WO2001035371A1

WO2001035371A1 - Method and apparatus for collecting high reliable traffic information using a probe car

Info

Publication number: WO2001035371A1
Application number: PCT/KR2000/001269
Authority: WO
Inventors: Heung-Soo Lee
Original assignee: Haiwool Co., Ltd.
Priority date: 1999-11-06
Filing date: 2000-11-06
Publication date: 2001-05-17
Also published as: KR20010045683A; AU1177701A; KR100663308B1

Abstract

The present invention relates to a method and apparatus for effectively collecting high reliable traffic information from the moving vehicles by reducing error data. The apparatus for collecting traffic information has sensors to detect passing-on-adjacent-lane cars, which sensors' effective range are adjusted less than the width of a lane on road. Also, if at least 2 sensors are located on the side of the probe car, the distance D between the sensors us adjusted from 5 to 30 centimeters, and preferably, from 10 to 20 centimeters, to detect same car driving on adjacent lane. Also, by the use of state transition of the passing-on-adjacent-lane car. Valid data is selected. Also, in order to eliminate non-vehicle noise, the length of passing-on-adjacent-lane car is produced and then, if the length is not included to a predetermined value, the length is regarded as a noise.

Description

METHOD AND APPARATUS FOR COLLECTING HIGH RELIABLE

TRAFFIC INFORMATION USING A PROBE CAR

TECHNICAL FIELD

The present invention relates to a method and apparatus for collecting traffic

information using a probe car. More particularly, the invention relates to a method and

apparatus for effectively collecting highly reliable traffic information about moving

vehicles by reducing errors in data.

BACKGROUND ART

Currently, traffic information is collected by means of CCTVs, traffic reporters and

sensors, which are located on roads. This information is processed at a traffic control

center and transmitted to all sorts of institutions or systems that provide traffic information.

However, sensors such as point detectors or beacons have low accuracy in

collecting traffic information, and the accuracy of traffic information is dependent on the

locations of sensors. That is, traffic information collected in areas with few sensors has

low accuracy.

Also, the installation cost of sensors is very high because they are set up on every

road where traffic information has to be collected. In addition, sensors have to be set up

on newly constructed roads and there are high costs for maintenance to prevent damage to the sensors due to aging.

Another method that has been proposed for collecting traffic information is to use

a probe car.

According to this method, when a probe car's position is reported to the traffic

control center, the average velocity of the probe car is determined by correlating the probe

car's reported position on the road at a given time with the probe car's last reported

position.

However, the disadvantages of this method are that an excessive amount of data

must be processed at the traffic control center and a considerable number of probe cars are

needed in order to improve the accuracy of velocity data.

To overcome the disadvantages of the above-mentioned method, an apparatus and

method using a probe car are proposed, which can collect traffic information about other

cars, process the collected traffic information, and transmit the processed traffic

information to the traffic control center.

Hereinafter, we will briefly examine the apparatus and method using probe car.

The traffic information collection apparatus for collecting traffic information using

a probe car is comprised of: (1) at least two scanning sensors located at a fixed distance on

at least one side of a probe car for collecting information about vehicles in adjacent lanes;

(2) a means of analyzing the output signals of scanning sensors to determine the velocity of

vehicles in adjacent lanes; (3) a means of generating traffic information about vehicles in adjacent lanes based on the velocity of those vehicles; and (4) a means of transmitting the

traffic information collected about vehicles in adjacent lanes. Preferably, the traffic

information collection apparatus for collecting traffic information further comprises a

means of measuring a probe car's velocity, a means of measuring a probe car's position,

and a means of generating traffic information for collection purposes by using the probe

car's velocity, the probe car's position, and information about traffic in adjacent lanes

whereby the traffic information is transmitted for collection purposes. Moreover, the

means of measuring the probe car's position is a global positioning system, and the means

of transmitting the traffic information about vehicles in adjacent lanes can be a mobile

phone such as a PCS, TRS or cellular phone. In addition, traffic information about vehicles

in adjacent lanes can comprise data on the number of passing cars in adjacent lanes and

their average velocity, as well as data on variations in the average velocities of passing cars

in adjacent lanes.

There may be only one sensor mounted on each probe car, and in this case the

sensor is mounted on the corner of the probe car. By processing the signals form the

sensor, the velocity of passing cars in adjacent lanes, the number of those cars, and

variations in the average velocities of those cars are measured.

