VEHICLE CONTROL SYSTEM

Abstract

Vehicle control system (10) installed in a vehicle (1), wherein the vehicle (1) has: a wheel (2); a steering wheel (3); and a steering mechanism (5) coupled to the steering wheel (3) and configured to turn the wheel (2) according to a rotation operation of the steering wheel (3), the vehicle control system (10) includes: a torque sensor (42) configured to detect a torque applied to a first position of the steering shaft (4) as a sensor-detected torque (MT); and a control device (100) configured to perform automatic steering control that turns the wheel (2), where: an upper frictional torque (TF) is an absolute value of the sensor-detected torque (MT) caused by a frictional force acting on the steering shaft (4) between the first position and the steering wheel (3) when the steering shaft (4) is rotated; a hands-on state is a driver state in which a driver of the vehicle (1) intervenes in the steering of the vehicle (1); A hands-off state is a driver state other than the hands-on state; and characterized in that the steering mechanism (5) is coupled to the steering wheel (3) by a steering shaft (4), and in that the control device (100) is further configured to perform the following: an estimation process for upper friction that repeatedly estimates the upper frictional torque (TF) based on the sensor-detected torque (MT) when the wheel (2) is turned; a limit value setting process that variably sets a limit value (TH) to be equal to or greater than an estimated value of the upper frictional torque (TF); and a driver state determination processing that determines that the driver state is the hands-on state when the absolute value of the sensor-detected moment (MT) is greater than the determination limit (TH), and determines that the driver state is the hands-off state when the absolute value of the sensor-detected moment (MT) is equal to or less than the determination limit (TH).

Description

[0001] The present invention relates to a vehicle control system installed in a vehicle. In particular, the present invention relates to a vehicle control system that performs automatic steering control.

[0002] JP 2008-290 679 A discloses a steering assistance device installed in a vehicle. The steering assistance device performs lane keeping control such that the vehicle travels along a lane. In particular, the steering assistance device calculates an assistance torque necessary for lane keeping control and controls an EPS (electric power steering) motor such that the assistance torque is applied to a steering mechanism.

[0003] The power steering system also has a function for determining whether or not a driver is performing a steering operation. This determination process considers both a frictional torque Tg, caused by gear friction in the steering mechanism, and a frictional torque Tf, caused by gear friction in the steering column. The frictional torques Tg and Tf are estimated beforehand through tests and given fixed values. If the assistance torque from the EPS motor is greater than the frictional torque Tg, and a steering torque detected by a torque sensor is greater than the frictional torque Tf, the power steering system determines that the driver is performing a steering operation.

[0004] JP 2014-172 610 A discloses a method for detecting an increase in gear friction within a rack-and-pinion EPS device. According to the method, a test current is supplied to a servo motor in the EPS device, and the resulting change in rotor position is detected. An increase in gear friction is detected based on a comparison between the detected change in rotor position and an expected value.

[0005] This section considers a vehicle control system that performs "automatic steering control," which automatically turns one of the vehicle's wheels. During the execution of automatic steering control, a driver can intervene in the vehicle's steering. A driver state in which the driver intervenes in the steering is referred to as a "hands-on state."

[0006] Determining whether the driver is in the hands-on state is crucial for automatic steering control. Consider, for example, a case where the vehicle control system issues a transition request. If the system completes automatic steering control before the driver has transitioned to the hands-on state, the vehicle may drift out of its lane. It is preferable for the system to complete automatic steering control only after confirming that the driver is in the hands-on state.

[0007] According to the technique disclosed in JP 2008-290 679 A above, the frictional torques caused by transmission friction in the steering mechanism and the steering column are taken into account when determining whether the driver's condition is the hand-on condition or not. The frictional torques are estimated in advance through tests and given fixed values.

[0008] However, the magnitude of the frictional torque varies depending on the temperature environment and the age of the parts. Therefore, if the frictional torque is assigned a fixed value, it is highly likely to deviate from the actual value. An increase in this deviation leads to a decrease in the accuracy of determining whether the driver's condition is the hand-on state or not.

