JP2023038220A - Posture estimation device, control method, program and storage medium - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a posture estimation device that can preferably estimate a fitting posture of a measurement device which is fitted to a movable body to the movable body.

SOLUTION: An on-board machine 1 acquires a group of measurement points obtained by measuring a section line with a rider 2 fitted to a vehicle and calculates a central line of the section line along a travel direction of the vehicle on the basis of the group of measurement points. The on-board machine 1 estimates a fitting posture of the rider 2 to the vehicle on the basis of a direction of the central line of the section line relative to a yaw-directional reference angle that is used by the rider 2 as a reference.

SELECTED DRAWING: Figure 15

COPYRIGHT: (C)2023,JPO&INPIT

Description

The present invention relates to technology for estimating the orientation of a measuring device.

2. Description of the Related Art Conventionally, there has been known a technique for estimating a position of a vehicle based on measurement data from measurement devices such as radar and cameras. For example, Patent Literature 1 discloses a technique of estimating a self-position by collating the output of a measurement sensor with the position information of a feature registered in advance on a map. Further, Patent Document 2 discloses a vehicle position estimation technique using a Kalman filter.

JP 2013-257742 A JP 2017-72422 A

The data obtained from the measuring device is the value of the coordinate system with the measuring device as the reference, and it depends on the attitude of the measuring device with respect to the vehicle. There is a need. Therefore, when the posture of the measuring device deviates, it is necessary to accurately detect the deviation and reflect it in the data of the measuring device or to correct the posture of the measuring device.

SUMMARY OF THE INVENTION The present invention has been made to solve the above-described problems, and is primarily to provide a posture estimation device capable of suitably estimating the mounting posture of a measuring device attached to a moving body. purpose.

According to a claim of the invention, there is provided a posture estimating apparatus in which a feature whose surface to be measured is parallel to a road surface on which a mobile object travels and which is formed along the road surface is attached to the mobile object. Acquisition means for acquiring a measurement point group measured by a measurement device, calculation means for calculating a line on the surface to be measured along the traveling direction of the moving body based on the measurement point group, and the measurement device an estimating means for estimating the attachment posture of the measuring device to the moving body based on the direction of the line with respect to the reference direction.

Further, the claimed invention is a control method executed by a posture estimating device, wherein a surface to be measured is parallel to a road surface on which a moving body travels, and a feature formed along the road surface is detected. an acquisition step of acquiring a measurement point group measured by a measuring device attached to the moving body; and a calculation of calculating a line on the surface to be measured along the traveling direction of the moving body based on the measurement point group. and an estimating step of estimating the attachment posture of the measuring device to the moving object based on the direction of the line with respect to the direction that the measuring device uses as a reference.

Further, the invention as claimed in claims is a program executed by a computer, wherein a surface to be measured is parallel to a road surface on which a mobile body travels and is formed along the road surface. Acquisition means for acquiring a measurement point group measured by a measurement device attached to a body; calculation means for calculating a line on the surface to be measured along the running direction of the moving body based on the measurement point group; The computer is caused to function as estimation means for estimating the attachment posture of the measuring device to the moving body based on the direction of the line with respect to the direction used as a reference by the measuring device.

1 is a schematic configuration diagram of a vehicle system; FIG. 3 is a block diagram showing a functional configuration of an in-vehicle device; FIG. FIG. 2 is a diagram showing the relationship between a vehicle coordinate system and a rider coordinate system represented by two-dimensional coordinates; FIG. 2 is a diagram showing the relationship between a vehicle coordinate system and a rider coordinate system represented by three-dimensional coordinates; FIG. 10 shows a bird's-eye view of the surroundings of the vehicle when the vehicle runs near a road sign that is a detection target feature when estimating the roll angle of the rider. The positional relationship between the road sign and the target measurement point group when the measurement surface of the road sign is viewed from the front is shown. FIG. 2 shows a target measurement point cloud and an extracted outer edge point cloud in the lidar coordinate system; FIG. The approximation line of each side of the quadrangle formed by the edge point cloud is shown. 4 is a flowchart showing the procedure of roll angle estimation processing; FIG. 10 shows a bird's-eye view of the surroundings of the vehicle when the vehicle runs near a road sign serving as a detection target feature when estimating a rider's pitch angle. 4 is a flowchart showing a procedure of pitch angle estimation processing; FIG. 10 is a bird's-eye view of the surroundings of the vehicle when the vehicle travels along a lane marking that serves as a detection target feature when estimating the yaw angle of the rider; FIG. 4 is a diagram showing a center point in a detection window with a circle; The angle formed by the center line of the marking line calculated for each detection window and the yaw direction reference angle is shown. 4 is a flowchart showing the procedure of yaw angle estimation processing; 4 is a flowchart showing a specific example of processing based on estimated values of roll angle, pitch angle, and yaw angle;

According to a preferred embodiment of the present invention, the posture estimating apparatus attaches to the moving object a feature whose surface to be measured is parallel to a road surface on which the moving object travels and which is formed along the road surface. Acquisition means for acquiring a measurement point group measured by the measuring device, calculation means for calculating a line on the surface to be measured along the traveling direction of the moving body based on the measurement point group, and the measurement device an estimating means for estimating the attachment posture of the measuring device to the moving object based on the direction of the line with respect to the reference direction. According to this aspect, the posture estimation device uses as a reference the feature whose surface to be measured is parallel to the road surface on which the mobile body travels and which is formed along the road surface, so that the mounting posture of the measuring device to the mobile body is suitable. can be estimated to

In one aspect of the posture estimation device, the acquisition means recognizes the position of the feature based on the feature information about the feature, so that the measurement point group measured by the measuring device at the position is obtained from the feature. obtained as a measurement point cloud of According to this aspect, the attitude estimation device grasps the position of a feature having a vertical plane along the travel path of the moving object or a vertical plane along the width direction of the travel path, and suitably obtains a measurement point cloud of the feature. can be obtained.

In another aspect of the posture estimation device, the estimation means estimates the posture of the measurement device in the yaw direction based on an angle between the line and a reference line in the yaw direction of the measurement device. With this aspect, the posture estimation device can preferably estimate the posture of the measurement device in the yaw direction.

In another aspect of the posture estimating apparatus, the estimating means averages angles formed between the reference line and the plurality of lines calculated by the calculating means based on the different measurement point groups, so that the measured Estimate the yaw attitude of the device. With this aspect, the posture estimation device can more accurately estimate the posture of the measurement device in the yaw direction.

In another aspect of the posture estimation device, the calculation means calculates a plurality of points on the surface to be measured along the traveling direction of the moving object based on the measured point group, and from the plurality of points, The line is calculated, and the estimating means calculates the angles formed by the reference line and the plurality of lines calculated by the calculating means based on the different measurement point groups, of the points used to calculate each of the lines. The yaw orientation of the measuring device is estimated by averaging based on the numbers. In this aspect, the posture estimating device can more accurately estimate the yaw direction posture of the measuring device by appropriately weighting the calculated lines in consideration of the accuracy of each line.

