Sensor unit, system and method for detecting magnetic marker

Abstract

A vehicle system (1S) is provided with: a sensor unit (1) which is provided with magnetic sensors (Cn) for detecting magnetic markers, and also a contactless displacement sensor (15) which measures displacement relative to the road surface where the magnetic marker is installed; a processing circuit which executes detection processing for detecting magnetic markers by performing processing on a magnetic measured value measured by multiple magnetic sensors (Cn); and a switching circuit (103) which switches detection processing depending on the displacement measured value measured by the contactless displacement sensor (15). It is possible to highly reliably detect magnetic markers by switching detection processing depending on the displacement measured value measured by the contactless displacement sensor (15).

Description

Sensor unit and system

[0001] The present invention relates to a sensor unit and system for detecting magnetic markers.

[0002] Conventionally, a magnetic marker detection system for a vehicle that uses magnetic markers laid on a road for vehicle control has been known (see, for example, Patent Document 1). Such a magnetic marker detection system is realized by combining magnetic markers laid on the road with a vehicle-side magnetic marker detection device including a magnetic sensor. For example, if a vehicle can detect magnetic markers laid along lanes on a road, various driving assistance functions such as automatic steering control and lane departure warning can be realized.

[0003] Japanese Patent Application Laid-Open No. 2005-202478

[0004] However, the conventional magnetic marker detection system has the following problem. That is, there is a risk that the reliability of magnetic marker detection may be impaired due to external magnetic disturbances acting on the magnetic sensor. For example, reinforced concrete bridges and tunnels that form roads have steel reinforcing plates and reinforcing bars strung throughout their interiors to ensure structural strength. Because iron is a magnetic material, structures such as bridges and tunnels can potentially be a source of large magnetic fields. While the residual magnetization of steel materials such as reinforcing bars is very small compared to that of magnets, the large volume of bridges and tunnels can generate magnetic fields that exceed the geomagnetic field, or these structures can act like a yoke to collect the geomagnetic field and generate relatively large magnetic fields. For example, the magnetic fields of various magnetic sources present on roads, such as bridges and tunnels, are one factor that reduces the reliability of magnetic marker detection.

[0005] The present invention has been made in consideration of the above-mentioned problems of the conventional art, and aims to provide a sensor unit and system that are useful for detecting magnetic markers with high reliability.

[0006] One aspect of the present invention is a sensor unit that is attached to a moving body to detect a magnetic marker, the sensor unit comprising one or more magnetic sensors that measure the magnetic field acting from the magnetic marker, and one or more non-contact displacement sensors that measure the displacement relative to the surface on which the magnetic marker is arranged.

[0007] One aspect of the present invention is a system for a mobile body to detect a magnetic marker, in which a sensor unit is attached to the mobile body and includes one or more magnetic sensors that measure the magnetic field acting from the magnetic marker, and one or more non-contact displacement sensors that measure the displacement relative to the surface on which the magnetic marker is arranged.

[0008] One of the technical features of the present invention is a sensor unit that incorporates a non-contact displacement sensor in addition to a magnetic sensor. This sensor unit can measure the magnetic field generated by a magnetic marker and can also measure the displacement relative to the surface on which the magnetic marker is disposed using the non-contact displacement sensor.

[0009] In the configuration of the present invention, when detecting a magnetic marker, the measurement results of the displacement relative to the surface on which the magnetic marker is disposed can be used. The measurement results of the displacement relative to the surface on which the magnetic marker is disposed are effective in eliminating external magnetic disturbances other than those originating from the magnetic marker, and are useful for improving the detection reliability when detecting the magnetic marker.

[0010] 1 is a front view of a vehicle equipped with a sensor unit. 2 is an explanatory diagram showing a vehicle traveling along a lane on which magnetic markers are installed. 3 is a block diagram showing the configuration of the sensor unit and main circuit. 4 is a flow diagram showing the flow of displacement measurement using a non-contact displacement sensor. 5 is an explanatory diagram showing the process of comparing a previous image with a latest image. 6 is a flow diagram showing the flow of marker detection processing. 7 is an explanatory diagram showing the change over time in the distribution waveform of magnetism in the vehicle width direction when passing a magnetic marker. 8 is an explanatory diagram showing the change over time in the distribution waveform of magnetic gradient in the vehicle width direction when passing a magnetic marker. 9 is an explanatory diagram of filter processing. 10 is a block diagram showing another sensor unit.

[0011] The present invention will be described in detail below with reference to the following examples. (Example 1) This example relates to an in-vehicle sensor unit 1 that detects magnetic markers 10 installed on a road. The details of this example will be described with reference to FIGS. 1 to 10.

[0012] This example illustrates a vehicle system 1S (FIGS. 1 and 2) that uses magnetic markers 10 to achieve driving assistance control for a vehicle 5. A vehicle 5 equipped with a sensor unit 1 can detect the magnetic markers 10 while traveling on a road on which the magnetic markers 10 are arranged. For example, the magnetic markers 10 are arranged along the center of a lane 100, which is the vehicle's driving area on the road. The vehicle system 1S is an example of a system that uses the magnetic markers 10 arranged along the lane 100 to achieve driving assistance control, such as lane keeping control.

[0013] The vehicle 5 constituting the vehicle system 1S is an example of a moving body. The vehicle 5 includes a sensor unit 1 for detecting a magnetic marker 10, a main circuit 20 for executing marker detection processing, a vehicle ECU 30 for executing driving assistance control, etc. Below, the magnetic marker 10 will be explained, followed by an explanation of the configuration and operation of the vehicle system 1S.