However, in the case that sensors are mounted on the side of a probe car, the

sensors detect all objects adjacent to the probe car and produce signals. Accordingly,

sensors produce not only signals due to adjacent cars but also due to pedestrians, electrical poles, guardrails, revetments or buildings (hereinafter abbreviated, "noise"), which signals

are useless for collecting traffic information.

In addition, referring to FIG. 1, Sensor SE-A on a probe car (100) detects a car

(110) driving in an adjacent lane, but Sensor SE-B on the probe car (100) detects a car

(120) driving in the leftmost lane. So, because the signal of Sensor SE-A and the signal

of Sensor SE-B are produced by cars in different lanes, the calculation of the velocity of

the car in an adjacent lane based on these signals is subject to error. Hence, the object

that Sensor SE-A and Sensor SE-B detect must be a single car driving in an adjacent lane.

Also, referring to FIG. 2, if Sensor SE-A and Sensor SE-B are located too far from

each other on the probe car, the signal of Sensor SE-A and the signal of Sensor SE-B can

be produced by different cars respectively. That is, Sensor SE-A detects one car (140)

and Sensor SE-B detects another car (130), so the calculation of the velocity of the car in

the adjacent lane based on these signals is also erroneous.

DISCLOSURE OF THE INVENTION

It is an object of the present invention to provide a method and apparatus for

improving the accuracy of collected traffic information by using probe cars.

More particularly, it is a primary object of the present invention to provide a

method and apparatus for detecting vehicles driving in an adjacent lane for the purpose of

collecting traffic information. Another object of the present invention is to provide a method and apparatus for

improving the accuracy of traffic information by eliminating noise from the traffic

information that is collected.

To achieve these objects, according to one aspect of the preferred embodiment of

the present invention, an apparatus for collecting traffic information by using probe car,

which comprises at least 2 scanning sensors, horizontally located on at least one side of

said probe car at predetermined distance, for detecting a plurality of passing-on-adjacent-

lane cars, a signal processing means for producing a passing-on-adjacent-lane car's

velocity by analyzing output signals of said scanning sensors and for producing a passing-

on-adjacent-lane car's data by the use of the passing-on-adjacent-lane car's velocity, means

for transmitting the passing-on-adjacent-lane car's data, wherein said scanning sensors,

each effective ranges of said scanning sensors are adjusted less than the width of a lane on

road is provided.

In the preferred embodiment of the present invention, the scanning sensors are

photo sensors and the scanning sensors comprise a lens respectively and the adjustment of

the effective ranges of said scanning sensors is performed by adjusting location of said lens.

Also, the effective ranges of said scanning sensors are adjusted within 2 meters.

According to the another preferred embodiment of the present invention, an

apparatus for collecting traffic information by using probe car, which comprises at least 2

scanning sensors, horizontally located on at least one side of said probe car at a predetermined distance, for detecting a plurality of passing-on-adjacent-lane cars, a signal

processing means for producing a passing-on-adjacent-lane car's velocity by analyzing

output signal of said scanning sensors and for producing a passing-on-adjacent-lane car's

data by the use of the passing-on-adjacent-lane car's velocity, means for transmitting the

passing-on-adjacent-lane car's data, wherein said scanning sensors, the predetermined

distance of said scanning sensors are between 5 centimeters and 30 centimeters is provided.

According to the another preferred embodiment of the present invention, an

apparatus for collecting traffic information by using probe car, which comprises means for

detecting a location and a velocity of said probe car, at least one sensor located on said

probe car, a signal processing means for producing a passing-on-adjacent-lane car's data

on a passing-on-adjacent-lane car, wherein the passing-on-adjacent-lane car's data

comprises at least a passing-on-adjacent-lane car's velocity, and for producing a collecting-

purpose traffic information by the use of the location and the velocity of said probe car,

and the passing-on-adjacent-lane car's velocity, means for transmitting the collecting-

purpose traffic information, wherein said scanning sensors, effective range of said scanning

sensor is adjusted less than the width of a lane on road, wherein the effective range of said

sensor is adjusted within 2 meters is provided.

