JP3007019B2 - Traffic flow measurement device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a traffic flow measuring device for detecting an abnormal event in a traffic flow of an automobile on a road.

[0002]

2. Description of the Related Art As an example of a traffic flow measuring device, there has been proposed a device which uses a TV camera as a sensor and processes an image obtained from the TV camera to measure a traffic condition on a road. Figure 18 shows the 1986 17th Image Engineering Conference, Yoshiaki Inoue and three others, "A trial of vehicle dynamics measurement" (16
It is a block diagram which shows the structure of the conventional traffic flow measuring device by the image processing described in -13).

In FIG. 18, reference numeral 1 denotes a TV camera as an image pickup device, 2 denotes a TV camera control circuit for controlling the TV camera 1, 3 denotes a frame memory for receiving digitized image pickup signals, and 4 denotes an extraction area of the vehicle. Reference numeral 5 denotes a vehicle detection circuit, 6 denotes a vehicle traveling position detection circuit, 7 denotes a vehicle movement amount detection circuit, and 8 denotes a vehicle information measurement circuit.

Next, the operation will be described. TV camera 1
Captures an image of the traveling vehicle at an arbitrary frame interval according to the TV camera control circuit 2. The frame memory 3 takes in data obtained by converting an image signal of the TV camera 1 into a digital signal. The window setting circuit 4 sets a window which is the size of the vehicle on the road on the screen. Vehicle detection circuit 5
Detects a vehicle by extracting features of the vehicle, such as a horizontal line and a vertical line, from the area set by the window setting circuit 4. The vehicle traveling position detection circuit 6 detects a traveling position on the screen of the vehicle detected by the vehicle detection circuit 5,
The vehicle movement detecting circuit 7 compares the running position with the running position of the preceding frame of the vehicle, detects a moving amount of the vehicle on the screen, and transmits the detected moving amount to the window setting circuit 4. The window setting circuit 4 moves the window according to the movement amount. The above operation is repeated to track the vehicle. When the vehicle disappears from the screen, the tracking ends, and the vehicle information measurement circuit 8
Measures the speed of the vehicle, the distance between the vehicle and the preceding vehicle, and the like. As described above, the traffic flow parameters of the traffic flow are measured.
Because of the resolution of the TV camera, like a distant road,
Even in the field of view of the camera, there are areas where it is difficult to detect a vehicle by image processing. Further, when monitoring over a wide area, a plurality of TV cameras and a traffic flow measuring device are used to cope with the problem. However, when the number of TV cameras is small, there is an area between the TV cameras and the TV camera that cannot be seen. In the conventional traffic flow measuring device, there are areas, such as these areas, which are so-called blind spots where vehicle detection and measurement are impossible.

[0005]

As described above, the conventional traffic flow measuring device is designed to monitor a wide area traffic condition when the sensors are installed at a certain interval in the target area. Not only point measurement sensors such as ultrasonic sensors and radar sensors, but also sensors that can measure a certain area such as image processing type, for example, the majority of the target area becomes a blind spot area and traffic conditions However, there was a problem that measurement was not possible. If the sensor interval is several hundred meters, a posteriori estimation is performed after a sufficient number of vehicles have been measured from a large number of sensors. There was a problem that cannot be achieved. The present invention has been made to solve the above problems, and provides a traffic flow measuring device capable of detecting a traffic condition in a blind spot area, particularly an abnormal event.

[0006]

According to a first aspect of the present invention, there is provided a traffic flow measuring device arranged at predetermined intervals on a road to detect the presence of a vehicle on the road. A first vehicle measuring unit and a second vehicle measuring unit corresponding to each sensor for detecting a vehicle traveling on a road based on an output of each sensor and measuring a traffic flow parameter relating to a traffic state. Traffic flow measuring device
Of the first vehicle calculated by the first vehicle measuring means corresponding to the first sensor
A predicting means for calculating a predicted value of a traffic flow parameter measured by a second vehicle measuring means corresponding to the second sensor after a predetermined time from a time at which the first measured value is measured, based on the measured value of Determining means for comparing the second measured value measured by the second vehicle measuring means with the predicted value and estimating a traffic state in an area where both sensors cannot be measured within an area between the two sensors. It is a thing.

[0007] A traffic flow measuring device according to a second aspect of the present invention comprises:
A plurality of sensors after a first sensor, a second sensor, and a third sensor, which are arranged at predetermined intervals on the road to detect the presence of a vehicle on the road, and travel on the road by the outputs of these sensors. In a traffic flow measuring device including a plurality of vehicle measuring means corresponding to each sensor for detecting a traffic vehicle and measuring a traffic flow parameter related to a traffic condition, a first sensor among the sensors is used as a reference, Prediction of a traffic flow parameter measured by each vehicle measuring means corresponding to each sensor after a predetermined time from the time when the measured value is measured, based on the measured value measured by the first vehicle measuring means corresponding to the first sensor The prediction means for calculating the value, each measurement value measured by each vehicle measurement means and each prediction value are compared, and the first sensor and the second sensor are set in an area between the first sensor and the second sensor. In the area where no sensor can measure It is obtained by a determination means for estimating the traffic conditions.

According to a third aspect of the present invention, there is provided a traffic flow measuring device,
A traffic flow measuring device according to claim 1 or claim 2, on the traffic flow the first sensor with respect to the traveling direction of the vehicle
It is arranged on the downstream side .

According to a fourth aspect of the present invention, there is provided a traffic flow measuring device.
The traffic flow measuring device according to claim 1 or 2, wherein the first sensor is disposed on a downstream side of the traffic flow with respect to the traveling direction of the vehicle .

According to a fifth aspect of the invention, there is provided a traffic flow measuring device.
Multiple sensors arranged at predetermined intervals on the road to detect the presence of vehicles on the road, and the output of each sensor detects vehicles traveling on the road to measure traffic flow parameters related to traffic conditions A traffic flow measuring device including a plurality of vehicle measuring means corresponding to each of the sensors, wherein a first sensor downstream of the traffic flow with respect to a traveling direction of the vehicle is used as a reference, and a second sensor corresponding to the first sensor is provided. Estimation of a traffic flow parameter measured by each vehicle detecting means corresponding to each sensor a predetermined time before the time when the first measured value is measured, based on the first measured value measured by the first vehicle detecting means Estimating means for calculating the value, and comparing the second measured value and the estimated value measured by each vehicle measuring means at the time when the estimated value is calculated, and determining the area where both sensors cannot be measured within the area between the two sensors. Means for estimating traffic conditions at It includes those were.

According to a sixth aspect of the present invention, there is provided a traffic flow measuring device,
The traffic flow measuring device according to any one of claims 1 to 5, wherein the judging means is in a region where each sensor cannot measure when an evaluation value indicating a result of the comparison exceeds a threshold value for abnormal event judgment. It is determined that an abnormal event has occurred.

[0012] The traffic flow measuring device according to the invention of claim 7 is:
4. The traffic flow measuring device according to claim 3, wherein the predicting means calculates a predicted arrival time of the vehicle measured by the second sensor on the downstream side from the speed of the vehicle passing through the first sensor, and the determining means calculates the predetermined arrival time. In this case, the degree of coincidence between the arrival time of the vehicle measured by the second sensor and the predicted arrival time is compared, and the traffic state in the unmeasurable region between the two sensors is estimated.

The traffic flow measuring device according to the invention of claim 8 is:
A first sensor that is disposed on the road at an upstream side of the traffic flow with respect to the traveling direction of the vehicle and detects the vehicle; and a first vehicle number that detects the vehicle number of the vehicle based on a detection signal of the first sensor. Reading means, a second sensor arranged downstream of the traffic flow to detect the vehicle, a second vehicle number reading means for detecting the vehicle number of the vehicle based on a detection signal of the second sensor, Prediction means for calculating a predicted arrival time measured by the second sensor from the speed and measurement time of the vehicle calculated from the detection signal of the sensor, and a second vehicle number reading means detects the vehicle number within the predicted arrival time. And determining means for detecting occurrence of an abnormal event in an area where measurement by both sensors is impossible in an area between both sensors when there is no such event.

According to a ninth aspect of the present invention, there is provided a traffic flow measuring device,
A first sensor disposed on the road with respect to the traveling direction of the vehicle on the upstream side of the traffic flow to detect the vehicle, a second sensor disposed on the downstream side of the traffic flow to detect the vehicle, The second speed measured by the second sensor is calculated based on the first speed of the vehicle measured by the first sensor, including the speed change between the two sensors, and the first speed at which the vehicle reaches the second sensor is calculated. Predicting means for determining the arrival time of the vehicle, including a required time calculated from the section length between the two sensors and the first speed, including a variation in the required time due to a speed change between the two sensors of the vehicle;
Comparing the degree of coincidence between the second speed and the first arrival time with respect to the third speed and the second arrival time of the vehicle detected by the sensor, and detects traffic in the unmeasurable region between the two sensors. It is provided with a judgment means for estimating the state.