[0014] The magnetic marker 10 is a flat sheet-like magnet with a diameter of 100 mm and a thickness of 1.5 mm. The magnetic marker 10 can be attached to the road surface 100S, for example, by adhesive or the like. The individual magnetic markers 10 are arranged, for example, at intervals of 2 m along the center of the lane 100. Here, the road surface 100S is an example of a surface on which the magnetic markers 10 are arranged. Instead of the sheet-like magnetic markers in this example, columnar magnetic markers with a diameter of 28 mm and a height of 20 mm may also be used. In the case of columnar magnetic markers, they may be arranged, for example, housed in holes drilled in the road surface 100S. The road surface 100S on which these columnar magnetic markers are embedded is also an example of a surface on which magnetic markers are arranged.

[0015] As shown in FIG. 3 , the sensor unit 1 is an in-vehicle unit that incorporates 15 magnetic sensors Cn (n is an integer between 1 and 15), a differential circuit Gm (m is an integer between 1 and 14), and a non-contact displacement sensor 15 that measures the amount of displacement relative to the road surface 100S. The sensor unit 1 has input and output ports. The input ports include at least an input port for the main circuit 20 to input a control signal. There are at least two types of output ports: the first output port is a 14-channel output port that outputs sensor signals (hereinafter referred to as magnetic signals) obtained by performing preprocessing (differential calculation) on magnetic measurement values ​​obtained by each magnetic sensor Cn; and the second output port is an output port for the sensor signal (hereinafter referred to as displacement signal) of the non-contact displacement sensor 15.

[0016] In the sensor unit 1, 15 magnetic sensors Cn are arranged in a straight line at equal intervals of 10 cm. This gives the sensor unit 1 an elongated rod shape as a whole. The sensor unit 1 is attached, for example, inside the front bumper of the vehicle 5 along the vehicle width direction. In the illustrated vehicle 5, the sensor unit 1 is attached at a height of 200 mm from the road surface 100S.

[0017] The magnetic sensor Cn is a highly sensitive MI sensor that detects magnetism using the well-known MI (Magneto Impedance Effect). The MI effect is an electromagnetic effect in which the impedance of a magnetosensitive material, such as an amorphous wire, changes sensitively in response to an external magnetic field. The MI sensor has detection sensitivity in the axial direction (longitudinal direction) of the amorphous wire, which is the magnetosensitive material.

[0018] In the rod-shaped sensor unit 1, magnetic sensors Cn are arranged along the longitudinal direction. Each magnetic sensor Cn is assembled so that the axial direction of the amorphous wire coincides with the longitudinal direction of the sensor unit 1. As described above, the sensor unit 1 is mounted along the vehicle width direction. Therefore, the axial direction of the amorphous wire of each magnetic sensor Cn of the sensor unit 1 is aligned with the vehicle width direction. The magnetic sensors Cn are highly sensitive sensors with a magnetic flux density measurement range of ±0.6 mT (millitesla) and a magnetic flux resolution within the measurement range of 0.02 μT. Furthermore, the magnetic sensors Cn are capable of magnetic measurement at a fast frequency of 3 kHz, making them suitable for high-speed vehicle driving. In this example, the sensor unit 1 is mounted on the vehicle 5 so that the magnetic sensor C1 is located on the left side of the vehicle 5 (the passenger side of a right-hand drive vehicle).

[0019] The differential circuit Gm is a circuit that performs differential calculation of magnetic measurement values ​​obtained by two magnetic sensors as preprocessing. Differential circuits Gm are individually provided corresponding to 14 pairs of adjacent two magnetic sensors among the 15 magnetic sensors Cn. The differential circuits Gm can calculate the magnetic gradient in the arrangement direction of the magnetic sensors (corresponding to the vehicle width direction). The differential calculation value obtained by the differential circuit Gm (magnetic gradient in the vehicle width direction) is output as a magnetic signal from the sensor unit 1. The sensor unit 1 has 14 output ports (not shown) that can simultaneously output the differential calculation values ​​obtained by the differential circuit Gm. The differential circuit Gm also functions as a sampling circuit that acquires magnetic measurement values ​​obtained by the magnetic sensors Cn. The differential circuit Gm, as a sampling circuit, acquires magnetic measurement values ​​at a 3 kHz frequency, at which the magnetic sensors Cn perform magnetic measurements.

[0020] The non-contact displacement sensor 15 is a sensor that detects displacement relative to the road surface 100S. The non-contact displacement sensor 15 is disposed, for example, adjacent to the magnetic sensor C8 located in the center of the sensor unit 1. In the configuration of this example, the non-contact displacement sensor 15 is disposed in a position corresponding to the gap between the magnetic sensors C7 and C8.

[0021] The non-contact displacement sensor 15 includes a light source element 151 of blue laser light, an image pickup element 153 that picks up an image of the road surface 100S, and a displacement measurement circuit 150 that processes the picked-up image. The light source element 151 is an element that irradiates a certain area on the road surface 100S with blue laser light. The image pickup element 153 is a camera that picks up an image of the area irradiated by the light source element 151. The image pickup element 153 is controlled to perform an image pickup operation in synchronization with the light emission operation of the light source element 151.

[0022] The displacement measurement circuit 150 is a circuit that performs displacement measurement on consecutive images captured in succession over time. The displacement measurement circuit 150 shifts an earlier image in time and overlays it on a later image in time to check the degree of match. The displacement measurement circuit 150 then measures the displacement of the vehicle 5 by identifying the amount of shift at which the earlier image most closely matches the later image. The displacement measured in this manner (displacement measurement values ​​such as the amount of displacement and direction of displacement) is output as a displacement signal from the non-contact displacement sensor 15. The specific details of the displacement measurement will be described in detail later.

