USRE40767E1 - Unmanned integrated optical remote emissions sensor (RES) for motor vehicles - Google Patents

Abstract

An unmanned integrated RES12integrates all of its components except the reflector22into a single console30that is positioned at the side of a road and has a CPU36that controls calibration, verification and data gathering. The RES's source32and receiver34are preferably stacked one on top of the other such that the IR beam24traverses a low and high path as it crosses the road14.This allows the RES to detect both low and high ground clearance vehicles. To maintain the vehicle processing and identification throughput, the speed sensor54and ALPR48,50detect the passing vehicles at steep angles, approximately 20 to 35 degrees. In a preferred system, a manned control center16communicates with a large number of the unmanned integrated RES to download emissions data, perform remote diagnostics, and, if necessary, dispatch a technician to perform maintenance on a particular RES.

Description

CROSS-REFERENCE TORELATED APPLICATION APPLICATIONS

This application is related to application Ser. No. 08/318,566, entitled “Optical Sensing Apparatus for Remotely Measuring Exhaust Gas Composition of Moving Motor Vehicles” filed Oct. 5, 1994 and assigned to Santa Barbara Research Corporation, the assignee of the present invention.This application is a continuation of U.S. patent application Ser. No.09/521,858, filed Mar.9,2000 (now abandoned), which is a Reissue of U.S. patent application Ser. No.08/739,487, filed Oct.26,1996(which issued as U.S. Pat. No.5,726,450 on Mar.10,1998). Additionally, this application is related to U.S. patent application Ser. No.08/318,566, filed Oct.5,1994, entitled “Optical Sensing Apparatus for Remotely Measuring Exhaust Gas Composition of Moving Motor Vehicles” (now U.S. Pat. No.5,591,975, ussued Jan.7,1997).

BACKGROUND OF THE INVENTION

1. Field of the Invention

This invention relates to the monitoring of environmental pollution, and more specifically to an unmanned integrated RES for remotely monitoring the exhaust gas composition of moving motor vehicles.

2. Description of the Related Art

Environmental pollution is a serious problem which is especially acute in urban areas. A major cause of this pollution is exhaust emissions from automotive vehicles. Official standards have been set for regulating the allowable amounts of pollutants species in automobile exhausts, and in some areas, periodic inspections or “smog checks” are required to ensure that vehicles meet these standards.

Anti-pollution devices which are required equipment on newer vehicles accomplish their intended purpose of reducing pollution in the vehicle exhaust to within prescribed levels. However, some older vehicles and special types of vehicles are exempt from inspections. Furthermore, some vehicle owners with mechanical expertise can perform whatever servicing is necessary to place their vehicles in condition to pass required inspections, and subsequently remove anti-pollution devices and/or return the vehicles with an attendant increase in pollutant emissions for normal use. The relatively small number of noncomplying vehicles generate a disproportionately large amount of pollution.

As a result, an anti-pollution program which depends entirely on mandatory periodic inspections performed at fixed facilities is inadequate. It is necessary to identify vehicles which are actually operating in violation of prescribed emission standards, and either require them to be placed in conformance with the standards or be removed from operation.

Manned RESs are now used to augment the periodic inspection program to identify vehicles that are in violation of the emission standards. In general, RES are a nonobtrusive and cost-effective means for identifying the high pollution emitting vehicles and notifying the owner to take corrective action in a timely manner. The Smog Dog™ RES produced by Santa Barbara Research Center, the assignee of the present invention, includes a source and a receiver that are mounted on respective tripods and positioned on opposite sides of a road, a video camera and speed sensor that are mounted on a tripod that is positioned about 50 feet up the road in the direction of oncoming traffic, a van that contains a computer, data storage, power sources, calibration gas, and a video monitor, and a technician.

The source projects an IR beam across the road to the receiver which continuously senses pollutant levels such as carbon monoxide (CO), carbon dioxide (CO₂), hydrocarbons (HC), water (H₂O), nitric oxide (NO_x) in the received IR beam. When a vehicle passes through the IR beam, a sensor triggers the receiver to write the pollutant levels for the vehicle's exhaust plume to a data file in the data storage. The beam is set at a height to detect either low profile vehicles (cars) or high profile vehicles (trucks), but not both. The video camera takes a picture of the passing vehicle and the computer executes a character recognition program to identify the plate, which is then appended to the data file. If the speed sensor determines that the vehicle's acceleration and/or speed exceed certain levels, indicating that the vehicle's emissions control equipment are disabled, the recorded data is invalidated.