[0009] One object of the present disclosure is to provide a technique relating to the vehicle control system that performs automatic steering control and can improve the accuracy of determining whether the driver is in a hands-on state or not. This object is achieved by a system having the features listed in claim 1. Advantageous embodiments are the subject of the dependent claims.

[0010] A first aspect is directed towards a vehicle control system with the features listed in claim 1, which is installed in a vehicle.

[0011] A second aspect further has the feature according to claim 2 in addition to the first aspect.

[0012] A third aspect further has the feature according to claim 3 in addition to the second aspect.

[0013] A fourth aspect further has the feature according to claim 4 in addition to the second aspect.

[0014] According to the present disclosure, the vehicle control system performs driver state determination processing based on a comparison between the sensor-detected torque and the determination limit. Furthermore, the vehicle control system variably sets the determination limit according to a variation of the upper friction torque. In particular, the vehicle control system repeatedly estimates the upper friction torque and sets the determination limit to be equal to or greater than the estimated value of the upper friction torque. Consequently, the accuracy of the driver state determination processing is improved. Brief description of the drawings Fig. Figure 1 is a conceptual diagram to illustrate the sketching of a vehicle control system according to an embodiment of the present disclosure; Fig. Figure 2 is a block diagram that represents a specific configuration example of a vehicle and the vehicle control system according to the embodiment of the present disclosure; Fig. Figure 3 is a block diagram representing a functional configuration of a control device of the control vehicle system according to the embodiment of the present disclosure; Fig. Figure 4 is a conceptual diagram to illustrate an inverse input friction estimation method in the embodiment of the present disclosure; Fig. Figure 5 is a conceptual diagram to explain the inverse input friction estimation method in the embodiment of the present disclosure; Fig. Figure 6 is a conceptual diagram to illustrate a normal input friction estimation method in the embodiment of the present disclosure; Fig. Figure 7 is a block diagram that represents an example of a functional configuration of a setting unit for a variable limit value according to the embodiment of the present disclosure; Fig. Figure 8 is a block diagram that provides a further example of the functional configuration of the setting unit for a variable limit value according to the embodiment of the present disclosure; Fig. Figure 9 is a flowchart that illustrates an example of a determination process for a variable limit value according to the embodiment of the present disclosure; Fig. 10 is a flowchart that represents a first example of a vehicle control system in the embodiment of the present disclosure; Fig. Figure 11 is a flowchart that represents a second example of the vehicle control system in the embodiment of the present disclosure; Fig. Figure 12 is a flowchart that represents a third example of the vehicle control system in the embodiment of the present disclosure; Fig. Figure 13 is a block diagram representing a functional configuration of the control device of the vehicle control system according to the embodiment of the present disclosure; Fig. Figure 14 is a conceptual diagram to illustrate a first example of parameter estimation processing in the embodiment of the present disclosure; Fig. Figure 15 is a conceptual diagram to illustrate a second example of parameter estimation processing in the embodiment of the present disclosure; and Fig. Figure 16 is a flowchart to explain the second example of parameter estimation processing in the embodiment of the present disclosure. Embodiments

[0015] Embodiments of the present disclosure are explained below with reference to the drawings. 1. Sketch of vehicle control system

[0016] Fig. Figure 1 is a conceptual diagram illustrating a sketch of a vehicle control system 10 according to the present embodiment. The vehicle control system 10 is installed in a vehicle 1 and controls an operation of the vehicle 1. In particular, the vehicle control system 10 performs at least “automatic steering control,” which automatically controls the steering of the vehicle 1. Steering the vehicle 1 means turning a wheel 2 of the vehicle 1.

[0017] For example, the vehicle control system 10 determines a target path PT (target trajectory), as in Fig. As shown in Figure 1, the vehicle control system 10 then executes the automatic steering control such that the vehicle 1 follows the target path PT. This automatic steering control is implemented in lane tracing assist (LTA) control, automated driving control, and the like. In the case of automated driving control, the vehicle control system 10 executes not only the automatic steering control but also "automatic driving control," which automatically controls the driving (acceleration and deceleration) of the vehicle 1.