[0023] As described above, the vehicle system 1S is a system for implementing driving assistance control such as lane keeping control. In addition to the sensor unit 1, the vehicle system 1 is configured to include a main circuit 20 that executes marker detection processing, a vehicle ECU 30 that controls a steering actuator, and the like. In the vehicle system 1, the differential circuit Gm of the sensor unit 1 and the main circuit 20 form a processing circuit that executes detection processing of the magnetic marker 10.

[0024] The result of the marker detection process by the main circuit 20 is output from the main circuit 20 and input to the vehicle ECU 30. As will be described in detail later, the result of the marker detection process includes the amount of lateral deviation of the vehicle 5 from the detected magnetic marker 10. The vehicle ECU 30 controls the steering actuator to adjust the steering angle so that this amount of lateral deviation approaches zero.

[0025] The main circuit 20 (FIG. 3) is a circuit that controls the sensor unit 1 and executes marker detection processing. The main circuit 20 includes an electronic board (not shown) on which a central processing unit (CPU) that executes various calculations, memory elements such as a read only memory (ROM) and a random access memory (RAM), and the like are mounted.

[0026] The main circuit 20 has input / output ports for communication with the sensor unit 1. The input ports include input ports for the 14-channel magnetic signals, input ports for the displacement signals, etc. The output ports include output ports for control signals to the sensor unit 1, output ports for the results of marker detection processing, etc.

[0027] The main circuit 20 includes data areas M1 to M14 (denoted as Mm where appropriate) and data areas H1 to H14 (denoted as Hm where appropriate) for storing time-series data, a filter processing circuit 205 that performs arithmetic processing, and a detection processing circuit 207. Furthermore, the main circuit 20 includes a switching circuit 203 that switches the detection processing for detecting the magnetic marker 10.

[0028] The data area Mm is a storage area for storing (saving) data (magnetic gradient in the vehicle width direction) represented by the 14-channel magnetic signals output as sensor signals from the sensor unit 1. The data area Hm is a storage area for storing filter output values ​​from the filter processing circuit 205.

[0029] The filter processing circuit 205 is a circuit that performs filter processing on the time series data stored in the data area Mm. The detection processing circuit 207 is a circuit that performs marker detection processing on the time series data (filter output values) stored in the data area Hm. The filter processing circuit 205 and the detection processing circuit 207 constitute a processing circuit that executes detection processing for the magnetic marker 10. The filter processing by the filter processing circuit 205 and the marker detection processing by the detection processing circuit 207 constitute detection processing for detecting the magnetic marker 10. The switching circuit 203 switches the detection processing for detecting the magnetic marker 10 by changing the filter characteristics of the filter processing circuit 205.

[0030] The filter processing circuit 205 is a circuit that performs filter processing for each channel on the 14-channel time-series data stored in the data area Mm. As described above, the filter processing circuit 205 sequentially stores the filter output values ​​of each channel in the data area Hm. This filter processing circuit 205 is provided with multiple types of filter characteristics, and the switching circuit 203 switches to one of the filter characteristics. Note that, as will be described in detail later, the switching circuit 203 changes the filter characteristics in the filter processing circuit 205 in accordance with the displacement measurement value obtained by the non-contact displacement sensor 15.

[0031] The detection processing circuit 207 is a circuit that executes marker detection processing to detect the magnetic marker 10. The marker detection processing is executed based on the time-series filter output values ​​stored in the data area Hm. This marker detection processing includes a process of measuring the amount of lateral deviation of the vehicle 5 from the detected magnetic marker 10.

[0032] Next, the details of (1) displacement measurement by the sensor unit 1 and (2) marker detection processing by the sensor unit 1 and the main circuit 20 will be described in order. (1) Displacement Measurement The details of displacement measurement will be described with reference to Figures 4 and 5, focusing on the operation of the displacement measurement circuit 150 that constitutes the non-contact displacement sensor 15. Figure 4 is a flow chart showing the flow of displacement measurement. Figure 5 is an explanatory diagram of how the previous image and the latest image are compared.

[0033] The displacement measurement circuit 150 controls the light source element 151 and the image capturing element 153 so that the light emission operation and the image capturing operation are executed in synchronization. The image capturing operation is executed at a frequency of, for example, 3 kHz. The image capturing element 153 captures an image of the area on the road surface 100S illuminated by the light source element 151 (S101). The light source element 151 is configured to illuminate the road surface 100S broadly.

[0034] The displacement measurement circuit 150 compares the previous captured image (previous image) acquired in the previous processing loop and stored in step S104 (described later) with the latest captured image (latest image) acquired in the immediately preceding step S101 (S102). Specifically, as shown in Fig. 5, the displacement measurement circuit 150 shifts the previous image in the x and y directions and checks the correlation (degree of agreement) with the latest image.

[0035] The displacement measurement circuit 150 calculates a correlation coefficient with the latest image for each position to which the previous image was shifted, and identifies the position to which the previous image was shifted that has the largest correlation coefficient. In this way, the displacement measurement circuit 150 identifies the position to which the previous image was shifted, thereby identifying the amount of displacement and the direction of displacement (measured displacement value) on the imaging surface (S103). Here, the amount of displacement is the length of a vector defined by the amount of shift in the x direction and the amount of shift in the y direction. The direction of displacement is the orientation of this vector.

[0036] After determining the amount and direction of displacement as described above, the displacement measurement circuit 150 stores and saves the latest image acquired in step S101 as a previous image (S104). By storing and saving the latest image in this manner, it becomes possible to refer to it as a previous image in the next processing loop. The displacement measurement circuit 150 also outputs a displacement signal representing the amount and direction of displacement determined in step S103 (S105) and inputs it to the main circuit 20.

[0037] (2) Marker Detection Processing The marker detection processing is a processing that is repeatedly executed at a frequency of 3 kHz by cooperative operations of the sensor unit 1 and the main circuit 20. The contents of this marker detection processing will be described with reference to the flowchart of FIG.