One drawback of the SMOG Dog™ and the other known RES systems is that the components, i.e. the sensor, receiver, video camera/speed sensor, and the van, are discrete parts that are positioned over a relatively large area. The source and receiver are positioned on opposite sides of the road. For safety purposes, they must be set back from the edges of the road. The video camera/speed sensor are positioned up the road such that their detection angles with respect to the passing vehicles is sufficiently shallow, approximately 3 degrees, to provide an accurate acceleration estimate and a high confidence of plate recognition. This can cause mismatch errors between the emissions readings and the plate recognition. Also, there must be enough room to park the van. These spatial requirements limit the applicability of the known RES systems. Furthermore, the discrete components are expensive because they require their own tripod, power supply, and alignment mechanisms.

Another drawback is that the known RES must be continuously manned by a technician, which is very expensive. After initial set up and alignment, the technician monitors the equipment to protect it from vandalism, performs required maintenance, and puts the system away at the end of the day. For example, the components may fall out of alignment due to the vibrations caused by passing vehicles, the various lenses may become occluded or the calibration gas may run out. Furthermore, the technician controls the data gathering process. The technician periodically places the RES in calibration mode, puffs a calibration gas into the IR beam to calibrate the system and evaluates the results displayed on the video monitor to accept or reject the calibration. Thereafter, the technician places the RES in data gathering mode, puffs the calibration gas, and compares the computed pollutant levels to the known levels of the calibration gas to accept or reject the verification of the calibration. During data gathering, the technician monitors both the signal levels of the exhaust plumes and the ambient air to determine whether the system has gone out of calibration or has a mechanical error. The technician also verifies the results of the plate recognition system.

U.S. Pat. No. 5,418,366 “IR-Based Nitric Oxide Sensor Having Water Vapor Compensation” issued May 23, 1995 discloses a specific receiver configuration having three channels for measuring a NO transmission, a water transmission, and a reference transmission, respectively, that are combined to give the effective NO transmission value. U.S. Pat. No. 5,210,702, “Apparatus for Remote Analysis of Vehicle Emissions” issued May 11, 1993 discloses a specific receiver configuration in which the ultraviolet beam is separated from the IR beam to sense the NO levels, and the IR beam is split into a plurality of components to measure CO, CO₂, HC and H₂O. Both systems use discrete source and receiver components placed on opposite sides of a road, a camera mounted on a tripod up the road, and a van for housing the control electronics, and require a technician to set the system up, calibrate the system, control the data gathering process, and pack it up at the end of the day.

In 1992 Remote Sensing Technologies (RST) experimented with a double-pass RES system called the RSD1000 in which a van housing both the source and the receiver and the video camera was suspended from a 20 foot boom. The IR beam was reflected off a mirror on the opposite side of the road back to the receiver. RST's system did not include the plate recognition or speed sensing capabilities, and never worked well enough for commercial exploitation. As a result, RST developed a one-pass system with the source and receiver on opposite sides of the road.

SUMMARY OF THE INVENTION

In view of the above problems, the present invention provides an unmanned integrated RES that reduces cost and simplifies operation.

This is accomplished by integrating each of the RES's components except the reflector into a single console that is positioned at the side of a road and providing a CPU that controls calibration, verification and data gathering. The source and receiver are preferably stacked one on top of the other such that the IR beam traverses a low and high path as it crosses the road. This allows the RES to detect both low and high ground clearance vehicles. To maintain the vehicle processing and identification throughput, the speed sensor and ALPR detect the passing vehicles at steep angles, approximately 20 to 35 degrees. In a preferred system, a manned control center communicates with a large number of the unmanned integrated RES to download emissions data, perform remote diagnostics, and, if necessary, dispatch a technician to perform maintenance on a particular RES.

These and other features and advantages of the invention will be apparent to those skilled in the art from the following detailed description of preferred embodiments, taken together with the accompanying drawings, in which:

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 is a diagram of an remote emissions sensing system in which a plurality of unmanned integrated RESs record vehicle emissions and communicate with a central unmanned control center;

FIG. 2 is a perspective view of one of the unmanned integrated RESs shown inFIG. 1;

FIG. 3 is a diagram of the source shown inFIG. 2;

FIG. 4 is a diagram of the receiver shown inFIG. 2;

FIG. 5 is a block diagram of the automated control processes executed by the control CPU shown inFIG. 2;

FIG. 6 is a flowchart illustrating the operation of the speed sensor shown inFIG. 2;

FIG. 7 is a flowchart illustrating the operation of the automated license plate reader (ALPR) shown inFIG. 2; and

FIG. 8 is a flowchart illustrating the coordination of the speed sensor and the ALPR shown in FIG.2.