In a preferred example, the features may be marking lines or curbs provided on the travel path. In another preferred example, the measuring device may be an optical scanning device that scans light.

In another aspect of the posture estimating device, the reference direction is a direction regarded as a running direction of the moving body in the coordinate system of the measuring device when the mounting posture does not deviate, and When there is a deviation in the mounting posture, a direction different from the running direction of the moving body is indicated according to the deviation. According to this aspect, the attitude estimation device can preferably estimate the attachment attitude of the measuring device to the moving object based on the calculated deviation of the reference direction with respect to the line on the measurement target surface.

According to another preferred embodiment of the present invention, there is provided a control method executed by a posture estimating device, in which the surface to be measured is parallel to and formed along the road surface on which the moving body travels. an acquiring step of acquiring a measurement point group obtained by measuring a feature with a measuring device attached to the moving body; and obtaining a line on the measurement surface along the running direction of the moving body based on the measurement point group. and an estimating step of estimating the attachment posture of the measuring device to the moving body based on the direction of the line with respect to the direction that the measuring device uses as a reference. By executing this control method, the attitude estimation device can suitably estimate the mounting attitude of the measuring device to the moving body.

According to another preferred embodiment of the present invention, there is provided a program executed by a computer, wherein a surface to be measured is parallel to a road surface on which a moving object travels, and a feature formed along the road surface is measured. an acquisition means for acquiring a measurement point group measured by a measuring device attached to the moving body; and a calculation for calculating a line on the surface to be measured along the running direction of the moving body based on the measurement point group. and the computer functions as estimation means for estimating the attachment posture of the measuring device to the moving object based on the direction of the line with respect to the direction that the measuring device uses as a reference. By executing this program, the computer can suitably estimate the attachment attitude of the measuring device to the moving object. Preferably, the program is stored in a storage medium.

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

[Outline configuration]
FIG. 1 is a schematic configuration diagram of a vehicle system according to this embodiment. The vehicle system shown in FIG. 1 includes an in-vehicle device 1 that performs control related to driving support of the vehicle, a lidar (Light Detection and Ranging, or Laser Illuminated Detection And Ranging) 2, a gyro sensor 3, an acceleration sensor 4, and a GPS. and a group of sensors such as the receiver 5 .

The in-vehicle device 1 is electrically connected to a group of sensors such as the lidar 2, the gyro sensor 3, the acceleration sensor 4, and the GPS receiver 5, and acquires output data from them. It also stores a map database (DB: DataBase) 10 that stores road data and feature information about features provided near roads. Then, the in-vehicle device 1 estimates the position of the vehicle based on the output data and the map DB 10 described above, and performs control related to driving assistance of the vehicle such as automatic driving control based on the position estimation result. Also, the vehicle-mounted device 1 estimates the posture of the rider 2 based on the output of the rider 2 and the like. Based on this estimation result, the in-vehicle device 1 performs processing such as correcting each measurement value of the point cloud data output by the rider 2 . The in-vehicle device 1 is an example of the "attitude estimation device" in the present invention.

The lidar 2 discretely measures the distance to an object existing in the outside world by emitting a pulsed laser over a predetermined angular range in the horizontal and vertical directions, and generates a three-dimensional point indicating the position of the object. Generate group information. In this case, the lidar 2 includes an irradiation unit that irradiates laser light while changing the direction of irradiation, a light receiving unit that receives reflected light (scattered light) of the irradiated laser light, and scan data based on light receiving signals output by the light receiving unit. and an output unit for outputting The scan data is generated based on the irradiation direction corresponding to the laser beam received by the light receiving unit and the distance to the object in the irradiation direction of the laser beam specified based on the above-mentioned light receiving signal, and is sent to the vehicle-mounted device 1. supplied. The rider 2 is an example of a "measuring device" in the present invention.

FIG. 2 is a block diagram showing the functional configuration of the vehicle-mounted device 1. As shown in FIG. The in-vehicle device 1 mainly has an interface 11 , a storage section 12 , an input section 14 , a control section 15 and an information output section 16 . These elements are interconnected via bus lines.

The interface 11 acquires output data from sensors such as the lidar 2 , the gyro sensor 3 , the acceleration sensor 4 , and the GPS receiver 5 and supplies the data to the control unit 15 . In addition, the interface 11 supplies a signal related to vehicle travel control generated by the control unit 15 to an electronic control unit (ECU) of the vehicle.

The storage unit 12 stores programs executed by the control unit 15 and information necessary for the control unit 15 to execute predetermined processing. In this embodiment, the storage unit 12 has a map DB 10 and rider installation information IL. The map DB 10 is a database containing, for example, road data, facility data, feature data around roads, and the like. The road data includes lane network data for route search, road shape data, traffic regulation data, and the like. The feature data includes information (for example, position information and type information) such as signboards such as road signs, road markings such as stop lines, road division lines such as white lines, and structures along the road. The feature data may also include highly accurate point group information of features for use in estimating the position of the vehicle. In addition, the map DB may store various data necessary for position estimation.

The rider installation information IL is information about the relative attitude and position of the rider 2 with respect to the vehicle at a certain reference time (for example, when there is no attitude deviation such as immediately after alignment adjustment of the rider 2). In this embodiment, the posture of the rider 2 and the like is represented by roll angle, pitch angle, and yaw angle (that is, Euler angle). The rider installation information IL may be updated based on the estimation result when the attitude estimation process of the rider 2, which will be described later, is performed.

The input unit 14 is a button, a touch panel, a remote controller, a voice input device, etc. for user operation, and accepts input specifying a destination for route search, input specifying automatic driving on and off, etc. . The information output unit 16 is, for example, a display, a speaker, or the like that outputs based on the control of the control unit 15 .

The control unit 15 includes a CPU that executes programs and the like, and controls the entire in-vehicle device 1 . The control unit 15 estimates the position of the vehicle based on the output signal of each sensor supplied from the interface 11 and the map DB 10, and performs control related to vehicle driving support including automatic driving control based on the estimated result of the vehicle position. etc. At this time, when the output data of the rider 2 is used, the control unit 15 sets the measurement data output by the rider 2 based on the orientation and position included in the installation information of the rider 2 recorded in the rider installation information IL. , the coordinate system based on the rider 2 is converted into a coordinate system based on the vehicle (hereinafter referred to as the "reference coordinate system"). Furthermore, in this embodiment, the control unit 15 estimates the current attitude of the rider 2 with respect to the vehicle (that is, at the time of the processing reference), calculates the amount of change in the attitude recorded in the rider installation information IL, and The measurement data output by the rider 2 is corrected based on the amount. Thereby, even if the attitude of the rider 2 deviates, the control unit 15 corrects the measurement data output by the rider 2 so as not to be affected by the deviation. The control unit 15 is an example of a "computer" that executes programs in the present invention.