[0038] In response to control by the main circuit 20, the sensor unit 1 performs magnetic detection at a frequency of 3 kHz (S201). As described above, in each magnetic sensor Cn, an amorphous wire (not shown) serving as a magnetosensitive body is arranged along the vehicle width direction. The magnetic field acting on the magnetic marker 10 in the vehicle width direction is directed toward the outside of the magnetic marker 10. Therefore, the direction of the magnetic field acting on each magnetic sensor Cn in the vehicle width direction is opposite depending on whether it is on the left or right side of the magnetic marker 10.

[0039] 7 illustrates an example of temporal changes in the distribution waveform of the magnetic field in the vehicle width direction of the magnetic measurement values ​​(magnetic intensity in the vehicle width direction) obtained by each magnetic sensor Cn constituting the sensor unit 1. In the figure, the traveling direction (time direction) of the vehicle 5 is defined from position p1 in the upper left to position p7 in the lower right. Position p4 is located directly above the magnetic marker 10. Positions p1 → p4 are the section approaching the magnetic marker 10, and positions p4 → p7 are the section moving away from the magnetic marker 10.

[0040] 7, the amplitude of the magnetic distribution waveform in the vehicle width direction at each position differs, but in all cases a zero cross Zc occurs corresponding to the position of the magnetic marker 10 in the vehicle width direction, and the waveform has two peaks with alternating positive and negative signs on both sides of the zero cross Zc. As the vehicle 5 passes the magnetic marker 10, the amplitude of the two-peak distribution waveform gradually increases as the vehicle 5 approaches the magnetic marker 10, and reaches a maximum at position p4 directly above the magnetic marker 10. Thereafter, the amplitude of the two-peak distribution waveform gradually decreases as the vehicle 5 moves away from the magnetic marker 10.

[0041] In the sensor unit 1, the magnetic measurement value of each magnetic sensor Cn is immediately input to a differential circuit Gm (FIG. 3). For example, the magnetic measurement values ​​of magnetic sensors C1 and C2 are input to differential circuit G1, and a differential calculation is performed to subtract the magnetic measurement value of C1 from the magnetic measurement value of C2 (S202). The differential circuit Gm performs a differential calculation to subtract the magnetic measurement value of magnetic sensor Cm (m is a natural number from 1 to 14) from the magnetic measurement value of magnetic sensor C(m+1).

[0042] The difference calculation value of the differential circuit Gm is the difference between the magnetic measurement values ​​of two adjacent magnetic sensors Cn in the sensor unit 1, and indicates the magnetic gradient in the vehicle width direction. The distribution waveform of the magnetic gradient in the vehicle width direction is a waveform in which small peaks with opposite polarities to the high peaks are adjacent to each other on both sides of the high peaks, as exemplified in each of the graphs at positions p1 to p7 in Figure 8.

[0043] The difference calculation in step S202 in Fig. 6 is effective in removing common magnetic noise that acts uniformly on each magnetic sensor Cn. Common magnetic noise is generated not only by the earth's magnetic field, but also by large magnetic sources such as railway bridges and other vehicles. In the case of a large magnetic source, the magnetic field loop from the north pole to the south pole becomes very large, so the magnetic field becomes nearly uniform at positions intermediate between the two poles, and the magnetic field acting on each magnetic sensor Cn becomes nearly uniform. The difference calculation in step S202 described above is also effective in removing such magnetic fields that act nearly uniformly on each magnetic sensor Cn.

[0044] The differential calculation values ​​(examples of calculation values ​​based on magnetic measurement values) by the differential circuit Gm are simultaneously output as 14-channel sensor signals by the sensor unit 1. In the main circuit 20, time-series data for each channel based on these sensor signals is stored in the data area Mm ( FIG. 3 ). When a new sensor signal is acquired, the main circuit 20 erases the oldest data in the data area Mm and sequentially shifts through the data in the data area Mm to create an empty space. The main circuit 20 then stores the data represented by the newly acquired sensor signal in that empty space. Through this storage operation by the data area Mm, each channel in the data area Mm stores time-series data of differential calculation values ​​(transverse magnetic gradient) over a predetermined period of time.

[0045] As shown in Figure 8, the time-series data of the difference calculation value (magnetic gradient in the vehicle width direction) in data area Mm shows an increase in amplitude as the vehicle 5 approaches the magnetic marker 10. The amplitude reaches a maximum at position p4, and the amplitude decreases as the vehicle moves away from the magnetic marker 10, forming a single-peak distribution. For example, the diagonal graph on the right side of Figure 8 is a graph showing the temporal change in the peak value of the magnetic gradient in the vehicle width direction at each of positions p1 to p7. In this graph, the axis pointing diagonally downward to the right defines the direction of travel (time direction), and the perpendicular axis defines the magnetic gradient in the vehicle width direction.

[0046] 6, the sensor unit 1 performs the displacement measurement described above with reference to FIGS. 4 and 5 using the non-contact displacement sensor 15. The displacement measurement results (displacement amount and displacement direction) are input as a displacement signal to the main circuit 20. This displacement signal is a signal that represents the displacement of the non-contact displacement sensor 15 relative to the road surface 100S, i.e., the displacement of the vehicle 5 relative to the road surface 100S.

[0047] When the main circuit 20 receives the displacement signal from the sensor unit 1, it changes the filter characteristics of the filter processing circuit 205 to filter characteristics that correspond to the amount of displacement contained in the displacement signal (S204). As described above, multiple types of filter characteristics are preset in the filter processing circuit 205. As will be described later, the main circuit 20 selects one of the multiple types of filter characteristics in accordance with the amount of displacement.

