Slashdot Mirror
GPS Used As Defence In Radar Speeding Case
James Thigpen writes "There is an article over at Ars Technica about an accused speeder contesting his speeding ticket based on his car's built-in GPS system's records. According to the article his car says he was going slower than the radar gun clocked him at. Contesting a ticket based on GPS data has never before been tested in court."
Plus it would be cool to have onboard footage of your driving for analysis and review.
But will he be able to produce the source code for the GPS when the police request it to check its accuracy?
Breathalyzer Source Code Revealed
Closed Source -> Charges Dismissed?
Comment removed based on user account deletion
GPS data was actually used recently in taiwan to prove a man's innocence. A truck driver A went into an accident with a motorcyclist B. A stayed and helped B up, and even paid cash. B said he's fine, so A drove off, only later to recevie a notice that B has filed a hit and run case against him. The court found A not guilty since the gps data showed that A stayed at the site for more than 15 mins.
The pretty large difference between his 'radar' speed, and his 'gps'(actual) speed was pretty large. IMHO this sets brings into question just about every speeding ticket ever given by radar gun.....
lets say that the gun is wrong 1% of the time, which in the case of a cop handing out tickets by hand is okay (imho) because there is human intervention, he (or she) can look at the thing, bang it on his hand a little, and shake the error off as a fluke.
The speed cameras on the 101 in scottsdale, arizona issue about 250 tickets daily. Thats 2.5 tickets daily that the gun gets wrong (the 1% figure was pulled from my ass, but I'm using it as an example). With THIS there is no human intervention at all (other than a pissed off commuter)..
grr...not sure where i'm going with this, I just REALLY hate it that humans are being taken out of (at least that little part) of the legal system. I don't want my fate decided by a computer!
NewslilySocial News. No lolcats allowed.
I have heard stories of people trying to submit GPS data logs in the past to prove they were not speeding. The judges would not accept the data because it could be considered suspect, particularly if you presented it on a USB dongle since the data is beyond easy to modify. Radar does have its issues, specifically if you are in a group of cars (have you ever been blamed for speeding when the car beside you was passing you). Unless you can provide data in a method that is deemed "un-crackable", I doubt it would be allowed. I could easily re-run the route that I was on when I got stopped, take the track log and modify the time stamps (if they are present and that depends on the GPS data stream you selected). You would want the time stamps to be there to compare against the time the police officer stated on the ticket. I have to take this jab at the judicial system though, despite the fact the the citations say you are not pleading guilt or innocence at the time of the infraction, you are pretty much labelled guilty, the police never lie in the courts point of view and if you claim you are innocent, you get treated as if you were guilty anyway. The only way I can see to defend yourself is to have the same set up in your car as the police do and have it display speed on the recording. Then again we saw recently what happened someone who tried that in Missouri.
"Don't be afraid of knowledge" Schwarzwald The Big-O Guy
From my understanding, and the contention of the officer, the GPS logs average speed. Which means that during a short period of time, the defendant could have greatly exceeded the speed limit (and was clocked by the officer at that time), while the average speed was far lower than that. In which case, both the cop and the defendant are correct, and the cop is till valid in giving the ticket...
On my system the GPS application stores its logs in a textfile which I can easily edit. It would be trivial for me to doctor the text file to contest any speeding ticket. I'm not sure that this is a good form of evidence.
I used the GPS defense when pulled over.
.75 mph.
In San Antonio, TX I was pulled over for doing 76 in a 75 zone. I successfully argued that the GPS was more accurate than the RADAR, when I said that it used "government satellite signals."
In fact, most police radar units are +/- 3mph. A consumer GPS speed indicator is typically accurate to within
When working in ship navigation systems (Laser Plot), I was involved in dumping track information from a ship to show that it was not in an area when a boating accident occurred.
The hacking issue is correct, one can always hack the data. The Cop can lie about the reading on the radar unit too. If it gets to 'real court' you have the standard issues of scientific reliability (Daubert test) and the authenticity of the data. In the late 90s, there was a case (in Georgia, I think) where a speeding conviction was thrown out because there was no reliability of the laser speed testing introduced.
Fight Spammers!
Eaton VORAD units, which are a phased-array anti-collision radar for trucks, have been used to provide evidence in favor of the truck driver. The VORAD units track individual car-sized targets, and provide range, range rate, and azimuth. Range and range rate are quite good; azimuth isn't that accurate. The control unit keeps track of recent events ten minutes before a collision, and also has speed info available. The latest versions can interface with GPS and other vehicle systems. This allows detailed accident reconstruction.
It's most useful where an accident resulted when someone drove in front of a truck. The VORAD record shows not just what the VORAD-equipped vehicle was doing, but what the other vehicles were doing.
In most jurisdictions, such traffic cases are considered civil and the standards for evidence are different than those of criminal cases or what you may see on 'Law & Order'. The judge is free to weight the officers evidence more highly than yours and presume it to be correct unless you can show overwhelmingly that it is not. Sort of like being guilty until proven innocent.
Furthermore, courts have quite a bit of latitude to allow or deny the admissibility of data as evidence. For example: Radar is quite accurate (it reads the speed of an object quite close to its actual speed) but not very selective (it might be reading the speed of something else, or interpret some RF noise as speed). Take the boilerplate testimony that an officer reads about 'calibrating the gun with a tuning fork' and all the b.s. about standards traceability. None of this is necessary, as the most common source of errors are due to poor selectivity. But it sure sounds great in court.
