The present invention relates to an override determination method for a driving assist device and a driving assist device.
An invention in which autonomous driving is cancelled when an intervention operation by an occupant is detected during autonomous driving is known from Japanese Laid Open Patent Application No. 2007-32645 (Patent Document 1). In the invention disclosed in Patent Document 1, autonomous driving is canceled when an operating torque applied to a steering wheel exceeds a threshold value.
However, the invention disclosed in Patent Document 1 does not take into consideration the steering angle of the steering wheel. It is necessary to change the determination threshold for the intervention operation depending on the steering angle, but this is not taken into consideration. For this reason, in the invention disclosed in Patent Document 1, there is the risk of an erroneous determination that an override has occurred, even though an intervention operation (override) by the occupant has not occurred.
In view of the problem described above, an object of the present invention is to provide an override determination method for a driving assist device that can prevent erroneous determination of an override by using a steering angle, and a driving assist device.
An override determination method for a driving assist device according to one aspect of the present invention comprises acquiring a first motor current for canceling out a self-aligning torque generated in tires mounted on a vehicle while executing driving assist control, using the first motor current to set a first threshold value that is greater than the first motor current, and determining that an override has occurred when a second motor current for controlling a turning angle of the tires in accordance with a steering angle is greater than the first threshold value.
By means of the present invention, erroneous determination of an override can be prevented.
Referring now to the attached drawings which form a part of this original disclosure.
Embodiments of the present invention are described below with reference to the drawings. In the descriptions of the drawings, identical parts have been assigned the same reference numerals, and the descriptions thereof have been omitted.
The driving assist device according to the present embodiment is used in vehicles having an autonomous driving function. An architecture of the autonomous driving according to the present embodiment will be described with reference to
When carrying out autonomous driving, ascertaining one's own location and ascertaining information on the surroundings of the vehicle are required. With these ascertainments, it becomes possible for the vehicle to automatically change lanes and to proceed in the appropriate direction at an intersection, to thereby reach the destination. The architecture for ascertaining the self-location and the architecture for ascertaining the information on the vehicle surroundings are indicated by the reference symbols 100-105 in
Sensors indicated by the reference symbol 100 in
More specifically, a plurality of cameras are provided, in the front, side, rear, the side-view mirrors, and the like, of the vehicle. The camera comprises an imaging element, such as a CCD (charge-coupled device), CMOS (complementary metal oxide semiconductor), and the like. The camera detects objects present in the vehicle surroundings (pedestrians, bicycles, two-wheeled vehicles, other vehicles, and the like) as well as information on the vehicle surroundings (road boundary lines, traffic lights, signs, pedestrian crossings, intersections, and the like).
The radar emits radio waves toward an object in front of the vehicle and measures the reflected waves, thereby measuring the distance and direction to the object. LIDAR (Laser Imaging Detection and Ranging) carries out scanning by means of laser light in the horizontal direction as well as the vertical direction, to thereby measure the position and shape of objects present in the vehicle surroundings.
In addition, a GPS receiver may be included in the sensors 100. The GPS receiver receives radio waves from a satellite to thereby detect position information about a vehicle on the ground (including latitude and longitude information). However, the method for detecting the vehicle's position information is not limited to a GPS receiver. For example, the position may be estimated using a method called odometry. Odometry is a method for estimating the vehicle position by calculating the amount and direction of movement of the vehicle in accordance with the rotation angle and the rotational angular velocity of the vehicle.
The information detected by the sensors 100 is transmitted to a controller (not shown) mounted on the vehicle and processed. An example of the controller is a general-purpose microcomputer equipped with a CPU (central processing device), memory, and an input/output unit. A computer program is installed in the microcomputer to cause it to function as the driving assist device. By executing the computer program, the microcomputer functions as a plurality of information processing circuits included in the driving assist device. Here, an example is shown in which the plurality of information processing circuits included in the driving assist device is realized by software, but the information processing circuits can of course comprise dedicated hardware for executing each of the information processes shown below.