[0048] The main circuit 20 inputs the time-series data of the magnetic gradient in the vehicle width direction for 14 channels stored in the data area Mm (m is an integer between 1 and 14) to the filter processing circuit 205, and executes filter processing for each channel (S205). This filter processing is a filtering process that blocks low-frequency components and passes high-frequency components. The filter processing circuit 205 calculates a filter output value by convolution of the time-series data in the data area Mm using an IIR (Infinite Impulse Response) filter, and stores the filter output value in the data area Hm (FIG. 3).

[0049] Here, the content and effect of the filtering process executed by the filtering circuit 205 will be explained. As described above, in the case of a large magnetic source such as a bridge or tunnel, the differential calculation in step S202 in FIG. 6 has a certain effect. However, even in the case of a large magnetic source, a gentle magnetic gradient occurs around the edges that become magnetic poles due to the magnetic field wraparound. The differential calculation in S202 described above removes uniform magnetism, and it is difficult to remove magnetic gradients.

[0050] When comparing a large magnetic source such as a bridge with a small magnetic source such as the magnetic marker 10, the distance between the magnetic poles is different, which results in a different rate of change of the magnetic gradient. That is, in a large magnetic source with a long distance between the magnetic poles, the distance it takes for the magnetic gradient of one magnetic pole to transition to the magnetic gradient of the other magnetic pole is long, so the change in the magnetic gradient is gradual and the frequency is low. On the other hand, in a small magnetic source such as the magnetic marker 10, the distance between the poles is short, so the change in the magnetic gradient is sudden, the rate of change is large, and the frequency is high. Filter processing that blocks low-frequency components can remove the magnetic gradient that changes slowly and has a small rate of change, which originates from a large magnetic source.

[0051] For example, when passing any one of the magnetic markers, the magnetic gradient in the vehicle width direction calculated by the difference calculation in step S202 changes as shown in the diagonal graph on the right side of the page in Figure 8. For example, when the vehicle 5 travels along the lane 100, ideally, a peak should occur each time the vehicle passes a magnetic marker 10. However, because the actual road environment includes magnetic sources such as bridges and tunnels, the ideal change in which a peak periodically appears each time the vehicle passes a magnetic marker 10 is not obtained, and it is highly likely that the change will be influenced by external magnetic disturbances as shown in Figure 9(a).

[0052] The filtering process by the filter processing circuit 205 is a process for cutting off low-frequency components from such a change in the magnetic gradient in the vehicle width direction (FIG. 9(a)). This filtering process makes it possible to make the change in the magnetic gradient in the vehicle width direction in FIG. 9(a) closer to an ideal change in which the above-mentioned peaks appear periodically, as shown in FIG. 9(b).

[0053] Here, we will explain the multiple types of filter characteristics set by the filter processing circuit 205 and the methods for setting (changing and selecting) the filter characteristics. The target of filter processing by the filter processing circuit 205 is, for example, time series data representing the temporal change in the magnetic gradient in the vehicle width direction as shown in Figure 9. As described above, the purpose of this filter processing is to remove the influence of the magnetic gradient generated at the end of a large magnetic source, such as a bridge or tunnel, which is a magnetic pole, and to make the magnetic gradient around the magnetic marker 10 more noticeable.

[0054] For example, as the vehicle speed increases, the time required to pass through an area near the edge of a large magnetic source, such as a bridge or tunnel, becomes shorter. Conversely, as the vehicle speed decreases, the time required to pass through this area becomes longer. Therefore, the temporal change in the magnetic gradient in the vehicle width direction in Figure 9 becomes higher in frequency as the vehicle speed increases, and becomes lower as the vehicle speed decreases.

[0055] The filter processing circuit 205, which performs filter processing to remove low-frequency components, is set with a plurality of filter characteristics with different cutoff frequencies. The filter processing circuit 205 can change the filter characteristics according to the vehicle speed. The filter characteristics include a high cutoff frequency filter characteristic corresponding to high-speed driving, a low cutoff frequency filter characteristic corresponding to low-speed driving, and a medium cutoff frequency filter characteristic corresponding to medium-speed driving. In step S204 ( FIG. 6 ) described above, the filter processing performed by the filter processing circuit 205 is changed to one of the three filter characteristics described above according to the displacement measured in step S203.

[0056] In step S205, the detection processing circuit 207 (main circuit 20) reads the time-series filter output values ​​(e.g., FIG. 9B) of each channel in the data area Hm and executes marker detection processing (S206). For example, the detection processing circuit 207 selects the channel with the largest fluctuation range among the channels of the time-series filter output values. The detection processing circuit 207 then applies, for example, threshold processing to the time-series filter output values ​​of the selected channel, i.e., the changes shown in FIG. 9B. By using threshold processing, it is possible to relatively easily detect peaks that appear periodically each time a magnetic marker 10 passes by, for example, from the changes in the time-series filter values ​​shown in FIG. 9B.

[0057] Furthermore, the detection processing circuit 207 reads out the time-series data of the magnetic gradient in the vehicle width direction stored in the data area Mm (see FIG. 3 ), including the data for each channel at the time when the magnetic marker 10 was detected, i.e., the data constituting the distribution waveform at time point p4 in FIG. 8 . The detection processing circuit 207 then identifies the vehicle width position of the peak value of the data constituting this distribution waveform. The position of this peak value in the vehicle width direction is the position directly above the magnetic marker 10 in the vehicle width direction when the sensor unit 1 passes over the magnetic marker 10. Therefore, based on the position of this peak value, it is possible to identify the amount of lateral deviation of the vehicle 5 relative to the magnetic marker 10. Note that the detection of the magnetic marker 10 and the detected amount of lateral deviation relative to the magnetic marker 10 are output from the main circuit 20 as the results of the marker detection processing. The results of the marker detection processing are input to the vehicle ECU 30 as described above and are used for lane keeping control, etc.