DETAILED DESCRIPTION OF THE INVENTION

The present invention provides an emissions sensing system that includes a plurality of unmanned integrated RES. A manned control center communicates with a large number of the RESs to download emissions data, perform remote diagnostics, and, if necessary, dispatch a technician to perform maintenance on a particular RES. The source, receiver, speed sensor, automated license plate reader (ALPR), gas canister, power supplies, and computer are integrated into a console that can be positioned at the side of a road either permanently or for an extended period of time. A reflector is positioned on the other side of the road to reflect the IR beam back to the receiver. The source and receiver are preferably stacked one on top of the other such that the IR beam traverses a low and high path as it crosses the road. This allows the RES to detect both low and high ground clearance vehicles. To maintain the vehicle processing and identification throughput of the known systems, the speed sensor and ALPR detect the passing vehicles at steep angles, approximately 20 to 35 degrees. This has the beneficial effect of reducing the number of mismatches between pollutant readings and vehicle identification. Furthermore, data gathering control including calibration, verification, and data gathering are automated. This eliminates the need for an on site technician, which further reduces cost.

As shown inFIG. 1, a remoteemissions sensing system10 includes a plurality of unmanned integratedRESs12 that are placed at different positions in a network ofroads14, amanned control center16 and a two-way communications channel18. The communications channel shown is a wire-to-wire telephone network. Alternately, a cellular or satellite network could be used.

TheRES12 and areflector22 are placed on opposite sides of theroad14 and aligned such that the RES'sIR beam24 is reflected back to theRES12. When avehicle26 passes through theIR beam24, theRES12 writes the pollutant levels from the vehicle'sexhaust plume28 to a data file and appends the vehicle's license plate number. If the vehicle's speed or acceleration are too high, indicating that the vehicle's emissions control have been disabled, the data is invalidated.

TheRESs12 are automated to maintain calibration and, if repeated recalibration fails, to notify thecontrol center16. The control center performs remote diagnostics to identify the cause of the calibration failure and, if possible, to correct the problem. Otherwise a technician is dispatched to theRES12. TheRESs12 periodically download the gathered emissions data to thecontrol center16.

As shown inFIG. 2, all of the components of theRES12, except for thereflector22, are enclosed in aconsole30, suitably5′ high,3′ wide, and2′ deep. Thereflector22 such as a piece of black aluminum that is opaque in the visible spectrum is attached to the guard rail at the side of the road, for example. Asource32 emits theIR beam24 that crosses the road and reflects off of thereflector22 back to areceiver34. Thereceiver34 samples the radiation levels in thebeam24 at various wavelengths. Because of the presence of NO, water vapor, CO₂, CO, HC and other molecular species within theexhaust plume28, theIR beam24 is partially absorbed at the various wavelengths when it passes through the plume. Acomputer36 includes a data processing central processing unit (CPU)38 that computes the composition of the ambient air, and when a vehicle passes by, computes the composition of theplume28 in terms of the percentage or concentrations of the constituents NO, CO₂, CO and HC based on the sampled radiation levels. The computation of the composition is well known in the art and is thus omitted.

Atrigger circuit40 in thereceiver34 provides a trigger signal when a vehicle passes through thebeam24. The circuit responds to the sequential condition of the received signal going to zero, “beam block” followed by the received signal returning to a valid emissions level, “beam unblock.” Placing thesource32 on top of thereceiver34 causes theIR beam24 to traverse anupper path42 across the road and to return along alower path44 to the receiver. As a result, the circuit will trigger on either low or high ground clearance vehicles. The trigger signal is fed to thedata processing CPU38 causing it to write the composition of the plume to a data file on ahard disk46.

A vehicle identification system identifies the passing vehicle and appends the identification to the data file. The currently preferred approach is an automated license plate reader (ALPR) system that includes avideo camera48 that takes a picture of the vehicle'slicense plate50 in response to the trigger signal and aidentification CPU52 that executes a character recognition algorithm to extract the plate number. Alternately, the vehicles could transmit identification codes that would be detected as the vehicles pass by the RES.

The video camera takes the picture at an angle ρ with respect to the road. The shallower the angle, the easier it is for the character recognition algorithm to extract the plate number. However, the shallower the angle, the farther the vehicle is past the RES when its plate is read. This increases the chance of mismatching the vehicle identification to the wrong data file. Furthermore, this reduces the number of vehicles that can be tested in a given time, i.e. the vehicle throughput.

An optional speed sensor system determines the acceleration of an oncoming vehicle and invalidates the subsequently measured data if the acceleration is too high. The speed sensor system preferably includes anoblique angle radar54 that detects oncoming vehicles and aCPU56 that computes the vehicle's acceleration. Alternately, a LIDAR system, piezeo or pneumatic cables, or an optical sensor could be used to measure the vehicle's acceleration. Similar to the video camera, the slant radar detects the oncoming vehicle at an angle with respect to the road. The shallower the angle, the more accurate the estimate of the acceleration using known techniques but the lower the vehicle throughput. As a result, the known ALPR and speed/acceleration algorithms are modified as shown inFIGS. 7 and 6, respectively, to enable steep angle detection.

