JPS6252359B2 - - Google Patents

Description

[Detailed description of the invention] The present invention relates to a vehicle sensing method.

Currently, various vehicle sensing devices such as loop type, radar type, and ultrasonic type have been developed, and it has become possible to measure vehicle speed and detect vehicles with high precision. However, as road information and the like are increasingly collected and utilized, a device having a single housing for sensing is by no means desirable. Further, the fact that two circuits, a receiving circuit and a transmitting circuit, are required is not desirable in terms of cost and maintenance.

The present invention focuses on these points, and by applying a correlation method to vehicle sensing and vehicle speed measurement, it is easier to minimize maintenance and costs, and by performing this processing using software, it is easier to integrate with other systems. This provides a device that facilitates joining and is highly expandable.

An example of this will be explained below with reference to the drawings. Reference numerals 1A and 1B are two microphones with strong directivity installed at a certain distance along the running direction, and receive the noise when a vehicle passes by. do. 2A and 2B are amplifiers, 3A and 3B are sample circuits that sample only a few milliseconds of data from the rise of the output, and 4 is a signal processing circuit that extracts the vehicle detection output from the correlation between the outputs of sample circuits 3A and 3B. It is.

In this case, as shown in FIG. 1, microphones 1A and 1B are delayed in the rise of the vehicle noise signal by the distance between microphones 1A and 1B. Fourth
xB and xA in the figure indicate the state. This signal is amplified by amplifiers 2A and 2B of the receiving circuit, respectively, and only data of several milliseconds from the rise of the noise level is sampled by sample circuits 3A and 3B. This is shown in xB' and xA' in Figure 5.

Next, the process proceeds to the process of calculating the correlation function in the signal processing circuit 4, but this method can be performed with analog data itself or with conversion to digital data. Here, we will discuss the case of digital sampling. Digital sample data xB′,
When we calculate the cross-correlation function of xA′, we obtain a result like φB′A′. That is, an output pulse is obtained only when microphones 1B and 1A detect signals generated from exactly the same object. Even if the background noise is high, it can be removed by taking the correlation, and even if there is noise other than cars in the microphones 1A and 1B at the same time, if the correlation is taken, a pulse will be generated at τ = 0 (measurement start time). This can be easily removed.

The correlation function extracts and compares (correlates) the power spectra of the signals, and generates a pulse only if the signals are from the same vehicle. Therefore, even if there are vehicles following each other as shown in FIG. 3, it is possible to identify them.

Furthermore, when calculating the cross-correlation function of the signals xB' and xA', the time until the pulse is generated is the time it takes to pass between the microphones 1B and 1A, and the velocity can be calculated from this.

As is clear from the above embodiment, the present invention provides two microphones at regular intervals in one lane in the driving direction, and measures the cross-correlation function from the frequency, power level, and time elements of the signals detected by the two microphones. , when noise other than cars enters, τ=
Since the pulse generated at 0 is removed and the output is obtained when there is a correlation, even if two or more vehicles pass between the microphones, each vehicle will be able to respond, and it will be possible to eliminate noise such as city noise. It also has the effect of being resistant to noise.

[Brief explanation of the drawing]

FIG. 1 is a layout diagram of a vehicle sensing microphone according to the present invention, FIG. 2 is a block diagram of a vehicle sensing method according to an embodiment of the present invention, FIG. 3 is a diagram showing the positional relationship of the vehicle with respect to the microphone, and FIG. 4 ~FIG. 6 is an output waveform diagram of each part in FIG. 2. 1A, 1B...Microphone, 2A, 2B...
Amplifier, 3A, 3B...Sample circuit, 4...Signal processing circuit.

Claims (1)

[Claims] 1 Two microphones are installed at regular intervals in one lane in the driving direction, and the cross-correlation function is measured from the frequency, power level, and time elements of the signals detected by these two microphones. A vehicle sensing method characterized in that a pulse generated at the start of measurement is removed and an output is obtained when there is a correlation.