[0058] As described above, the sensor unit 1 in the vehicle system 1 of this embodiment includes a magnetic sensor and a non-contact displacement sensor. This sensor unit 1 can measure the amount of displacement of the vehicle 5 relative to the road surface 100S on which the magnetic markers 10 are arranged.

[0059] The non-contact displacement sensor 15 in this example is a sensor that measures the displacement of the vehicle 5. The non-contact displacement sensor 15 measures displacement by shifting an earlier image in time among temporally consecutive images, overlaying it on a later image, and examining the degree of match. Instead of shifting the entire image, it is also possible to divide the image into regions and shift each region to examine the degree of match. Furthermore, the non-contact displacement sensor 15 in this example assumes only translational movement without rotation between temporally consecutive images. It is also possible to examine rotation in addition to translational movement. In this case, it becomes possible to measure changes in the direction of the vehicle.

[0060] In the vehicle system 1 of this example, the filter characteristics of the filter processing, which is part of the marker detection processing, are changed according to the amount of displacement of the vehicle 5. By changing the filter characteristics according to the amount of displacement of the vehicle 5 in this way, appropriate filter processing can be applied according to the vehicle speed, thereby improving the reliability of detection of the magnetic marker 10.

[0061] In this example, the main circuit 20 that performs filtering and marker detection processing is provided separately from the sensor unit 1. Alternatively, a sensor unit in which the main circuit 20 is integrally incorporated may be used. Alternatively, a sensor unit in which the data area Mm of the main circuit 20 and the filtering circuit 205 are integrally incorporated may be used. In this case, the sensor signal becomes the filter output value for each channel.

[0062] In this example, three types of filter characteristics are set in the filter processing circuit 205. Alternatively, a stepless filter characteristic may be set according to the displacement or speed of the vehicle 5. In this case, more appropriate filter processing can be performed according to the displacement, etc., of the vehicle 5.

[0063] The filtering performed by the filtering circuit 205 in this example is a high-pass filter that passes high-frequency components and blocks low-frequency components. It is also possible to lower the cutoff frequency that blocks low frequencies or to set the filter's cutoff characteristics more gradual. A gradual cutoff characteristic increases the degree of freedom in filter design and potentially reduces the computational load required for filtering. Instead of a high-pass filter, a band-pass filter may be used that passes frequency components within a range corresponding to the magnetic changes that occur when passing through the magnetic marker 10 and blocks frequency components higher and lower than that range.

[0064] The detection process in this example includes filtering of a signal representing a temporal change in the calculated difference value (time-series data of the magnetic gradient in the vehicle width direction). Here, the calculated difference value is an example of a magnetic measurement value obtained by a magnetic sensor or a calculated value based on the magnetic measurement value. The filtering may be filtering of a signal representing a temporal change in the magnetic measurement value obtained by a magnetic sensor. It may also be filtering of a signal representing a spatial change in the calculated difference value. Possible spatial changes include, for example, changes depending on the position of the vehicle 5 in the traveling direction.

[0065] The time interval for acquiring magnetic measurement values ​​may be changed so that the positions at which magnetic measurement values ​​are acquired by the magnetic sensor are spaced apart at a fixed distance. For example, the magnetic sensor may be configured to perform magnetic measurement each time the displacement measured by the non-contact displacement sensor reaches a predetermined distance, such as 10 mm, 20 mm, or 50 mm. In this case, sequence data may be acquired according to position, such as magnetic data in the vehicle width direction. In this case, the sequence data represents spatial changes. While the magnetic sensor performs magnetic measurement at a sufficiently fast period, the period for acquiring the sensor signal on the main circuit 20 side may be set to be different from the measurement period. For example, the main circuit 20 may be configured to acquire a sensor signal from the sensor unit 1 each time the vehicle 5 travels a predetermined distance.

[0066] In this example, the magnetic sensor Cn is sensitive in the vehicle width direction, but it may also be sensitive in the traveling direction or in the vertical direction. Furthermore, it may also be sensitive in two axial directions, for example, the vehicle width direction and the traveling direction, or in two axial directions, for example, the traveling direction and the vertical direction, or it may be sensitive in three axial directions, for example, the vehicle width direction, the traveling direction, and the vertical direction. By using a magnetic sensor sensitive in multiple axial directions, it is possible to measure the magnitude of the magnetic field as well as the direction of magnetic action, and generate a magnetic vector. It is also possible to distinguish between the magnetic field of the magnetic marker 10 and external magnetic disturbances by using the difference between the magnetic vectors and the rate of change of that difference in the traveling direction.

[0067] One-dimensional filtering with respect to the vehicle's traveling direction (time direction) is illustrated. Alternatively, or in addition, spatial filtering may be performed on magnetic changes in a two-dimensional space defined by the vehicle's traveling direction (time direction) and the vehicle's width direction to remove disturbance magnetism. A spatial filter may also be applied to magnetic changes in a two-dimensional space defined by the vehicle's width direction and vertical direction. Furthermore, a space-time filter may be applied to magnetic changes in a space-time domain that combines the vehicle's traveling direction (time direction) with this two-dimensional space to remove disturbance magnetism.

[0068] In this example, the sensor unit 1 has magnetic sensors arranged in a straight line, but a sensor unit having only one magnetic sensor is also conceivable. Alternatively, in this example, a configuration has been illustrated in which one non-contact displacement sensor is incorporated into a sensor unit having magnetic sensors arranged in a straight line, but multiple non-contact displacement sensors may also be incorporated. For example, as shown in FIG. 10 , one non-contact displacement sensor 15 may be incorporated at each end of the sensor unit 1. The non-contact displacement sensors 15 at both ends are arranged on the straight line on which the magnetic sensors are arranged, or on a line parallel to this straight line.

