JPH1090413A - Vehicle position detecting device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To accurately detect the relative position of a vehicle by removing noise from the detected signals of markers provided on the surface of a road. SOLUTION: A linear light emitter-receiver 2 is provided below a vehicle 1 so as to receive reflected signals from lane markers 3 provided on the surface of a road. When a plurality of intensity peaks exist in the detected signals, whether or not intensity peaks exist at nearly same positions at the preceding sampling time and the succeeding sampling time and, when no peak exists, the detected signals are discriminated as noise. Even when intensity peaks exist continuously in time, the detected signals are discriminated as noise when the intensity widths of the peaks remarkably different from a reference width. Even when only one intensity peak is discriminated, the signal is discriminated as noise when the positional change of the peak with time is such variation that is impossible from the running speed of the vehicle.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

The present invention relates to a vehicle position detecting device,
In particular, the present invention relates to an apparatus for detecting a vehicle position by detecting a marking means (marker) provided on a road surface.

[0002]

2. Description of the Related Art Hitherto, there has been proposed an apparatus for detecting a marker such as a light reflecting tape provided on a road surface by a detector mounted on a vehicle and specifying a position of the vehicle on the road surface.

For example, Japanese Unexamined Utility Model Publication No. 1-109910 discloses a configuration in which light is projected onto a reflective tape laid on a road surface and the reflected light is received to calculate a displacement from a reference guide position. I have.

[0004]

However, on an actual road, a light-reflective material (for example, a water film or glass) other than the light-reflective tape may be present in the vicinity of the light-reflective tape. There is a problem that light is erroneously detected as reflected light from the original light reflecting tape.

Further, even when there is no such light-reflective substance, the distance between the photodetector of the vehicle and the light-reflecting tape changes due to the presence of unevenness on the actual road surface, which causes erroneous detection. was there.

SUMMARY OF THE INVENTION The present invention has been made in view of the above-mentioned problems of the prior art, and has as its object to eliminate the influence of noise and suppress erroneous detection due to unevenness of a road surface to accurately detect a vehicle position. It is an object of the present invention to provide a device capable of performing such operations.

[0007]

In order to achieve the above object, a first invention is to provide a road marking means provided on a road surface,
It is characterized by comprising a vehicle-mounted detecting means for detecting the lane marking means at a predetermined sampling time, and a processing means for removing noise included in a signal detected by the detecting means based on a vehicle speed and the sampling time. When the relative position between the vehicle and the lane marking means changes, the detection signal also changes, but the amount of change in the detection signal within the sampling time is so large as to be physically impossible considering the vehicle speed of the vehicle. In this case, the detection signal can be determined to be noise that is not a regular detection signal by the lane marking means.

In a second aspect of the present invention, a road marking means provided on a road surface, a vehicle-mounted detecting means for detecting the road marking means at a predetermined sampling time, and a detecting means based on a pitch amount of the vehicle. And processing means for removing noise contained in the signal. When the size of the lane marking means is constant, the magnitude of the detection signal is also constant. Therefore, in principle, by looking at the intensity width of the detection signal, it can be determined whether it is the detection signal of the lane marking means or the noise, but the distance between the detection means and the lane marking means has changed. In such a case, the signal intensity also changes, so that noise cannot be accurately removed simply by monitoring the intensity width of the detection signal. Therefore, in the present invention, noise is reliably removed by considering the pitch amount of the vehicle.

[0009]

Embodiments of the present invention will be described below with reference to the drawings.

FIG. 1 shows a conceptual configuration of the present embodiment. An optical transceiver 2 is provided before and after the vehicle 1, and transmits light at a predetermined timing toward a lower portion of the vehicle,
The reflected light is received. Note that transmission and reception can use visible light or infrared light. On the other hand, a lane marker 3 is provided on the road surface along the traveling direction of the runway. The lane marker 3 is formed of a so-called recursive reflective material, and reflects incident light in the incident direction. Therefore,
When the lane marker 3 exists in the beam area of the optical transceiver 2, light emitted from the optical transceiver 2 is reflected by the lane marker 3, and the reflected light is received. On the other hand, when the vehicle 1 travels at a position away from the lane marker 3 and the lane marker 3 deviates from the beam area of the optical transceiver 2, the optical transceiver 2 cannot receive the reflected light from the lane marker 3. Therefore, the relative position of the vehicle 1 to the lane marker 3 can be detected based on the intensity of the reflected light.