Such a controller also includes an ECU (Electronic Control Unit). Normally, a plurality of ECUs are provided in a vehicle. In addition to travel control of the vehicle, ECUs are also used for controlling the stereo, air conditioner, and the like.
The information detected by the sensors 100 is localized to fit the detected region (reference symbol 103 in
The information detected by the sensors 100 and map information are merged, and an environment recognition unit 104 in the controller generates a world model. The world model here means information on the surrounding environment on a digital map that combines static map information and a high-precision map with dynamic position information, such as self-location information, other vehicle information, pedestrian information, and the like.
The high-precision map will be described. A high-precision map contains road information, such as the number of lanes of a road, road width information, information on road undulations, and the like, and information such as road signs indicating the speed limit, one-way streets, and the like, road markings, such as pedestrian crossings, partition lines, and the like. Furthermore, the high-precision map may include facility information such as road structures (for example, traffic lights, telephone poles) and buildings. These pieces of high-precision map information are provided in an HD map 102 shown in
Various data, such as road information and facility information, are not necessarily limited to data acquired from the HD map 102, and may be acquired using vehicle-to-vehicle communication or road-to-vehicle communication. If various data such as road information and facility information are stored in an externally installed server, the controller may acquire these data from the cloud at any time by communication. In addition, the controller may periodically acquire the latest map information from an externally installed server and update the map information being held in the controller.
An object recognition unit 105 in the controller generates recognition information about objects around the vehicle, which is generated based on information detected by the sensors 100, and generates a local model. The local model includes, as the object recognition information, other vehicle information, pedestrian information, and the like. The other vehicle information includes speeds, traveling directions, travel lanes, and the like, of other vehicles. The pedestrian information includes attributes (adult or child), orientations of faces, directions of travel, and the like, of pedestrians. The local model generated by the object recognition unit 105 is used for generating the world model.
A travel control architecture for autonomous driving will be described next with reference to reference symbols 106-111 in
The user sets a destination using a navigation device 101 (Navigation) (reference symbol 106 in
An action determination unit 108 in the controller determines the action to be taken when traveling autonomously along a route that is set using the information generated by the environment recognition unit 104 and the object recognition unit 105. In addition, the action determination unit 108 determines whether to advance or stop the host vehicle. For example, the vehicle is stopped if the traffic light is red and the vehicle is driven if the traffic light is green. In addition, the action determination unit 108 determines the timing for operating a turn signal and the timing for operating the steering wheel when changing lanes.
Next, the controller reads the HD map 102 and the local model generated by the object recognition unit 105, and plans a drive zone (reference symbol 109 in
The controller then sets a trajectory along the drive zone (reference symbol 110 in
The controller then controls the attitude of the vehicle to drive the vehicle along the trajectory at an appropriate attitude (reference symbol 111 in
Details of a motion control 111 will be described with reference to
An AD (Autonomous Driving) layer 111a shown in
Information used for autonomous driving is input from the AD layer 1lIa to an arbitration 111d (AD input 111b). Information relating to the occupant's voluntary driving operation is also input to the arbitration 111d (MD input 111c). Information relating to the occupant's voluntary driving operation include the occupant's operations of the steering wheel, the accelerator pedal, and the brake pedal.
If an intervention operation by an occupant is carried out during autonomous driving, the autonomous driving is switched to manual driving. Switching from autonomous driving to manual driving is called override. The arbitration 111d monitors the AD input 111b and the MD input 111c and switches from autonomous driving to manual driving, or conducts arbitration such that a process related to the autonomous driving does not conflict with a process related to the manual driving. Override has been described as switching from autonomous driving to manual driving above, but is not limited to this. Override may be defined as the occupant's intervention operation itself during autonomous driving.
Switching from autonomous driving to manual driving (override) means that all driving authority is transferred, usually to the occupant. However, in the present embodiment, an override does not necessarily mean that all driving authority is transferred to the occupant. For example, if the occupant's intervention operation is a slight operation, it cannot be definitively determined that the occupant's intention is to request transfer of all driving authority. Thus, if the occupant's intervention operation is detected during autonomous driving (if it is determined that an override has occurred), the arbitration 111d calculates an override amount indicating the proportion of the occupant's intervention operation. Then, the arbitration 111d conducts arbitration such that a process related to the autonomous driving does not conflict with a process related to the manual driving based on the calculated override amount.