[0069] If the two non-contact displacement sensors 15 arranged at both ends of the sensor unit 1 each measure displacement, it becomes possible to measure the change in tilt of the sensor unit 1, i.e., the rotational component. In this case, it is possible to measure not only translational movement defined by the displacement amount and displacement direction, but also displacement accompanied by rotation. By measuring the displacement of the sensor unit 1 with high accuracy in this way, it becomes possible to individually identify the displacement of each magnetic sensor constituting the sensor unit 1. Alternatively, a non-contact displacement sensor may be provided individually corresponding to each magnetic sensor so that the displacement of each magnetic sensor can be identified. If the displacement of each magnetic sensor can be identified, for example, when applying filter processing to time-series data of the magnetic gradient in the vehicle width direction, the filter characteristics can be changed for each channel. Note that in this example, the channel corresponds to the differential circuit Gm. However, if the differential circuit Gm is omitted, a channel may be provided for each magnetic sensor and the filter characteristics may be changed for each channel.

[0070] Furthermore, a sensor unit in which magnetic sensors are arranged two-dimensionally may be employed. In this case, the number of non-contact displacement sensors incorporated in the sensor unit may be one, but preferably two or more. Using two non-contact displacement sensors arranged at different positions allows for the determination of not only the displacement amount and displacement direction of the sensor unit, but also the rotation of the sensor unit. Accurate determination of the displacement of the sensor unit allows for accurate determination of the displacement of each magnetic sensor, thereby enabling more appropriate processing of the data derived from each magnetic sensor. In this example, the magnetic gradient in the vehicle width direction is generated by differential calculation for the magnetic sensors Cn arranged in the vehicle width direction. Alternatively, or in addition, the magnetic sensors Cn may be arranged in the vehicle's traveling direction, and the magnetic gradient in the traveling direction may be determined by differential calculation for the two magnetic sensors arranged in the traveling direction.

[0071] In this example, the focus is on the displacement amount among the displacement measurements obtained by the non-contact displacement sensor, and the filter characteristics of the filter processing circuit 205 are changed according to the displacement amount, thereby switching the detection process for detecting the magnetic marker. It is also possible to switch the detection process according to the displacement direction in addition to or instead of the displacement amount. Furthermore, it is also possible to switch the detection process taking into account the rotational component of the magnetic sensor. The meaning of "displacement" in descriptions such as the displacement of a vehicle, which is a moving body, and the displacement of a magnetic sensor includes physical quantities for identifying the displacement, such as the displacement amount and displacement direction.

[0072] In the configuration of this example, the magnetic measurement and marker detection process are performed at a fixed time period of 3 kHz. Instead of a time period, the magnetic measurement and marker detection process may be performed each time the displacement detected by the non-contact displacement sensor reaches 5 cm. In this case, the displacement measurement by the non-contact displacement sensor may be repeatedly performed at a sufficiently fast time period of, for example, 10 kHz. The displacement measurement period may be set to, for example, a period corresponding to the movement of a vehicle traveling at high speed of 0.2 cm to 1 cm.

[0073] In this way, if magnetic measurement and marker detection processing are performed each time the vehicle displaces a predetermined amount (predetermined distance), the changes illustrated in the right-hand diagram of Fig. 8 and the changes illustrated in Fig. 9 are positional changes rather than temporal changes. Positional changes are the same whether the vehicle speed is high and the displacement measured by the non-contact displacement sensor is large or low and the displacement measured by the non-contact displacement sensor is small. Therefore, if the change is positional, there is less need to switch the filter characteristics depending on the displacement measured by the non-contact displacement sensor.

[0074] On the other hand, when comparing expressways and ordinary roads, for example, there are often differences in size, such as the size of bridges and the scale of tunnels. If the size of a structure that can be a large magnetic source, such as a bridge, differs, the magnetic gradient generated at the end of the structure, which becomes a magnetic pole, may differ. Therefore, it is possible to switch the characteristics of filter processing for suppressing the influence of the magnetic gradient that can be generated at the end of a large magnetic source, such as a bridge, between expressways with large bridges and tunnels and ordinary roads with small bridges and tunnels. Here, the distinction between traveling on an expressway and an ordinary road may be made based on, for example, the result of identifying the road being traveled using a map matching function of the navigation system, or based on the amount of displacement per unit time (i.e., vehicle speed) or driving pattern.

[0075] Although specific examples of the present invention have been described in detail as examples, these examples merely disclose examples of the technology encompassed by the claims. Needless to say, the scope of the claims should not be interpreted as being limited by the configurations, numerical values, etc. of the specific examples. The claims encompass technologies that are obtained by variously modifying, changing, or appropriately combining the specific examples using publicly known technology and the knowledge of those skilled in the art.

[0076] REFERENCE SIGNS LIST 1 Sensor unit Cn Magnetic sensor 1S Vehicle system (system) 10 Magnetic marker 100S Road surface (surface) 15 Non-contact displacement sensor 150 Displacement measurement circuit 151 Light source element 153 Imaging element 20 Main circuit 203 Switching circuit 205 Filter processing circuit 207 Detection processing circuit 30 Vehicle ECU 5 Vehicle

Claims

1. A sensor unit attached to a moving object for detecting a magnetic marker, One or more magnetic sensors that measure the magnetic field acting from the magnetic marker; A sensor unit comprising: one or more non-contact displacement sensors that measure displacement relative to a surface on which the magnetic marker is disposed.

2. In claim 1, the non-contact displacement sensor comprises a light source element that irradiates light onto the surface, an image pickup element that picks up an image of the surface in response to light emission by the light source element, and a displacement measurement circuit that measures the displacement of the vehicle using the image picked up by the image pickup element; The displacement measurement circuit is a sensor unit that measures displacement by identifying the amount of shift that most closely matches one of two captured images when the other captured image is shifted in position.