[Coordinate system conversion]
The three-dimensional coordinates indicated by each measurement point of the three-dimensional point cloud data acquired by the rider 2 are expressed in a coordinate system (also referred to as a "rider coordinate system") based on the position and orientation of the rider 2, It is necessary to convert to a coordinate system based on the position and orientation of the vehicle (also called a "vehicle coordinate system"). Here, the conversion between the lidar coordinate system and the vehicle coordinate system will be described.

FIG. 3 is a diagram showing the relationship between the vehicle coordinate system and the rider coordinate system represented by two-dimensional coordinates. Here, the vehicle coordinate system has the center of the vehicle as the origin, and has a coordinate axis " _xb " along the traveling direction of the vehicle and a coordinate axis " _yb " along the lateral direction of the vehicle. The lidar coordinate system also has a coordinate axis “x _L ” along the front direction of the rider 2 (see arrow A2) and a coordinate axis “y _L ” along the lateral direction of the rider 2 .

Here, when the yaw angle of the rider 2 with respect to the vehicle coordinate system is “L _ψ ” and the position of the rider 2 is [L _x , _Ly ] ^T , the measurement point [x _b (k), y _b (k)] ^T is converted to the coordinates [x _L (k), y _L (k)] ^T in the lidar coordinate system by the following equation (1) using the rotation matrix “C _ψ ” converted.

一方、ライダ座標系から車両座標系への変換は、回転行列の逆行列（転置行列）を用いればよい。よって、ライダ座標系で取得した時刻ｋの計測点［ｘ_Ｌ（ｋ）、ｙ_Ｌ（ｋ）］^Ｔは、以下の式（２）により車両座標系の座標［ｘ_ｂ（ｋ）、ｙ_ｂ（ｋ）］^Ｔに変換することが可能である。

On the other hand, the inverse matrix (transposed matrix) of the rotation matrix may be used for conversion from the lidar coordinate system to the vehicle coordinate system. Therefore, the measurement point [x _L (k), y _L (k)] ^T at time k acquired in the rider coordinate system is obtained by the following equation (2), and the coordinates [x _b (k), y _b (k)] can be transformed into ^T.

図４は、３次元座標により表された車両座標系とライダ座標系との関係を示す図である。ここでは、座標軸ｘ_ｂ、ｙ_ｂに垂直な座標軸を「ｚ_ｂ」、座標軸ｘ_Ｌ、ｙ_Ｌに垂直な座標軸を「ｚ_Ｌ」とする。

FIG. 4 is a diagram showing the relationship between the vehicle coordinate system and the rider coordinate system represented by three-dimensional coordinates. Here, the coordinate axis perpendicular to the coordinate axes _xb and _yb is " _zb ", and the coordinate axis perpendicular to the coordinate axes _xL and _yL is " _zL ".

_The roll angle of the rider 2 with respect to the vehicle coordinate system is “ _{L φ} ”, the pitch angle is “L _θ ”, and the yaw angle is _“ L _ψ ”. When the position is “L _y ” and the position on the coordinate axis z _b is “L _z ”, the measurement point [x _b (k), y _b (k), z _b ( k)] ^T is the following equation (3) using the direction cosine matrix “C” represented by the rotation matrices “C _φ ”, “C _θ ”, and “C _ψ ” corresponding to roll, pitch, and yaw to coordinates [x _L (k), y _L (k), z _L (k)] ^T in the lidar coordinate system.

一方、ライダ座標系から車両座標系への変換は、方向余弦行列の逆行列（転置行列）を用いればよい。よって、ライダ座標系で取得した時刻ｋの計測点［ｘ_Ｌ（ｋ）、ｙ_Ｌ（ｋ）、ｚ_Ｌ（ｋ）］^Ｔは、以下の式（４）により車両座標系の座標［ｘ_ｂ（ｋ）、ｙ_ｂ（ｋ）、ｚ_ｂ（ｋ）］^Ｔに変換することが可能である。

On the other hand, the inverse matrix (transposed matrix) of the direction cosine matrix may be used for conversion from the rider coordinate system to the vehicle coordinate system. Therefore, the measurement point [x _L (k), y _L (k), z _L (k)] ^T at the time k acquired in the rider coordinate system is obtained by the following equation (4), the coordinate [x _b (k), y _b (k), z _b (k)] ^T.

以下では、ライダ座標系から車両座標系へ変換するための式（４）に用いられるロール角Ｌ_φ、ピッチ角Ｌ_θ、ヨー角Ｌ_ψの各推定方法について順に説明する。

Methods for estimating roll angle L _φ , pitch angle L _θ , and yaw angle L _ψ used in equation (4) for transforming from the rider coordinate system to the vehicle coordinate system will be described in order below.

[Estimation of roll angle]
First, a method for estimating the roll angle _Lφ of the rider 2 will be described. The in-vehicle device 1 selects a square (rectangular) road sign having a plane perpendicular to the travel direction of the travel road (that is, the front of the travel road) as a feature to be detected by the rider 2 ("detection The roll angle _Lφ of the rider 2 is estimated by calculating the inclination of the road sign based on the measured point group measured by the rider 2 .

FIG. 5 shows a bird's-eye view of the surroundings of the vehicle when the vehicle runs near the road sign 22 which is the detection target feature Ftag when estimating the roll angle of the rider 2 . In FIG. 5, a dashed circle 20 indicates a range in which a feature can be detected by the lidar 2 (also referred to as a "lidar detection range"), and a dashed frame 21 detects a road sign 22 that is a detection target feature Ftag. shows the detection window for

In this case, the vehicle-mounted device 1 refers to the map DB 10 to recognize that a rectangular road sign 22 having a plane perpendicular to the travel direction of the travel path exists within the rider detection range, and detects the road sign. A detection window indicated by a dashed frame 21 is set based on the feature data related to 22 . In this case, the in-vehicle device 1 stores, for example, the type information of the road sign serving as the detection target feature Ftag in advance, and determines whether or not the road sign of the type indicated by the type information exists within the rider detection range. is determined by referring to the feature data in the map DB 10 . Note that when the feature data in the map DB 10 includes information on the size and/or shape of the feature, the in-vehicle device 1 determines the size and/or shape of the road sign 22 serving as the detection target feature Ftag. The shape information may be used to determine the size and/or shape of the detection window. Then, the in-vehicle device 1 detects the measurement point group (also referred to as “target measurement point group Ptag”) existing within the set detection window among the measurement point groups measured by the rider 2 as the measurement point group of the road sign 22 . Extract as

6A to 6C show target measurement point group Ptag and scanning line 23 of laser light by rider 2 (that is, point irradiated with laser light) when the measured surface of road sign 22 is viewed from the front. line connecting series). For ease of explanation, the case where the initial attitude angles L _φ , L _θ , and L _ψ of the rider 2 are 0 degree is taken as an example. Here, FIG. 6A shows the relationship between the object measurement point group Ptag and the scanning line 23 of the rider 2 when there is no shift in the roll direction of the rider 2, and FIGS. shows the relationship between the object measurement point group Ptag and the scanning line 23 of the rider 2 when there is a shift ΔL _φ in the roll direction. Here, FIG. 6B shows the relationship between the object measurement point group Ptag and the scanning line 23 of the rider 2 in the _yb - _zb plane of the vehicle coordinate system, and FIG. 2 shows the relationship between the object measurement point group Ptag and the scanning line 23 of the rider 2 in the y _b '-z _b ' plane of . For convenience of explanation, it is assumed that the longitudinal direction of the road sign 22 is parallel to the _yb axis and the vehicle is on a flat road when the rider 2 is not displaced. .

