JP6855759B2 - Self-driving vehicle control system - Google Patents

Description

The present invention relates to a control system for an autonomous vehicle.

When the self-position of the vehicle is estimated by D-GPS positioning when automatic driving is not performed, the difference between the estimated self-position and the predetermined reference point is calculated, and the difference is less than or equal to the threshold value. There is a known autonomous driving vehicle that allows the start of automatic driving and prohibits the start of automatic driving when the difference is larger than the threshold value (see, for example, Patent Document 1).

Japanese Unexamined Patent Publication No. 2000-029522

However, if the estimated self-position is inaccurate, it is not possible to accurately determine whether to allow or prohibit the start of automatic driving. Therefore, it is necessary to appropriately evaluate the accuracy of the estimated self-position and perform control based on the evaluation result. Patent Document 1 does not disclose this point at all.

According to the present invention, a control system for an automatically driven vehicle, a GPS receiver configured to receive a signal from a GPS satellite, and an internal sensor configured to detect the running state of the vehicle. And a camera configured to acquire image data around the vehicle, a road information storage device configured to associate road information with position information and store it in advance, and a reference correlation value. A reference correlation value storage device that is configured to be associated with information and stored in advance, a notification device that is configured to notify the driver of a manual operation preparation request during automatic operation, and an electronic control unit. Further, an acquired road surface image is generated from the self-position estimation unit configured to estimate the self-position of the vehicle based on the outputs of the GPS receiver and the internal sensor, and the image data acquired by the camera. , A road surface image generation unit configured to generate a pseudo road surface image from the road information at the acquisition position where the camera has acquired the image data, and a correlation value between the generated road surface image and the pseudo road surface image. Based on the result of comparison between the correlation value calculation unit configured to calculate the reference correlation value at the acquisition position and the reference correlation value at the acquisition position during automatic operation. Control of an automatically driven vehicle, comprising an electronic control unit comprising a notification control unit configured to control the notification device to execute or stop the notification of the manual driving preparation request to the driver. The system is provided.

It is possible to appropriately evaluate the estimation accuracy of the self-position during automatic operation and appropriately control the execution or stop of the notification of the manual operation preparation request.

It is a block diagram of the control system of the autonomous driving vehicle of the Example by this invention. It is the schematic explaining the external sensor of the Example by this invention. It is a figure which shows the passenger room of the Example by this invention. It is a block diagram of the electronic control unit of the Example by this invention. It is a conceptual diagram for demonstrating the notification control of the manual operation preparation request in the Example by this invention. It is a figure which shows the map of the reference correlation value. It is a diagram which shows an example of a reference correlation value. It is a diagram for demonstrating the notification control of the manual operation preparation request in the Example by this invention. It is a flowchart which shows the notification control routine of the manual operation preparation request of an Example by this invention.

FIG. 1 shows a block diagram of a control system for an autonomous vehicle according to an embodiment of the present invention. Referring to FIG. 1, the control system of the automatically driven vehicle according to the present invention includes an external sensor 1, a GPS receiver 2, an internal sensor 3, a road information storage device 4, a navigation system 5, a storage device 6, and various actuators 7. , An operating device 8, a reference correlation value storage device 9, a notification device 10, and an electronic control unit (ECU, Electronic Computer Unit) 20.

The external sensor 1 is configured to detect information on the outside or surroundings of the own vehicle. The external sensor 1 includes at least one of a lidar (Laser Imaging Detection and Ranking), a radar (Radar), and a camera. In the embodiment according to the present invention, as shown in FIG. 2, the external sensor 1 includes a rider 1a, a radar 1b, and a camera 1c.

The rider 1a is a device that uses a laser beam to detect a road on which the vehicle is traveling or an external obstacle. In the example shown in FIG. 2, four riders 1a are attached to bumpers at the four corners of the vehicle V, respectively. The rider 1a sequentially irradiates the laser beam toward the entire circumference of the own vehicle V, measures the distance from the reflected light to the obstacles on the road and around the road, and measures the road and the obstacles around the entire circumference of the own vehicle V. Is detected in the form of a three-dimensional image. The three-dimensional images of roads and obstacles detected by the rider 1a are transmitted to the electronic control unit 20. On the other hand, the radar 1b is a device that detects an obstacle outside the own vehicle V by using radio waves. In the example shown in FIG. 2, four radars 1b are attached to bumpers at the four corners of the vehicle V, respectively. The radar 1b emits radio waves from the radar 1b around the own vehicle V, and measures the distance from the reflected wave to an obstacle around the own vehicle V. The obstacle information detected by the radar 1b is transmitted to the electronic control unit 20. In the example shown in FIG. 2, the camera 1c includes a front camera provided inside the windshield of the vehicle V. The front camera 1c photographs the front of the own vehicle V in color or monochrome, and the color or monochrome photographing information by the front camera 1c is transmitted to the electronic control unit 20.