A reference model 111e is a function for calculating the attitude of the vehicle (particularly the attitude when turning). The reference model 111e calculates the yaw moment, yaw rate, and slip angle. Specifically, the reference model 111e acquires the curvature of the road on which the vehicle will travel in the future. The curvature is acquired using a camera image, for example. A curve in front of the vehicle obtained from the camera image may be approximated to an arc, and the reciprocal of the radius of this arc may be used as the curvature. In general, curvature is defined as the reciprocal of the curvature radius. The reference model 111e uses the vehicle speed, the lateral force generated in the vehicle (lateral positional displacement), and attitude angle, and the like to thereby calculate the yaw moment, yaw rate, and slip angle for appropriately traveling on a curve.
A body motion control 111f executes a feedback control so as to be able to obtain the yaw moment, yaw rate, and slip angle calculated by the reference model 111e. A wheel motion control 111g outputs a turn control command to an ECU that controls the front wheels and the rear wheels so as to be able to obtain the yaw moment, yaw rate, and slip angle calculated by the reference model 111e.
Finally, the controller controls various actuators (accelerator pedal actuator, brake actuator, steering actuator) (reference symbol 111h in
Next, a steering wheel mechanism mounted on a vehicle will be described with reference to
A steer-by-wire system is applied as the steering wheel mechanism in the present embodiment. As shown in
In the steer-by-wire system, the steering angle of the steering wheel 2 is detected by a steering angle sensor (not shown). The steering angle detected by the steering angle sensor is output to the ECU 10. The ECU 10 calculates the turning angle of the turning wheels 6a, 6b based on the steering angle detected by the steering angle sensor. The ECU 10 supplies a motor current to the turning actuator 4, thereby driving the turning actuator 4, such that the turning angle becomes the calculated turning angle. For example, a DC brushless motor can be used as the turning actuator 4. By means of the steer-by-wire system, design flexibility such as the arrangement of various devices (such as the steering wheel 2) increases, which may contribute to size reduction.
In autonomous driving, since the occupant does not operate the steering wheel 2 the operation of the steer-by-wire system is as follows. The ECU 10 supplies a motor current to the turning actuator 4, thereby driving the turning actuator 4, in accordance with a command from the wheel motion control 111g (refer to
In autonomous driving in which the steer-by-wire system is employed, it is not essential to rotate the steering wheel 2 in accordance with the turning of the turning wheels 6a, 6b. However, it is preferable to rotate the steering wheel 2 in accordance with the turning of the turning wheels 6a, 6b. If the steering wheel 2 does not rotate even though the turning wheels 6a, 6b are turning, the occupant may feel a sense of discomfort. Thus, the ECU 10 controls a steering actuator 3 in accordance with the turning of the turning wheels 6a, 6b, thereby rotating the steering wheel 2. The occupant's discomfort is thereby reduced. The ECU 10 is one of a plurality of ECUs provided in the vehicle.
One example of an override determination method will be described next with reference to
The straight line L1 shown in
The self-aligning torque increases as the steering angle increases. Thus, the controller increases the motor current for canceling out the self-aligning torque as the steering angle increases.
σ0 in
A straight line L2 shown in
In automatic turning, there are cases when the occupant intervenes. The occupant's intervention operation referred to here means that the occupant operates the steering wheel 2 during automatic turning. If the occupant operates the steering wheel 2 during automatic turning, motor current corresponding to the steering angle related to the occupant's operation is supplied to the turning actuator 4. Motor current corresponding to the steering angle related to the occupant's operation is a current for controlling the turning angle of the turning wheels 6a, 6b. The controller detects the motor current during automatic turning and determines whether the detected motor current is greater than the threshold L2. If the detected motor current is greater than the threshold L2, the controller determines that an occupant's intervention operation has occurred, that is, that an override has occurred.