According to a tenth aspect of the present invention, there is provided a traffic flow measuring device arranged at predetermined intervals on a road to detect the presence of a vehicle on the road, and a first sensor and a second sensor for detecting the presence of a vehicle on the road. A traffic flow measuring device including a first vehicle measuring unit and a second vehicle measuring unit corresponding to each sensor for detecting a vehicle traveling on a road based on an output and measuring a traffic flow parameter relating to a traffic state. Measuring the first measurement value based on the first measurement value calculated by the first vehicle measurement means corresponding to the first sensor arranged on the upstream side of the traffic flow with respect to the traveling direction of the vehicle A predicting means for calculating a predicted value of the traffic flow parameter measured by the second vehicle measuring means corresponding to the second sensor after a predetermined time from the performed time, and a second sensor arranged downstream of the traffic flow. Measured by the second vehicle measuring means When a part of the measurement value of No. 2 is not measured, the possibility of the measurement value abnormality of the second sensor is estimated from the measurement value measured by the second vehicle measurement means and the inter-vehicle distance predicted by the prediction means. Judgment means.

A traffic flow measuring device according to an eleventh aspect of the present invention comprises a first sensor and a second sensor arranged at predetermined intervals on a road to detect the presence of a vehicle on the road, A traffic flow measuring device including a first vehicle measuring unit and a second vehicle measuring unit corresponding to each sensor for detecting a vehicle traveling on a road based on an output and measuring a traffic flow parameter relating to a traffic state. A second sensor disposed on the downstream side of the traffic flow based on the speed and the inter-vehicle distance of the vehicle measured by the first vehicle measuring means corresponding to the first sensor disposed on the upstream side of the traffic flow with respect to the traveling direction of the vehicle; Prediction means for calculating a predicted value of the inter-vehicle distance of the vehicle measured by the second vehicle measurement means corresponding to the sensor of the second type, and prediction of the speed, arrival time and inter-vehicle distance of the vehicle measured by the second vehicle measurement means Matching each predicted value of the means Evaluated, in which a determination means for estimating the traffic conditions at the measurement impossible areas of both sensors in the region between the two sensors.

According to a twelfth aspect of the present invention, there is provided a traffic flow measuring device arranged at predetermined intervals on a road to detect the presence of a vehicle on the road, and a first sensor and a second sensor for detecting the presence of a vehicle on the road. A traffic flow measuring device including a first vehicle measuring unit and a second vehicle measuring unit corresponding to each sensor for detecting a vehicle traveling on a road based on an output and measuring a traffic flow parameter relating to a traffic state. A second sensor based on the speeds and arrival times of the two vehicles calculated by the first vehicle measuring means corresponding to the first sensor disposed on the upstream side of the traffic flow with respect to the traveling direction of the vehicle Prediction means for calculating a predicted speed and a predicted arrival time of each vehicle measured by the second vehicle measurement means corresponding to the vehicle, and a following vehicle based on the predicted speed and the predicted arrival time at which the preceding vehicle of each vehicle reaches the second sensor. Predicted speed and The measured arrival time is corrected, the measured value measured by the second vehicle measuring means is compared with the corrected predicted speed and the predicted arrival time, and traffic in an area where both sensors cannot be measured within an area between the two sensors. Judgment means for estimating the state.

According to a thirteenth aspect of the present invention, there is provided a traffic flow measuring device, comprising: a first sensor and a second sensor arranged at predetermined intervals on a road to detect the presence of a vehicle on the road; A traffic flow measuring device including a first vehicle measuring unit and a second vehicle measuring unit corresponding to each sensor for detecting a vehicle traveling on a road based on an output and measuring a traffic flow parameter relating to a traffic state. Predictive measurement in which a second sensor disposed downstream of a traffic flow among a plurality of vehicles measured by a first sensor disposed upstream of a traffic flow with respect to a traveling direction of a vehicle measures within a predetermined time Predicting means for calculating a value, and associating the relationship between the measured value and the predicted value with an evaluation value for a plurality of vehicles in which the measured value measured by the second sensor and the predicted measured value match, based on the evaluation value Of both sensors within the area between them It is obtained by a determination means for estimating the traffic condition in the measurement impossible range.

According to a fourteenth aspect of the present invention, there is provided a traffic flow measuring device according to any one of the first to thirteenth aspects.
In the measuring device, when the judgment means detects an abnormal event
A control command signal is issued to measure the location of the abnormal event.
And an abnormal event imaging means for imaging the occurrence point of the abnormal event in accordance with the control command signal of the determining means.

According to a fifteenth aspect of the present invention, in the traffic flow measuring device according to the fourteenth aspect, the abnormal event imaging means includes a first sensor and a second sensor, an ITV camera, a millimeter wave camera, It is a system such as an infrared camera or a satellite capable of communicating by capturing an image with a camera.

[0021]

According to the first aspect of the present invention, the second sensor corresponding to the second sensor after a predetermined time from the time when the first measurement value is measured.
The predicted value of the traffic flow parameter measured by the vehicle measuring means is calculated by the predicting means, and the second value measured by the second vehicle measuring means is calculated by the predicting means.
The measured value and the predicted value are compared by the judging means to estimate the traffic condition in the area where the sensors cannot be measured within the area between the two sensors.

According to the second aspect of the present invention, the time at which the measured value is measured based on the measured value measured by the first vehicle measuring means corresponding to the first sensor based on the first sensor. After a predetermined time from, the predicted value of the traffic flow parameter measured by each vehicle measuring means corresponding to each sensor is calculated by the predicting means, and each measured value measured by each vehicle measuring means is compared with each predicted value by the determining means. And estimating a traffic condition in an area where the first sensor and the second sensor cannot measure within an area between the first sensor and the second sensor.

According to a third aspect of the present invention, in the traffic flow measuring device according to the first or second aspect, the first sensor is disposed upstream of the traffic flow with respect to the traveling direction of the vehicle. The second sensor is arranged on the downstream side of the traffic flow, and based on the first measurement value measured by the first vehicle measurement means corresponding to the first sensor, from the time when the first measurement value is measured After a predetermined time, the predicted value of the traffic flow parameter measured by each vehicle measuring means corresponding to each sensor is calculated by the predicting means, and the second measured value measured by each vehicle measuring means is compared with the predicted value by the determining means. And estimating the traffic condition in an area where both sensors cannot measure within an area between the two sensors.

According to a fourth aspect of the present invention, in the traffic flow measuring device according to the first or second aspect, the first sensor is disposed downstream of the traffic flow with respect to the traveling direction of the vehicle . Based on the first measurement value measured by the first vehicle measurement means corresponding to the first sensor, each vehicle measurement means corresponding to each sensor measures after a predetermined time from the time at which the first measurement value was measured. The predicted value of the traffic flow parameter to be calculated is calculated by the predicting means, and the second measured value measured by each vehicle measuring means is compared with the predicted value. Estimate the traffic condition of.

According to the fifth aspect of the present invention, the first sensor detected by the first vehicle detecting means corresponding to the first sensor is based on the first sensor on the downstream side of the traffic flow with respect to the traveling direction of the vehicle. On the basis of the measured value of 1, the estimated value of the traffic flow parameter measured by each vehicle detecting means corresponding to each sensor a predetermined time before the time when the first measured value is measured is calculated by the estimating means, The second measurement value measured by the vehicle measurement means and the estimated value are compared by the determination means, and a traffic state in an area where both sensors cannot be measured within an area between the two sensors is estimated.

According to a sixth aspect of the present invention, in the traffic flow measuring device according to any one of the first to fifth aspects, the evaluation means indicating that the evaluation value indicating the comparison result exceeds the abnormal event determination threshold value. Then, it is determined that an abnormal event has occurred in the area where measurement of each sensor is impossible.

According to a seventh aspect of the present invention, in the traffic flow measuring device according to the third aspect, the predicting means measures the speed of the vehicle which is measured by the second sensor on the downstream side from the speed of the vehicle passing through the first sensor. The predicted arrival time is calculated, the degree of coincidence between the arrival time of the vehicle measured by the second sensor within the predetermined time and the predicted arrival time is compared by the determination means, and traffic in the unmeasurable area between the two sensors is determined. Estimate the state.