According to the another preferred embodiment of the present invention, an

detecting a location said probe car, at least one sensor located on said probe car, a pulse generator for producing pulse whenever said probe car runs at predetermined distance, a

counter for counting the output of said pulse generator, a signal processing means for

producing a passing-on-adjacent-lane object's data on a passing-on-adjacent-lane object,

wherein the passing-on-adjacent-lane object's data comprises at least a passing-on-

adjacent-lane object's velocity, and for producing a collecting-purpose traffic information

by the use of the location and the velocity of said probe car, and the passing-on-adjacent-

lane object's data, means for transmitting the collecting-purpose traffic information,

wherein said signal processing means produce the passing-on-adjacent-lane object's length,

on the basis of the output of said sensor and the output of said counter and if the passing-

on-adjacent-lane object's length is not included to predetermined range, manipulate the

passing-on-adjacent-lane object's length as a noise is provided.

According to another aspect of the preferred embodiment of the present invention,

a method for collecting traffic information by using probe car, said method comprising the

steps of: detecting a passing-on-adjacent-lane object by the use of at least 2 sensors,

horizontally located on at least one side of said probe car at a predetermined distance,

deciding validity of the sensor's output signal by the use of state transition of the passing-

on-adjacent-lane object, producing a passing-on-adjacent-lane object's velocity by the use

of the valid output signal, producing a collecting-purpose traffic information that

comprises the passing-on-adjacent-lane object's velocity and transmitting the collecting-

purpose traffic information is provided. In the preferred embodiment of the present invention, the method further

comprises the step of detecting said probe car's location, and the collecting-purpose traffic

information further comprising said probe car's location and velocity.

According to the another preferred embodiment of the present invention, a method

for collecting traffic information by using probe car, said method comprising the steps of:

detecting a passing-on-adjacent-lane object by the use of at least 2 sensors, horizontally

located on at least one side of said probe car at a predetermined distance, deciding validity

of the sensor's output signal by the use of state transition of the passing-on-adjacent-lane

object, producing a passing-on-adjacent-lane object's velocity by the use of the valid

output signal, producing a collecting-purpose traffic information that comprises the

passing-on-adjacent-lane object's velocity and transmitting the collecting-purpose traffic

information is provided.

a computer readable medium storing instructions for carrying out the method is provided.

BRIEF DESCRIPTION OF DRAWINGS

FIG. 1 and FIG. 2 illustrate errors that occur in information collection by a probe

car (100) when a passing vehicle is detected in an adjacent lane;

FIG. 3 illustrates the adjustment of the effective range of scanning sensors SE-A

and SE-B in accordance with the preferred embodiment of the present invention; FIG. 4a and 4b illustrate the effective range of scanning sensors located on a probe

car;

FIG. 5 is a schematic representation of the apparatus for collecting traffic

information in accordance with the preferred embodiment of the present invention;

FIG. 6 is a flow chart showing one example of the method by which the apparatus

collects traffic information;

FIG. 7 shows the case of a vehicle passing in an adjacent lane;

FIG. 8 shows the logical processing of signals from the first and second sensor in

the case of a vehicle passing in an adjacent lane;

FIG. 9 shows examples of different possible situations in which vehicles pass in an

adjacent lane;

FIG. 10 illustrates the procedure for calculating the velocity of a vehicle passing in

an adjacent lane in a case similar to that of Situation (a) shown in FIG. 9;

FIG. 11 illustrates the procedure for calculating the velocity of a vehicle in an

adjacent lane in a case similar to that of Situation (e) shown in FIG. 9;

FIG. 12 is a schematic representation of the apparatus for collecting traffic

information in accordance with another preferred embodiment of the present invention;

FIG. 13a and FIG. 13b show an example of the operation of the pulse generator;

FIG. 14 shows the criteria for setting the length limitations of passing vehicles in

an adjacent lane; FIG. 15 is a flowchart showing the method for collecting traffic information in

accordance with the preferred embodiment of the present invention.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

Hereinafter, the preferred embodiment of the present invention will be described

with reference to accompanying drawings.

and SE-B in accordance with the preferred embodiment of the present invention.

Though Photo sensors, image-detecting sensors, speed-gun type sensors are

available to probe car 100, FIG. 3 is for illustrating the adjustment of effective range if the

adopted sensors are photo sensors.

Referring to FIG. 3, left-hand side represents the subject to be detected(hereinafter

referred as to 'subject side'), and right-hand side represents detecting part of sensors

(hereinafter referred as to 'film side').

Here, the size of pixel that is clearly viewed is defined as the circle of least

confusion and the circle of least confusion is represented as a circle.

In FIG. 3, the subject side is adjusted to be on A and film side is adjusted to be on

A'. Hence A' corresponding to A of the subject side is on the film side, if the subject is on

A, the image of subject on A is clearly focused on A'.