FIG. 2 is a block diagram showing the configuration of this embodiment. A signal detected by a line sensor (hereinafter abbreviated as line sensor 2) in the optical transceiver 2 is supplied to an electronic control unit ECU4 having a microcomputer and a memory, and the detection signals are sequentially stored in the memory. The memory needs to have a capacity sufficient to store data for at least three consecutive sampling times. As the line sensor 2, for example, a CCD sensor or the like can be used. FIG. 3 schematically shows a detection area of the line sensor 2, and an arrow indicated by a in FIG. 3 has a length sufficient to cross the lane marker 3 in the detection area. Therefore, when the lane marker 3 is present in the detection area of the line sensor 2, a signal whose signal intensity peaks at a position directly above the lane marker 3 in the line sensor 2 and becomes almost zero at other positions, that is, an intensity A signal having a single peak is obtained (an ideal state without noise).

In addition to the line sensor 2, a vehicle speed sensor 5 for detecting the speed of the vehicle and a pitch amount sensor 6 for detecting the pitch of the vehicle are provided, and supply detection signals to the ECU 4. The ECU 4 removes a noise component included in the detection signal from the line sensor 2 based on the input vehicle speed signal, the sampling timing at the line sensor 2 and the pitch amount, and extracts only the reflection signal from the lane marker 3. Detect the relative position of the vehicle. Then, the corrected steering amount is determined based on the detected relative position, and is supplied to the steering actuator 7. Since the line sensors 2 are provided before and after the vehicle, the ECU 4 can simultaneously detect the attitude angle of the vehicle from the relative positions of the front and rear.

FIG. 4 shows a processing flowchart of the ECU 4. First, the ECU 4 determines whether a plurality of intensity peaks exist in the detection signal supplied from the line sensor 2 (S101). In this determination, since only one lane marker 3 exists in the vehicle width direction on the road surface, in principle, the detection signal from the line sensor 2 has only one intensity peak value, and in the case of a plurality, the noise Is considered to be mixed. If there is no plurality of intensity peaks, that is, if only a single intensity peak is detected, then a time-series continuous detection signal is read from the memory, and the amount of change in the intensity peak position is determined by a predetermined threshold value. Th1 is compared (S102). This is because even if the intensity peak value is single, if the change amount is a change that cannot be determined from the current running state of the vehicle, the signal is considered to be noise. Hereinafter, this S10
The determination process 2 will be described in detail.

FIG. 5 shows an example of the detection signal stored in the memory of the ECU 4. In the figure, the horizontal axis indicates the position of the line sensor 2, and (a) indicates the time tn.
(B) shows the time tn +
(C) is a detection signal at time tn + 2. At time tn, the intensity peak 100 exists at a certain position, at time tn + 1, the intensity peak 102 exists at another position, and at time tn + 2, the intensity peak 100 exists again at almost the same position as time tn. And The change amount of the intensity peak is x
It is the amount indicated by. Assuming that both the intensity peaks 100 and 102 indicate the reflected signal from the normal lane marker 3, the vehicle has moved by x during the time (tn + 1-tn). Therefore, when the vehicle speed at the time tn to tn + 1 is low and such movement is physically impossible, the intensity peak 102 is not a reflected signal from the normal lane marker 3 but a certain noise (for example, a road surface). (A reflection signal from a puddle).

FIG. 6 shows a specific relationship between the amount of change, the vehicle speed, and the sampling time. In the figure, the horizontal axis is the sampling time (ms), and the vertical axis is the change amount X (c
m). The broken line in (I), the solid line in (II), and the one-dot chain line in (III) indicate vehicle speeds of 30 km / h, 60 km / h, 9 km, respectively.
This is the threshold value th1 at 0 km / h. For example, when the vehicle speed is 30 km / h and the sampling time is 10 ms, the threshold value th1 is about 5 cm, and when the actually detected change amount is within 5 cm, it is possible considering the state of the vehicle. Since it is the movement amount, it can be determined that the signal is a reflection signal from the normal lane marker 3, but the actual change amount is 5c.
If it exceeds m, the amount of movement is impossible due to the state of the vehicle, so it is determined to be noise. That is, in FIG. 6, areas below the boundaries indicated by (I), (II), and (III) are areas where normal signals are determined, and areas above are boundaries where noise is determined.