According to the invention of claim 8, the first sensor is disposed on the road at the upstream side of the traffic flow with respect to the traveling direction of the vehicle to detect the vehicle, and the first sensor detects the vehicle based on the detection signal of the first sensor. 1
Under the detected vehicle number of the vehicle in vehicle number reader, traffic flow
The second sensor is disposed on the downstream side to detect the vehicle, the vehicle number of the vehicle is detected by the second vehicle number reading means based on the detection signal of the second sensor, and calculated from the detection signal of the first sensor. The predicting means calculates the predicted arrival time measured by the second sensor from the vehicle speed and the measured time, and when the second vehicle number reading means does not detect the vehicle number within the predicted reaching time, the determining means determines the both. An occurrence of an abnormal event in a region where both sensors cannot be measured within a region between the sensors is detected.

According to the ninth aspect of the present invention, the second sensor on the downstream side of the traffic flow is connected to the first sensor based on the first speed of the vehicle measured by the first sensor on the upstream side of the traffic flow. A second speed measured by the second sensor including a speed change between the sensor and the second sensor is calculated, and a first arrival time at which the vehicle reaches the second sensor is calculated by a prediction unit in a section between the two sensors. Chief and first
The required time calculated from the speed of the vehicle is determined to include a variation in the required time due to a speed change between the two sensors of the vehicle, and the third speed and the second arrival time of the vehicle detected by the second sensor are determined. The determination means compares the degree of coincidence between the second speed and the first arrival time with respect to, and estimates a traffic state in an unmeasurable region between the two sensors.

According to the tenth aspect of the present invention, the first predicting means calculates the first value based on the first measurement value calculated by the first vehicle measuring means corresponding to the first sensor on the upstream side of the traffic flow. After a predetermined period of time from the time when the measured value is measured, the predicted value of the traffic flow parameter measured by the second vehicle measuring means corresponding to the second sensor is calculated. When a part of the second measurement value measured by the second vehicle measurement means corresponding to the second sensor on the downstream side of the traffic flow is not measured by the determination means, the second vehicle measurement means measures The possibility of abnormality of the measurement value of the second sensor is estimated from the measured value thus obtained and the inter-vehicle distance predicted by the prediction means.

According to the eleventh aspect of the present invention, based on the vehicle speed and the inter-vehicle distance measured by the first vehicle measuring means corresponding to the first sensor on the upstream side of the traffic flow, the second vehicle on the downstream side of the traffic flow is determined.
The predicting means calculates a predicted value of the inter-vehicle distance of the vehicle measured by the second vehicle measuring means corresponding to the sensor, and predicts the speed, arrival time and inter-vehicle distance of the vehicle measured by the second vehicle measuring means. The degree of coincidence with each predicted value of the means is evaluated by the judging means, and a traffic state in an area where the sensors cannot be measured within the area between the two sensors is estimated.

According to the twelfth aspect of the present invention, the predicting means is based on the speeds and arrival times of the two vehicles calculated by the first vehicle measuring means corresponding to the first sensor on the upstream side of the traffic flow. Calculates the predicted speed and predicted arrival time of each vehicle measured by the second vehicle measuring means corresponding to the second sensor, and determines the predicted speed and predicted time at which the preceding vehicle of each vehicle reaches the second sensor by the determining means. The predicted speed and the predicted arrival time of the following vehicle are corrected based on the arrival time, and the measured value measured by the second vehicle measuring means is compared with the corrected predicted speed and the predicted arrival time. Estimate the traffic condition in the unmeasurable area.

According to the thirteenth aspect of the present invention, the predictive measurement in which the second sensor downstream of the traffic flow measures within a predetermined time among the plurality of vehicles measured by the first sensor upstream of the traffic flow. The value is calculated by the predicting means, and the relationship between the measured value and the predicted value is evaluated for a plurality of vehicles in which the measured value measured by the second sensor and the predicted measured value measured by the second sensor match by the determining means. Then, based on the evaluation value, a traffic state in a region where both sensors cannot be measured in a region between the two sensors is estimated.

According to the fourteenth aspect, from the first aspect,
The traffic flow measuring device according to claim 13.
When an abnormal event is detected, a control command signal is issued so as to measure the location of the occurrence of the abnormal event, and the abnormal event imaging means images the location of the occurrence of the abnormal event based on the control command signal.

According to a fifteenth aspect of the present invention, in the traffic flow measuring device according to the fourteenth aspect, the abnormal event imaging means includes a first sensor and a second sensor, an ITV camera, and a millimeter wave camera.
When an abnormal event is detected by using a system such as an infrared camera or a satellite capable of communicating by imaging with an infrared camera , an image of the point where the abnormal event occurs is taken.

[0036]

【Example】

Embodiment 1 FIG. FIG. 1 is a configuration diagram of the first embodiment. In FIG. 1, reference numeral 9 denotes a road, and reference numeral 10 denotes a vehicle. Numeral 11 denotes a target area for estimating a traffic condition, 12a to 12d denote image processing sensors for measuring the vehicle 10, and points 13a to 13d are representative measurement points, and a region indicated by a dotted line is a measurable region. Reference numerals 14a to 14d denote vehicle measuring means for measuring values such as vehicle speed, arrival time, traffic density, traffic congestion, and vehicle type from the outputs of the sensors 12a to 12d, and measuring the average speed, large vehicle mixing ratio, the number of vehicles, etc. from these values. Find the value. In addition, as the vehicle measuring means 14a to 14d, for example, the following moving object measuring device described in JP-A-5-6494 can be used. That is,
The existence area of the vehicle along the movement route of the vehicle is set in advance from the image data obtained by capturing the vehicle with the TV camera, and data indicating the vehicle is extracted from the image data in the existence area, and the data is displayed in time series as a data time series image. And accumulates continuously at a constant cycle. Then, the state quantity of the vehicle is measured by determining the presence of the vehicle moving in the preset existence area based on the obtained data time-series image.

15a to 15d are vehicle measuring means 14a to 1
A memory for storing the measurement values obtained in 4d for a predetermined time, 1
Numeral 6 denotes a predicting means which selects and reads necessary data from the memories 15a to 15d, and a time T after a predetermined time has elapsed from the time when the vehicle measuring means 14a to 14d corresponding to the sensors 12b to 12d measured the measured value. Or the sensor 12b
A predetermined time Δ from time T after a time determined from the measurement result of the passing time and the vehicle speed measured by 12d to 12d elapses
By t, the measurement values measured by the vehicle measurement units 14b to 14d corresponding to the sensors 12b to 12d are predicted. A memory 17 stores a predicted value predicted by the prediction means 16 for a predetermined time. Numeral 18 denotes a judging means, and the predicted values detected by the sensors 12b to 12d and the sensors 12b to
By comparing the measured value with the measurement value measured by 12d, the traffic state in the area where the sensors 12b and 12c cannot be measured in the target area 11 is estimated to detect an abnormal event.

Next, the operation will be described. In FIG. 1, sensors 12a to 12d monitor respective measurement points or measurement areas, and output a signal when the vehicle 10 passes or while the vehicle 10 passes. Next, vehicle measurement means 1
4a to 14d recognize the passage or presence of the vehicle 10 by processing the output signals of the sensors 12a to 12d, and determine the traffic flow parameters such as vehicle speed, arrival time, traffic density, traffic congestion, and vehicle type, and the average obtained from these values. The measured values such as the speed, the mixing ratio of large vehicles, the number of vehicles, etc.
Save to 5d. The memories 15a to 15d discard the stored measurement values after holding them for a certain period of time. The prediction unit 16 selects and reads out measurement values required for prediction from all or a part of the memories 15a to 15d at regular time intervals or continuously, and obtains a specific sensor necessary for estimating the traffic condition of the target area 11. The measurement values corresponding to 12b to 12d are predicted.
The predicted value is stored in the memory 17 for a certain period of time until it is used for the judgment by the judgment means 18 after a certain time. Finally,
The determination unit 18 compares the measured values stored in the memories 15b to 15d with the predicted values stored in the memory 17 of the prediction unit for each of the sensors 12b to 12d to determine the traffic condition of the target area 11. Estimate and detect abnormal events.

Embodiment 2 FIG. FIG. 2 is a configuration diagram of the second embodiment when two sensors are used for estimating a traffic state.
In FIG. 2, a predicting unit 16 uses a measurement value of the vehicle 10 measured by a sensor 12 b disposed on the upstream side of the traffic flow with respect to the traveling direction of the vehicle 10 to a sensor 12 c disposed on the downstream side of the traffic flow. Predict the arrival time. And
The judgment means 18 compares the predicted value to the sensor 12c with the measurement value actually measured by the sensor 12c,
The traffic condition in an area where the sensors 12b and 12c cannot be measured in the area between the points b and 12c is estimated.