Here, first examine the film side's image of B that locates by Q2 ahead from the subject side. B' of the film side corresponding to B of the subject side is not on the film

side and image BM (the image of B on the film side) has a regular size. Accordingly,

image on the film side is not vivid. However, as appreciated in FIG. 3, if the error range

(i.e., the length of BM) on the film side becomes smaller than the size of the circle of least

confusion, the image can be regarded as a vivid image.

Also, the film side's image CM corresponding to C that locates by Ql behind from

the subject side's focus A is not a pixel and has a regular size, so the image is not vivid.

However, if the size of CM is smaller than the size of the circle of least confusion, the

image can be regarded as a vivid image.

Here, on the subject side, there are two surfaces where film side's error range is

same with a diameter of the circle of least confusion, and the distance between these two

surfaces is called 'depth of focus' (hereinafter abbreviated as 'DOF'). Especially, the

distance between point A and point B (i.e., Ql) is called 'front DOF', the distance between

point A and point C (i.e., Q2) is called 'behind DOF' and the length of front DOF is

generally shorter than the one of behind DOF.

Accordingly, by adjusting effective range of sensors (i.e., distance from the sensors

to subject side's focus) less than the width of a lane on road and lens's DOF same with the

width of a lane on road, noise to be occurred by detecting vehicles on another lane, not on

adjacent lane, can be eliminated.

More particularly, noise can be eliminated by adjusting effective range of sensors as follows. In FIG. 3, let distance from the surface including point B to lens be minimum

detection range Dl and distance from the surface including point C to lens be maximum

detection range D2. The minimum detection range of sensors Dl, as shown in FIG. 4a,

preferably becomes within the distance that probe car get near to the passing-on-adjacent-

lane vehicle (for example, 10cm). Also, the maximum detection range of sensors D2

becomes within the distance that probe car can go far away from the passing-on-adjacent-

lane vehicle. That is, if the width of a lane on road is Droad and the minimum width of a

vehicle is Dear, the maximum detection range of sensors D2 is preferably adjusted to be 2

X(Droad-Dcar)(as a preferable embodiment, maximum detection range is adjusted within

2 meters). Here, effective range of sensors is the distance between minimum detection

range and maximum detection range. The photo sensor in FIG. 3 is a direct-reflecting

photo sensor and comprises a light receiving part and a light concentrating part. The light

receiving part has a lens to collect a reflected light from the subject to be detected. In

order to adjust the sensor having this constitution, controlling light intensity by a volume

switch attached to the sensor is a general way to adjust the effective range of sensor.

However, in this case, if the size of the subject is large, any subject at a long distance can

be detected and, as a result, error can be occurred. Accordingly, it is a right way to

restrict effective range to a desirable range by first fixing the detection intensity of sensor

and then adjusting DOF by the use of the lens in a light concentrating part.

Here, DOF is the range where the detected object can be viewed clearly on that location when focusing on any point of detected subject. The smaller size of lens and the

shorter focusing range, the deeper DOF.

Since the size of lens in a photo sensor is regular, DOF can be narrowed with

controlling focal distance as a consequence of moving the location of lens forward or

backward. As a result, the detection range of sensor is restricted within a desirable scope.

Generally, the width of a lane on road is about 2.5 meters, so to restrict effective

range of sensor within 2 meters is preferable to only detect vehicle on adjacent lane. As

mentioned above, by detecting vehicle on adjacent lane as a consequence of adjusting

effective range of sensor, error shown in FIG. 1 can be eliminated. Also, detection signal

caused by a vehicle that is out of effective range of sensor does not come up with a

threshold value. Accordingly, if the signals from the first sensor SE-A and the second

sensor SE-B are not enough with a predetermined threshold value, signal processing part in

accordance with the present invention regards the signal as a noise.

On the other side, the noise caused by the situation shown in FIG. 2 can be

eliminated by narrowing the distance between the first sensor SE-A and the second sensor

SE-B.

Generally, the distance D between front vehicle and rear vehicle on the same lane

is larger than 20 centimeters. Accordingly, by narrowing the distance D within 20

centimeters, the situation shown in FIG. 2 can be prevented. However, if the distance D

is excessively narrowed, reflected light from the detected object may affect another sensor, so malfunction in sensor can be occurred. Also, if the surface of the detected vehicle is

not flat, the detection sequence between front sensor and rear sensor can be changed.

