The Lowdown on Lidar

I confess I was expecting to not care for this. I&#x27;m not sure how any of it should be surprising, as it seemed fairly straight forward analysis of the tools.<p>That said, this was a ridiculously fun read. More detailed than expected for some parts, and always informative. I&#x27;m glad it did not come across as anti speed detectors. Really more of a &quot;know the limitations.&quot;

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&#x27;t believe it&#x27;s not implanted at least in all interstate freeways.

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&#x27;ll usually throw it out. Remember to ask the officer(after he has given you your ticket) what model laser&#x2F;radar detector he was using, and at what distance he measured you at. Anything over 800&#x2F;1000 ft(this is really close, and rarely do they get you this close) will usually automatically be struck down if you question LIDAR&#x27;s validity at that distance. That is a case you will win every time, if you do it right. The court cannot prove it&#x27;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

As a college intern at a national lab my boss&#x27; 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&#x27;t going to have to pay the fine but the judge wasn&#x27;t going to find him not guilty because he wasn&#x27;t going to risk invalidating all speed gun findings. I don&#x27;t recall if it was lidar but it&#x27;s funny the stories are so similar. It&#x27;s hard to use technology to mess with engineers. :)

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:<p>1) Just to demonstrate that a nonuniform sweep is amenable to detection or processing, consider this picture, with the correct estimate in green: <a href="http://imgur.com/FsPAwCn" rel="nofollow">http:&#x2F;&#x2F;imgur.com&#x2F;FsPAwCn</a><p>2) Now imagine the device is more sophisticated. Consider it has a very large resolution around the incoming pulse: <a href="http://imgur.com/z8UrhUw" rel="nofollow">http:&#x2F;&#x2F;imgur.com&#x2F;z8UrhUw</a> 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&#x27;=t-v*dt&#x2F;c).<p>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&#x27;t present many threats for error.

When I was a teenager I tried to fight a speeding ticket, and prepared a presentation like this with several good points (I think).<p>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 &quot;guilty&quot;.<p>I&#x27;m surprised this guy was allowed to present his whole argument and it actually convinced a judge.

Can anyone tell me why speeding tickets aren&#x27;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.<p>Irrespective of the rights or wrongs of speed limit laws, I do believe that consistent and efficient enforcement can only be a good thing.

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.

Anecdotally, I heard of an astronomer in a similar situation, who attempted to prove the radar gun used to &quot;catch&quot; him speeding was incapable of measuring the speed accurately enough (presenting info on the radar gun&#x27;s sampling properties and on sampling theory; I forget the specifics). Supposedly the judge maintained the speeding violation but reduced the man&#x27;s fine for &quot;entertaining the court&quot;.

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

In Michigan, a speeding ticket is a civil infraction, not a criminal offense, so the legal standard is <i>preponderance of the evidence</i>, not <i>beyond a reasonable doubt</i>, as the slide deck says. I suspect it&#x27;s the same in California.

I have lidar jammers on my car, but Arizona doesn&#x27;t use a lot of lidar. (Motors and sometimes bored DPS will, but typically it&#x27;s Ka band radar via Stalker Duals.) Not really knowing how lidar worked, except that a &quot;laser&quot; 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&#x27;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.<p>I don&#x27;t think I&#x27;ve ever successfully jammed one, since again, AZ is mostly Ka, but it&#x27;s pretty neat to know that they can work as advertised (since they&#x27;re built in to the front of the car most likely to be targeted by lidar).

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: <a href="http://www.wired.com/2010/12/swedish-speed-camera-pays-drivers-to-slow-down/" rel="nofollow">http:&#x2F;&#x2F;www.wired.com&#x2F;2010&#x2F;12&#x2F;swedish-speed-camera-pays-drive...</a>

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

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&#x27; or less in some jurisdictions - but contrast with overall lack of case law.<p>Could be of value to anyone wishing to fight a ticket.

Perhaps anyone can explain me why Lidar is not dangerous to sight?<p>Whenever you see anyone working with lasers they&#x27;d wear protective glasses. Now I realise Lidar uses infrared wavelength which is not visible to us, but that does not mean it&#x27;s not dangerous, or does it?

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&#x27;s a significant distance. I&#x27;m assuming the cop was hiding or trying to stay hidden.

A lot of this is common knowledge now. You can check out guys of LIDAR who research these types of guns &amp; how accurate their findings are. This is also why cops should be taught to do several speed readings &amp; to point at the vehicle, not the headlights. Also, depending on the paint, it can also cloak the vehicle&#x27;s speed or make it take longer to get back to the gun.

Next time I get a parking ticket, I&#x27;ll be sure to investigate the temperature stability of the parking meter&#x27;s clock.

I&#x27;m surprised the angle of measurement never came up. Isn&#x27;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.

I&#x27;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.

Well, he fought with half science: he came up with a set of testable hypotheses, but didn&#x27;t actually run any of the tests. I guess that&#x27;s as scientific as cosmology :-)

So regardless of the science behind it, what is the successful legal forum &#x2F; mechanism to raise doubt about the accuracy of the technology being used by police?

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

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

Well the author may be a clever computer scientist but he knows nothing about physics and LIDAR equipment. Pathetic I&#x27;m afraid.

Why do HN mods keep changing the title from meaningful ones, which make the reader interested in the content, to generic, vague ones? You&#x27;ll see that there&#x27;s definitely a correlation between the moment that HN mods change the title and the post dropping off the top frontpage spot.

Don&#x27;t Google&#x27;s self driving cars use LIDAR.

Exec summary: a geek&#x2F;nerd gets a speeding ticket driving a porsche. 5 years later he&#x27;s still doing powerpoints on it rather than getting a life

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