FIG. 3 is a flowchart showing the operation of the prediction means 16. 2 and 3, in step 19, if the vehicle 12 is measured by the sensor 12b, the measured value is input from the memory 15b to the prediction means 16. In step 20, a predicted value measured by the downstream sensor 12c is predicted based on the measured value input from the memory 15b,
In step 21, the prediction result is stored in the memory 17, and the process is terminated (step 22). When measured vehicle speed V _b at time t ₁ by the sensor 12b, time t ₂₀ the velocity V _b is the vehicle to the sensor 12c when the constant is measured reaches becomes Equation (1). t ₂₀ = t ₁ + L / V _b (1) L in Expression (1) is the distance of the target area 11. Therefore, in the sensor 12c, the arrival time t ₂₀ -Δt to t ₂₀ + Δt
It is predicted that the vehicle of _Vb + ΔV will be measured from the vehicle speed _Vb− ΔV during the period. Here, Δt and ΔV are predetermined allowable fluctuation widths for the fluctuation of the speed and the measurement error, respectively.

FIG. 4 is a flowchart showing the operation of the judgment means 18. The following processing is performed at a predetermined time interval, for example, 1
Run every second. In FIG. 2 and FIG.
In step 3, the predicted values are read from the memory 17 in order from the earliest predicted arrival time. Next, at step 24, the current time t
And when performs processing in accordance the time t and the expected arrival time t ₂₀ -.DELTA.t a predicted value read current to the relationship between t ₂₀ + Delta] t. If the current time t is within the prediction range, in step 25, the measurement time is stored in the memory 15c from t ₂₀ −Δt to t.
₂₀ + Delta] t, and the vehicle speed to find the measured values such that V _b + [Delta] V from V _b - [Delta] V. If these measured values are present, it is determined as “normal” in step 26, and the predicted value is removed from the memory 17 (step 2).
7) Yes. If there is no measured value, nothing is performed, and the process returns to step 23. Next, when the time t exceeds t ₂₀ + Δt, it is determined that the vehicle corresponding to the measured value has not reached the predicted arrival time, and it is determined as “abnormal” in step 28, and the predicted value is stored in the memory 17. (Step 29). If the time t does not reach t ₂₀ −Δt, the processing at the time t is completed (step 3).
0).

As described above, in the first and second embodiments, the measured value measured by each vehicle measuring means is compared with the predicted value predicted by the predicting means, and the traffic in the unmeasurable area within the target area is compared. The state can be estimated.

In the first embodiment, the vehicle is read as a vehicle group consisting of a plurality of vehicles detected at a fixed time, and the presence of the vehicle is defined as the number of vehicles or the traffic density, and the vehicle speed is defined as the average vehicle speed. A vehicle group consisting of a plurality of vehicles measured during a certain period of time can be targeted.

Embodiment 3 FIG. FIG. 5 is a configuration diagram of the third embodiment, which estimates a traffic state based on a plurality of sensors. In FIG. 5, reference numerals 31 and 32 denote target areas for performing auxiliary estimation necessary for estimating the traffic state of the area 11. Reference numerals 33 to 35 denote prediction means,
31 and 32. 36 to 38 are primary judgment means,
They correspond to the areas 11, 31, and 32, respectively. Reference numeral 39 denotes a secondary determining means for determining the state of the area 11 from the results of the plurality of primary determining means 36 to 38. In addition, the prediction means 33-
For example, the method according to the first embodiment can be used as 35 and the primary determination means 36 to 38.

In the third embodiment, the prediction means 1 shown in the first embodiment is used.
6 and the judgment means 18 are realized by integrating the plurality of prediction means 16 and judgment means 18 shown in the second embodiment. That is, the data of the sensors 12c and 12d is predicted from the data of the sensors 12a and 12b. The secondary determining means 39 determines the traffic condition of the area 11 according to, for example, the following rules. First, the primary judgment means 36
When the area 11 is determined to be normal, the secondary determination means 39 determines that the area 11 is “normal”. Second, if the primary judging means 36 does not judge that the area 11 is normal, the judgment is suspended until the primary judging means 38 evaluates the corresponding prediction result. If it is determined that the sensor 12c is normal, it is determined that the sensor 12c has made an erroneous determination due to omission in detection, and the secondary determination unit 39 sets the status to "normal". Third, the primary judgment means 37 is normal and the primary judgment means 36
If it is determined that the sensor 12b is abnormal, it is determined that the sensor 12b has made an erroneous determination due to detection omission.
9 is “normal”. Fourth, in cases other than the above,
In the secondary judgment means 39, "abnormal" is determined.

As a judgment rule, in the second and third cases, "slightly normal" or the like is notified to the effect that the judgment cannot be completely made only by the information of the sensor. You can also prompt.
Lastly, in this example, prediction and determination using four sensors are performed, but prediction and determination using a plurality of sensors installed upstream or downstream of these may be performed.

Embodiment 4 FIG. FIG. 6 is a configuration diagram of the fourth embodiment. In FIG. 6, sensors 12a to 12c are sensors for measuring the congestion state at the measurement point, eliminating the congestion, and measuring the number of passing vehicles. The sensor 12b is a target of prediction, and estimates the traffic state of the estimation target area 11 from the prediction result. Vehicle counting and vehicle speed measurement may not be performed during traffic congestion because the distance between vehicles is small. However, it is often possible to detect the presence of a vehicle or traffic congestion even during traffic congestion. This is especially true when using image processing sensors.

The operation will be described. In FIG. 6, it is assumed that traffic congestion is first measured by the sensors 12b and 12c as a precondition. First traffic jam at time t ₁ at the sensor 12c is to be measured. Next, the sensor 12c starts measuring the number of vehicles. In the prediction means 16, the sensor 12
From traffic congestion is measured in c, for example, the time at which the vehicle of a predetermined number n which satisfies the equation (2) is measured that has passed through the sensor 12c and t _2, from t ₂ t ₂ + delta
It is predicted that t is the time at which congestion elimination is measured by the sensor 12b. N ₁ = N _c − (t ₂ −t ₁ ) × (n−n _i ) (2) N ₁ and N _c are the average number of vehicles in the area 11 at the time of congestion and non-congestion, respectively. _ni is the area 11 from the sensor 12b during the traffic jam.
The average number of vehicles flowing into the vehicle per unit time is preset.

The judging means 18 monitors the measurement of the congestion elimination for the sensor 12b until the time t ₂ + Δt predicted by the predicting means 16, and when the congestion elimination is measured by the predicted time, it is determined to be “normal”. If not, it is determined to be "abnormal" and the result of the determination is output immediately. Finally, Δ
Although a predetermined fixed value may be used as t, if there is no traffic congestion on the upstream side of the area 11, it is considered that the traffic congestion will be resolved earlier than the predicted time, so the sensor 12a measures the traffic congestion or the passage of the vehicle. If not, the value of Δt is reduced according to the value, and the determination is made early.

As described above, when the traffic congestion is measured by each of the upstream and downstream sensors, the time of the congestion elimination of the upstream sensor is predicted and then compared with the actually measured congestion elimination time. By performing the above, it is possible to estimate whether or not an abnormal event has occurred during the traffic jam, thereby preventing erroneous determination of the abnormal event in the early stage of traffic jam resolution.

Embodiment 5 FIG. FIG. 7 is a configuration diagram of the fifth embodiment. In FIG. 7, the sensors 12 are sequentially arranged from the upstream side of the traffic flow.
a, 12b and 12c are arranged. 40 is an estimating means, which is a sensor 12b based on the measured values of the sensors 12a to 12c.
Estimate the measured value for. FIG. 8 is an explanatory diagram for explaining the operation of FIG. FIG. 8A shows that the estimating means 40 uses the memory 1
The area 41 measured by the sensor 12c from time t to time t-Δt read from 5c is shown. FIG. 8 (b) shows an area 42 in which the estimated time at which the vehicle measured by the sensor 12c would have passed the sensor 12b has a time Δt and a width determined by the allowable error range when the vehicle speed is substantially constant. Means 18
Read from the memory 15b.

Next, the operation will be described. 7 and 8
In the estimation means 40, at time t, the sensor 1
The time t represented by the area 41 in the measured value for 2c
From the time t−Δt are stored in the memory 15
Read from c. A plurality of vehicles corresponding to the read measurement values will be referred to as a vehicle group b. Next, vehicle group b
Is estimated by the sensor 12b at a time corresponding to the speed as indicated by the area 42 and stored in the memory 15b. The area 42 is determined, for example, as follows. If the vehicle speed V _c is substantially constant, the estimated time t ₁₀ to the vehicle which has been measured is passed through the sensor 12b at time t ₂ by the sensor 12c is considered to be the formula (3). t ₁₀ = t ₂ −L / V _c (3) Therefore, the area 42 is an area obtained by adding a width determined by the time Δt and the allowable error range to the curve represented by the equation (3). Can decide.