As described above, the magnitude of the change is compared with the threshold in S102, and if the change is within the threshold th1, it is determined that the signal is normal and the relative position of the vehicle 1 is calculated ( S103). On the other hand, if the threshold value th1 is exceeded, the noise is determined and the intensity peak 102 is removed (S105). Note that the relative position of the vehicle can be specifically detected from the position of the intensity peak. That is, if the reflection signal is received at the center of the line sensor 2 when the vehicle 1 passes right above the lane marker 3, the relative displacement of the vehicle 1 with respect to the lane marker 3 corresponds to the deviation amount from the center of the intensity peak as it is. Because you do. However, as the intensity peak position, there is a method in which the position is determined based on the edge portion, a method in which the position is determined based on the midpoint between the rising edge and the falling edge, and any of them may be used. When the intensity peak 102 is determined to be noise, the detection signal at the time tn + 1 cannot be used after all, but in this case, the times tn and tn + 2 before and after the detection signal are used.
It is conceivable to interpolate the signal at the time tn + 1 with the average value of the intensity peak positions at. Of course, another method may be used as the interpolation method, or the steering may be maintained as it is without calculating the relative position at time tn + 1.

On the other hand, if YES in S101, that is, if there are a plurality of intensity peaks in the detection signal, it is necessary to distinguish between a normal signal and noise because one of them is considered to be noise. Therefore, the ECU 4 reads out the detection signals before and after one sampling timing for each intensity peak, and determines whether or not the intensity peak exists at that position (S10).
4). This determination is made when there are multiple intensity peaks.
This is based on the fact that the intensity peaks that exist continuously in time series are more certain, and the intensity peaks that suddenly exist are likely to be noise.

FIG. 7 shows another example of the detection signal stored in the memory of the ECU 4. In the figure, the horizontal axis indicates the position of the line sensor 2, and (a) indicates the time t.
n is the detection signal, and (b) is the time tn that is the target of processing.
(C) is the detection signal at time tn + 2.
At time tn, the intensity peak 100 exists at a certain position, and at time tn + 1, the intensity peak 104
Exists at another position, and at time tn + 2, the intensity peak 100
At the same position as time tn again.
In this case, since the intensity peak 100 exists at almost the same position at the time tn and the time tn + 2, it is highly likely that the intensity peak 100 is a reflection signal from the normal lane marker 3, but the intensity peak 104 is the time tn and the time tn + Time tn + 1 which does not exist in 2
Since there is only an abrupt presence only in, the possibility of noise is high. Therefore, the ECU 4 determines that the intensity peak 104 is noise (S10).
5). Although the intensity peak 100 is likely to be a legitimate signal, in the present embodiment, the following processing is further performed to ensure its reliability.

That is, the ECU 4 determines in step S104 that YE
If S, that is, if the intensity peaks are successively present at substantially the same position in time, then the width of the intensity peak is determined (S106). Specifically, the determination is made as to whether the absolute value of the difference (Lref-L) between the reference width Lref determined according to the pitch amount of the vehicle and the width L of the detected intensity peak is within the threshold th2. It depends on whether or not. Since the width of the lane marker 3 is known, the width L0 of the intensity peak, which is the reflected signal from the lane marker 3, also has a constant value. Therefore, if the width of the intensity peak is significantly different from L0, it is noise that is not a reflected signal from the lane marker 3. However, since the intensity of the reflected signal from the lane marker 3 depends on the distance between the line sensor 2 and the road surface, L0 itself changes under the influence of the vehicle pitch. Therefore, the ECU 4 calibrates L0 based on the detection signal from the pitch amount sensor 6, calculates the reference width Lref, and calculates the difference from the detected intensity peak width.