TheRES12 includes a number of secondary components that are required to support the data gathering process. Apower supply58 supplies power to thesource32,receiver34,video camera48,radar54, and the CPUs. A pair offans60 cool the electrical systems in theRES12. A pair ofvents62 vent the source and calibration gas to the atmosphere. Anexternal computer port64 allows a service technician to connect a laptop computer to theRES12 to access the CPUs and perform diagnostics.

An automated control system controls the data gathering process for theRES12. The primary function of the control system is to maintain calibration so that the recorded data is reliable. Agas canister66 contains calibration gas that has a known composition of pollutants. When actuated, thegas canister66 emits a puff of calibration gas in front of the source. This is used to both recompute the calibration curves and to verify the calibration.

TheRES12 can lose calibration for a number of reasons. First, the ambient conditions can change. For example, the CO₂levels typically rise during the day, the HC levels near industrial plants will also rise during the day, heavy traffic will increase the background pollutant levels, and rain will destroy the IR signature. Second, mechanical problems such as the gas bottle being empty, the source being worn out, or a stolen reflector will result in a loss of calibration. Another common source of signal degradation is a dirty receiver lens. In known systems, when the technician notices signal degradation he manually cleans the lens on the receiver. In the automated RES, amulti-position lens cover68 is placed in front of the receiver lens, and indexed when the signal levels degrade.

Acontrol CPU70, as detailed inFIG. 5, automates the calibration, verification, and data gathering processes by controlling the actuation of thegas canister66 and the indexing of themulti-position lens cover68 and monitoring the compositions of the exhaust plume and ambient air. When repeated attempts to calibrate the system fail, theCPU70 sends a help mess over thecommunications channel18 shown inFIG. 1 via acommunications port72.

As shown inFIG. 3, thesource32 includes anIR source74, preferably a broadband IR source such as a glow bar, that has a significant IR radiation output in the range of approximately 3 micrometers to approximately 6 micrometers. TheIR source74 provides abeam24 that may optionally be passed through a chopper76 (nominally 200 cycles per second) and a beam former78, such as a parabolic reflector. In the preferred embodiment, the receiver34 (shown in detail inFIG. 4) uses solid state detectors which must be turned on and off in order to detect the radiation levels. As a result, thechopper76 is positioned in the path of the IR beam to block and unblock the beam and thereby switch the detectors on and off.

In the preferred embodiment, thechopper76 is positioned at the IR source, which enables the system to distinguish infrared radiation emitted by the source from that emitted by the vehicle exhaust. When the chopper blocks the beam, the receiver measures the infrared radiation emitted from the plume. The data processing CPU calculates the peak-to-peak signal which removes the quiescent levels of the receiver as well as the interference from the vehicle exhaust. Thus, the measurements of the transmission levels are more accurate. Alternately, thechopper76 can be positioned at the receiver. However, in this configuration the constituent measurements can be distorted by irradiance from the plume itself.

As shown inFIG. 4, thereceiver34 includes themulti-positioned lens cover68 that is periodically indexed to provide a clean surface for receiving theIR beam24. Themulti-position lens cover68 is preferably an IR transmissive sheet on a roller. The IR beam is applied to a plurality n of narrow band filters80, where n is equal to a number of measurement channels. Eachfilter80 is selected so as to pass a predetermined narrow band of wavelengths to an associated one of a plurality ofIR detectors82. The IR detectors include photosensitive elements which are integrally fabricated on a substrate. The elements are preferably photoconductive and formed of mercury cadmium telluride (HgCdTe or HCT), whereas the substrate is cadmium zinc telluride (CdZnTe).

Eachdetector82 outputs an electrical signal corresponding to the radiation level at its wavelength to anamplifier84. An n channel analog to digital (A/D)converter86 digitizes the amplified signals and outputs them to thedata processing CPU38 shown inFIG. 2. Asuitable cooler88, such as a thermo-electric (TE) device, is employed for cooling those types ofIR detectors82 which are required to be cooled to an operating point that is below ambient temperature.

Abeam integrator lens90 is preferably placed between thelens cover68 and thefilters80 to homogenize thebeam24 after propagation through the plume to remove the spatial and temporal variations of the constituent concentrations so that the detected signals are synchronized. The optical intensity or energy that is incident on thephotodetectors82 is substantially uniform throughout the cross-section of thehomogenized beam24. This ensures that the same homogenized or averaged scene is sensed by thephotodetectors82, and substantially increases the accuracy of the measurement by reducing the spatial and temporal variance of the constituent concentrations by over an order of a magnitude. The beam integrator lens enables synchronous operation of the photodetectors.