As shown in FIG. 6A, when there is no shift in the roll angle of the rider 2, the rider 2 scans the road sign 22 in parallel with the lateral direction (that is, the horizontal direction or the longitudinal direction). In this case, the number of point groups in the vertical direction of the target measurement point group Ptag is the same number (six in FIG. 6A) regardless of the position in the horizontal direction.

On the other hand, when the roll angle of the rider 2 is deviated, the scanning line is inclined with respect to the road sign 22 . Here, the longitudinal direction of the road sign 22 is not tilted with respect to the vehicle, and the scanning line of the rider 2 is tilted with respect to the vehicle. As shown, the yb _- axis and the longitudinal direction of the road sign 22 are parallel, and the scanning line of the rider 2 is tilted with respect to the yb _- axis. On the other hand, when observed in the reference coordinate system, as shown in FIG. 6C, the _yb ' axis and the scanning line of the rider 2 are parallel, and the longitudinal direction of the road sign 22 is aligned with the _yb ' axis. and tilt.

Next, processing after extraction of the target measurement point group Ptag will be described. First, the in-vehicle device 1 determines the received light intensity for each measurement point of the target measurement point group Ptag using a predetermined threshold value, thereby excluding the measurement points corresponding to the received light intensity that is equal to or less than the predetermined threshold value from the target measurement point group Ptag. do. Then, the vehicle-mounted device 1 extracts a point group forming an outer edge (also referred to as an "outer edge point group Pout") from the target measurement point group Ptag after exclusion processing based on the received light intensity. For example, the in-vehicle device 1 extracts, as the outer edge point group Pout, the measurement points of the target measurement point group Ptag that do not have adjacent measurement points in at least one of the up, down, left, and right directions.

FIG. 7A shows the target measurement point group Ptag in the lidar coordinate system, and FIG. 7B shows the target measurement point group Ptag after excluding measurement points corresponding to received light intensities equal to or lower than a predetermined threshold. show. FIG. 7(C) shows the outer edge point group Pout extracted from the target measurement point group Ptag shown in FIG. 7(B). As shown in FIGS. 7A and 7B, of the measurement points shown in FIG. 7A, the received light intensity is is equal to or less than a predetermined threshold, and is excluded from the target measurement point group Ptag. In addition, as shown in FIGS. 7B and 7C, measurement points that do not have adjacent measurement points in at least one of the vertical and horizontal directions are extracted as the outer edge point group Pout.

Next, the in-vehicle device 1 calculates the inclination of the road sign 22 in the longitudinal direction from the outer edge point group Pout. Here, since the outer edge point group Pout forms a quadrangle, the outer edge point group Pout is classified for each side of this quadrangle, and from the outer edge point group Pout classified for each side, a regression analysis method such as the least squares method Calculate the slope of the approximation straight line for each side.

FIG. 8(A) shows a diagram obtained by extracting the outer edge point group Pout forming the upper side of the quadrangle and drawing a straight line by the method of least squares. FIG. 8B shows a diagram in which the outer edge point group Pout forming the base of the quadrangle is extracted and a straight line is drawn by the method of least squares. FIG. 8(C) shows a diagram obtained by extracting the outer edge point group Pout forming the left side of the quadrangle and drawing a straight line by the method of least squares. FIG. 8(D) shows a diagram in which the outer edge point group Pout forming the right side of the quadrangle is extracted and a straight line is drawn by the method of least squares.

In this case, the vehicle-mounted device 1 calculates each straight line shown in FIGS. 8A to 8D, and calculates the inclinations "φ ₁ " to "φ ₄ " of each straight line with respect to the _yb ' axis. Then, the in-vehicle device 1 calculates the average of these inclinations "φ ₁ " to "φ ₄ " as the inclination "φ" of the road sign 22, as shown in the following equation (5).

ここで、四角形の左辺及び右辺は、上辺及び底辺に対して約９０°ずれているため、式（５）では、四角形の左辺及び右辺を構成する外縁点群Ｐｏｕｔから求めた直線の傾きφ_３及びφ_４に対して「π／２」だけ補正している。

Here, since the left and right sides of the quadrangle are deviated from the top and bottom sides by about 90°, in equation (5), the inclination of the straight line obtained from the outer edge point group Pout that constitutes the left and right sides of the quadrangle is φ ₃ and _φ4 are corrected by "π/2".

Preferably, when obtaining the inclination φ, the in-vehicle device 1 may weight the inclinations φ ₁ to φ ₄ based on the number of the outer edge point groups Pout forming each side of the quadrangle. In this case, the number of outer edge point groups Pout forming the upper side of the quadrangle is "n ₁ ", the number of outer edge point groups Pout forming the base of the quadrangle is "n ₂ ", and the number of outer edge point groups Pout forming the left side of the quadrangle is Assuming that the number is “n ₃ ” and the number of the outer edge point group Pout forming the right side of the quadrangle is “n ₄ ”, the in-vehicle device 1 performs weighted averaging based on the following formula (6) to obtain the slope φ calculate.

他の好適な例では、車載機１は、傾きφを求める場合、ライダ２の水平方向の対象計測点群Ｐｔａｇの間隔「Ｉ_Ｈ」と垂直方向の対象計測点群Ｐｔａｇの間隔「Ｉ_Ｖ」とに基づいて、各傾きφ_１～φ_４に重み付けをしてもよい。ここで、水平間隔Ｉ_Ｈ及び垂直間隔Ｉ_Ｖは、それぞれ図７（Ｂ）に示される間隔となる。この場合、車載機１は、垂直間隔Ｉ_Ｖが長いほど、四角形の上辺及び底辺の傾きφ_１及びφ_２の精度が低くなり、水平間隔Ｉ_Ｈが長いほど、四角形の左辺及び右辺の傾きφ_３及びφ_４の精度が低くなるとみなし、以下の式（７）に示すように、水平間隔Ｉ_Ｈ及び垂直間隔Ｉ_Ｖの逆数を用いて重み付け平均を行うことで、傾きφを算出する。

In another preferred example, when the in-vehicle device 1 obtains the inclination φ, the interval “I _H ” between the group of target measurement points Ptag in the horizontal direction of the rider 2 and the interval “ _IV ” between the group of target measurement points Ptag in the vertical direction are Each of the slopes φ ₁ to φ ₄ may be weighted based on and. Here, the horizontal interval _IH and the vertical interval _IV are the intervals shown in FIG. 7B. In this case, the vehicle-mounted device 1 reduces the accuracy of the slopes φ1 and _φ2 of the top and bottom sides of the rectangle as the vertical interval _IV increases, and decreases the slopes φ1 and _φ2 of the left and right sides of the rectangle as the horizontal interval _IH increases. ₃ and φ ₄ are considered to be less accurate, and the slope φ is calculated by performing a weighted average using the reciprocal of the horizontal interval _IH and the vertical interval _IV as shown in Equation (7) below.