In fact, calibrating a radar gun with a tuning fork is a good demonstration of its susceptibility to AM noise. An ideal radar gun should only measure frequency shift due to the Doppler effect and reject the sort of modulation that a tuning fork creates. After all, the instantaneous velocity of its tines is dependant on its amplitude and the average velocity is zero (unless you throw it). But no court would hear such an argument, as it would undermine their entire traffic enforcement/revenue collection program. And, as a civil case, they are not required to consider it.
Have gnu, will travel.
Yup. On most people's home projects list that's the one right after "Get microwave to stop blinking 12:00."
The contest for ages has been to rescue liberty from the grasp of executive power. -- Daniel Webster
You see, Instantaneous velocity is the limit of the average velocity where the time of averaging tends to zero.
In other words, the value of f'(t0), where the position x is x = f(t) at a given time t0.
Or in other words, angle of the tangent of the curve x = f(t) in the given time t0.
Now, if your argument is that "a GPS device cannot give the measure of the instantaneous velocity because it does not sample fast enough to get a really good approximation of the curve x = f(t) and hence, the value of f'(t0)", then you could be right because 1Hz is not really a high sampling rate. But you could have said so
The (analog) speedometer in most cars measure speed by measuring the RPMS of the gear box and multiplying by gear ratios and tire size: they normally do that with a continuous measuring (springs and coils), and what they measure is a good approximation of the instantaneous velocity of the vehicle. A good analog speedometer is somewhat reliable, especially if the scale is correct(*)
(*) their scale is not linear like you see in a normal car: but exponential, so it should be like: and this is why they have a "sweet calibration spot" (normally near the top of the dial; have you already thought about why they make 1.2l-engine cars with 220 km/h marking in the speedometer [a speed they usually don't achieve even in freefall
DISCLAIMER: I was a software developer for a road engineering company for one and a half year.
It's better to be the foot on the boot than the face on the pavement. ~~ tkx Kadin2048
It's probably far too late to get a mod up on this, but I thought I'd add it to the knowledge base of the Internet in case somebody decides to Google it one day.
Simply put: This is not going to work.
The system is rigged against fighting speeding tickets. Even if you've got the money to pay for evidence collection, expert witnesses, and everything else -- BEFORE your trial -- you'll still lose. The justice system will protect the police from having even one ticket investigated, because it calls into question other tickets the officer may have written using the same or similar equipment; a very large expense. It just won't happen.
Here's a TRUE story, as related to me by my friend who drives commercial truck:
My friend was pulled over by a police officer for "speeding" and given a ticket for 75 in a 65. Only one small problem here: The area in Ohio where he received the ticket was absolutely flat, and his truck is GOVERNED at 68. Exceeding 68 miles per hour on a flat road is literally IMPOSSIBLE for his truck, so says the manufacturer of the engine and the manufacturer of the vehicle. Understandably, my friend was very upset at receiving such an obviously bogus ticket, and decided to fight it.
Nice thing for my friend, engines in big trucks have computers to track fuel usage, speed, etc. over time. Getting the data from the engine is a matter of taking it to the service center, hooking up a computer, and getting a printout. He obtained this printout and showed it to me; it's so simple my grandmother could easily see his truck hadn't gone over 68 at any point during the data recording. The dates were clearly marked; it showed on the day in question, the truck did not go anywhere near 75.
Armed with this and people willing to testify that the truck's governor was functional and the printout was accurate, he attempted to fight the ticket. He was informed that he would have to pay all of the trial costs up front ($10,000) before the trial began, and even if he won, he wouldn't be able to get reimbursed for this expense. So basically, it came down to a choice: Swallow pride and pay the $350 ticket, or pay $10,000 to prove he was in the right and get the ticket voided on the basis of the evidence.
Sadly, but wisely, my friend opted for the former. Proving his case was not worth the extra $9650 it would cost to do so.
Take note: Traffic court is rigged against regular people. If the highwaymen in blue try to rob you, just give up the money; losing your time, energy, and sanity over government sponsored theft will just victimize you more.
I work for the radar company that made the radar gun the cop used. I don't have all the information about what happened, but I have a hard time believing the GPS is more accurate.
Radar guns are certified regularly, which is most often a pretty simple accuracy test (but very well could have been a full diagnostic), so it's doubtful the radar gun was malfunctioning (iirc, those guns have an internal lockout in case of malfunction).
Also, remember that we're talking basically the speed of light here, with some minor latency for the unit to process the Doppler shift. Radar's pretty much instaneous, within miliseconds, at least.
Now, that's not to say that the officer didn't make an error. Radar's not an exact tool--b/c the beam is so wide, you can pick up a lot of things and an untrained officer can get some misleading speeds.
At the same time, remember that most traffic officers do this all day, at least five days a week. They make mistakes just like anybody else, but they're rare. And for that matter, officers are trained to use the radar as a confirmation of their own judgement of how fast the vehicle's moving. And since they're doing it all day long every day, they can tell you within a mph or two how fast a car is going just by looking at it.
Again, I'm not pretending to have all the information, but if it came down to trusting GPS or trusting the radar, I'd trust the radar. It's just a simpler tool, with less hoops to jump through (and fewer things to go wrong).
Disclaimer: I'm in marketing for Decatur Electronics. But for what it's worth, I use Linux on my machine at home, hehehe.