The GPS receiver 2 is configured to receive signals from three or more GPS satellites and thereby detect the absolute position of the own vehicle V (for example, the latitude and longitude of the own vehicle V). The absolute position information of the own vehicle V detected by the GPS receiver 2 is transmitted to the electronic control unit 20.

The internal sensor 3 is configured to detect the traveling state of the own vehicle V. The traveling state of the own vehicle V is represented by at least one of the speed, acceleration, and posture of the own vehicle. The internal sensor 3 includes one or both of a vehicle speed sensor and an IMU (Inertial Measurement Unit). In the embodiment according to the present invention, the internal sensor 3 includes a vehicle speed sensor and an IMU. The vehicle speed sensor detects the speed of the own vehicle V. The IMU is equipped with, for example, a three-axis gyro and a three-direction acceleration sensor, detects the three-dimensional angular velocity and acceleration of the own vehicle V, and detects the acceleration and attitude of the own vehicle V based on them. The traveling state information of the own vehicle V detected by the internal sensor 3 is transmitted to the electronic control unit 20.

The road information storage device 4 is configured to store road information in advance in association with position information. The road information includes, for example, information on the shape of the road (for example, the width of the road, the type of the curve and the straight line portion, the curvature of the curve, the slope of the road, the intersection, the type of the confluence and the branch point, and the like). In one example, road information and location information are stored in advance in the form of map information. This map information also includes building information (building shape, etc.).

The navigation system 5 is configured to guide the own vehicle V to the destination input to the navigation system 5 by the driver of the own vehicle V. The navigation system 5 calculates a target route to the destination based on the current position information of the own vehicle V and the road information of the road information storage device 4. Information on the target route of the own vehicle V is transmitted to the electronic control unit 20.

The storage device 6 stores a three-dimensional image of an obstacle detected by the rider 1a and a road map dedicated to autonomous driving created based on the detection result of the rider 1a. Three-dimensional images and road maps of these obstacles are updated constantly or regularly.

The actuator 7 is a device for controlling the traveling operation of the own vehicle V in response to a control signal from the electronic control unit 20. The traveling operation of the vehicle V includes driving, braking and steering of the vehicle V. The actuator 7 includes at least one of a drive actuator, a braking actuator, and a steering actuator. In the embodiment according to the present invention, the actuator 7 includes a drive actuator, a braking actuator, and a steering actuator. The drive actuator controls the output of the engine or electric motor that provides the driving force of the vehicle V, thereby controlling the driving operation of the vehicle V. The braking actuator operates the braking device of the vehicle V, thereby controlling the braking operation of the vehicle V. The steering actuator operates the steering device of the vehicle V, thereby controlling the steering operation of the vehicle V.

The operating device 8 is configured to be operated by the driver of the vehicle. The operation device 8 includes, for example, at least one of an operation button, a switch, a lever, a touch panel, a voice recognition device (microphone), a steering wheel, an accelerator pedal, and a brake pedal. In the example shown in FIG. 3, the operating device 8 includes an operating button 8a. Further, in the example shown in FIG. 3, the operation button 8a is provided on the steering STR. A signal indicating that the operating device 8 has been operated is transmitted to the electronic control unit 20.

The reference correlation value storage device 9 is configured to associate the reference correlation value with the position information and store it in advance. The reference correlation value will be described later.