FIG. 8 shows an example in which a plurality of intensity peaks are continuously present in time. (A) is a detection signal at times tn, tn + 1, and tn + 2, and it is assumed that the intensity peaks 100, 106, and 108 exist at any time. On the other hand, (b) shows the reference intensity peak ref of the lane marker 3 calculated from the pitch amount of the vehicle 1.
(It is not necessary to actually create such a waveform, and the ECU 4 only needs to calculate only the reference width Lref of the reference intensity peak ref.). The ECU 4 calculates a plurality of intensity peak widths in (a), and calculates a difference from the reference width Lref. The width of the intensity peak 100 is almost the same as the reference width, and the difference is the threshold th2.
As described below, the width of the intensity peak 106 is significantly smaller than the reference width Lref, and the width of the intensity peak 108 is
Since the difference is significantly larger than f, the absolute value of the difference exceeds the threshold value th2. Therefore, the intensity peaks 106 and 108 are both determined to be noise (S105), and the intensity peak 10 out of the plurality of intensity peaks is determined.
It is determined that only 0 is a normal signal (S107).
In addition, as a case where a plurality of intensity peaks exist consecutively in time, for example, a case where a metal rod is present as a falling object on the road substantially in parallel with the lane marker 3 is considered. As long as are not the same, both can be clearly distinguished in this embodiment.

As described above, in the present embodiment, whether the signal is normal or noise is determined based on the vehicle speed, the sampling time, and the pitch amount of the vehicle, and the signal determined as noise is not used. Can be detected.

In this embodiment, the case where one lane marker 3 exists in the width direction of the road surface is shown. However, when a plurality of lane markers 3 are laid, the determination processing in S101 needs to be changed according to the number of laid lanes. Needless to say.

In this embodiment, the amount of change in the intensity peak is evaluated based on the vehicle speed and the sampling time. However, the amount may be evaluated in consideration of the curvature of the road surface and the steering amount. For example, when the traveling road is a curved road, the relative displacement of the vehicle with respect to the lane marker 3 can greatly change even at the same vehicle speed as compared with a straight road, and therefore, the threshold value th for the change amount determination
It is conceivable to increase 1 according to the curvature.

In the present embodiment, when the intensity peak is singular, whether the noise is noise is determined by comparing the amount of change with a predetermined threshold th1 (S101,
S102) Even if YES is determined in S102, the process of S106, that is, the process of determining whether or not the difference value between the width of the intensity peak and the reference width is within threshold value th2 may be added. Good. By this processing, for example, even when the metal bar exists parallel to the lane marker and overlaps the lane marker, it is possible to prevent erroneous calculation of the relative position.

Further, in this embodiment, the case where the relative position of the vehicle is detected using the light reflection type lane marker has been described. However, the gist of the present invention is that the noise included in the detection signal is detected by the vehicle speed, the sampling timing, and the pitch amount. Since it is distinguished by using the same, it can be applied to a magnetic field detection type or a radio wave type such as a millimeter wave.

[0026]

As described above, according to the present invention, it is possible to remove the influence of noise, suppress erroneous detection due to irregularities on the road surface, and detect the vehicle position with high accuracy.

[Brief description of the drawings]

FIG. 1 is a conceptual configuration diagram of an embodiment of the present invention.

FIG. 2 is a configuration block diagram of the embodiment.

FIG. 3 is an explanatory diagram showing a detection range of the line sensor of the embodiment.

FIG. 4 is a processing flowchart of an ECU of the embodiment.

FIG. 5 is an explanatory diagram of the same embodiment when the intensity peak is singular.

FIG. 6 is a graph showing a relationship between a vehicle speed, a sampling time, and an amount of change in the embodiment.

FIG. 7 is an explanatory diagram of the same embodiment when there are a plurality of intensity peaks.

FIG. 8 is an explanatory diagram of a reference peak used for noise determination when there are a plurality of intensity peaks according to the embodiment.

[Explanation of symbols]

Reference Signs List 1 vehicle, 2 optical transceiver (line sensor) 3 lane marker, 4 ECU (electronic control device), 5 vehicle speed sensor, 6 pitch sensor, 7 steering actuator.

Claims (2)

[Claims] 1. A lane marking provided on a road surface, a vehicle-mounted detector for detecting the lane marking at a predetermined sampling time, and a signal detected by the detector based on a vehicle speed and the sampling time. A vehicle position detecting device, comprising: processing means for removing noise. 2. A lane marking means provided on a road surface, a vehicle-mounted detecting means for detecting the lane marking means at a predetermined sampling time, and a noise included in a signal detected by the detecting means based on a pitch amount of the vehicle. A vehicle position detecting device, comprising: