lidar but it's funny the stories are so similar. It's hard to use technology to mess with engineers. :)

aroch 12y ago
Lower courts (even in the Military) can't really establish any meaningful precedence
reustle 12y ago
If you were really gung-ho about getting it marked as not guilty, could you appeal it and move up the chain?

ylhert 12y ago

LIDAR sucks and there is no judicial notice on it in California. If you ever get a LIDAR ticket, do a trial by written declaration (or TBD). Just state that you dispute the validity and accuracy of the LIDAR reading and they'll usually throw it out. Remember to ask the officer(after he has given you your ticket) what model laser/radar detector he was using, and at what distance he measured you at. Anything over 800/1000 ft(this is really close, and rarely do they get you this close) will usually automatically be struck down if you question LIDAR's validity at that distance. That is a case you will win every time, if you do it right. The court cannot prove it's accuracy at these ranges and they know it. Also, rule of thumb, all CHP officers have laser only on their bikes now. source: I used to have a Lancer Evolution

gknoy 12y ago
>> Remember to ask the officer(after he has given you your ticket) what model laser/radar detector he was using, and at what distance he measured you at. That's an interesting point. Would that be admissible, though? Wouldn't you saying "He said it was at X feet" would be considered hearsay in court, versus a cop saying "I took a reading at X/2 feet"?
- ylhert 12y ago
  He/she should normally write it on the ticket(pretty sure they have to, it's always been there on tickets I've seen), but if you can get him/her to say it out loud, it's evidence that you can probably use against him/her. Also you should record your interactions with all officers. My final not a lawyer advice on beating speeding tickets: shut your mouth and don't answer the officers questions. If the officer asks how fast you were going, nothing you can say helps you, especially "I don't know". Just do your best to be polite, but firm, and don't give him anything more than a nice attitude
  - lawnchair_larry 12y ago
    This isn't always the best strategy. If you're unlikely to make a court date, being remorseful and apologetic has a decent chance of getting you a reduction from the officer. If you play hardball, so will he, then you have no choice on going to court if you want a reduction. If you're in serious trouble, STFU is the correct choice, but it's not necessarily the most economical for a moving violation.

ape4 12y ago

Don't Google's self driving cars use LIDAR.

diminoten 12y ago
They probably take all of these things into account. If they didn't, I suspect Google's self driving cars wouldn't use LIDAR.

kudu 12y ago

Why do HN mods keep changing the title from meaningful ones, which make the reader interested in the content, to generic, vague ones? You'll see that there's definitely a correlation between the moment that HN mods change the title and the post dropping off the top frontpage spot.

Karunamon 12y ago
Agreed - the original title, about a guy beating a LIDAR ticket, was much more descriptive than this headline, which effectively says "This article is about LIDAR".
- shangxiao 12y ago
  It was the fact that the guy was Joshua Bloch which got me to read the presentation - and I'm totally glad I did! I would never have read it if it was just named "The Lowdown on Lidar"
  - puredemo 12y ago
    So appeals to authority work really well on you..
    - gwern 12y ago
      BTW, your downvotes are an opportunity to reflect on why your comment was obviously wrong and counterproductive.
- lnanek2 12y ago
  Yeah, I had to read a lot of slides wondering why it was repeating the same thing over and over before I realized he was trying to beat a ticket. It definitely isn't a general purpose tech primer like the new title makes it sound, it's just focused on covering an error situation that gives you a legal out.
  - DanBC 12y ago

jstalin 12y ago

In Michigan, a speeding ticket is a civil infraction, not a criminal offense, so the legal standard is preponderance of the evidence, not beyond a reasonable doubt, as the slide deck says. I suspect it's the same in California.

dtrizzle 12y ago
In CA, infractions are crimes so the reasonable doubt standard applies. See Cal. Pen. Code section 16. http://law.onecle.com/california/penal/16.html http://law.onecle.com/california/penal/16.html
- jervisfm 12y ago
  Wait a minute, if that's true won't it imply that traffic violations/infractions convictions get you a criminal record ?
  - dtrizzle 12y ago
    Yes. Infractions are crimes so yes. For the most part, they are not treated as crimes. You cannot be jailed based on an infraction. Most job applications I've seen typically explicitly exclude requiring information about past infraction violations.

mrfusion 12y ago

When I was a teenager I tried to fight a speeding ticket, and prepared a presentation like this with several good points (I think). Come court day, I got all dressed up, stood up when called and started my arguments ... and the judge simply cut me off after two sentences and said "guilty". I'm surprised this guy was allowed to present his whole argument and it actually convinced a judge.

jc4p 12y ago
From the beginning, it sounds like he's presenting this to a group of people invited by his lawyer, and the lawyer took care of the actual ticket resolution.
timdierks 12y ago
This presentation was not to a court; see slide 51 for the resolution: his attorney bargained it down to a zero-points violation before trial.
scotty79 12y ago
I'm sure it was valuable lesson for you that law is not concerned with the truth as much as it is with order.
driverdan 12y ago
Did you appeal? You can always appeal your first court hearing. The key is to talk to the prosecutor beforehand. They will almost always plea it down.
- mrfusion 12y ago
  This was in Maryland. It seemed you went right to the courtroom and talked to a judge along with 40 other people. There was no prosecutor present as far as I could tell. I've fought them in Connecticut later and there you only meet with a prosecutor and they warn you all over the place that if you don't take the deal and try to go to court the judge can significantly increase the fine :-(

supernova87a 12y ago

So regardless of the science behind it, what is the successful legal forum / mechanism to raise doubt about the accuracy of the technology being used by police?