The notification device 10 is configured to notify the driver. This notification includes, for example, at least one of visual, auditory, and sensory notifications. When the notification is a visual notification, the notification device 10 includes a display device that is visible to the driver. The display device includes, for example, at least one such as a display and a lamp. The displays include, for example, a meter display 10a1 provided on the instrument panel IP behind the steering STR, a head-up display 10a2 provided on the instrument panel IP above the meter display 10a1, and a vehicle, as shown in FIG. The center display 10a3 provided on the instrument panel IP at substantially the center in the width direction is included. The meter display 10a1 displays, for example, vehicle speed, engine speed, remaining fuel, remaining battery, and the like. The center display 10a3 is used by the navigation system 5 to display a target route and a guide to the destination (such as turning left at the next intersection). When the notification device 10 includes a display, the notification device 10 notifies the driver by displaying character information or image information on the display. On the other hand, the lamp includes, for example, an LED (light emitting diode) 10a4 attached to the steering STR, as shown in FIG. When the notification device 10 includes a lamp, the notification device 10 notifies the driver by turning on or blinking the lamp.

On the other hand, when the notification is an auditory notification, the notification device 10 includes, for example, a speaker. The speaker includes, for example, a pair of speakers 10b provided at the top of the instrument panel IP, as shown in FIG. When the notification device 10 includes a speaker, the notification device 10 notifies the driver by emitting at least one of a voice and an alarm sound from the speaker.

When the notification is a sensory notification, the notification device 10 includes, for example, a vibrator that vibrates the driver. The vibrator includes, for example, a vibrator 10c incorporated in the driver's seat, as shown in FIG. When the notification device 10 includes a vibrator, the notification device 10 notifies the driver by operating the vibrator.

In the embodiment according to the present invention, the notification device 10 includes at least one of a meter display 10a1, a head-up display 10a2, a center display 10a3, a steering STR LED 10a4, a speaker 10b, and a vibrator 10c. The notification device 10 is controlled based on a control signal from the electronic control unit 20.

The electronic control unit 20 is a computer including a ROM (Read Only Memory), a RAM (Random Access Memory), a CPU (Central Processing Unit), and the like, which are connected to each other by a bidirectional bus. In the embodiment according to the present invention, as shown in FIG. 4, the electronic control unit 20 includes an operation control unit 21, a notification control unit 22, a self-position estimation unit 23, a road surface image generation unit 24, and a correlation value calculation unit 25. Be prepared.

The driving control unit 21 is configured to control vehicle driving. Specifically, in the embodiment according to the present invention, when automatic driving should be performed, the driving control unit 21 controls the actuator 7 to control the driving, braking, and steering, which are the traveling operations of the vehicle V. On the other hand, when the manual operation should be performed, the operation control unit 21 ends the automatic operation. In manual driving, the driving, braking, and steering, which are the traveling operations of the vehicle V, are performed by the driver.

On the other hand, the notification control unit 22 is configured to control the notification device 10 to execute and stop the notification to the driver. In the embodiment according to the present invention, the notification control unit 22 determines whether or not automatic operation is possible during manual operation. For example, when the external condition of the vehicle V can be reliably detected by the external sensor 1, it is determined that automatic driving is possible. On the other hand, when the external condition of the vehicle V cannot be reliably detected by the external sensor 1, it is not determined that automatic driving is possible. When it is determined that automatic driving is possible, the notification control unit 22 controls the notification device 10 to notify the driver that automatic driving is possible. Next, when the driver operates the operation device 8, the operation control unit 21 determines that the automatic operation should be performed, and starts the automatic operation.

Further, in the embodiment according to the present invention, the notification control unit 22 determines whether or not it is difficult to continue the automatic operation during the automatic operation. For example, when the external condition of the vehicle V cannot be reliably detected by the external sensor 1, it is determined that it is difficult to continue the automatic driving. On the other hand, when the external condition of the vehicle V can be reliably detected by the external sensor 1, it is not determined that it is difficult to continue the automatic driving. When it is determined that it is difficult to continue the automatic operation, the notification control unit 22 controls the notification device 10 to notify the driver of the automatic operation end request. That is, the driver is required to operate the operating device 8 in order to end the automatic operation. Next, when the driver operates the operation device 8, the operation control unit 21 determines that the manual operation should be performed, ends the automatic operation, and starts the manual operation.

When the operation device 8 includes the operation button 8a, the operation for ending the automatic operation is that the driver presses the operation button 8a during the automatic operation. Alternatively, when the operating device 8 includes a steering wheel, the operation for ending the automatic driving is that the driver operates the steering wheel by a predetermined threshold or more during the automatic driving. When the operating device 8 includes an accelerator pedal, the operation for ending the automatic driving is that the driver depresses the accelerator pedal by a predetermined threshold or more during the automatic driving. When the operating device 8 includes a brake pedal, the operation for ending the automatic driving is that the driver depresses the brake pedal by a predetermined threshold or more during the automatic driving.