また、車載機１は、四角形の各辺を構成する点群の数ｎ_１～ｎ_４及び垂直間隔Ｉ_Ｖ及び水平間隔Ｉ_Ｈの両方を加味した重み付け平均により傾きφを算出してもよい。この場合、車載機１は、以下の式（８）に基づき重み付け平均を行うことで、傾きφを算出する。

Further, the in-vehicle device 1 may calculate the inclination φ by a weighted average taking into account the numbers n ₁ to n ₄ of the point groups forming each side of the quadrangle and both the vertical interval _IV and the horizontal interval _IH . In this case, the vehicle-mounted device 1 calculates the slope φ by performing weighted averaging based on the following equation (8).

また、車載機１は、ジャイロセンサ３及び加速度センサ４の出力に基づき車体のロール角「φ_０」を算出し、式（５）～（８）のいずれかにより算出した傾きφとの差分をとることで、ライダ２のロール角Ｌ_φを算出する。この場合、車載機１は、以下の式（９）に基づきライダ２のロール角Ｌ_φを算出する。

Further, the in-vehicle device 1 calculates the roll angle “φ ₀ ” of the vehicle body based on the outputs of the gyro sensor 3 and the acceleration sensor 4, and calculates the difference from the inclination φ calculated by any of the formulas (5) to (8). Then, the roll angle _Lφ of the rider 2 is calculated. In this case, the vehicle-mounted device 1 calculates the roll angle _Lφ of the rider 2 based on the following equation (9).

これにより、車載機１は、車両の傾きの影響を排除したライダ２のロール角Ｌ_φを好適に算出することができる。

As a result, the vehicle-mounted device 1 can suitably calculate the roll angle _Lφ of the rider 2 excluding the influence of the tilt of the vehicle.

Preferably, considering the possibility that the road sign 22, which is the detection target feature Ftag, is tilted with respect to the horizontal direction, the measurement error, etc., the vehicle-mounted device 1 calculates the roll angle _Lφ of the rider 2 as described above. , a number of road signs serving as the detection target feature Ftag, and average them. As a result, the vehicle-mounted device 1 can suitably calculate the likely inclination _Lφ of the rider 2 in the roll direction.

In addition, since the roll angle φ ₀ of the vehicle body can be regarded as 0° on average, it becomes unnecessary to obtain the roll angle φ ₀ of the vehicle body by lengthening the above-described averaging time. For example, when obtaining N sufficiently large inclinations φ, the vehicle-mounted device 1 may determine the roll angle L _φ of the rider 2 by the following equation (10).

上述のＮは、車体ロール角φ_０の平均が略０°とみなすことができるサンプル数であり、例えば実験等に基づき予め定められる。

The above-mentioned N is the number of samples for which the average of the vehicle body roll angles φ ₀ can be regarded as approximately 0°, and is determined in advance based on, for example, experiments.

FIG. 9 is a flowchart showing the procedure of roll angle estimation processing.

First, the in-vehicle device 1 refers to the map DB 10 and sets a detection window for the detection target feature Ftag around the vehicle (step S101). In this case, the in-vehicle device 1 selects a square road sign having a plane perpendicular to the travel direction of the travel path as the detection target feature Ftag, and selects the road sign existing within a predetermined distance from the current position. It specifies from map DB10. Then, the in-vehicle device 1 sets a detection window by acquiring position information and the like related to the identified road sign from the map DB 10 .

Next, the in-vehicle device 1 acquires the measurement point group of the rider 2 within the detection window set in step S101 as the point group of the detection target feature Ftag (that is, the target measurement point group Ptag) (step S102). Then, the in-vehicle device 1 extracts the outer edge point group Pout forming the outer edge of these point groups from the target measurement point group Ptag acquired in step S102 (step S103).

Then, the in-vehicle device 1 calculates the lateral inclination φ of the outer edge point group Pout extracted in step S103 (step S104). In this case, as described with reference to FIGS. 8A to 8D, the in-vehicle device 1 obtains the inclination φ _{1 ∼φ4} is obtained, and the slope φ is calculated using any one of the formulas (5) to (8).

After that, the in-vehicle device 1 calculates the roll angle _Lφ of the rider 2 by averaging the inclination φ and/or subtracting it from the roll angle _φ0 of the vehicle body (step S105). In the above-described averaging process, the vehicle-mounted device 1 acquires a plurality of inclinations φ by executing the processes of steps S101 to S104 on a plurality of detection target features Ftag, and calculates the average value of the acquired plurality of inclinations φ. is calculated as the slope L _φ .

[Estimation of pitch angle]
Next, a method for estimating the pitch angle _Lθ of the rider 2 will be described. The in-vehicle device 1 regards a square-shaped road sign having a plane perpendicular to the width direction of the travel road (that is, a plane facing laterally to the travel road) as a detection target feature Ftag, and determines the inclination of the road sign to the rider 2. The pitch angle of the rider 2 is estimated by calculating based on the measurement point group output by .

FIG. 10 shows a bird's-eye view of the surroundings of the vehicle when the vehicle runs near the road sign 24, which is the detection target feature Ftag when estimating the pitch angle of the rider 2. FIG. In FIG. 10, a dashed line circle 20 indicates a lidar detection range in which a feature can be detected by the lidar 2, and a dashed line frame 21 indicates a detection window set for a road sign 24 that is a detection target feature Ftag.

In this case, the in-vehicle device 1 refers to the map DB 10 to determine that a rectangular road sign 24 having a plane perpendicular to the width direction of the travel path exists within the rider detection range indicated by the dashed circle 20. A detection window indicated by a dashed frame 21 is set based on the positional information about the road sign 24 and the like. In this case, the in-vehicle device 1 stores, for example, the type information of the road sign serving as the detection target feature Ftag in advance, and determines whether or not the road sign of the type indicated by the type information exists within the rider detection range. is determined by referring to the feature data in the map DB 10 . Then, the in-vehicle device 1 extracts the measurement point group existing within the set detection window from the measurement point group output by the rider 2 as the target measurement point group Ptag.