darkmighty 12y ago

I think he missed a couple important points. The sweep effect, shaking and aiming are more amenable to signal processing than it seems, depending on the resolution and noise in the system. I tried to illustrate in pictures: 1) Just to demonstrate that a nonuniform sweep is amenable to detection or processing, consider this picture, with the correct estimate in green: http://imgur.com/FsPAwCn http://imgur.com/FsPAwCn 2) Now imagine the device is more sophisticated. Consider it has a very large resolution around the incoming pulse: http://imgur.com/z8UrhUw http://imgur.com/z8UrhUw In this case the beam width is actually beneficial, and so are large distances. The sweep effect would cause a shift in the peak intensity shown in the graph, but the object depth frustrum should be clear across samples (and dislocating at t'=t-v*dt/c). 3) Officers can be trained to recognize situations unforable, regardless of distance. One fast car passing a slow one would make it easy to confuse the equipment. A clear desert however doesn't present many threats for error.

Trombone5 12y ago
I think its funny how he claims location sampling at different times is not a measurement of speed. It's strange how he mentions the sampling speed, but then pretends it's 1/3 second. Basically what he is attacking is the worst design of a speed gun that he could imagine from the public information. A more fair challenge would be to find sufficient error from a competent design of a speed gun, but then it wouldn't be as self serving.

dzhiurgis 12y ago

Perhaps anyone can explain me why Lidar is not dangerous to sight? Whenever you see anyone working with lasers they'd wear protective glasses. Now I realise Lidar uses infrared wavelength which is not visible to us, but that does not mean it's not dangerous, or does it?

bicubic 12y ago
Non-visible lasers pose a much bigger hazard than visible. Intense visible light triggers an involuntary defence called the blink reflex whereby we blink before the retina absorbs enough energy to get damaged. Lasers outside the visible range don't trigger the blink reflex and can destroy the retina before you know what happened. Lidar operates at relatively low power and relies on very sensitive detection in a very narrow wavelength range to identify reflections.

xchaotic 12y ago

Exec summary: a geek/nerd gets a speeding ticket driving a porsche. 5 years later he's still doing powerpoints on it rather than getting a life

trhway 12y ago

now if only somebody published a similarly powerful defense on "traffic control device violation" :)

Shivetya 12y ago

How does the officer confirm which vehicle the LIDAR selected? I have never been clear on this point.

njharman 12y ago
I believe they have "sights" and are aimed like a gun.
- icelancer 12y ago
  Cops routinely shoot innocent bystanders.
john_b 12y ago
Having fought a LIDAR ticket before, the officer simply claims that he or she targeted your car and the judge admits the claim as a fact without question. Any attempts on your part to question this will be met with a reminder that the officer is Certified and Trained and you are just a random person with no qualifications to question the officer's abilities. You can ask the officer if there was a wind that day, and if so how heavy. You can ask if the officer used a tripod or held the gun freehanded. Even if the officer admits that there was a heavy wind, the gun was used freehanded, and he or she pinged you in traffic at 950 ft, the judge will still accept the officer's word that your car, and only your car, was targeted. Why? Because it's easier for the court, because they know you can't prove without a doubt that the officer targeted a different car, and because it is an important source of income for the city or county which has summoned you. It's a scam, but it's one which Americans in general are comfortable with because it gives them the illusion of safety.

Marcus10110 12y ago

Next time I get a parking ticket, I'll be sure to investigate the temperature stability of the parking meter's clock.

Relaxx 12y ago

I'm surprised the angle of measurement never came up. Isn't it the case that a reading has to be head-on for it to be accurate? It seems pretty obvious that at higher angles the observed distance delta will be smaller.

disillusioned 12y ago

I have lidar jammers on my car, but Arizona doesn't use a lot of lidar. (Motors and sometimes bored DPS will, but typically it's Ka band radar via Stalker Duals.) Not really knowing how lidar worked, except that a "laser" pulse hits the car and reads back a speed, this article was great because it explains that the measurement process is far slower than you'd think (0.3 seconds!) which is easily enough time for my jammer to detect the inbound infrared and start spamming light on that frequency, which is what they do. I don't think I've ever successfully jammed one, since again, AZ is mostly Ka, but it's pretty neat to know that they can work as advertised (since they're built in to the front of the car most likely to be targeted by lidar).

tempestn 12y ago
I've read that in some jurisdictions police have started confiscating jammers and even charging users with obstruction of justice, on the theory that while owning them is explicitly legal, using them prevents police from carrying out their duties. Do you know what the history in AZ is like?

hellbreakslose 12y ago

Great thread loved it... But thats why we GREEKS invented another patent that works. You won't get a fine if you are 15-20mph over the limit, EVER!

Jimny22 12y ago

Well the author may be a clever computer scientist but he knows nothing about physics and LIDAR equipment. Pathetic I'm afraid.

BHSPitMonkey 12y ago
Given that (1) you're an expert in these areas and (2) you found no points in the author's work worth refuting, I guess it follows that the author is actually an expert as well.