The notification of the automatic operation end request is performed as follows, for example. That is, in one example, character information such as "Please end automatic operation" is displayed on displays such as the meter display 10a1, the head-up display 10a2, and the center display 10a3. In another example, the speaker 10b emits voice information such as "Please end the automatic operation". In yet another example, the above-mentioned character information and voice information are duplicated and provided to the driver. When the driver is notified using the plurality of items of the notification device 10 in this way, the driver can more reliably understand or recognize the notification.

Further, in the embodiment according to the present invention, the notification control unit 22 is configured to control the notification device 10 to notify the driver of the manual operation preparation request. That is, the driver is required to be ready for manual operation.

In the embodiment according to the present invention, the preparatory state for manual operation includes at least a hands-on state. The hands-on state includes, for example, a state in which the driver puts his / her hand on the steering STR, a state in which the driver holds the steering STR, and the like. In another embodiment (not shown), the preparatory state for manual driving includes, in addition to the hands-on state, a state in which the driver puts his / her foot on the brake pedal, a state in which the driver's line of sight points toward the front of the vehicle, and the like. ..

The notification of the manual operation preparation request is performed in the same manner as the automatic operation end request, for example. That is, in one example, character information such as "Prepare for manual operation" is displayed on displays such as the meter display 10a1, the head-up display 10a2, and the center display 10a3. In another example, the speaker 10b emits voice information such as "Prepare for manual operation". In yet another example, the above-mentioned character information and voice information are duplicated and provided to the driver.

Next, the control of execution and stop of the notification of the manual operation preparation request in the embodiment according to the present invention will be described with reference to FIG. Referring to FIG. 5, first, the camera 1c acquires image data around the vehicle V during automatic driving. Next, the road surface image generation unit 24 generates a road surface image from this image data. Hereinafter, this road surface image will be referred to as an acquired road surface image. When the position where the camera 1c has acquired the image data is referred to as an acquisition position, the road surface image generation unit 24 further generates a road surface image from the road information at the acquisition position. Hereinafter, this road surface image will be referred to as a pseudo road surface image. Here, the acquisition position is a position estimated by the self-position estimation unit 23. The self-position estimation unit 23 is configured to estimate the self-position (for example, absolute position) of the vehicle V based on the outputs of the GPS receiver 2 and the internal sensor 3. Since the self-position is estimated based on not only the output of the GPS receiver 2 but also the output of the internal sensor 3, the estimation accuracy of the self-position is improved. On the other hand, the road information is stored in the road information storage device 4 described above.

Next, the correlation value calculation unit 25 calculates the correlation value between the acquired road surface image and the pseudo road surface image. This correlation value represents the degree of similarity between the acquired road surface image and the pseudo road surface image, and is calculated in the form of a numerical value between zero and one.

Further, the correlation value calculation unit 25 calculates the reference correlation value at the acquisition position. Here, the reference correlation value is stored in the reference correlation value storage device 9. In the examples according to the present invention, the reference correlation value takes the form of a range from the lower limit value CVm to the upper limit value CVM. Hereinafter, the range from the lower limit value CVm to the upper limit value CVM is referred to as a reference range. This reference range is stored in the reference correlation value storage device 9 as a function of position, that is, latitude and longitude, as shown in FIG. 6, for example. In another embodiment (not shown), the reference correlation value takes the form of a position-dependent value.

In the embodiment according to the present invention, the reference range of a certain position, that is, the lower limit value CVm and the upper limit value CVM are calculated based on the correlation value when the vehicle travels the certain position a plurality of times. Specifically, the lower limit value CVm and the upper limit value CVM are calculated using the average value CVave and the standard deviation σ of the correlation values at the respective positions. In one example, the lower limit CVm is calculated in the form of (CVave-2σ). In another example, the upper limit CVM is calculated in the form (CVave + σ).

FIG. 7 shows an example of the relationship between the mileage of the vehicle V and the reference correlation value when the vehicle V travels along a predetermined route from a predetermined starting point. In FIG. 7, the mileage on the horizontal axis represents a position on the route. As shown in FIG. 7, the reference correlation value, that is, the reference range determined by the lower limit value CVm and the upper limit value CVM varies depending on the mileage or position of the vehicle V. That is, the lower limit CVm at one position is smaller and the lower limit CVm at another position is larger. Also, the upper limit CVM at one position is larger and the upper limit CVM at another position is larger. Further, the variation of the correlation value at one position is smaller, and the variation of the correlation value at another position is larger. The magnitude of the variation in the correlation value is represented by the width of the reference range (CVM-CVm).