Accordingly, the distance D is wide enough not to affect each other.

According to the preferred embodiment, the distance D is between 5 centimeters

and 30 centimeters, more preferably, between 10 centimeters and 20 centimeters.

As mentioned above, if the distance D and the effective range of sensor are

adjusted, noise can be eliminated from the signals generated by each sensor.

In addition, detecting signal has noises that can not be utilized to produce traffic

information and these noises that are shown in FIG. 1 and FIG. 2 include the one caused by

passing-on-adjacent-lane vehicle and another one caused by pedestrians, electric poles,

streetlamps, trees, guardrails and buildings. The above-mentioned method is for

eliminating the former noise.

Hereinafter, examine method for effectively collecting high reliable traffic

information by analyzing the moving vehicle.

Producing traffic information by the use of state transition of passing-on-adjacent-

lane vehicle

FIG. 5 is a schematic representation of the apparatus for collecting traffic

Referring to FIG. 5, the apparatus for collecting traffic information comprises clock generator 210, first sensor SE-A 220, second sensor SE-B 230, counter 240, signal

processing part 250, memory 260 and transmitting part 270.

Clock generator 210 generates clock signal. Counter 240 receives clock signal

CLK and counts this signal. First sensor SE-A and second sensor SE-B for detecting

passing-on-adjacent-lane vehicle are located on the side of probe car and transmit each first

detecting signal SE-AP and second detecting signal SE-BP to signal processing part 250.

Signal processing part 250 is coupled to counter 240, first sensor SE-A, second sensor SE-

B, memory 260 and transmitting part 270. Memory 260 stores traffic information

produced by signal processing part 250 and transmitting part 270 transmits traffic

information produces by signal processing part 250 (or stored in memory 260) to traffic

control center. Transmitting part 270 is preferably mobile communication device such as

CDMA cellular phone and TRS phone.

FIG. 6 is a flow chart showing one example of the method by which the apparatus

collects traffic information.

Referring to FIG. 6, at step 501, signal processing part 250 receives first detecting

signal SE-AP and second detecting signal SE-BP from sensors and at step 502, determines

state transition of passing-on-adjacent-lane vehicle.

Also, at step 503, signal processing part 250 determines if detecting signals are

valid b the use of state transition information. If detecting signals are valid, proceeds to step 504 and signal processing part 250 produce the velocity of passing-on-adjacent-lane

vehicle by the use of output value of counter 240 at when the edges of first detecting signal

SE-AP and second detecting signal SE-BP generate. Also, at step 503, if detecting

signals are invalid, returns to step 501.

At step 504, by the use of produced velocity of passing-on-adjacent-lane vehicle,

signal processing part 250 produces traffic information and, if necessary, stores the traffic

information to memory 260. And at step 506, traffic information is transmitted to traffic

control center.

FIG. 7 shows the case of a vehicle passing in an adjacent lane.

Referring to FIG. 7, a passing-on-adjacent-lane vehicle has 4 kinds of state

transitions for the detecting line of first sensor SE-A and second sensor SE-B of probe car

100. At SO state, passing-on-adjacent-lane vehicle is on the left side of first sensor SE-A

or on the right side of second sensor SE-B and is not detected by each sensor.

At S1 state, passing-on-adjacent-lane vehicle is detected by first sensor SE-A but

is not detected by second sensor SE-B. Also, at state S2, passing-on-adjacent-lane

vehicle is detected by first sensor SE-A and second sensor SE-B. At state S3, passing-

on-adjacent-lane vehicle is not detected by first sensor SE-A but is detected by second

sensor SE-B. Finally, At SO state, passing-on-adjacent-lane vehicle is on the left side of

first sensor SE-A or on the right side of second sensor SE-B and is not detected by each sensor.

the case of a vehicle passing in an adjacent lane. If passing-on-adjacent-lane vehicle is

detected by first sensor SE-A and second sensor SE-B, detecting signal becomes 'HIGH'

level. Otherwise becomes 'LOW level.

When passing-on-adjacent-lane vehicle drives on any side of probe car 100,

passing-on-adjacent-lane vehicle makes above-mentioned transitions sequentially.

Examine this sequence in detail with FIG. 9.

adjacent lane.

For the purpose of explanation, first sensor SE-A is located on the front of probe

car 100 and second sensor SE-B is located on the rear of probe car 100.