Next, the in-vehicle device 1 estimates the pitch angle of the rider 2 by performing the same process as the roll angle estimation method on the extracted target measurement point group Ptag.

Specifically, first, the in-vehicle device 1 determines the received light intensity for each measurement point of the target measurement point group Ptag by using a predetermined threshold value, and measures the measurement points corresponding to the received light intensity that is equal to or less than the predetermined threshold value. Exclude from point group Ptag. Then, the in-vehicle device 1 extracts the outer edge point group Pout forming the outer edge from the target measurement point group Ptag after the exclusion process based on the received light intensity. Next, the in-vehicle device 1 calculates the inclinations “θ ₁ ” to “θ ₄ ” of the approximation line of each side of the quadrangle formed by the outer edge point group Pout by a regression analysis technique such as the least squares method. Then, the in-vehicle device 1 calculates the inclination “θ” of the road sign 22 by performing averaging processing based on the inclinations θ ₁ to θ ₄ . Note that the in-vehicle device 1 determines at least one of the numbers n ₁ to n ₄ of the point groups forming each side of the quadrangle, the vertical interval I _V and the horizontal interval I _H as in formulas (6) to (8). You may calculate inclination (theta) by the weighted average which added consideration.

In addition, the in-vehicle device 1 calculates the pitch angle “θ ₀ ” of the vehicle body based on the outputs of the gyro sensor 3 and the acceleration sensor 4, and calculates the pitch angle L _θ of the rider 2 by taking the difference from the inclination θ. . In this case, the vehicle-mounted device 1 calculates the inclination L _θ of the rider 2 in the pitch direction based on the following equation (11).

これにより、車載機１は、車両の傾きの影響を排除したライダ２のピッチ角Ｌ_θを好適に算出することができる。また、好適には、検出対象地物Ｆｔａｇである道路標識２４が水平方向に対して傾いている可能性や計測誤差等を勘案し、車載機１は、上記のライダ２のピッチ角Ｌ_θの算出を、検出対象地物Ｆｔａｇとなる多くの道路標識に対して実施し、それらを平均化するとよい。これにより、車載機１は、確からしいライダ２のピッチ角Ｌ_θを好適に算出することができる。ここで、車体のピッチ角θ_０は平均すると０°と見なせるため，上述の平均化の時間を長くすることにより、車体のピッチ角度θ_０を取得する必要がなくなる。例えば、Ｎ個分の傾きθを取得した場合、ライダ２のピッチ角Ｌ_θは、以下の式（１２）により表される。

As a result, the in-vehicle device 1 can suitably calculate the pitch angle L _θ of the rider 2 excluding the influence of the inclination of the vehicle. In addition, preferably, considering the possibility that the road sign 24, which is the detection target feature Ftag, is tilted with respect to the horizontal direction, the measurement error, etc., the vehicle-mounted device 1 determines the pitch angle L _θ It is preferable to perform the calculation for many road signs serving as the detection target feature Ftag and average them. As a result, the vehicle-mounted device 1 can suitably calculate the probable pitch angle L _θ of the rider 2 . Here, since the pitch angle θ ₀ of the vehicle body can be regarded as 0° on average, it becomes unnecessary to obtain the pitch angle θ ₀ of the vehicle body by lengthening the above-described averaging time. For example, when N inclinations θ are obtained, the pitch angle L _θ of the rider 2 is expressed by the following equation (12).

FIG. 11 is a flowchart showing the procedure of pitch angle estimation processing.

First, the in-vehicle device 1 refers to the map DB 10 and sets a detection window for the detection target feature Ftag around the vehicle (step S201). In this case, the in-vehicle device 1 selects a square road sign having a plane perpendicular to the width direction of the travel path as the detection target feature Ftag, and selects the above-described road sign existing within a predetermined distance from the current position. It specifies from map DB10. Then, the in-vehicle device 1 sets a detection window by acquiring position information and the like related to the identified road sign from the map DB 10 .

Next, the in-vehicle device 1 acquires the measurement point group of the rider 2 within the detection window set in step S201 as the point group of the detection target feature Ftag (that is, the target measurement point group Ptag) (step S202). Then, the in-vehicle device 1 extracts the outer edge point group Pout forming the outer edge of the point group from the point group of the detection target feature Ftag acquired in step S202 (step S203).

Then, the in-vehicle device 1 calculates the inclination θ in the longitudinal direction of the outer edge point group Pout extracted in step S203 (step S204). After that, the in-vehicle device 1 calculates the inclination L _θ of the rider 2 in the pitch direction by averaging the inclination θ and/or subtracting it from the pitch angle θ ₀ of the vehicle body (step S205). In the above-described averaging process, the in-vehicle device 1 calculates a plurality of slopes θ by executing the processes of steps S201 to S204 on a plurality of target measurement point groups Ptag, and uses the average value of these slopes as the slope L _θ . calculate.

[Estimation of Yaw Angle]
Next, a method for estimating the yaw angle L _ψ of the rider 2 will be described. The in-vehicle device 1 regards a marking line such as a white line as a detection target feature Ftag, and calculates the direction of the center line of the marking line based on the measurement point group measured by the rider 2, thereby obtaining the yaw angle L _ψ of the rider 2. presume.

FIG. 12 shows a bird's-eye view of the surroundings of the vehicle when the vehicle travels along the division line that serves as the detection target feature Ftag when estimating the yaw angle of the rider 2 . In this example, a continuous line 30, which is a continuous lane marking, exists on the left side of the vehicle, and a broken line 31, which is an intermittent lane marking, exists on the right side of the vehicle. An arrow 25 indicates the reference direction of the yaw direction when the rider 2 is not displaced in the yaw direction, and is the direction obtained by rotating the coordinate axis "x _L " of the rider 2 by the yaw angle L _Ψ of the rider 2 . , coincide with the coordinate axis “x _b ” along the direction of travel of the vehicle. That is, the yaw angle L _Ψ of the rider 2 indicates a yaw angle (also referred to as a "yaw direction reference angle") that is considered to be the running direction of the vehicle if there is no deviation. On the other hand, the arrow 26 indicates the yaw direction reference angle when the rider 2 deviates in the yaw direction ΔL _Ψ , and unless the yaw direction reference angle is corrected, it will not match the running direction of the vehicle. The yaw direction reference angle is stored in advance in the storage unit 12 or the like, for example.

In this case, for example, the map DB 10 contains lane marking information indicating discrete coordinate positions of the continuous line 30 and discrete coordinate positions of the dashed line 31 . Then, the in-vehicle device 1 refers to the lane marking information, and finds the closest coordinate position from a position a predetermined distance (for example, 5 m) away from the vehicle in each of the left front, left rear, right front, and right rear directions of the vehicle. A rectangular area centered on the extracted coordinate position is set as a detection window indicated by broken-line frames 21A to 21D.