The judging means 18 first reads out a plurality of measured values included in the area 42 estimated by the estimating means 40 from the memory 15b. A plurality of vehicles corresponding to the read measurement values will be referred to as a vehicle group a. Next, for each of the vehicle group a and the vehicle group b, an average value and a variance of the vehicle speed, the number of vehicles, and the mixing ratio of large vehicles at that time are obtained. Finally, the average value and the variance are compared, and if the difference is equal to or smaller than a threshold value of a predetermined value, it is determined to be “normal”, and otherwise, it is determined to be “abnormal”.

As described above, based on the measured values of the vehicle measuring means for a plurality of sensors, the measured values of the vehicle measuring means corresponding to one of the sensors before a predetermined time or a predetermined time depending on the measurement result has elapsed. Is estimated and compared with the measurement value of the vehicle measurement means corresponding to the other sensor before the predetermined time or the predetermined time depending on the measurement result elapses, to thereby determine the area not included in the area of the two sensors. Traffic conditions can be estimated. In addition, the average and variance of the measured vehicle speed, the number of vehicles, and the percentage of large vehicles mixed at that time are obtained and compared, and when the comparison value exceeds a set threshold value, it is included in the measurement area of the two sensors. It can be determined that an abnormal event has occurred in a non-existent area. Furthermore, the occurrence of an abnormal event in a region not included in the measurement region of the two sensors can be detected based on the degree of coincidence between the sensor located downstream and the arrival time of the vehicle measured within a certain time by the vehicle measurement unit.

Embodiment 6 FIG. FIG. 9 is a configuration diagram of the sixth embodiment. In FIG. 9, reference numerals 43 and 44 denote sensors 12 respectively.
a vehicle number reading means for reading a license plate when the vehicle 10 is measured in steps b and 12c;
a and 44a and vehicle number recognition means 43b and 44b for recognizing the license plate and outputting license plate information. The memories 15b and 15c store the sensor 12b,
The measured values from the vehicle measuring means 14b and 14c corresponding to 12c and the license plate information from the vehicle number reading means 43 and 44 are stored. If the sensors 12b and 12c are cameras and the resolution and installation conditions are sufficient to be used for license plate recognition, they can be shared as the reading sensors 43a and 44a of the vehicle number reading means 43 and 44. it can.

Next, the operation will be described. In FIG. 9, the vehicle number reading means 43 and 44 have sensors 12b and 12c.
The license plate is read when the vehicle is measured at. The prediction unit 16 predicts the range of the predicted arrival time to the sensor 12c from the speed and the measurement time for each vehicle measured by the sensor 12b, and stores the predicted arrival time range in the memory 17 together with the vehicle information. In the judgment means 18, the memory 1
7, a set of a plurality of measured values and license plate information whose current time has reached the range of the predicted arrival time is read out from the memory 15.
From c, a search is made for a vehicle with a matching car number. If the search is successful, it is determined that the vehicle has passed the area 11 without any abnormality, and the set of the measured value and the license plate information is deleted from the memory 17. Finally, a set of the measured value and the license plate information whose current time exceeds the predicted arrival time is searched from the memory 17, and if the search is successful, it is determined that an abnormal event has occurred in the area.

As described above, the first sensor on the upstream side of the traffic flow detects the vehicle, the vehicle number is detected by the first vehicle number reading means based on the detection signal of the first sensor, and the downstream of the traffic flow is detected. When the second vehicle number reading means detects the vehicle number by the detection signal of the second sensor on the side, and the second vehicle number reading means does not detect the vehicle number within the predicted arrival time calculated by the prediction means. In addition, it is possible to estimate the traffic condition in an area where both sensors cannot measure within the area between both sensors.

Embodiment 7 FIG. 10 to 14 are explanatory diagrams of the seventh embodiment. In the seventh embodiment, the prediction and the determination of the traffic state by the prediction means and the determination means are performed by fuzzy inference. 2 and FIGS. 10 to 14, the prediction means 16 evaluates the existence probability of the vehicle from the rules f ₀ , f ₁ , the measurement time, the traffic density, and the predicted inter-vehicle distance for evaluating the difference between the predicted value with respect to the speed and the arrival time. Rules f ₂ , f ₃ , and f ₄ , and rules f ₅ and f ₆ for evaluating the difference between the predicted value and the inter-vehicle distance between the preceding vehicle and the following vehicle.

FIG. 10 is an explanatory diagram of the rule f ₀ . FIG.
In FIG. 10 and FIG. 10, f ₀ is a two-dimensional membership function of time and vehicle speed measured by the sensor 12c for evaluating a small difference between the speed and the predicted value of the arrival time.
The closer the value called “certainty” is to 1.0, the smaller the deviation. While passing through the area 11, the sensor 1
If the vehicle speed _Vb measured at 2b is substantially constant,
The arrival time t ₂ is as predicted by the equation (1) in the first embodiment. Here, since "accelerated vehicle arrives early", attention is paid to the speed change in sensors 12b and 12c.
When the vehicle speed measured by the sensor 12c is in the greater than the speed V _b set to be faster arrival time t _2, f ₀ is lowered right around the set 45 of prediction value predicted by equation (1) Can be defined by a two-dimensional membership function having a spread of the certainty factor distribution. Further, FIG. 11 is an explanatory diagram of f _1. f ₁ is a two-dimensional membership functions to assess the abundance of deviation between the predicted value of the velocity and the arrival time, where the simple confidence, and be defined by the formula (4). f ₁ = 1.0−f ₀ (4)

FIGS. 12A, 12B and 12C are explanatory diagrams of rules f ₂ , f ₃ and f ₄ , respectively. 2 and 12, when the sensor 12c (1) predicts that a vehicle at the current time will arrive, (2) measures a traffic density sufficient to determine that a vehicle is present. ,
If (3) the inter-vehicle distance is predicted at that time by the prediction means 16, even if the arrival time or the vehicle speed of the vehicle is not accurately detected or measured, the vehicle can be predicted as expected. It is determined that there is a high possibility of reaching, and it is presumed that measurement omission has occurred due to the sensor. Therefore, the above linguistic expressions (1) to (3) are converted into rules f ₂ ,
f ₃ and f ₄ are represented by membership functions as shown in FIGS. 12 (a), 12 (b) and 12 (c).

[0061] FIG. 13 is an explanatory diagram of the rule f _5. Let db be the inter-vehicle distance between the vehicle a detected by the sensor 12b and another vehicle b running in front of it, and _let D _c0 be the predicted inter-vehicle distance when these are measured by the sensor 12c. If the speed difference of the vehicle speed V _b of the vehicle speed V _b and the vehicle b of the vehicle a same inter-vehicle distance D _b and D _c0, since the vehicle b is the inter-vehicle distance becomes shorter when higher than the vehicle a, V _a, V _b , _Db and sensor 12
The two-dimensional membership function between Dc and the inter-vehicle distance measured by _c has a distribution that is downward-sloping to the right as illustrated. here,
The closer the certainty factor is to 1.0, the closer the inter-vehicle distance is to the predicted inter-vehicle distance. Further, it can be defined in the same and which the rules f ₆ to assess the distance to the following vehicle.

FIG. 14 is an explanatory diagram of fuzzy inference used in the judgment means 18 of FIG. In FIG. 14, -1 means "abnormal", 1 means "normal", and 0 means "indefinite", and fuzzy inference is performed by applying a method generally called a min-max rule. 46 is a rule f ₀ , f ₅ , f
_This is a membership function representing “normal” corresponding to the logical sum of ₆ . 47 correspond to the rule f _1,
This is a membership function that indicates "abnormal". 4
8 corresponds to the logical sum of the rules _{f 2, f 3, f 4} , a membership function representing that "somewhat good". The height of each membership function is determined by the certainty factor of the corresponding rule, and the maximum value is indicated by a dotted line. Reference numeral 49 denotes a method of estimating the normality of the target area from the logical sum of the membership functions.

Next, the operation will be described. 2 and 1
In FIG. 0 to FIG. 14, the predicting means 16 reads the measured value of the vehicle measured by the sensor 12b from the memory 15b at regular time intervals, and calculates the distance between vehicles based on the measured values of the vehicles before and after the vehicle read from the memory 15b. Is calculated. Next, the sensor 1 is measured using the membership function of each of the rules f ₀ to f ₆ with respect to the measured value of each vehicle, the distance between vehicles, and the like.
The distribution of the value measured in 2c is predicted, and the prediction corresponding to each rule is stored in the memory 17. At this time, if you prepare standard distribution patterns corresponding to each rule,
Only the offset value of the pattern may be actually stored in the memory.