The determination mechanism will now be described. The motor current when the occupant operates the steering wheel 2 is larger than the motor current during autonomous driving. The mechanism by which motor current is supplied to the turning actuator 4 by means of the operation of the steering wheel 2 by the occupant in a steer-by-wire system has already been described. In the present embodiment, the threshold L2 is set as a value larger than the motor current during autonomous driving. For this reason, without the occupant's intervention operation during automatic turning, the motor current during automatic turning usually does not exceed the threshold L2. Accordingly, if the motor current exceeds the threshold L2, it means that the occupant's intervention operation has likely occurred, that is, that an override has likely occurred. Thus, if the detected motor current is greater than the threshold L2, the controller determines that and override has occurred. By using the threshold L2, erroneous determination of an override can be prevented. Erroneous determination referred to here means a case in which an operation of the steering wheel 2 carried out automatically is erroneously determined to be an operation by the occupant.
The advantages attributable to the coefficient κ0 being different from the coefficient κ1 will be described next. Previously it was stated that σ0 and σ1 are eigenvalues determined in accordance with the mechanism of the steer-by-wire system. By combining such eigenvalues and coefficients having different values depending on the steering angle, the setting flexibility of the threshold L2 increases. As a result, it is possible to set a threshold value in consideration of the individual occupant's operation.
In addition, the coefficient κ1 is set smaller than the coefficient κ0. As a result, the determination sensitivity is enhanced in regions where the steering angle is large. If the coefficient κ1 is greater than the coefficient κ0, it is possible that the motor current will not exceed a threshold L2 even if the occupant operates the steering wheel 2 significantly. By setting the coefficient κ1 to be smaller than the coefficient κ0, such possibility can be eliminated.
The override amount (hereinafter appropriately referred to as OD amount) will be described next with reference to
In
A straight line L3 shown in
As shown in
On the other hand, as shown in
In addition, as shown in
In
One example of a method for setting the threshold values will be described next with reference to the flow chart of
In Step S101, the controller acquires an offset amount (Go) of the motor current. The offset amount is the current supplied to the turning actuator 4 that controls the turning angle of the turning wheels 6a, 6b when the steering angle of the steering wheel 2 is zero (refer to
The process proceeds to Step S103, and the controller acquires a motor current (σ1) for canceling out the self-aligning torque (refer to
The process proceeds to Step S105, and the controller determines the coefficients κ0, κ1 for setting the threshold value L2 used for the override determination (refer to
The process proceeds to Step S107, and the controller sets a value (κ0×σ0) obtained by multiplying σ0 by a coefficient co as the threshold value when the steering angle is zero. In addition, the controller sets a value (κ1×σ1) obtained by multiplying σ1 by a coefficient κ1 as the threshold value when the steering angle is 100 degrees. The controller connects a value (κ0×σ0) obtained by multiplying σ0 by the coefficient κ0 and a value (κ1×σ1) obtained by multiplying σ1 by the coefficient κ1 to thereby set the threshold L2 (refer to
The process proceeds to Step S109, and the controller sets a value (κ2×σ0) obtained by multiplying σ0 by a coefficient κ2 as the threshold value when the steering angle is zero. In addition, the controller sets a value (κ3×σ1) obtained by multiplying σ1 by a coefficient κ3 as the threshold value when the steering angle is 100 degrees. The controller connects a value (κ2×σ0) obtained by multiplying σ0 by the coefficient κ2 and a value (κ3×σ1) obtained by multiplying σ1 by the coefficient κ3 to thereby set the threshold L3 (refer to
One example of an override determination method will be described next with reference to the flow chart of
In Step S201, the controller detects the motor current supplied to the turning actuator 4 during automatic turning.