3. 2. The sensor unit according to claim 1, wherein a plurality of magnetic sensors are arranged in a straight line, and at least two of the plurality of non-contact displacement sensors are disposed on a line parallel to the straight line.

4. 2. The sensor unit according to claim 1, wherein a plurality of magnetic sensors are arranged two-dimensionally, and at least two of the plurality of non-contact displacement sensors are disposed at different positions.

5. 2. The sensor unit according to claim 1, comprising a plurality of magnetic sensors and a plurality of non-contact displacement sensors, wherein the non-contact displacement sensors are individually arranged for each of the plurality of magnetic sensors so that the displacement of each of the magnetic sensors can be individually measured.

6. In any one of claims 1 to 5, a sensor unit comprising: a processing circuit that processes magnetic measurement values ​​obtained by the one or more magnetic sensors to perform a detection process for detecting a magnetic marker; and a switching circuit that switches the detection process depending on the displacement measurement values ​​obtained by the one or more non-contact displacement sensors.

7. A processing circuit according to any one of claims 3 to 5, which performs a detection process for detecting a magnetic marker by processing magnetic measurement values ​​obtained by the plurality of magnetic sensors; a switching circuit that switches the detection process in response to displacement measurement values ​​obtained by the plurality of non-contact displacement sensors; the detection process includes a process performed on a time series of magnetic measurement values ​​obtained by one of the plurality of magnetic sensors; The switching circuit is a sensor unit configured to identify the displacement of each of the magnetic sensors constituting the plurality of magnetic sensors based on the displacement measurement values ​​by the plurality of non-contact displacement sensors, and to change the processing applied to the time-series magnetic measurement values ​​in accordance with the displacement of each magnetic sensor individually for each magnetic sensor.

8. The magnetic sensor according to any one of claims 1 to 5, further comprising a sampling circuit for acquiring magnetic measurement values ​​obtained by the one or more magnetic sensors, The sampling circuit is configured to change the time interval for acquiring magnetic measurement values ​​in response to displacement measurement values ​​obtained by the one or more non-contact displacement sensors.

9. The magnetic sensor according to any one of claims 3 to 5, further comprising a sampling circuit for acquiring magnetic measurement values ​​in parallel by the plurality of magnetic sensors, The sampling circuit is a sensor unit configured to identify the displacement of each of the magnetic sensors constituting the plurality of magnetic sensors based on displacement measurement values ​​by the plurality of non-contact displacement sensors, and to be able to change the time interval for acquiring magnetic measurement values ​​according to the displacement of each magnetic sensor individually for each magnetic sensor.

10. 10. The sensor unit according to claim 8 or 9, wherein the sampling circuit is configured to change the time interval at which the magnetic measurement values ​​are acquired so that the positions at which the magnetic measurement values ​​are acquired can be arranged at constant distance intervals.

11. According to claim 6 or 7, the detection process includes filtering a signal representing a temporal or spatial change in a magnetic measurement value obtained by the one or more magnetic sensors or a calculated value based on the magnetic measurement value, The switching circuit is configured to switch the detection processes by changing a filter characteristic of the filter process in response to a displacement measurement value by the one or more non-contact displacement sensors.

12. A system for detecting a magnetic marker by a moving object, comprising: One or more magnetic sensors that measure the magnetic field acting from the magnetic marker; A system in which a sensor unit is attached to the moving body and includes one or more non-contact displacement sensors that measure displacement relative to a surface on which the magnetic markers are arranged.

13. 13. A system according to claim 12, comprising: a processing circuit that processes magnetic measurement values ​​by the one or more magnetic sensors to execute a detection process for detecting a magnetic marker; and a switching circuit that switches the detection process depending on the displacement measured by the one or more non-contact displacement sensors.

14. According to claim 13, the detection process includes filtering a signal representing a temporal or spatial change in a magnetic measurement value obtained by the one or more magnetic sensors or a calculated value based on the magnetic measurement value, The system is configured such that the switching circuit switches the detection processes by changing filter characteristics of the filter processes in response to displacement measurements by the one or more non-contact displacement sensors.

15. 13. The apparatus according to claim 12, further comprising a sampling circuit for acquiring magnetic measurement values ​​obtained by the one or more magnetic sensors, The sampling circuit is configured to vary the time interval at which magnetic measurement values ​​are acquired in response to displacement measurements by the one or more non-contact displacement sensors.

16. 16. The system of claim 15, wherein the sampling circuitry is configured to vary the time interval at which magnetic measurements are taken so that the locations at which magnetic measurements are taken can be arranged at regular distance intervals.

17. 15. The moving body according to claim 13 or 14, wherein the moving body is a vehicle traveling on a road, The switching circuit is a system capable of switching the detection process depending on the type of road on which the moving object is traveling.

18. A method for detecting a magnetic marker by a moving object, comprising: The moving body includes: One or more magnetic sensors that measure the magnetic field acting from the magnetic marker; a sensor unit including one or more non-contact displacement sensors that measure a displacement relative to a surface on which the magnetic marker is disposed; A detection process for detecting a magnetic marker by processing a magnetic measurement value by the one or more magnetic sensors, A magnetic marker detection method, which switches the detection process depending on the displacement measured by the one or more non-contact displacement sensors.

19. In claim 18, the detection process includes filtering a signal representing a temporal or spatial change in a magnetic measurement value by the one or more magnetic sensors or a calculated value based on the magnetic measurement value, A magnetic marker detection method, which switches the detection process by changing the filter characteristics of the filter process depending on the displacement measurement value by the one or more non-contact displacement sensors.