In the judgment means 18, first, from the memory 17,
The current time is selected and read out the prediction as true for the lower limit from the upper limit of the estimated time distribution of the rule f _0. Next, for the prediction of the distribution of each measurement value corresponding to the read rules f ₀ to f ₆ , the sensor 12 stored in the memory 15 c
By applying each measurement value to c, a certainty factor is obtained. next,
According to the min-max rule, the membership functions 46 to
The degree of certainty with respect to 48 is calculated, the membership function 49 is synthesized, and the normality or abnormal degree of the traffic condition in the area is estimated and output from the position of the center of gravity of the graphic of the logical sum 49.

As described above, the two sensors 12b, 1
A first arrival time at which the vehicle arrives at the sensor 12c is determined, including a variation in required time due to a speed change between 2c, and a sensor for the third speed and the second arrival time of the vehicle detected by the sensor 12c By comparing the degree of coincidence between the second speed measured by the sensor 12c and the first arrival time by the determination means, it is possible to estimate the traffic state in the unmeasurable region between the sensors 12b and 12c.

When a part of the measured value measured by the vehicle measuring means 14c corresponding to the sensor 12c on the downstream side of the traffic flow is not measured by the judging means 18, the vehicle measuring means 14c
The possibility of abnormality in the measured value of the sensor 12c can be estimated from the measured value measured by c and the inter-vehicle distance predicted by the predicting means 16.

Further, a prediction value of the inter-vehicle distance of the vehicle measured by the vehicle measurement means 14c corresponding to the sensor 12c on the downstream side of the traffic flow is calculated by the prediction means, and the speed and arrival time of the vehicle measured by the vehicle measurement means 14c are calculated. The control unit 18 evaluates the degree of coincidence between the two sensors 12b and 12c with the predicted value of the prediction unit 16 with respect to the inter-vehicle distance.
It is possible to estimate the traffic condition in an area where both sensors 12b and 12c cannot measure.

The seventh embodiment is an example of a combination of rules from f ₀ to f ₆ , and is not limited to this combination in order to obtain the effect of the present invention. Also, the method and shape of the set are not limited to the min-max rule and the triangle, and other methods having the same effect can be used. Furthermore, it is not always necessary to use a fuzzy set as a means for realizing each rule. For example, it is possible to use a neural network by learning a function corresponding to all or some rules from many measured values.

Embodiment 8 FIG. In the present embodiment, the possibility that a certain vehicle will catch up with a preceding vehicle and start following travel is predicted from the predicted value by the prediction means. In FIG.
The two vehicles a and b are sequentially and successively detected by the sensor 12b, and the predicted times of reaching the sensor 12c are t _a and t _b
And At this time, if t _b <t _a + T, it is determined that the vehicle b follows the vehicle a and follows the vehicle a, and the measured value is corrected as in the equations (5) and (6). . t _b = t _a + T (5) V _b = V _a (6) where T is a time delay when the vehicle is following.
V _a and V _b are the vehicle speeds of the vehicles a and b. In addition, in a case where there are enough lanes on the road and both so-called following and overtaking can occur, it is desirable to leave both the corrected value and the measured value before the correction.

Embodiment 9 FIG. FIG. 15 is a configuration diagram of the ninth embodiment. In FIG. 15, reference numeral 50 denotes a measurement value selection unit that selects and reads a measurement value applicable to the estimation result from the memory 15 b;
An evaluation unit 51 calculates an evaluation value of the correctness of the association.
Reference numeral 2 denotes an associating unit for comparing and sequentially correcting the evaluation values to determine a correct association; 53, a memory for temporarily storing the results of the association; An abnormal event determination unit that estimates and determines the occurrence of an event. The determination means 55 is constituted by 50 to 54.

Next, the operation will be described. FIG. 16 is an explanatory diagram illustrating operations of the evaluation unit 51 and the association unit 52. In FIG. 15, sensors 12a, 12b, and 12c are sequentially arranged from the upstream side of the traffic flow as in the fourth embodiment. The estimating means 40 is the same as that shown in FIG. 7 of the fifth embodiment, and outputs the measured values of the vehicle group b included in the range of the area 42 in FIG. Next, the operation will be described. 8, 15 and 16, the measurement value selection unit 50 selects and reads a measurement value applicable to the estimation result from the memory 15 b based on the estimation result by the estimation unit 40. The plurality of vehicles that have been read are referred to as a vehicle group a. The evaluation unit 51 associates, among the combinations of the respective vehicles of the vehicle group a with the respective vehicles of the vehicle group b read by the estimating means 40, with respect to the pair for which the evaluation value has not been calculated by the current time. Calculate the evaluation value in case of Associating unit 5
2 is a memory 53 for the calculated set of associations.
Is added to the memory 53 when there is no inconsistency as compared with the set of associations stored in the memory, and when there is an inconsistent set, the set with the lower evaluation value is deleted.

Next, a method of associating will be described with reference to FIG. The upper columns 56a to 56d correspond to the respective vehicles of the vehicle group a, and the horizontal axis represents the estimated arrival time at which the vehicle arrives at the sensor 12c by the equation (1). The lower group 57a-57d is a vehicle group b
, And the horizontal axis represents the measurement time. Line segment 58a
５８58e indicate the correspondence of the vehicles, and the shorter the length, the more correctly the correspondence. When a new vehicle measurement value 57a is input at a certain time, the evaluation unit 51 calculates the evaluation values of the associations 58a and 58b. The evaluation value is calculated, for example, by an evaluation function that increases when the difference between the calculated estimated arrival time and the measurement time and the difference in vehicle speed are small. Here, even when a plurality of associations occur as in 58c and 58d, for example, 56c is measured as two vehicles due to a large vehicle, or conversely, as one vehicle because the inter-vehicle distance is short. If it is known from the measured values that there is a possibility that the association will be made, it is determined that both associations are not inconsistent. Otherwise, the association with the lower evaluation value is deleted.

For example, an average value of the evaluation values is determined by the determining unit 54 for the association set for the predetermined time stored in the memory 53, and when the average value is equal to or less than a certain value, the traffic in the area 11 is determined. It is determined that the state is abnormal. As a method of the determination, it is better to make the determination based on the difference between the maximum value, the minimum value, and the variance, instead of the average value, depending on the property of the evaluation function to be used.

As described above, of the plurality of vehicles measured within a predetermined time with reference to the downstream sensor and the vehicles measured by the upstream sensor, the estimation unit estimates the plurality of vehicles measured. The plurality of vehicles that match the set values are associated with each other by the measured values and the evaluation values determined by the distance between the front and rear vehicles and the time interval. Then, it is possible to detect the occurrence of an abnormal event in a region where both sensors cannot measure within a region between the two sensors based on the respective evaluation values of the plurality of vehicles estimated by storing the evaluated values.

Embodiment 10 FIG. FIG. 17 is a configuration diagram of the tenth embodiment. In FIG. 17, reference numerals 59a to 59d denote sensors composed of an ITV camera which can be used for estimating a traffic condition and which can be controlled by zooming, and are installed on a pan head which can control attitude and move. Reference numerals 60a to 60d denote sensors composed of an ITV camera or an infrared camera. The angle of view is set so that the target area 11 of each of the sensors 59a to 59d has no blind spot. 61
Is a guide rail for moving the sensor 59a. 62
Reference numerals a to 62c denote control means for controlling the sensors 59a and 60a to 60d, which are selected by a selection means 67 described later to control the sensors 59a and 60a to 60d. 63 is a satellite equipped with a camera, camera control means and a communication device,
It monitors a place where the target area is not obstructed from the sky like a tunnel.

The vehicle measuring means 14a to 14d are the same as those in the first embodiment, and each measured value is obtained from the output of each sensor 59a to 59d. The memories 15a to 15d have functions similar to those of the first embodiment. Numerals 64a to 64c denote prediction means having the same functions as the prediction means 16 of the first embodiment. Reference numerals 65a to 65c denote determination means having the same functions as the determination means 18 of the first embodiment. Reference numeral 66 denotes a terrestrial communication device which transmits a photographing request to the artificial satellite 63 in accordance with a command from a selecting means 67 described later. 67 is a selection means, and each determination means 65
Each sensor 59 in the area where abnormality is detected by a to 65c
a to 59d, 60a to 60d and various photographing means of the artificial satellite 63 are selected, and the control means 62a to 62c and the communication device 66 are activated.