If the motor current I1 detected in Step S201 is less than or equal to the threshold value L2 (NO in Step S203), the controller determines the override amount to be 0 (Step S209). In addition, at this time, the controller determines that an override has not occurred (refer to
If the detected motor current I2 is between the threshold value L2 and the threshold value L3 (NO in Step S205), the controller calculates the override amount that is greater than 0 and less than 1, in accordance with the magnitude of the motor current I2 (step S211). The closer the motor current I2 is to the threshold value L3, the closer the override amount is to one. On the other hand, the closer the motor current I2 is to the threshold value L2, the closer the override amount is to zero. In addition, at this time, the controller determines that an override has occurred (refer to
If the detected motor current I3 is greater than the threshold value L3 (YES in Step S205), the controller determines the override amount to be 1 (Step S207). In addition, at this time, the controller determines that an override has occurred (refer to
(Action and Effects)
As described above, the following actions and effects can be achieved by means of the driving assist device according to the present embodiment.
The driving assist device comprises a steering angle sensor that detects the steering angle of the steering wheel 2 installed in a vehicle and a controller that determines an occupant's override with respect to driving assist control that is being executed in the vehicle based on the steering angle detected by the steering angle sensor. The controller acquires a motor current (first motor current) for canceling out the self-aligning torque generated in tires (refer to
As described above, the self-aligning torque increases as the steering angle increases. For this reason, if a determination threshold for an override is set without using the steering angle, an operation of the steering wheel 2 carried out automatically may be erroneously determined as an occupant's operation. However, in the present embodiment, the controller sets the threshold value L2 in accordance with the magnitude of the steering angle to thereby determine the override. The determination accuracy of the override is thereby improved. In other words, erroneous determination of the override can be prevented.
As shown in
The threshold value L2 is set by adding an offset value of the current supplied to the turning actuator 4 when the steering angle of the steering wheel 2 is zero (refer to
The controller sets a threshold value L3 (second threshold value) that is greater than the threshold value L2 (refer to
If the second motor current is greater than the threshold value L3, the controller may make the override amount larger than when the second motor current is smaller than the threshold value L3. As described above, if the second motor current is greater than the threshold value L2 and smaller than the threshold value L3, the override amount is calculated as a value that is greater than 0 and less than 1 (at this stage, this value is referred to as a first override amount). When the second motor current is greater than the threshold value L3, the controller makes the override amount larger than when the second motor current is smaller than the threshold value L3, which means that the override amount is calculated to be larger than the first override amount.
The coefficient for setting the threshold value L2 differs depending on the magnitude of the steering angle. Specifically, when the steering angle is zero, the controller sets a value (κ0×σ0) obtained by multiplying σ0 by a coefficient κ0 as the threshold value. When the steering angle is 100 degrees, the controller sets a value (κ1×σ1) obtained by multiplying σ1 by a coefficient κ1 as the threshold value. σ0 and σ1 are eigenvalues determined in accordance with the mechanism of the steer-by-wire system. By combining such eigenvalues and coefficients having different values depending on the steering angle, the setting flexibility of the threshold L2 increases. As a result, it is possible to set a threshold value in consideration of the individual occupant's operation.
In the present embodiment, a steer-by-wire system is employed. In the steer-by-wire system, the steering wheel 2 and the turning wheels 6a, 6b (front wheels) are mechanically separated.
Each of the functions described in the embodiments above may be implemented by means of one or more processing circuits. The processing circuits include programmed processing devices, such as processing devices including electrical circuits. In addition, the processing circuits include devices such as circuit components and application-specific integrated circuits (ASIC) implemented to execute the described functions.
Embodiments of the present invention have been described above, but the descriptions and figures that form part of this disclosure should not be understood as limiting the present invention. From this disclosure, various alternative embodiments, examples, and operating techniques will be apparent to those skilled in the art.
In the above-described embodiment, a steer-by-wire system was described but the present invention is not limited thereto. For example, the present invention can be applied to an electric power steering system (EPS).
This application is a U.S. national stage application of International Application No. PCT/JP2019/041135, filed on Oct. 18, 2019.
Filing Document | Filing Date | Country | Kind |
---|---|---|---|
PCT/JP2019/041135 | 10/18/2019 | WO |