Next, the in-vehicle device 1 extracts, from the measurement point group measured by the rider 2, points on the road surface and having a reflection intensity equal to or greater than a predetermined threshold within the detection window as a target measurement point group Ptag. Then, the in-vehicle device 1 obtains the center point of the target measurement point group Ptag for each scanning line, and calculates a straight line (see arrows 27A to 27D) passing through these center points for each detection window as the center line of the marking line. .

FIG. 13A is a diagram showing the center point of the detection window of the dashed frame 21A with a circle, and FIG. 13B is a diagram showing the center point of the detection window of the dashed frame 21B with a circle. be. Also, in FIGS. 13A and 13B, scanning lines 28 of laser light by the lidar 2 are clearly shown. In addition, since the laser beam of the lidar 2 incident on the marking line is emitted obliquely downward, the distance between the scanning lines becomes shorter as the vehicle approaches the vehicle.

As shown in FIGS. 13A and 13B, the in-vehicle device 1 calculates the center point for each scanning line 28 . The center point may be the position obtained by averaging the coordinate positions indicated by the measurement points for each scanning line 28, or may be the middle point between the leftmost and rightmost measurement points on each scanning line 28. It may be a measuring point existing in the middle of the above. Then, the in-vehicle device 1 calculates center lines (see arrows 27A and 27B) of the partition lines for each detection window from the calculated center points by a regression analysis technique such as the least squares method.

Then, the in-vehicle device 1 calculates angles formed by the center line of the lane marking calculated for each detection window and the yaw direction reference angle (see arrow 26 in FIG. 12) as inclinations "ψ ₁ " to "ψ ₄ ." . FIGS. 14A to 14D are diagrams showing the slopes ψ ₁ to ψ ₄ , respectively. As shown in FIGS. 14A to 14D, the inclinations ψ ₁ to ψ ₄ correspond to angles formed by arrows 27A to 27D pointing the center lines of the compartment lines and arrow 26 pointing the yaw direction reference angle. .

Then, the vehicle-mounted device 1 averages the slopes ψ ₁ to ψ ₄ to calculate the slope ψ as shown in the following equation (13).

また、車載機１は、区画線の中心線の算出に用いた中心点の数に基づき傾きψ_１～ψ_４を重み付けしてもよい。この場合、傾きψ_１となる区画線の中心線の算出に用いた中心点の数を「ｎ_１」、傾きψ_２となる区画線の中心線の算出に用いた中心点の数を「ｎ_２」、傾きψ_３となる区画線の中心線の算出に用いた中心点の数を「ｎ_３」、傾きψ_４となる区画線の中心線の算出に用いた中心点の数を「ｎ_４」とすると、車載機１は、以下の式（１４）に基づき傾きψを算出する。

Further, the vehicle-mounted device 1 may weight the slopes ψ ₁ to ψ ₄ based on the number of center points used to calculate the center line of the lane marking. In this case, the number of center points used to calculate the center line of the plot line with a slope of ψ ₁ is “n ₁ ”, and the number of center points used to calculate the center line of the plot line with a slope of ψ ₂ is “n ₂ ”, the number of center points used to calculate the center line of the plot line with a slope of ψ ₃ is “n ₃ ”, and the number of center points used to calculate the center line of the plot line with a slope of ψ ₄ is “n ₄ ”, the vehicle-mounted device 1 calculates the inclination ψ based on the following equation (14).

これにより、車載機１は、少ない中心点により算出した区画線（図１２では破線３１）の中心線の傾きの重みを相対的に低くし、傾きψを的確に求めることができる。よって、図１３（Ａ）の検出状況に対して、図１３（Ｂ）の検出状況では重みが低くなる。なお、図１２～図１４の説明では、同時に４箇所の区画線に対する傾きψ_１～ψ_４を算出する例について示したが、少なくとも１箇所の区画線に対する傾きを算出すればよい。

As a result, the in-vehicle device 1 relatively lowers the weight of the inclination of the center line of the division line (broken line 31 in FIG. 12) calculated from a small number of center points, and can accurately obtain the inclination ψ. Therefore, the weight is lower in the detection state of FIG. 13(B) than in the detection state of FIG. 13(A). In the description of FIGS. 12 to 14, an example of calculating the slopes ψ ₁ to ψ ₄ with respect to the four lane markings at the same time was shown, but the slope with respect to at least one lane marking may be calculated.

Next, a method for calculating the yaw angle L _ψ of the rider 2 from the inclination ψ will be described. Since demarcation lines such as white lines are drawn along the road, the yaw angle of the vehicle body can be considered to be substantially constant (that is, parallel to the travel road). Therefore, unlike the method of estimating the roll angle and the pitch angle, the in-vehicle device 1 calculates the inclination ψ as the deviation ΔL _ψ of the yaw angle of the rider 2 (that is, “ΔL _ψ =ψ”). However, during normal driving, there are lane changes and body swaying within the lane. Therefore, the in-vehicle device 1 preferably executes a process of calculating the inclination of the center line for many lane markings and averages it. As a result, the in-vehicle device 1 can suitably calculate the probable yaw angle deviation ΔL _ψ of the rider 2 . For example, when N inclinations ψ are acquired, the yaw angle L _ψ of the rider 2 is expressed by the following equation (15).

ここで、「Ｌ_Ψ０」は、ヨー角にずれが生じていないときのヨー角Ｌ_Ψ（即ちヨー方向基準角）を示す。

Here, "L _Ψ0 " indicates the yaw angle L _Ψ (that is, the yaw direction reference angle) when there is no deviation in the yaw angle.

FIG. 15 is a flowchart showing the procedure of yaw angle estimation processing.

First, the in-vehicle device 1 refers to the map DB 10 and sets one or more detection windows for the lane markings (step S301). Then, the in-vehicle device 1 acquires the measured point group of the rider 2 within the detection window set in step S301 as the point group of the lane marking (that is, the target measured point group Ptag) (step S302).

Then, the in-vehicle device 1 calculates the center line of the lane markings within each set detection window (step S303). In this case, the vehicle-mounted device 1 obtains the center point of the target measurement point group Ptag for each scanning line of the lidar 2, and calculates the center line of the marking line for each detection window based on the least squares method or the like from the center point. Then, the in-vehicle device 1 calculates the angle ψ between the yaw angle reference angle stored in advance in the storage unit 12 or the like and the center line of the partition line (step S304). In this case, if two or more detection windows are set in step S301, the vehicle-mounted device 1 obtains the angle between the yaw angle reference angle and the center line of the lane marking for each detection window, and averages these angles. Alternatively, the above angle ψ is calculated by performing weighted averaging based on the number of center points.

Then, the in-vehicle device 1 averages a plurality of angles ψ calculated by executing steps S301 to S304 for different lane markings, and calculates the yaw angle L _ψ of the rider 2 based on the equation (15) ( step S305).