Next, the operation will be described. In FIG. 17, when the determination unit 65b determines that the target region 11 is abnormal, the selection unit 67 selects one or more of the control units 62a to 62c capable of photographing the region 11. Alternatively, the selection means 67 selects the artificial satellite 63 and transmits a photographing request through the communication device 66 together with the geographical coordinates of the area 11. The selected control means activates or controls the attitude of the sensors 59b, 59c or the sensors 60a, 60b corresponding to the position of the area 11. Each of the sensors 59b, 59c, 60a, and 60b sends the captured video to a control center or the like. Alternatively, when an artificial satellite is selected, the artificial satellite magnifies and shoots the requested coordinate point and transmits an image. The communication device 66 receives the transmitted video and transmits it to a control center or the like.

As described above, a plurality of ITV cameras or infrared cameras are installed in a region not included in the measurement region of the two sensors, or an artificial satellite system having a camera, its control means, and a communication system is used. When an abnormal event in the area is detected, by selecting a camera or system capable of photographing the point of occurrence of the abnormal event from these cameras and systems, and controlling to photograph the area, two sensors are measured. An image of an abnormal event that has occurred in a region that is not included in the region can be obtained.

[0079]

According to the first aspect of the present invention, by comparing the second measured value measured by the second vehicle measuring means with the predicted value by the judging means, the two sensors in the area between the two sensors are compared. It is possible to estimate the traffic condition in the unmeasurable area.

According to the second aspect of the present invention, the measured value measured by each vehicle measuring means and the predicted value are compared by the judging means, so that the measured value can be obtained within the area between the first sensor and the second sensor. It is possible to estimate a traffic condition in an area where the first sensor and the second sensor cannot measure.

According to the third aspect of the present invention, the second measurement value measured by each vehicle measurement means and the predicted value are compared by the judgment means, so that the measurement of both sensors is impossible in the area between the two sensors. The traffic condition in the area can be estimated.

According to a fourth aspect of the present invention, in the traffic flow measuring device according to the first or second aspect, by comparing the second measured value measured by each vehicle measuring means with the predicted value, It is possible to estimate a traffic state in an area where both sensors cannot measure within an area between the two sensors.

According to the fifth aspect of the present invention, the second measurement value measured by each vehicle measurement means and the estimated value are compared by the judgment means, so that the measurement of both sensors is impossible in the area between the two sensors. The traffic condition in the area can be estimated.

According to a sixth aspect of the present invention, in the traffic flow measuring device according to any one of the first to fifth aspects, the determining means determines whether or not the result of the comparison value exceeds the abnormal event determination threshold value. It can be determined that an abnormal event has occurred in an area where measurement is impossible for each sensor.

According to a seventh aspect of the present invention, in the traffic flow measuring device according to the third aspect, the degree of coincidence between the arrival time of the vehicle measured by the second sensor within a predetermined time by the determination means and the predicted arrival time is determined. By comparing the two, it is possible to estimate the traffic state in the unmeasurable region between the two sensors.

According to the eighth aspect of the present invention, the predicting means calculates the predicted arrival time measured by the second sensor from the vehicle speed and the measured time calculated from the detection signal of the first sensor,
When the second vehicle number reading means does not detect the vehicle number within the predicted arrival time, the determination means makes a determination, thereby detecting an abnormal event in an area between the two sensors that cannot be measured by the two sensors. The occurrence can be detected.

According to the ninth aspect, the degree of coincidence between the second speed and the first arrival time with respect to the third speed and the second arrival time of the vehicle detected by the second sensor is determined. By performing the comparison, it is possible to estimate the traffic state in the unmeasurable region between the two sensors.

According to the tenth aspect, part of the second measurement value measured by the second vehicle measurement means corresponding to the second sensor on the downstream side of the traffic flow is not measured by the judgment means. At this time, by comparing the measured value measured by the second vehicle measuring means with the inter-vehicle distance predicted by the predicting means, the second
It is possible to estimate the possibility of a measurement value abnormality of the sensor.

According to the eleventh aspect, the degree of coincidence of the vehicle speed, arrival time, and inter-vehicle distance measured by the second vehicle measuring means with each predicted value of the predicting means is evaluated by the judging means. Accordingly, it is possible to estimate a traffic state in an area where both sensors cannot measure within an area between both sensors.

According to the twelfth aspect of the present invention, the predicting means calculates the predicted speed and predicted arrival time of each vehicle measured by the second vehicle measuring means corresponding to the second sensor, and the judging means calculates the predicted arrival time of each vehicle. The predicted speed and predicted arrival time of the following vehicle are corrected based on the predicted speed and predicted arrival time at which the preceding vehicle reaches the second sensor, and the measured values measured by the second vehicle measuring means and the corrected predicted speed and predicted arrival time are corrected. By comparing with the above, it is possible to estimate a traffic state in an area where both sensors cannot measure within an area between both sensors.

According to the thirteenth aspect of the present invention, for a plurality of vehicles in which the measured value measured by the second sensor and the predicted measured value measured by the second sensor coincide with each other, the difference between the measured value and the predicted value is determined. By associating the relationship with the evaluation value, it is possible to estimate a traffic state in an area where both sensors cannot measure within an area between the two sensors.

According to the fourteenth aspect, from the first aspect,
The traffic flow measuring device according to claim 13.
Then, when an abnormal event is detected, by issuing a control command signal so as to measure the location of the occurrence of the abnormal event, the abnormal event imaging means can measure the location of the occurrence of the abnormal event.

According to a fifteenth aspect of the present invention, in the traffic flow measuring device according to the fourteenth aspect, the abnormal event imaging means includes a first sensor and a second sensor, an ITV camera, and a millimeter wave camera.
By using an infrared camera or a system such as an artificial satellite which can communicate by imaging with a camera, when an abnormal event is detected, it is possible to image the point where the abnormal event occurs.

[Brief description of the drawings]

FIG. 1 is a configuration diagram illustrating the invention of a first embodiment.

FIG. 2 is a configuration diagram when two sensors according to a second embodiment are used.

FIG. 3 is a flowchart showing an operation of a prediction unit 16;

FIG. 4 is a flowchart showing the operation of the judgment means 18.

FIG. 5 is a configuration diagram illustrating the invention of Embodiment 3.

FIG. 6 is a configuration diagram illustrating the invention of Embodiment 4.

FIG. 7 is a configuration diagram illustrating the invention of Embodiment 5.

FIG. 8 is an explanatory diagram of the operation of the sixth embodiment.

FIG. 9 is a configuration diagram illustrating the invention of Embodiment 7.

FIG. 10 is an explanatory diagram of the invention of the eighth embodiment.

FIG. 11 is an explanatory diagram of the invention of the eighth embodiment.

FIG. 12 is an explanatory diagram of the invention of the eighth embodiment.

FIG. 13 is an explanatory diagram of the invention of the eighth embodiment.

FIG. 14 is an explanatory diagram of the invention of the eighth embodiment.

FIG. 15 is a configuration diagram illustrating the invention of the ninth embodiment.

FIG. 16 is an explanatory diagram of the operation of the ninth embodiment.

FIG. 17 is a configuration diagram illustrating the invention of Embodiment 10.

FIG. 18 is a configuration diagram of a conventional traffic flow measuring device.

[Explanation of symbols]

9 roads, 10 vehicles, 11 areas, 12a to 12d,
59a to 59d, 60a to 60d, 63 sensors, 14
a to 14d vehicle measuring means, 16, 64a to 64c predicting means, 18, 55, 65a to 65c determining means, 40
Estimating means, 43, 44 Vehicle number reading means.

────────────────────────────────────────────────── ─── Continuation of the front page (56) References JP-A-6-195588 (JP, A) JP-A-6-76195 (JP, A) JP-B-45-3433 (JP, B1) JP-B-50- 38520 (JP, B1) (58) Field surveyed (Int. Cl. ⁷ , DB name) G08G 1/00-1/16

Claims (15)

