JP6306429B2 - Steering control system - Google Patents

Description

  One aspect of the present invention relates to a steering control system that performs steering control of a vehicle.

  Conventionally, as a steering control system, a lane line detection unit that detects the shape of a lane line in front of the vehicle (for example, a roadway outer line, a roadway center line, and a lane boundary line), and a vehicle front detected by the lane line detection unit And a steering control unit that executes steering control using the lane marking shape are known. For example, a detection device such as that described in Patent Document 1 may be used for detecting the lane marking in front of the vehicle. In the detection apparatus described in Patent Document 1, it is intended to detect a lane line shape in front of a vehicle by detecting a lane line position based on a road image captured by an in-vehicle camera (sensor). .

JP-A-6-225308

  By the way, in the vehicle steering control, it is generally preferable to calculate a point where the vehicle travels after a predetermined time (for example, after 2 to 3 seconds), and it is required to grasp the lane marking shape in front of the vehicle. It is done. However, in the steering control system using the detection device as described above, for example, when the lane marking ahead of the vehicle is shielded by a preceding vehicle, the lane marking shape ahead of the vehicle is accurately recognized only by the recognition technology using the sensor. May be difficult. Such a problem is particularly likely to occur when a vehicle travels on a highway where a necessary detection range tends to be wide.

  This invention is made | formed in view of the said situation, and makes it a subject to provide the steering control system which can improve the estimation precision of the division line shape in front of a vehicle.

  A steering control system according to an aspect of the present invention is a steering control system that performs steering control of a vehicle, and is detected by a lane line recognition sensor that detects a lane line position of a road on which the vehicle travels, and a lane line recognition sensor. A storage unit for storing the lane line position, and a lane line shape in front of the vehicle is estimated based on the lane line position stored in the storage unit, and vehicle steering control is executed based on the estimated lane line shape in front of the vehicle A lane line position detected by the lane line recognition sensor when the vehicle travels in the first travel zone from the first point behind the travel direction to the current location. When the control unit fits the multi-order curve to the first section lane line position stored in the storage unit, the standard deviation between the obtained multi-order curve and the first section lane line position is smaller than a predetermined threshold value. The multi-order curve Estimating the Luo vehicle front compartment line shape.

  In this steering control system, when a multi-order curve is fitted to a lane line position detected in a travel section where the vehicle has already traveled, and the standard deviation is smaller than a predetermined threshold, the lane line ahead of the vehicle from the multi-order curve Estimate the shape. Therefore, for example, even when the lane marking in front of the vehicle is shielded, it is possible to suppress the estimation accuracy of the lane marking shape in front of the vehicle from being deteriorated, and the estimation accuracy can be improved.

  In addition, the steering control system according to one aspect of the present invention provides a lane marking detected by a lane marking recognition sensor when the vehicle travels in a second travel zone from a second point closer to the current location than the first location to the current location. When the position is set as the second section lane line position, the control unit fits the multi-order curve to the second section lane line position stored in the storage unit when the standard deviation is greater than or equal to the threshold, and the obtained multi-order curve When the standard deviation between the curve and the second section lane marking position is smaller than the threshold value, the lane marking shape in front of the vehicle may be estimated from the multi-order curve.

  When fitting a multi-degree curve to the first section lane line position, if a section having a large curvature change, such as a boundary between a straight line and a curve, is included in the first traveling section, the standard deviation increases. A large standard deviation means that a fitting shift is large, and in this case, there is a possibility that the lane marking shape in front of the vehicle cannot be predicted correctly. On the other hand, according to the steering control system according to one aspect of the present invention, when the standard deviation is equal to or greater than the threshold value, it is detected in the second traveling section from the second point closer to the current location than the first location to the current location. The lane marking shape ahead of the vehicle is estimated based on the second segment lane marking position. Therefore, for example, a section with a large curvature change is excluded from a traveling section to be referred to at the time of fitting, and thus it is possible to cope with a large curvature change.

  According to one aspect of the present invention, it is possible to provide a steering control system that can improve the estimation accuracy of a lane marking in front of a vehicle.

It is a schematic structure figure showing a steering control system concerning one embodiment. It is explanatory drawing explaining the estimation method of the lane marking shape by the steering control system of FIG. It is another explanatory drawing explaining the lane marking shape estimation method by the steering control system of FIG. It is a flowchart explaining the estimation method of the lane marking shape by the steering control system of FIG.

  Hereinafter, embodiments according to one aspect of the present invention will be described in detail with reference to the drawings. In the following description, the same reference numerals are used for the same or corresponding elements, and duplicate descriptions are omitted.

  FIG. 1 is a schematic configuration diagram illustrating a steering control system according to an embodiment. A steering control system 1 shown in FIG. 1 is mounted and used in a vehicle 30 such as a commercial vehicle or a passenger vehicle including a truck and a bus. The steering control system 1 is a system for performing steering control of the vehicle 30. For example, when the driver becomes unable to drive on a highway, the vehicle 30 is stopped as necessary to stop the vehicle 30 on the shoulder. Steering control is performed to guide to the shoulder while changing lanes.

  First, the configuration of each part of the steering control system 1 will be described. As shown in FIG. 1, the steering control system 1 includes a lane marking recognition sensor 11, a storage unit 13, a driver state detection unit 15, a steering actuator 17, and a control unit 19 as main components. .

  The lane marking recognition sensor 11 is a laser sensor attached to, for example, an outer surface on the front side of the vehicle 30 (for example, a side mirror or a bumper), and detects the lane marking position of the road on which the vehicle 30 travels. A lane line is a roadway outer line, a roadway center line, a lane boundary line, or the like drawn using a white or yellow paint to divide a road (lane). This is the position of the lane markings around (left and right). The lane marking recognition sensor 11 is not limited to a laser sensor, and may use image recognition such as an in-vehicle camera.

  The storage unit 13 is a storage medium (memory) for storing the lane line position detected by the lane line recognition sensor 11. In the present embodiment, the storage unit 13 is a first section that is a lane line position detected by the lane line recognition sensor 11 when the vehicle 30 travels in a first traveling section from the first point behind the traveling direction to the current location. The lane line position is stored. In the present embodiment, as an example, the first point is set as a point 100 m behind the current location, and the latest traveling section of the past 100 m is set as the first traveling section. As the storage unit 13, a semiconductor memory, a magnetic storage device, or the like can be used, or an external storage device (for example, a hard disk) can be used.

  The driver state detection unit 15 is a camera that is attached to, for example, the interior of the vehicle 30 and captures an image including the driver's face. The steering actuator 17 is an actuator including, for example, an electric motor, a gear, and the like, and changes the steering angle of the vehicle 30 in accordance with an instruction from the control unit 19.

  The control unit 19 is an electronic control unit (ECU) configured by a computer including, for example, a CPU (Central Processing Unit), a ROM (Read Only Memory), and a RAM (Random Access Memory). The control unit 19 is connected to the lane marking recognition sensor 11, the storage unit 13, the driver state detection unit 15, and the steering actuator 17.

  The control unit 19 sequentially determines whether or not the driver is driving normally based on the image captured by the driver state detection unit 15, and when determining that the driver is unable to drive, the control unit 19 moves the vehicle 30 to the road shoulder. The steering control to guide is executed. This steering control is performed by the control unit 19 controlling the steering actuator 17. And in the control part 19, in the case of the said steering control, the lane line shape ahead of a vehicle is estimated based on the 1st area lane line position memorize | stored in the memory | storage part 13, and the lane line shape ahead of the estimated vehicle is estimated. The steering control of the vehicle 30 is executed based on the above (the method for estimating the lane marking in front of the vehicle will be described below).

  In such steering control, in general, it is preferable to estimate a lane line shape to a point where the vehicle 30 travels after 2 to 3 seconds. For example, it is assumed that the vehicle 30 is traveling at 100 km / h. In this case, it is preferable to estimate the lane marking shape up to a point about 85 m away.

  Next, the operation of the steering control system 1 will be described with reference to FIGS. 2 and 3 are explanatory diagrams for explaining a method for estimating a lane line shape by the steering control system 1 of FIG. 1, and FIG. 4 is a flowchart for explaining a lane line method by the steering control system 1 of FIG. .

  In the steering control system 1, for example, when the vehicle 30 travels on an expressway, as described above, when the vehicle 30 travels in the first travel section (past 100 m), it is detected by the lane marking recognition sensor 11. The first section lane marking position 21 (see FIG. 2) that is the lane marking position is sequentially stored in the storage unit 13. The first section lane line position 21 stored in the storage unit 13 is updated whenever the vehicle 30 travels and the lane line recognition sensor 11 newly detects the lane line position, and is always the latest data. .

  And when it determines with the driver having become unable to drive | operate based on the image imaged by the driver state detection part 15, the lane line shape ahead of a vehicle is based on the 1st area lane line position 21 memorize | stored in the memory | storage part 13. Based on the estimated lane marking shape in front of the vehicle, steering control for guiding the vehicle 30 to the road shoulder is executed. Thereafter, the vehicle 30 is stopped by controlling a brake actuator (not shown).

ｘ_ｉ：検出された区画線位置
ｙ_ｉ：二次曲線で表される区画線位置 Here, a method of estimating the lane marking shape will be described. First, as shown in FIG. 2, a quadratic curve is fitted to the first section marking line position 21. Specifically, as shown in FIG. 3, when it is assumed that n detection points are stored in the storage unit 13 as data representing the first section lane marking position 21, for the n detection points, For example, a quadratic curve is fitted using a least square method. When the standard deviation σ between the obtained secondary curve 23 and the first section lane marking position 21 is smaller than a predetermined threshold S, the lane marking shape ahead of the vehicle is estimated from the secondary curve 23. For example, assuming that the lane marking shape in front of the vehicle is also located on the obtained quadratic curve 23, the lane marking shape in front of the vehicle is estimated. Here, the standard deviation σ is calculated by the following formula 1.

x _i : detected lane line position y _i : lane line position represented by a quadratic curve

  On the other hand, if the standard deviation σ is equal to or greater than the threshold S as a result of the fitting, the fitting is performed while shortening the traveling section to be fitted by a predetermined value (for example, 2 m) until the standard deviation σ is smaller than the threshold S. Do it recursively. That is, as a result of fitting the quadratic curve to the first section lane line position 21, if the standard deviation σ is equal to or greater than the threshold S, the second traveling section from the second point closer to the current location to the current location than the first location A quadratic curve is fitted to the second section lane line position. The second section lane line position is a lane line position detected by the lane line recognition sensor 11 when the vehicle 30 travels in the second traveling section.

  For example, if the standard deviation σ is equal to or greater than the threshold value S as a result of fitting a quadratic curve to n detection points (first section lane line position 21) in the first traveling section of 100 m in the past, A quadratic curve is fitted to n-1 detection points (second section lane marking positions) in the traveling section (when the detection interval of the lane marking recognition sensor 11 is 2 m). And when standard deviation (sigma) is smaller than threshold value S, the lane marking shape ahead of a vehicle is estimated from the obtained quadratic curve. That is, in this example, the second point is a point 98 m behind the current location, and the latest traveling section of the past 98 m is the second traveling section.

  The second section lane line position can be obtained from the first section lane line position 21 stored in the storage unit 13. Furthermore, when the standard curve σ is equal to or greater than the threshold S as a result of fitting the quadratic curve to the second section lane line position, the traveling section is further shortened and the fitting is performed again. For example, a quadratic curve is fitted to n-2 detection points in the third traveling section of the past 96 m.

  As described above, the steering control system 1 predicts the future lane line shape (front of the vehicle) by fitting a quadratic curve to the lane line positions detected in the past (rear of the vehicle). Further, the length of the travel section to be referred to at the time of fitting is determined according to the standard deviation σ. In a portion where the curvature of the road is constant, such as in the middle of a straight line or curve, the standard deviation σ does not increase even if fitting is performed for a long traveling section. However, in a portion where the curvature is changing, such as at the beginning of the curve or at the end of the curve, the standard deviation σ increases when fitting to a long traveling section. When the standard deviation σ is large, it means that the displacement of the fitting is large. In this case, there is a possibility that the lane line shape in front of the vehicle cannot be predicted correctly. Therefore, until the standard deviation σ becomes smaller than the threshold value S, the fitting of the quadratic curve is performed recursively while shortening the travel section to be referred to at the time of fitting.

  Next, a processing procedure by the control unit 19 when estimating the lane marking shape in front of the vehicle will be described with reference to the flowchart of FIG.

  First, the control unit 19 determines whether or not the first section lane line position 21 detected by the lane line recognition sensor 11 in the first traveling section of the past 100 m is stored in the storage unit 13 (S11). As a result of the determination, if it is determined that it is stored, the process proceeds to S12, and if it is determined that it is not stored, the process ends.

  In S12, the control part 19 sets the distance Xm of the travel area referred at the time of fitting to 100 m as an initial value. Then, the control unit 19 fits a quadratic curve to the lane line position in the past Xm (S13), and calculates a standard deviation σ between the lane line position in the past Xm and the obtained quadratic curve (S14). When the standard deviation σ calculated in S14 is smaller than the threshold value S (YES in S15), the control unit 19 estimates the lane marking shape ahead of the vehicle from the quadratic curve obtained in S13 (S17), and ends the process. To do.

  On the other hand, when the standard deviation σ calculated in S14 is greater than or equal to the threshold S (NO in S15), the control unit 19 determines whether or not the distance X is less than a predetermined lower limit distance (here, 30 m). (S16). When the distance X is less than 30 m, the lane marking shape ahead of the vehicle is estimated from the quadratic curve obtained in S13 (S17), and the process ends. On the other hand, when the distance X is 30 m or more, the distance X is reduced by a predetermined value (here, 2 m), and the process returns to S13.

  Through these series of processes, after fitting with the initial value of the distance X being 100 m, the fitting of the quadratic curve is repeatedly performed while decreasing the distance X by a predetermined value until the distance X becomes less than 30 m. And when standard deviation (sigma) becomes smaller than threshold value S, the lane marking shape ahead of a vehicle is estimated based on the quadratic curve obtained by the said fitting. Further, when the distance X becomes less than 30 m during the process, the lane marking shape in front of the vehicle is estimated based on the quadratic curve obtained at that time.

  As described above, according to the steering control system 1, since the lane line shape in front of the vehicle is estimated based on the lane line shape in the travel section in which the vehicle 30 has already traveled, the lane line in front of the vehicle is shielded. However, the estimation accuracy is not deteriorated, and the estimation accuracy can be improved. In particular, according to the steering control system 1, it can be suitably used on an expressway in which a curve with a constant curvature often continues for a long time, and the lane marking shape ahead of the vehicle can be accurately estimated. As a result, the steering control of the vehicle 30 can be suitably performed.

  Further, according to the steering control system 1, when the standard curve σ is equal to or greater than the threshold value S as a result of fitting a quadratic curve to the first section lane line position 21, the current location is determined from the second location closer to the current location than the first location. The lane marking shape ahead of the vehicle is estimated based on the second lane marking position detected in the second traveling zone until. Therefore, it is possible to exclude a section having a large curvature change from the travel section referred to at the time of fitting, and it is possible to cope with a change in curvature. That is, when there is a section with a large curvature change in the first traveling section, a section ahead of the vehicle is obtained from a quadratic curve fitted to the second section line position of the second traveling section excluding the section with a large curvature change. The line shape will be estimated.