In a presently preferred embodiment of this invention there are six spectral measurements channels. These are an NO spectral channel (having afilter80 with a passband centered on 5.26 μm), an H₂O spectral channel (having afilter80 with a passband centered on 5.02 μm), a first reference, or CO₂spectral channel (having afilter80 with a passband centered on 4.2 μm), a CO spectral channel (having afilter80 with a passband centered on 4.6 μm), a HC spectral channel (having afilter80 with a passband centered on 3.3 μm) and a second reference (REF) spectral channel (having afilter80 with a passband centered on 3.8 μm). Additional channels to measure other pollutants can also be added if desired.

In general, the NO spectral channel is located near resonant absorption peaks in the vicinity of 5.2 μm; the water vapor spectral channel is in a region of strong water absorption where fundamental lines do not saturate; the first reference spectral channel is employed for normalizing the pollutants to the normal combustion products, i.e., CO₂; and the second reference (REF) spectral channel is provided at a region in which no atmospheric or automotive emissions gases absorb.

The REF spectral channel compensates the other five spectral channels for variations caused by: (a) fluctuations in the output of theIR source74 shown inFIG. 3 during the passage of the vehicle; (b) particulate matter in the form of road dust; (c) particulate matter in theexhaust gas plume28; (d) infrared radiation emitted from the exhaust plume, and any other factors that may reduce the amount of illumination reaching thedetectors82. The REF spectral channel thus operates to provide a baseline output which is independent of the molecular species (NO, H₂O, CO₂, CO and HC) being measured. The output of the second REF spectral channel is used to normalize, such as by dividing, the five molecular species spectral channels.

FIG. 5 is a flowchart of the automated control process executed by thecontrol CPU70 shown inFIG. 2 in cooperation with themanned control center16 shown in FIG.1. Once the RES is set up, theCPU70 boots the system to a calibration mode (step92) and uses the speed and acceleration data provided by theCPU56 to determine whether a vehicle is approaching (step94). If so, theCPU70 waits (step96) until no vehicles are in range and performs a calibration (step98). TheCPU70 directs the gas canister to emit a puff of calibration gas so that the data processing CPU uses the radiation levels for the various pollutants and their known concentrations to recompute a set of calibration curves. Thereafter, theCPU70 switches to a measurement mode (step100).

Once in measurement mode, theCPU70 again determines whether a vehicle is approaching (step102), waits until no vehicle is in range (step104), and performs a puff-in-vehicle (PIV) test (step106) to verify the calibration. TheCPU70 directs the gas canister to emit another puff of calibration gas so that the data processing CPU uses the calibration curves to compute a composition for the calibration gas (step108). If the composition deviates from a known reference composition of the calibration gas then the calibration is rejected. If calibration has failed repeatedly (step110), for example 10 times in a row, theCPU70 directs the RES to notify the control center (step112). Otherwise theCPU70 repeats steps92 through108 to recalibrate the system and verify the calibration. When the composition calculated instep108 is close enough to the reference composition, the calibration is accepted and data collection initiated (step114). The data processing CPU will generate an error code 9999 when the data, i.e. the sensed radiation levels, is no good. Random and infrequent bad data is expected as part of the sensing process. However, a high percentage of bad data is indicative of a either a system problem such as an occluded lens, beam misalignment or mechanical problems in the source or the system being out of calibration. TheCPU70 monitors the data (step116), and if the frequency of error codes exceeds a threshold, initiates recalibration by returning control to step104. Otherwise, the data processing CPU continues gathering data (step118).

Because the ambient conditions can change over time, theCPU70 periodically verifies the last calibration (step120) by returning control to step102. The system continues gathering data (step122) in the measurement mode while theCPU70 monitors the data processing CPU to make sure that it is sampling the radiation levels and computing compositions (step124). If not, theCPU70 assumes that the system software has failed, power cycles the system (step126) to reboot the software, performs a calibration (step128), and determines whether the calibration was effective (step130). If power cycling has restored the system, control returns to step102 to verify the calibration. Otherwise, theCPU70 causes the RES to notify the control center (step112).

If the data processing CPU is receiving and processing the data instep124, theCPU70 monitors the ambient signal levels (radiation levels or compositions) (step132). If the ambient signal levels are close enough to a set of reference values (step134), for example, the values measured at the last calibration, then data gathering continues atstep114. If the signal levels have deviated, theCPU70 indexes thelens cover68 shown inFIG. 4 (step136). Oftentimes signal deviation is due to dirt or exhaust building up on the lens of the receiver. Thereafter, theCPU70 checks to determine whether the ambient signal levels have been corrected (step138). If so, the data processing CPU continues gathering data (step114). Otherwise, theCPU70 performs a calibration (step140) and a PIV (step142). If the calibration is accepted, data gathering continues. If not, the RES notifies the control center (step112).