The case (A) is, if probe car 100 passes passing-on-adjacent-lane vehicle ahead, a

sequence of the state transition of passing-on-adjacent-lane vehicle and the state transition

S0→S1 →S2→S3→S0 was sequentially made.

The case (E) is, if passing-on-adjacent-lane vehicle passes probe car 100 ahead, a

S0→S3→S2→S1 →S0 was sequentially made.

The case (B), (C), (D), (F), (G) and (H) represent that probe car 100 and passing- on-adjacent-lane vehicle run neck and neck.

In the preferable embodiment, for the purpose of simplification of data processing,

case (A) and (D) is processed as a valid data. However, it is easily appreciated by

someone skilled in the art that proper velocity calculation algorithm is applied to another

state transition.

an adjacent lane in a case similar to that of Situation (a) shown in FIG. 9.

Referring to FIG. 10, first detecting signal SE-AP becomes rising edge at time

TA1 and subsequently second detecting signal SE-BP becomes rising edge at time TB1.

Also, first detecting signal SE-AP represents dropping edge at time TA2 and subsequently

second detecting signal SE-BP represents dropping edge at time TB2.

Here, the interval time tr between rising edge of first detecting signal and second

detecting signal is TA1-TB1.

Accordingly, relational velocity of passing-on-adjacent-lane vehicle for probe car

100 is produced as following formula 1:

V. r>

where D is distance between sensors.

Also, in a preferable embodiment, detection-delaying time tr is calculated by counting the output clock of clock generator 210. That is, detection-delaying time tr is

calculated by subtracting output of counter 240 at time TBl from output of counter 240 at

time TA1 and multiplying by 1 clock time.

In case (A) as shown in FIG. 9, relational velocity of passing-on-adjacent-lane

vehicle is negative value.

Like this, relational velocity Vr of passing-on-adjacent-lane vehicle is produced

and then by the use of velocity Vp of probe car, velocity Vo of passing-on-adjacent-lane

vehicle is produced as following formula 2:

v ^τ o =v p +v r

Here, it is apparent to someone skilled in the art that the method for producing

velocity of probe car is diverse: (1) counting pulse per unit time generated by sensor at a

wheel axle, and (2) producing positioning transition per unit time by the use of GPS

(Global Positioning System).

adjacent lane in a case similar to that of Situation (e) shown in FIG. 9.

Likewise in FIG. 10, relational velocity of passing-on-adjacent-lane vehicle is

produced by the use of formula 1. In FIG. 11, differently in FIG. 10, relational velocity

Vr of passing-on-adjacent-lane vehicle is positive value. Also, velocity Vo of passing-on-

adjacent-lane vehicle can be produced by formula 2. FIG. 12 is a schematic representation of the apparatus for collecting traffic

information in accordance with another preferred embodiment of the present invention.

Referring to FIG. 12, compared to FIG. 5, the apparatus for collecting traffic

information further comprises pulse generator 290 and counter 280.

Pulse generator 290 produce pulse whenever wheel of vehicle rotates. Counter

280 counts pulses generated by generator 290. The output of counter 280 is applied to

signal processing part 250.

FIG. 13a and FIG. 13b show an example of the operation of the pulse generator.

Whenever the wheel rotates 1/32 rotation as shown in FIG. 13a, one pulse is generated as

shown in FIG. 13b. Accordingly, vehicle moves forward by St per 1 cycle of pulse signal

Cs, so velocity of probe car can be produced by the use of output value of counter.

Subsequently, examine the method for eliminating noise caused by non-vehicle.

Elimination of non-vehicle noise

At first, velocity of passing-on-adjacent-lane object is produced by above-

mentioned method. Here, passing-on-adjacent-lane object may be vehicle or not. In

order to discriminate vehicle and non-vehicle, the length of passing-on-adjacent-lane

object is produced as following formula 3:

|V I* Where, S represents the length of passing-on-adjacent-lane object, Vr represents

the velocity of passing-on-adjacent-lane object and ts represents detection-delaying time as

above-mentioned in FIG. 10 or FIG. 11.

Like this, the length S of passing-on-adjacent-lane object is produced and then if

the length is shorter than predetermined length (in preferable embodiment, 3-5 meters),

this passing-on-adjacent-lane object is regarded as a non-vehicle. If passing-on-adjacent-

lane object is recognized as a non-vehicle, the data caused by the object is abandoned.