Referring to FIG. 5 again, the notification control unit 22 then compares the correlation value with the reference correlation value or the reference range, and controls the execution and stop of the notification of the manual operation preparation request based on the comparison result. Specifically, when the correlation value is out of the reference range, the notification of the manual operation preparation request is executed. On the other hand, when the correlation value is within the reference range, the notification of the manual operation preparation request is stopped.

In FIG. 8, an example of the correlation value when the vehicle V travels along the above-mentioned route is shown by a curve X. As shown in FIG. 8, when the correlation value is within the reference range, the notification of the manual operation preparation request is stopped, and when the correlation value is outside the reference range, the notification of the manual operation preparation request is executed.

In the examples according to the invention, the reference correlation value or reference range depends on the position, as described with reference to FIG. On the other hand, the correlation value represents the estimation accuracy of the self-position. Therefore, the estimation accuracy of the self-position at one position is lower, and the estimation accuracy of the self-position at another position is higher. In addition, the variation in the estimation accuracy of the self-position at a certain position is smaller, and the variation in the estimation accuracy of the self-position at another position is larger. Furthermore, considering that road information (for example, road shape information) differs depending on the position, the estimation accuracy of the self-position on a road having a certain shape is lower, and the estimation accuracy of the self-position on a road having another shape is lower. It means that the estimation accuracy of the self-position is higher. In addition, the variation in the estimation accuracy of the self-position on the road having a certain shape is smaller, and the variation in the estimation accuracy of the self-position on the road having another shape is larger.

On the other hand, in the embodiment according to the present invention, as described above, the reference correlation value or the reference range is calculated from the conventional vehicle running result. Therefore, the reference correlation value or reference range at a certain position represents the accuracy of the previous estimation of the self-position at the certain position.

Then, when the correlation value of this time is out of the reference range, it can be seen that the estimation accuracy of the self-position this time is different from the estimation accuracy of the self-position before. On the other hand, when the correlation value of this time is within the reference range, it can be seen that the estimation accuracy of the self-position this time is about the same as the estimation accuracy of the self-position before.

Therefore, in the embodiment according to the present invention, when the current correlation value is out of the reference range, the driver is notified of the manual operation preparation request. On the other hand, when the correlation value this time is within the reference range, the driver is not notified of the manual operation preparation request.

If the self-position estimation accuracy is lower, it may be difficult to continue the automatic operation. That is, there is a possibility that the driver is notified of the automatic operation end request. In the embodiment according to the present invention, the driver is notified of the manual operation preparation request when the correlation value is out of the reference range. Therefore, the next time the automatic operation end request is notified, the driver can promptly end the automatic operation and start the manual operation.

Further, the manual operation preparation request does not request the driver to end the automatic operation. That is, the driver can take the manual operation ready state while continuing the automatic operation. Therefore, it is limited that the driver feels annoyed.

As described above, the manual operation preparation request requires the driver to take the preparation state for manual operation while continuing the automatic operation. Therefore, the manual operation preparation request is completely different in nature from the automatic operation end request that requires the end of the automatic operation.

Further, in the embodiment according to the present invention, when the correlation value returns to the reference range when the manual operation preparation request is notified, the notification of the manual operation preparation request is stopped. Therefore, it is limited that the driver feels annoyed.

Moreover, in the embodiment according to the present invention, the reference correlation value, that is, the reference range is set according to the position. Therefore, the estimation accuracy of the self-position is appropriately evaluated according to the position or road information, and the notification of the manual driving preparation request is appropriately controlled according to the position or road information.

As described above, in the embodiment according to the present invention, the electronic control unit 20 includes a self-position estimation unit 23 configured to estimate the self-position of the vehicle V based on the outputs of the GPS receiver 2 and the internal sensor 3. A road surface image generation unit 24 configured to generate an acquired road surface image from the image data acquired by the camera 1c and to generate a pseudo road surface image from the road information at the acquisition position where the camera 1c acquired the image data. The correlation value calculation unit 25 configured to calculate the correlation value between the acquired road surface image and the pseudo road surface image and the reference correlation value at the acquisition position, and the correlation value and the reference at the acquisition position during automatic operation. The notification control unit 22 is configured to control the notification device 10 based on the comparison result with the correlation value to execute or stop the notification of the manual operation preparation request to the driver.