Here, a specific example of processing based on the estimated values of the roll angle, pitch angle, and yaw angle described above will be described. FIG. 16 is a flow chart showing a specific example of processing based on the estimated roll angle, pitch angle, and yaw angle.

First, the in-vehicle device 1 calculates any of the roll angle L _φ , the pitch angle L _θ , or the yaw angle L _ψ of the rider 2 by executing the processing of the flowcharts of FIGS. It is determined whether or not (step S401). When the in-vehicle device 1 calculates any of the roll angle _Lφ , the pitch angle _Lθ , or the yaw angle _Lψ of the rider 2 (step S401; Yes), the calculated angle is recorded in the rider installation information IL. It is determined whether or not the angle has changed by a predetermined angle or more (step S402). The above-mentioned threshold value is a threshold value for determining whether or not the measurement data of rider 2 can continue to be used by correcting the measurement data of rider 2 in step S404, which will be described later. set.

Then, if the calculated angle has changed by a predetermined angle or more from the angle recorded in the rider installation information IL (step S402; Yes), the vehicle-mounted device 1 uses the output data of the target rider 2 (that is, obstacle detection). and vehicle position estimation, etc.), and the information output unit 16 outputs a warning to the effect that alignment adjustment must be performed again for the target rider 2 (step S403). As a result, the use of the measurement data of the rider 2, whose attitude and position have significantly deviated due to an accident or the like, can reliably prevent a decrease in safety or the like.

On the other hand, if the calculated angle has not changed by a predetermined angle or more from the angle recorded in the rider installation information IL (step S402; No), the vehicle-mounted device 1 changes the calculated angle from the angle recorded in the rider installation information IL. Each measurement value of the point cloud data output by the rider 2 is corrected based on the amount of change in (step S404). In this case, the vehicle-mounted device 1 stores, for example, a map or the like indicating the amount of correction of the measured value with respect to the amount of change described above, and refers to the map or the like to correct the above-described measured value. Alternatively, the measured value may be corrected using a value of a predetermined ratio of the amount of change as the correction amount of the measured value.

As described above, the in-vehicle device 1 according to the present embodiment displays a road sign having a plane perpendicular to the traveling direction of the travel path of the vehicle or a plane perpendicular to the width direction of the travel path to the rider attached to the vehicle. Acquire the target measurement point group Ptag measured by 2. Then, the in-vehicle device 1 adjusts the attachment attitude of the rider 2 to the vehicle based on the inclination of the road sign in the longitudinal direction with reference to the coordinate system based on the installation information of the rider 2, which is calculated from the target measurement point group Ptag. presume. As a result, the in-vehicle device 1 can accurately estimate the attachment attitude of the rider 2 to the vehicle in the roll direction and the pitch direction.

Further, as described above, the in-vehicle device 1 according to the present embodiment obtains a measurement point group obtained by measuring the lane markings by the rider 2 attached to the vehicle, and based on the measurement point group, moves along the traveling direction of the vehicle. Calculate the center line of the zoning line. Then, the in-vehicle device 1 estimates the attachment attitude of the rider 2 to the vehicle based on the direction of the center line of the lane marking with respect to the yaw direction reference angle that the rider 2 uses as a reference. Thereby, the on-vehicle device 1 can suitably estimate the mounting posture of the rider 2 on the vehicle in the yaw direction.

[Modification]
Modifications suitable for the embodiment will be described below. The following modifications may be combined and applied to the embodiment.

(Modification 1)
In the description of step S303 in FIGS. 13 and 15, the in-vehicle device 1 calculates the center line of the marking line by calculating the center point in the width direction of the marking line within the detection window. Alternatively, the vehicle-mounted device 1 may extract the right end point or left end point of the scanning line within the detection window, and calculate a straight line passing through the right end or left end of the partition line. Also by this, the vehicle-mounted device 1 can suitably identify the line parallel to the travel path and calculate the angle ψ.

(Modification 2)
In the yaw angle estimation processing described with reference to FIGS. 12 to 15, the vehicle-mounted device 1 regards the section line as the detection target feature Ftag. Instead of this, or in addition to this, the vehicle-mounted device 1 may regard a curbstone or the like as the detection target feature Ftag. In this way, the in-vehicle device 1 regards any arbitrary feature formed along the running road surface with the surface to be measured parallel to the road surface as the detection target feature Ftag, not limited to the lane markings, as shown in FIGS. The yaw angle estimation process described with reference to FIG. 15 may be executed.

(Modification 3)
In step S404 of FIG. 16, the in-vehicle device 1 corrects each measurement value of the point cloud data output by the rider 2, and instead of correcting each measurement value, the roll angle and pitch calculated by the processing of the flowchart of FIG. 9, FIG. 11, or FIG. Each measurement value of the point cloud data output by the rider 2 may be converted into the vehicle coordinate system based on the angle and the yaw angle.

In this case, the in-vehicle device 1 uses the calculated roll angle L _φ , pitch angle L _θ , and yaw angle L _ψ to convert each measurement value of the point cloud data output by the rider 2 to the rider coordinates based on Equation (4). system may be converted into the vehicle body coordinate system, and vehicle position estimation, automatic driving control, and the like may be executed based on the converted data.

As another example, if each rider 2 is provided with an adjustment mechanism such as an actuator for correcting the posture of each rider 2, the vehicle-mounted device 1 may perform rider installation instead of the processing of step S404 in FIG. Control may be performed to drive the adjustment mechanism so as to correct the attitude of the rider 2 by the amount of deviation from the angle recorded in the information IL.

(Modification 4)
The configuration of the vehicle system shown in FIG. 1 is an example, and the configuration of the vehicle system to which the present invention is applicable is not limited to the configuration shown in FIG. For example, in the vehicle system, instead of having the vehicle-mounted device 1, the electronic control device of the vehicle may execute the processes shown in FIGS. 9, 11, 15, 16 and the like. In this case, the rider installation information IL is stored, for example, in a storage unit within the vehicle, and the electronic control unit of the vehicle is configured to be able to receive output data from various sensors such as the rider 2 .

(Modification 5)
The vehicle system may include multiple riders 2 . In this case, the in-vehicle device 1 estimates the roll angle, pitch angle, and yaw angle of each rider 2 by executing the processing of the flowcharts of FIGS. 9, 11, and 15 for each rider 2. .

1 vehicle-mounted device 2 rider 3 gyro sensor 4 acceleration sensor 5 GPS receiver 10 map DB

Claims (1)

Acquisition means for acquiring a measurement point group obtained by measuring a feature whose surface to be measured is parallel to a road surface on which the mobile body travels and which is formed along the road surface, by a measuring device attached to the mobile body. ,
a calculation means for calculating a line on the surface to be measured along the running direction of the moving body based on the measurement point group;
estimating means for estimating a mounting posture of the measuring device to the moving object based on the direction of the line with respect to the direction that the measuring device uses as a reference;
A posture estimation device having

Images