Also, as shown in FIG. 14, the detectable part on the scanning line that has same height of

sensors located on probe car 100 is divided into more than 3 parts. More particularly, the

detectable part can be divided into the fore part ahead of front wheel Lf, the center part

between front wheel and back Lc wheel and the rear part in the rear of back wheel Lr, and

if aluminum wheel is installed on the wheel, detecting signal can be detected from the

aluminum wheel. Accordingly, in order to detect one pulse per one vehicle, detecting

signal must be detected from the center part Lc and any signal that is detected from another

part has to be treated as a noise. So, the length that is used to discriminate vehicle and

non-vehicle must be determined not to include any signal detected from the fore part Lf oτ

rear part Lr.

FIG. 15 is a flowchart showing the method for collecting traffic information in

accordance with the preferred embodiment of the present invention. Referring to FIG. 15, at step 601, signals are inputted by sensors. And at step

602, the state of passing-on-adjacent-lane object is decided. The state of passing-on-

adjacent-lane object is one of 4 states as shown in FIG. 7. The state of passing-on-

adjacent-lane object that is varying at time is recorded and at step 603, validity is

determined on the ground of recorded state transition. For example, state transition such

as S0→S1 →S2→S3 or S0→S3→S2→S1 is determined as valid data. At step 603, if

the state transition is determined as valid, proceeds to step 604 and the velocity of passing-

on-adjacent-lane object is produced. The velocity of passing-on-adjacent-lane object is

produced by the use of the relational velocity of passing-on-adjacent-lane object and the

velocity of probe car. If, at step 603, the state transition is determined as invalid, data

caused by that state transition is abandoned and return to step 601. At step 605, the

length of passing-on-adjacent-lane object is produced. The length of passing-on-

adjacent-lane object, as shown in formula 3, is produced by the use of detection-delaying

time ts and relational velocity of passing-on-adjacent-lane object. And at step 606, by

determining if the length of passing-on-adjacent-lane object is shorter than the

predetermined length, whether the object is vehicle or non-vehicle is determined. If the

object is non-vehicle, returns to step 601, otherwise proceeds to step 607. At step 607,

collecting-purpose traffic information including at least one selected from the group

consisting of the number of passing-on-adjacent-lane vehicle and the velocity is produced

and, if necessary, is stored in memory. Here, collecting-purpose traffic information preferably comprises data about the velocity of probe car, position and time. And then, at

step 608, collecting-purpose traffic information is transmitted to traffic control center.

The present invention is not limited to the above-mentioned embodiments, and

diverse modifications of the present invention are possible to someone skilled in the art

INDUSTRIAL APPLICABILITY

According to the present invention, in adjusting effective range of sensors located

on the probe car and in analyzing signals from the sensors, with determining if passing-on-

adjacent-lane object is vehicle or non-vehicle by the use of state transition and the length

of the object, the accuracy of traffic information is improved.

More particularly, in order to operate new signaling system (which system's signal

scheme automatically cooperates with traffic flow) controlling signal lamps, detection in

traffic volume and waiting row at intersection must be preceded. As existing method for

detecting in waiting row at intersection, there are loop detector and CC-TV detector.

However, these existing methods have defects such that installation cost and maintenance

cost are very high.

On the contrary to this, the present invention detects the velocity and position by

scanning passing-on-adjacent-lane vehicles, so the present invention can detect not only

waiting row at intersection but also traffic volume. Accordingly, method and apparatus

for collecting traffic information in accordance with the present invention can supply more accurate traffic information at less cost.

Claims

1. An apparatus for collecting traffic information by using probe car comprising:

at least 2 scanning sensors, horizontally located on at least one side of said probe

car at predetermined distance, for detecting a plurality of passing-on-adjacent-lane cars;

a signal processing means for producing a passing-on-adjacent-lane car's velocity

by analyzing output signals of said scanning sensors and for producing a passing-on-

adjacent-lane car's data by the use of the passing-on-adjacent-lane car's velocity;

means for transmitting the passing-on-adjacent-lane car's data,

wherein said scanning sensors, each effective ranges of said scanning sensors are

adjusted less than the width of a lane on road.

2. The apparatus as recited in claim 1, wherein said scanning sensors are photo

sensors.

3. The apparatus as recited in claim 2, wherein said scanning sensors comprise a

lens respectively and the adjustment of the effective ranges of said scanning sensors is

performed by adjusting location of said lens.

4. The apparatus as recited in claim 1, wherein the effective ranges of said

scanning sensors are adjusted within 2 meters.