(57) [Claims] 1. A first sensor and a second sensor which are arranged at predetermined intervals on a road to detect the presence of a vehicle on the road, and travel on the road based on outputs of the sensors. In a traffic flow measuring device including a first vehicle measuring unit and a second vehicle measuring unit corresponding to each of the sensors for detecting the vehicle and measuring a traffic flow parameter related to a traffic state, the first sensor and the second sensor Based on the first measurement value calculated by the corresponding first vehicle measurement means, the second vehicle corresponding to the second sensor after a predetermined time from the time when the first measurement value was measured. A prediction unit that calculates a predicted value of the traffic flow parameter measured by the measurement unit; and a second measurement value that is measured by the second vehicle measurement unit and the predicted value. Is impossible to measure with both sensors A traffic flow measuring device comprising: a judging means for estimating a traffic condition in an area. 2. A plurality of sensors after a first sensor, a second sensor, and a third sensor, which are arranged at predetermined intervals on a road to detect the presence of a vehicle on the road, In a traffic flow measuring device comprising a plurality of vehicle measuring means corresponding to each of the sensors for detecting the vehicle traveling on the road by output and measuring a traffic flow parameter relating to a traffic state, The first sensor is used as a reference, and based on the measurement value measured by the first vehicle measurement means corresponding to the first sensor, each of the sensors is determined after a predetermined time from the time when the measurement value is measured. The prediction means for calculating the predicted value of the traffic flow parameter measured by the corresponding vehicle measurement means, and the measured values measured by the vehicle measurement means are compared with the predicted values.
In the area between the first sensor and the second sensor,
A traffic flow measurement device comprising: a sensor for determining a traffic condition in an area where measurement by the second sensor is impossible and a second sensor. 3. The traffic flow measuring device according to claim 1, wherein the first sensor is disposed upstream of the traffic flow with respect to the traveling direction of the vehicle. . 4. The traffic flow measuring device according to claim 1, wherein the first sensor is arranged in a direction in which the vehicle travels.
Traffic flow measuring device, characterized in that in which was placed downstream of the traffic flow Te. 5. A plurality of sensors arranged at predetermined intervals on the road to detect the presence of vehicles on the road, and the output of each of the sensors detects the vehicle traveling on the road. In a traffic flow measuring device including a plurality of vehicle measuring means corresponding to each of the sensors for measuring a traffic flow parameter relating to a traffic condition, a first sensor on a downstream side of the traffic flow with respect to a traveling direction of the vehicle is referred to. And the first measured by the first vehicle detecting means corresponding to the first sensor.
Estimating means for calculating an estimated value of the traffic flow parameter measured by each of the vehicle detecting means corresponding to each of the sensors a predetermined time before the time when the first measured value is measured, based on the measured value of And comparing the second measured value measured by each of the vehicle measuring means and the estimated value at the time when the estimated value is calculated, in an area where the two sensors cannot measure within the area between the two sensors. A traffic flow measuring device comprising: a judgment means for estimating a traffic condition of a vehicle. 6. The traffic flow measuring device according to claim 1, wherein the determination means indicates a result of the comparison.
A traffic flow measuring device characterized in that it is determined that an abnormal event has occurred in an unmeasurable region of each sensor when the evaluation value exceeds a threshold value for determining an abnormal event. 7. The traffic flow measuring device according to claim 3, wherein the predicting means calculates a predicted arrival time of the vehicle measured by the downstream second sensor from the speed of the vehicle passing through the first sensor. Comparing the degree of coincidence between the arrival time of the vehicle and the predicted arrival time measured by the second sensor within a predetermined time by the determination means, and determines the traffic state in the unmeasurable area between the two sensors. A traffic flow measuring device characterized by being estimated. 8. A first sensor which is disposed on a road upstream of a traffic flow with respect to a traveling direction of a vehicle and detects the vehicle, and a vehicle number of the vehicle is determined by a detection signal of the first sensor. a first vehicle number reading means for detecting, under the traffic flow
A second sensor disposed on the upstream side for detecting the vehicle, second vehicle number reading means for detecting a vehicle number of the vehicle based on a detection signal of the second sensor, and detection of the first sensor Prediction means for calculating a predicted arrival time measured by the second sensor from the speed and measurement time of the vehicle calculated from the signal; and the second vehicle number reading means detects the vehicle number within the predicted arrival time A traffic flow measuring device comprising: a determination unit configured to detect occurrence of an abnormal event in a region where the sensors cannot be measured within a region between the sensors when the detection is not performed. 9. A first sensor disposed on a road upstream of a traffic flow with respect to a traveling direction of the vehicle to detect the vehicle, and a first sensor disposed downstream of the traffic flow to detect the vehicle. Calculating a second speed measured by the second sensor including a second sensor and a speed change between the two sensors based on the first speed of the vehicle measured by the first sensor; A first arrival time at which the vehicle reaches the second sensor is set to a required time calculated from the section length between the two sensors and the first speed, and a speed change between the two sensors of the vehicle is calculated. Prediction means for determining the change in the required time including the variation in the required time, and the second speed and the first arrival with respect to the third speed and the second arrival time of the vehicle detected by the second sensor. Comparison between time and time is not possible and measurement between both sensors is not possible A traffic flow measuring device provided with a judging means for estimating a traffic condition in an area of noh. 10. A first sensor and a second sensor which are arranged at predetermined intervals on a road to detect the presence of a vehicle on the road, and travel on the road based on outputs of the sensors. In a traffic flow measuring device including a first vehicle measuring means and a second vehicle measuring means corresponding to each of the sensors for detecting a traffic flow parameter relating to a traffic condition by detecting the vehicle, On the other hand, based on the first measurement value calculated by the first vehicle measurement means corresponding to the first sensor arranged on the upstream side of the traffic flow, from the time when the first measurement value is measured After a predetermined time, the second
Predicting means for calculating a predicted value of the traffic flow parameter measured by the second vehicle measuring means corresponding to the sensor of
When a part of the second measurement value measured by the second vehicle measurement means corresponding to the second sensor disposed downstream of the traffic flow is not measured, the second vehicle measurement means A traffic flow measurement device comprising: a determination unit configured to estimate a possibility of a measurement value abnormality of the second sensor based on a measured value measured and an inter-vehicle distance predicted by the prediction unit. 11. A first sensor and a second sensor which are arranged at predetermined intervals on a road to detect the presence of a vehicle on the road, and travel on the road based on outputs of the sensors. In a traffic flow measuring device including a first vehicle measuring means and a second vehicle measuring means corresponding to each of the sensors for detecting a traffic flow parameter relating to a traffic condition by detecting the vehicle, On the other hand, from the speed of the vehicle and the inter-vehicle distance measured by the first vehicle measuring means corresponding to the first sensor arranged on the upstream side of the traffic flow,
A predicting means for calculating a predicted value of an inter-vehicle distance of the vehicle measured by the second vehicle measuring means corresponding to the second sensor disposed on the downstream side of the traffic flow; and The measured vehicle speed, arrival time, and inter-vehicle distance are evaluated for the degree of coincidence with the respective predicted values of the predicting means, and traffic in an area where the two sensors cannot be measured within an area between the two sensors. A traffic flow measuring device comprising: a judgment unit for estimating a state. 12. A first sensor and a second sensor which are arranged at predetermined intervals on a road to detect the presence of a vehicle on the road, and travel on the road based on outputs of these sensors. In a traffic flow measuring device including a first vehicle measuring means and a second vehicle measuring means corresponding to each of the sensors for detecting a traffic flow parameter relating to a traffic condition by detecting the vehicle, On the other hand, based on the speed and arrival time of the two vehicles calculated by the first vehicle measuring means corresponding to the first sensor disposed on the upstream side of the traffic flow, the second sensor is associated with the second sensor. The second
Prediction means for calculating a predicted speed and a predicted arrival time of each of the vehicles measured by the vehicle measurement means, and a prediction of a following vehicle based on the predicted speed and the predicted arrival time of the preceding vehicle of each of the vehicles reaching the second sensor. Correcting the speed and the predicted arrival time, comparing the measured value measured by the second vehicle measuring means with the corrected predicted speed and the predicted arrival time,
A traffic flow measuring device comprising: a judging means for estimating a traffic state in an area where the two sensors cannot measure, in an area between the two sensors. 13. A first sensor and a second sensor which are arranged at predetermined intervals on a road to detect the presence of a vehicle on the road, and travel on the road based on the outputs of the sensors. In a traffic flow measuring device including a first vehicle measuring means and a second vehicle measuring means corresponding to each of the sensors for detecting a traffic flow parameter relating to a traffic condition by detecting the vehicle, On the other hand, of the plurality of vehicles measured by the first sensor arranged on the upstream side of the traffic flow, the predicted measurement value measured by the second sensor arranged on the downstream side of the traffic flow within a predetermined time is calculated. Predicting means for calculating, and relating the plurality of vehicles whose measured values measured by the second sensor match the predicted measured values, the relationship between the measured values and the predicted values with an evaluation value, Based on the values, A traffic flow measurement device comprising: a determination unit configured to estimate a traffic state in a region where the two sensors cannot measure, in a region between sensors. 14. The method according to claim 1, wherein
In the traffic flow measurement device described in the above, the judgment means detects an abnormal event.
When detecting, measure the point of occurrence of the above abnormal event
Issues a control command signal, traffic flow measurement apparatus and a abnormal event imaging means for capturing an occurrence point of the abnormal event by the control command signal of the determination unit. 15. The traffic flow measuring device according to claim 14, wherein the abnormal event imaging means is capable of communicating by imaging with the first sensor and the second sensor, an ITV camera, a millimeter wave camera, an infrared camera, or a camera. A traffic flow measuring device, which is a sensor for an artificial satellite system or the like.