Lidar has a wide range of applications; one use is in traffic enforcement and in particular speed limit enforcement, gradually replacing radar after 2000. Current devices are designed to automate the entire process of speed detection, vehicle identification, driver identification and evidentiary documentation.
Jeremy Dunn (Laser Technology Inc.) developed a police lidar device in 1989, and in 2004 10% of U.S. sales of traffic enforcement devices were lidar rising to 30% in 2006, given the advantages of lidar it appears likely that the majority of current sales are lidar, although sophisticated radar units are still being sold.
Current units combine five operations; speed detection; operator viewing, even under adverse conditions; imaging synchronised with speed detection; acquisition of court ready evidence; downloading of evidence to an external device. They can operate in automatic mode either attended or unattended.
Radar has wide signal beam divergence, so that an individual vehicle cannot be targeted, requiring significant operator skill, training and certification in order to visually estimate speed so as to locate an offender in a traffic stream, and offenders may use the defence that some other vehicle was offending. Radar will register the speed of any object in its field, for example a tree swaying or an airplane passing overhead.
Lidar has a narrow beam, and easily targets an individual vehicle, thereby eliminating the need for visual estimation, and some models can record an image of the license plate at the same instant as recording the speed violation. Speed estimation takes less than half a second, which, together with the narrow, targeted beam, results in offending vehicles having little warning even when using an evasion device. Lidar can measure the distance between vehicles to detect 'too close' (tailgating) infringements. The speed of a vehicle in the shadow of another vehicle cannot be measured.
The US Department of Transportation National Highway Traffic Safety Administration (NHTSA) has issued specifications for lidar devices, a conforming products list, and guidelines regarding implementation of traffic enforcement.
A typical NHTSA approved device weighs less than two kilogram, is battery powered, has speed detection accuracy +2 km/h and -3 km/h, distance accuracy +- 0.3 metres at 90 metres, and minimum range 300 metres. Devices must be capable of meeting these accuracy standards while exposed to ambient temperatures between -30 °C and 60 °C, relative humidity of 90% at 37 °C and normal urban road ambient electromagnetic radiation. The range of speeds required to be accurately detected is 16 km/h to 320 km/h. Speeding violations are required to be documented by the device with a recorded image showing license plate, location, speed, date, time and operator identification, some units identify the driver by image and record the direction of travel. The light emitted is required to be in the infrared range, meet eye safety standards, and have pulse repetition less than one kHz with beam divergence less than 5 milliradians.
Vehicle registration plates are an important part of traffic enforcement and in most jurisdictions the government holds a monopoly on the their manufacture, although this may be contracted out. Normally it is illegal for private citizens to modify, make and affix their own plates, as this is equivalent to forging an official document. California plates are required to be 6 in tall and 12 in wide, a usual standard, and have a reflective surface that is particularly sensitive to infrared light, which enables it to be imaged at night, enables Automatic License Plate Recognition, allows LIDAR devices to receive a strong reflective signal return, and have tamper-resistant markings.
Some jurisdictions do not require a front license plate on automobiles and many do not require them on certain vehicles such as motorcycles. Police generally prefer to detect from the front while observing oncoming traffic, which also enables the offender to be waved over and avoid the risks of high speeds required to catch up to the vehicle.
A number of jurisdictions prohibit any methods to thwart speed limit enforcement, and lidar manufacturers endeavour to stay ahead of detection avoidance measures.
Current lidar devices have a horizontal beam width of one meter at 300 meters, compared to the registration plate width of 30 cm, ensuring that little of the signal is scattered to following vehicles. Detecting the LIDAR signal in advance is difficult as the tight beam, short signal duration and targeting of individual vehicles minimizes scatter of the LIDAR signal to following or adjacent vehicles. Modifying the vehicle to deflect, absorb or jumble the signal is difficult, as it is typically the registration plate that is targeted. Modifying the registration plate is easily detected and may not be legal. Returning a false separate signal will be detected by current police lidar models and may not be legal, depending on the jurisdiction.
A typical NHTSA approved lidar device emits 30 ns pulses of laser light with wavelength 905 nm and 50 milliwatts of power with 3 milliradian beam divergence. The power is sufficiently low to ensure no ocular damage occurs. At 905 nm wavelengths, IEC 60825-1 Edition 2.0 allows a maximum energy per pulse of 0.5uJ.
Light travels approximately 30 cm per ns so each pulse has a length of about nine metres. At a target distance of 300 metres the light pulses take 2,000 ns to complete the round trip. The time interval between pulses is no less than one million ns, providing time to make a distance estimation from each pulse. Up to several hundred pulse readings are taken over a period less than half a second and used to estimate the change in distance over time, thereby estimating vehicle speed. Returning light is filtered to exclude light not in the wavelength range 899 nm to 909 nm. An internal proprietary algorithm rejects inaccurate readings; detection avoidance methods usually attempt to overload the filter and persuade the error rejection algorithm to incorrectly reject a reading.
An expert operator will use the viewing capability to select a likely offender prior to speed detection, this has the advantage that minimum signal is scattered to warn following vehicles equipped with lidar signal detection devices, this is not so important on sparsely trafficked roads and a lower capability lidar device may be used. Once a likely offender is detected and the registration plate targeted the operator triggers speed detection which includes acquisition of evidence, an audible tone indicates a good return signal. To produce an accurate reading the operator must focus the pulse on a single point for the duration of the read. At long range this is accomplished through the use of a stationary tripod to steady the aim. For speed detection any part of the vehicle may be targeted, although the registration plate is highly preferred.
Normal weather conditions have negligible impact on device performance but may impede operator ability to target a vehicle. This includes occasions when the sun is directly behind the target vehicle, nighttime, or when the device is used within a stationary vehicle with a soiled windshield, in which case the signal might be scattered. Heavy weather may reduce the range of the device and in particular heavy fog will render it unusable.
When used within a moving vehicle, the device measures the relative speed of the police and target vehicle. Police are required to follow the offending vehicle for 200 metres and have a certified speedometer, largely negating advantages of the device.
Like radar, lidar is subject to cosine error effect.
Sweeping the device while taking a reading so that (particularly at long range where angular separation between targets is slight) returning pulses from more than one target creates a false reading. Sweeping along the side of a vehicle may also cause false readings.
A false reading is produced when the pulse reflects off for example a wing mirror, hits a stationary reflective object and then returns to reflect off the mirror a second time.
These errors are largely eliminated when current devices are expertly used.
In 2008, the D.C. Superior Court upheld the admissibility of lidar evidence in its jurisdiction. In addition to expert testimony, the court noted that it factored scientific publications into its decision:
The Court conducted an extensive four-day Frye [Daubert] hearing... [in which it] considered such issues as the basic science of laser technology, the technical methodology of, and theoretical challenges to, the reliability of radar guns... including the possibility of other "pulses" in the vicinity of use, difficulties in target identification, possible errors caused by vehicle license plates, windshield glass, shape, and color, and potential malfunction of the device. The Court also took judicial notice of at least six scientific publications on the subject in various journals of interest, together with two police-related studies in Florida, one New Jersey [study], and one independent study in Florida on this and similar radar devices, all of which met the standards set forth by [the] National Highway Safety Administration...
The court also noted that not a single court had conducted full-blown hearings on the issue that found LiDAR unreliable, while more than a dozen jurisdictions had decided that lidar is reliable.