When the RES notifies the control center (step112), a technician at the control center executes remote diagnostics over the communications channel to identify the problem (step144). If the system can be fixed remotely (step146), control is returned toCPU70 to gather data. Otherwise, a service technician is dispatched to the RES (step148).

In order to maintain the same vehicle throughput as the known RES systems, theintegrated RES radar54 andvideo camera48 shown inFIG. 2 must detect the approaching and passing vehicles, respectively, at a steep angle, approximately 20 to 35 degrees. As shown inFIG. 6, theCPU56 computes the apparent speed measured by theradar54 shown inFIG. 2 (step150) and then corrects for what is called “cosine error” (step152) by multiplying the apparent speed by the cosine of the detection angle (cos θ) to produce an accurate reading of the oncoming vehicle's true speed. Instep154, the CPU computes the vehicle's acceleration. The vehicle's speed and acceleration are used to determine whether the vehicle's emissions systems are disabled and can be used to predict when the vehicle should trigger data acquisition to reduce mismatch between recorded emissions and the identified license plate as detailed in FIG.8.

As shown inFIG. 7, the preferred ALPR system deskews the picture of the vehicle's license plate to compensate for the steep detection angle prior to executing a character recognition algorithm. When triggered, thevideo camera48 takes a picture of the passing vehicle's license plate at a step angle (ρ), approximately 20 to 35 degrees (step156). The system's CPU52 (shown inFIG. 2) digitizes the picture into a digital image and transforms the skewed image into a normalized image, as if the picture had been taken at a shallow angle of approximately zero degrees (step158). The steep-to-shallow angle transformation may be achieved using an affine transformation, for example.

The CPU then executes a correlation algorithm on the first character in the normalized image to generate a correlation value for each character in an alpha-numeric set and selects the character with the highest correlation value (step160). Thereafter, the correlation value of the selected character is compared to a recognition threshold, e.g. 90% (step162). If the correlation value is less than the threshold, recognition is rejected (step164). If the correlation value exceeds the threshold, the character is written into the ALPR file which is appended to the recorded emissions data file (step166). The correlation algorithm is repeated for each character in the license plate until all the characters have been recognized or rejected (step168). If only one or two of the characters in the license plate are rejected, the plate may still be uniquely identifiable. If so, the partial plate can be appended to the emissions data and recorded. However, if too many characters in the entire license plate are rejected, then the entire plate recognition is rejected and the recorded emissions data is not reported (step170).

A common problem is known RES systems is a mismatch between the recorded emissions data and the license plate, i.e. the wrong car is matched to the offending emissions. The steep angles used by the radar and video camera reduce the frequency of mismatches to some extent by confining the area in which they look for a passing vehicle. As illustrated inFIG. 8, the mismatch frequency can be further reduced by combining the speed and acceleration information provided by the radar with the trigger signal. Instep172, the data processing CPU computes the speed and acceleration of an approaching vehicle as described in FIG.6. The CPU uses this information and the distance to the vehicle to estimate a time-to-trigger range(step174). When the vehicle passes through the IR beam, the CPU records the trigger time (step176) and determines whether it falls within the time-to-trigger range (step178). If the trigger falls within the range, the CPU merges the emissions data with the license plates (step180). Otherwise, the CPU invalidates the data (step182).

While several illustrative embodiments of the invention have been shown and described, numerous variations and alternate embodiments will occur to those skilled in the art. Such variations and alternate embodiments are contemplated, and can be made without departing from the spirit and scope of the invention as defined in the appended claims.

Claims (21)

1. An unmanned optical emissions sensor for sensing a gas mixture composition of an exhaust plume of a motor vehicle travelling along a road , comprising:

a source for radiating a beam of light radiation along a path across a road such that the beam passes through the exhaust plume of a passing vehicle and otherwise passes through ambient air;

a receiver for sampling radiation levels at a plurality of predetermined wavelengths from the beam;

a canister for emitting a puff of calibration gas in the path of the beam between the source and the receiver, said calibration gas having a known reference composition of gas which absorbs radiation at the predetermined wavelengths;

a data processing computer for computing a gas mixture composition from the sensed sampled radiation levels in accordance with stored calibration curves;

a trigger device that produces a trigger signal when a vehicle passes through the beam causing the data processing computer to record the gas mixture composition of the vehicle's exhaust plume for a period of time;

an automated control computer that

a) calibrates the data processing computer by directing the canister to emit a puff of calibration gas, whereby the data processing computer recomputes computes the calibration curves in accordance with the known reference composition;

b) verifies the calibration by directing the canister to emit a puff of calibration gas, whereby the data processing computer computes a test composition from the radiation levels and accepts the calibration when the test composition is close enough to the known reference composition and otherwise rejects the calibration and initiates recalibration a new calibration; and

c) monitors the gas mixture composition of the ambient air to control recalibration calibration of the data processing computer; and

a vehicle identification device that responds to the trigger signal by recording a vehicle identification for the passing vehicle.