5. An apparatus for collecting traffic information by using probe car comprising:

car at a predetermined distance, for detecting a plurality of passing-on-adjacent-lane cars;

by analyzing output signal of said scanning sensors and for producing a passing-on-

means for transmitting the passing-on-adjacent-lane car's data,

wherein said scanning sensors, the predetermined distance of said scanning

sensors are between 5 centimeters and 30 centimeters.

6. The apparatus as recited in claim 5, wherein the predetermined distance of said

scanning sensors is between 10 centimeters and 20 centimeters.

7. The apparatus as recited in claim 6, wherein each effective ranges of said

scanning sensors are adjusted less than the width of a lane on road.

8. An apparatus for collecting traffic information by using probe car comprising:

means for detecting a location and a velocity of said probe car;

at least one sensor located on said probe car; a signal processing means for producing a passing-on-adjacent-lane car's data on a

passing-on-adjacent-lane car, wherein the passing-on-adjacent-lane car's data comprises at

least a passing-on-adjacent-lane car's velocity, and for producing a collecting-purpose

traffic information by the use of the location and the velocity of said probe car, and the

passing-on-adjacent-lane car's velocity;

means for transmitting the collecting-purpose traffic information,

wherein said scanning sensors, effective range of said scanning sensor is adjusted

less than the width of a lane on road.

9. The apparatus as recited in claim 8, wherein said sensor is a photo sensor

comprising lens and the adjustment of the effective range of said sensor is performed by

adjusting location of said lens.

10. The apparatus as recited in claim 9, wherein the effective range of said sensor

is adjusted within 2 meters.

11. An apparatus for collecting traffic information by using probe car comprising:

means for detecting a location said probe car;

at least one sensor located on said probe car;

a pulse generator for producing pulse whenever said probe car runs at predetermined distance;

a counter for counting the output of said pulse generator;

a signal processing means for producing a passing-on-adjacent-lane object's data

on a passing-on-adjacent-lane object, wherein the passing-on-adjacent-lane object's data

comprises at least a passing-on-adjacent-lane object's velocity, and for producing a

collecting-purpose traffic information by the use of the location and the velocity of said

probe car, and the passing-on-adjacent-lane object's data;

means for transmitting the collecting-purpose traffic information,

wherein said signal processing means produce the passing-on-adjacent-lane

object's length, on the basis of the output of said sensor and the output of said counter and

if the passing-on-adjacent-lane object's length is not included to predetermined range,

manipulate the passing-on-adjacent-lane object's length as a noise.

12. The apparatus as recited in claim 11, wherein effective range of said sensor is

adjusted less than the width of a lane on road.

13. The apparatus as recited in claim 11, wherein said sensor is a photo sensor.

14. The apparatus as recited in claim 13, wherein said sensor comprises a lens and

the adjustment of the effective range of said sensor is performed by adjusting location of said lens.

15. The apparatus as recited in claim 11, wherein the number of said sensor is

more than 2 and said sensors are located on the side of said probe car at a predetermined

distance, wherein the predetermined distance of said sensors are between 5 centimeters and

30 centimeters.

16. The apparatus as recited in claim 15, wherein the predetermined distance of

said scanning sensors is between 10 centimeters and 20 centimeters.

17. A method for collecting traffic information by using probe car, said method

comprising the steps of:

detecting a passing-on-adjacent-lane object by the use of at least 2 sensors,

horizontally located on at least one side of said probe car at a predetermined distance;

deciding validity of the sensor's output signal by the use of state transition of the

passing-on-adjacent-lane object;

producing a passing-on-adjacent-lane object's velocity by the use of the valid

output signal;

producing a collecting-purpose traffic information that comprises the passing-on-

adjacent-lane object's velocity; and transmitting the collecting-puφose traffic information.

20. The method as recited in claim 19, wherein said step of producing the passing-

on-adjacent-lane object's velocity by the use of the valid output signal comprises:

producing a passing-on-adjacent-lane object's relational velocity on the basis of

the output signal of the sensors;

detecting said probe car's velocity; and

producing the passing-on-adjacent-lane object's velocity by the use of the passing-

on-adjacent-lane object's relational velocity and said probe car's velocity.

21. The method as recited in claim 20, further comprising the step of detecting said

probe car's location, and the collecting-puφose traffic information further comprising said

probe car's location and velocity.

22. A computer readable medium storing instructions for carrying out the method

as recited in claim 17 to claim 21.