FIG. 9 shows a routine for executing the above-described notification control of the manual operation preparation request according to the embodiment of the present invention. This routine is executed by interrupts at predetermined set time intervals. With reference to FIG. 9, in step 100, it is determined whether or not the vehicle V is in automatic driving. When the vehicle V is not in automatic driving, that is, in manual driving, the processing cycle is terminated. When the vehicle V is in automatic driving, the process proceeds to step 101 to generate an acquired road surface image and a pseudo road surface image. In the following step 102, the correlation value is calculated. In the following step 103, the reference range is calculated from the map of FIG. In the following step 104, it is determined whether or not the correlation value is out of the reference range. When the correlation value is out of the reference range, the process proceeds to step 105, and the notification of the manual operation preparation request is executed. On the other hand, when the correlation value is within the reference range, the process proceeds to step 106, and the notification of the manual operation preparation request is stopped.

The logic used by the correlation value calculation unit 25 to calculate the correlation value is the same as the logic used to calculate the reference correlation value. Therefore, when the logic used by the correlation value calculation unit 25 to calculate the correlation value is changed, it is necessary to recalculate the reference correlation value.

Next, another embodiment according to the present invention will be described. In another embodiment according to the present invention, the electronic control unit 20 further includes a reliability calculation unit that repeatedly calculates an automatic operation reliability value that expresses the degree of reliability of automatic operation in the form of a numerical value during automatic operation. When the automatic driving reliability value is small, the degree of reliability of automatic driving is lower than when the automatic driving reliability value is large. Further, in another embodiment according to the present invention, the notification control unit 22 controls the notification device 10 when the automatic operation reliability value is smaller than the predetermined preparation request value, and sends the manual operation preparation request to the driver. Notice.

Then, in another embodiment according to the present invention, the notification control unit 22 determines that the automatic driving reliability value is smaller than the preparation request value when the correlation value is outside the reference range, and the correlation value is within the reference range. When it is, it is judged that the automatic operation reliability value is larger than the preparation request value. Therefore, the notification of the manual operation preparation request is executed when the correlation value is outside the reference range, and the notification of the manual operation preparation request is stopped when the correlation value is within the reference range.

In another embodiment according to the present invention, the notification control unit 22 controls the notification device 10 automatically when the automatic operation reliability value is smaller than the preparation request value and the end request value set. Notify the driver of the end-of-operation request. That is, the driver is notified of the manual operation preparation request before the automatic operation end request is notified. Therefore, when the automatic operation end request is notified, the driver can promptly end the automatic operation and start the manual operation.

1c Camera 2 GPS receiver 3 Internal sensor 4 Road information storage device 9 Reference correlation value storage device 10 Notification device 20 Electronic control unit 22 Notification control unit 23 Self-position estimation unit 24 Road surface image generation unit 25 Correlation value calculation unit V Vehicle

Claims (1)

A control system for self-driving vehicles
GPS receivers that are configured to receive signals from GPS satellites,
An internal sensor that is configured to detect the running condition of the vehicle,
With a camera that is configured to acquire image data around the vehicle,
A road information storage device that is configured to associate road information with position information as an absolute position and store it in advance.
A reference correlation value storage device configured to associate a reference correlation value with position information as an absolute position and store it in advance, and a reference correlation value storage device.
A notification device configured to notify the driver of a manual operation preparation request during autonomous driving, and
It is an electronic control unit
A self-position estimation unit configured to estimate the vehicle's self-position as an absolute position based on the outputs of the GPS receiver and the internal sensor.
A road surface image generation unit configured to generate an acquired road surface image from the image data acquired by the camera and to generate a pseudo road surface image from the road information at the acquisition position where the camera acquired the image data. ,
A correlation value representing the degree of similarity between the acquired road surface image and the pseudo road surface image, and a correlation value calculating unit configured to calculate the reference correlation value at the acquired position.
During automatic operation, the notification device is controlled based on the comparison result between the correlation value and the reference correlation value which is the reference value of the correlation value at the acquisition position, and the driver is notified of the manual operation preparation request. A notification control unit that is configured to run or stop,
With an electronic control unit equipped with
A control system for self-driving vehicles.

Images