2. The unmanned optical emissions sensor ofclaim 1, further comprising:

a multi-position lens cover on the receiver, said automated control computer indexing the position of the lens cover when the gas mixture composition of the ambient air deviates from an ambient reference level by more than a specified threshold and initiates recalibration a new calibration if the deviation remains greater than the specified threshold.

3. The unmanned optical emissions sensor ofclaim 1, wherein the automated control computer monitors the gas mixture composition of the vehicle's exhaust plume to control reverification verification of the calibration and initiation of a new calibration.

4. The unmanned optical emissions sensor ofclaim 1, wherein the automated control computer monitors a time from the last calibration and when the time exceeds a mandatory recalibration period it initiates another a new calibration.

5. The unmanned optical emissions sensor ofclaim 1, wherein the automated control computer monitors the data processing computer and power cycles the emissions sensor when the data processing computer fails to produce gas mixture compositions.

6. The unmanned optical emissions sensor ofclaim 1, further comprising:

a manned control center; and

a communications channel for communication between the automated control computer and the manned control center, said automated control computer responding to repeated calibration rejections by transmitting a help message to the manned control center, which in turn responds by performing diagnostics to determine a cause for the calibration rejection and then either remedy remedies the cause remotely or dispatch dispatches a technician to remedy the cause on site.

7. The unmanned optical emissions sensor ofclaim 1, further comprising:

a vehicle detector for sensing an oncoming a vehicle and computing its acceleration, said data processing computer disabling the recordation a recording of the gas mixture composition of the vehicle's exhaust plume when the acceleration exceeds a threshold.

8. The unmanned optical emission sensor ofclaim 7, wherein the vehicle detector compute's computes the vehicle's speed and computes the time-to-trigger range from the vehicle's measured speed and acceleration, said data processing computer disabling the recordation a recording of the gas mixture composition of the vehicle's exhaust plume when triggering the trigger signal occurs outside the time-to-trigger range.

9. The unmanned optical emissions sensor ofclaim 7, wherein said source and said receiver are placed on the same side of the a road, further comprising:

a reflector that is positioned on the other side of the road such that the beam emitted by the source reflects off of the reflector and back to the receiver.

10. The unmanned optical emissions sensor ofclaim 9, further comprising:

a single console that contains comprises the source, the receiver, the canister, the data processing computer, the automated control computer, the vehicle identification device, and the vehicle detector.

11. The unmanned optical emissions sensor ofclaim 10, wherein the vehicle identification device comprises an automated license plate reader (ALPR) that reads the vehicle's license at an angle of at least 20 degrees and said vehicle detector senses the oncoming vehicle at an angle of at least 20 degrees to maintain a vehicle throughput .

12. The unmanned optical emissions sensor ofclaim 10, wherein one of said source and said receiver is positioned above the other so that the beam traverses the road in a low path in one direction and in a high path in the other direction so that the trigger device will trigger on both high and low ground clearance vehicles.

13. An integrated optical emissions sensor for sensing a gas mixture composition of an exhaust plume of a motor vehicle travelling along a road , comprising:

a single console that is positioned at one side of the road a detection space;

a vehicle detector in said console for sensing an oncoming a vehicle and computing its acceleration;

a source in said console for radiating a beam of light along a path across the road such that the beam passes through the exhaust plume of a passing the vehicle and otherwise passes through ambient air;

a reflector that is positioned on the other side of the road detection space such that the beam reflects off of the reflector and back to the console;

a receiver in said console sampling radiation levels at a plurality of predetermined wavelengths from the beam;

a data processing computer in said console for computing a gas mixture composition from the sensed sampled radiation levels in accordance with stored calibration curves;

a canister in said console for emitting a puff of calibration gas in the path of the beam between the source and the receiver to recompute compute the calibration curves;

a trigger device in said console that produces a trigger signal when a the vehicle passes through the beam causing the data processing computer to record the gas mixture composition of the vehicle's exhaust plume for a period of time, said data processing computer disabling the recordation of the composition of the vehicle's exhaust plume when the acceleration exceeds a threshold ; and

a vehicle identification device in said console that responds to the trigger signal by recording a vehicle identification for the passing vehicle.