  The preferred embodiments of the present invention have been described above. However, the present invention is not limited to the above-described embodiments. The present invention can be modified without departing from the scope described in the claims or applied to other embodiments. May be.

  For example, in the above embodiment, the lane line shape in front of the vehicle is estimated on the assumption that the lane line shape in front of the vehicle is located on the obtained quadratic curve 23, but the lane line shape in front of the vehicle is obtained from the obtained secondary curve 23. For example, the lane marking shape ahead of the vehicle may be estimated after further correction or the like is added to the quadratic curve 23.

  In the above embodiment, the configuration in which the steering control is performed using the estimated lane marking shape in front of the vehicle as it is has been described. However, any configuration may be used as long as the steering control is performed based on the estimation result. For example, the steering control system 1 further includes an in-vehicle camera, detects a lane line shape in front of the vehicle based on an image captured by the in-vehicle camera in the vicinity of the estimated lane line position in front of the vehicle, Steering control may be performed using a lane marking.

  In the above-described embodiment, as an example of the steering control, the example in which the vehicle 30 is guided to the road shoulder when the driver becomes unable to drive on the highway has been described, but the content of the steering control is not limited thereto. Alternatively, instead of or in addition to this, other steering control may be performed. For example, tracking control (lane keeping) for a lane marking on an expressway may be performed.

  In the above-described embodiment, the configuration in which fitting is performed using a quadratic curve has been described. However, a cubic or higher-order curve may be used. Moreover, in the said embodiment, although detected and estimated about both the left and right lane markings of the vehicle 30, you may detect and estimate any one of the left and right lane markings.

  In the above embodiment, the storage unit 13 is described as storing the first section lane line position 21 detected in the first travel section (past 100 m), but the storage unit 13 is detected at least in the first travel section. What is necessary is just to memorize | store the lane line position made, and may memorize | store the lane line position detected in the area longer than a 1st driving | running | working area.

  In the above embodiment, the driver state detection unit 15 has been described as an example of an in-vehicle camera. However, the driver state detection unit 15 may detect information necessary for determining the driving state of the driver. For example, it may be a device that is directly attached to the driver and detects the biological information of the driver, or a device that detects operation information.

DESCRIPTION OF SYMBOLS 1 ... Steering control system, 11 ... Compartment line recognition sensor, 13 ... Memory | storage part, 15 ... Driver state detection part, 17 ... Steering actuator, 19 ... Control part, 21 ... 1st area division line position, 23 ... Quadratic curve, 30: Vehicle.

Claims (2)

A steering control system for performing steering control of a vehicle,
A lane marking recognition sensor for detecting a lane marking position of a road on which the vehicle travels;
A storage unit for storing a lane line position detected by the lane line recognition sensor;
A controller that estimates a lane line shape in front of the vehicle based on a lane line position stored in the storage unit, and executes steering control of the vehicle based on the estimated lane line shape in front of the vehicle. ,
When the lane line position detected by the lane line recognition sensor when the vehicle travels in the first travel zone from the first point in the running direction to the current location is set as the first zone lane line position,
The control unit fits a multi-order curve to the first section lane line position stored in the storage unit, and a standard deviation between the obtained multi-order curve and the first section lane line position is a predetermined threshold value. If smaller than, the lane marking shape ahead of the vehicle is estimated from the multi-order curve. When the lane line position detected by the lane line recognition sensor when the vehicle travels in the second travel zone from the second point closer to the current location than the first location to the current location is set as the second segment lane line position. ,
When the standard deviation is greater than or equal to the threshold, the control unit fits the multi-order curve to the second section lane line position stored in the storage unit, and the obtained multi-order curve and the second curve The steering control system according to claim 1, wherein when the standard deviation from the section lane line position is smaller than the threshold value, the lane line shape in front of the vehicle is estimated from the multi-order curve.

Images