14. The unmanned optical emissions sensor ofclaim 13, wherein the vehicle identification device comprises an automated license plate reader (ALPR) that reads the vehicle's license at an angle of at least 20 degrees and said vehicle detector senses the oncoming vehicle at an angle of at least 20 degrees to maintain a vehicle throughput .

15. The unmanned optical emissions sensor ofclaim 13, wherein one of said source and said receiver is positioned above the other so that the beam traverses the road detection space in a low path in one direction and in a high path in the other direction so that the trigger device will trigger on both high and low ground clearance vehicles .

16. The unmanned optical emissions sensor ofclaim 13, wherein the vehicle detector compute's computes the vehicle's speed and computes a time-to-trigger range from the vehicle's measured speed and acceleration, said data processing computer disabling the recordation a recording of the gas mixture composition of the vehicle's exhaust plume when triggering the triggering signal occurs outside the time-to-trigger range.

17. The unmanned optical emissions sensor ofclaim 13, wherein said calibration gas has a known reference composition of gas which absorbs radiation at the predetermined wavelengths, further comprising an automated control computer that

a) calibrates the data processing computer by directing the canister to emit a puff of calibration gas, whereby the data processing computer recomputes computes the calibration curves in accordance with the known reference composition;

b) verifies the calibration by directing the canister to emit a puff of calibration gas, whereby the data processing computer computes a test composition from the radiation levels and accepts the calibration when the test composition is close enough to the known reference composition and otherwise rejects the calibration and initiates recalibration a new calibration; and

c) monitors the gas mixture composition of the ambient air to control recalibration calibration of the data processing computer.

18. A remote emissions sensing system sensing gas mixture compositions of exhaust plumes for motor vehicles traveling along a network of roads, comprising:

a plurality of unmanned integrated optical emissions sensors positioned at different places in the network on a side of the road, each emissions sensor comprising:

a console;

a vehicle detector in said console for sensing an oncoming a vehicle and computing its acceleration;

a source in said console for radiating a beam of light along a path across the road such that the beam passes through the exhaust plume of a passing the vehicle and otherwise passes through ambient air;

a reflector that is positioned on the other side of the road such that the beam reflects off of the reflector and back to the console;

a receiver in said console that samples radiation levels at a plurality of predetermined wavelengths from the beam;

a data processing computer in said console for computing a gas mixture composition from the sensed sampled radiation levels in accordance with stored calibration curves;

a canister in said console for emitting a puff of calibration gas in the path of the beam between the source and the receiver, said calibration gas having a known reference composition of gas which absorbs radiation at the predetermined wavelengths;

a trigger device in said console that produces a trigger signal when a the vehicle passes through the beam causing the data processing computer to record the gas mixture composition of the vehicle's exhaust plume for a period of time, said data processing computer invalidating the recordation of the composition of the vehicle's exhaust plume when the acceleration exceeds a threshold ;

an automated control computer that

a) calibrates the data processing computer by directing the canister to emit a puff of calibration gas, whereby the data processing computer recomputes computes the calibration curves in accordance with the known reference composition;

b) verifies the calibration by directing the canister to emit a puff of calibration gas, whereby the data processing computer computes a test composition from the radiation levels and accepts the calibration when the test composition is close enough to the known reference composition and otherwise rejects the calibration and initiates recalibration a new calibration; and

c) monitors the gas mixture composition of the ambient air to control recalibration calibration of the data processing computer; and

a vehicle identification device in said console that responds to the trigger signal by recording a vehicle identification for the passing vehicle;

a manned control center; and

a communications channel for communication between the emissions sensors and the manned control center, said emissions sensors responding to repeated calibration rejections by transmitting a help message to the manned control center, which in turn responds by performing diagnostics to determine a cause for the calibration rejection and then either remedy remedies the cause remotely or dispatch dispatches a technician to remedy the cause on site.

19. The unmanned optical emissions sensor system ofclaim 18, wherein the vehicle detector compute's of the emissions sensors computes the vehicle's speed and computes a time-to-trigger range from the vehicle's measured speed and acceleration, said data processing computer disabling the recordation a recording of the gas mixture composition of the vehicle's exhaust plume when triggering the trigger signal occurs outside the time-to-trigger range.

20. The unmanned optical emissions sensor system ofclaim 18, wherein the vehicle identification device of the emissions sensors comprises an automated license plate reader (ALPR) that reads the vehicle's license at an angle of at least 20 degrees and said vehicle detector senses the oncoming vehicle at an angle of at least 20 degrees to maintain a vehicle throughput .

21. The unmanned optical emissions sensor system ofclaim 18, wherein one of said source and said receiver of said emission sensors is positioned above the other so that the beam traverses the road in a low path in one direction and in a high path in the other direction so that the trigger device will trigger on both high and low ground clearance vehicles .

Images