[0018] Fig. Figure 2 is a block diagram that represents a specific configuration example of the vehicle 1 and the vehicle control system 10 according to the present embodiment.

[0019] Vehicle 1 is equipped with a wheel 2, a steering wheel 3, a steering shaft 4, a steering mechanism 5, and a steering column 6. The steering wheel 3 is an operational element used by the driver to perform a steering operation. One end of the steering shaft 4 is coupled to the steering wheel 3, and its other end is coupled to the steering mechanism 5. The steering mechanism 5 turns the wheel 2 in accordance with a rotational operation of the steering wheel 3. Specifically, the steering mechanism 5 includes a pinion coupled to the steering shaft 4, a rack that engages with the pinion, and a tie rod connecting the rack to the wheel 2. A rotation of the steering wheel 3 is transmitted to the pinion via the steering shaft 4. A rotational movement of the pinion is converted into a linear movement of the rack, thereby changing the steering angle θ of the wheel 2.

[0020] The vehicle control system 10 includes an EPS device (Electric Power Steering) 20, a driving device 30, a sensor group 40, an HMI unit (Human Machine Interface) 50, a driving environment information acquisition device 60 and a control device (controller) 100.

[0021] The EPS device 20 incorporates an electric motor and turns the wheel 2 by means of a rotation of the electric motor. For example, the electric motor is coupled to the rack by a conversion mechanism. When a rotor of the electric motor rotates, the conversion mechanism converts the rotational motion into a linear motion of the rack, thereby changing the steering angle θ of the wheel 2. It should be noted that the EPS device 20 is not limited to a rack support type. For example, the EPS device 20 can be a column support type. An operation of the EPS device 20 is controlled by the control device 100.

[0022] The driving device 30 includes a drive device and a braking device. The drive device is an energy source that generates a driving force. The drive device is exemplified by an internal combustion engine and an electric motor. The braking device generates a braking force. An operation of the driving device 30 is controlled by the control device 100.

[0023] Sensor group 40 detects a state of the vehicle 1. For example, sensor group 40 includes a steering wheel angle sensor 41, a torque sensor 42 and a steering angle sensor 43.

[0024] The steering wheel angle sensor 41 detects a steering wheel angle MA, which is a steering angle of the steering wheel 3. The steering wheel angle sensor 41 outputs information specifying the steering angle MA to the control device 100.

[0025] The torque sensor 42 detects a torque exerted on the steering shaft 4. Specifically, the torque sensor 42 is installed at a predetermined position (first position) of the steering shaft 4 and detects a torque exerted on this predetermined position. The torque detected by the torque sensor 42 is hereinafter referred to as a "sensor-detected torque MT". The torque sensor 42 outputs information specifying the sensor-detected torque MT to the control device 100.

[0026] The steering angle sensor 43 detects the steering angle θ of the wheel 2. For example, the steering angle sensor 43 calculates the steering angle θ from a rotation angle of the electric motor of the EPS device 20. The steering angle sensor 43 outputs information specifying the steering angle θ to the control device 100.

[0027] The HMI unit 50 is an interface for providing the driver with information and receiving information from the driver. Specifically, the HMI unit 50 includes an input device and an output device. The input device is exemplified by a touch panel, a switch, a microphone, and the like. The output device is exemplified by a display, a loudspeaker, and the like.

[0028] The driving environment information acquisition device 60 acquires "driving environment information ENV," which specifies a driving environment for vehicle 1. The driving environment information ENV includes position information, map information, environmental situation information, vehicle status information, etc. The position information is information that specifies the position of vehicle 1 and is acquired, for example, by using a GPS (Global Positioning System). The map information specifies a lane configuration and road shape. The environmental situation information is information that specifies a situation around vehicle 1 and is acquired by using an external sensor such as a camera, a LiDAR, radar, and the like. For example, the environmental situation information includes information about a surrounding vehicle and a white line around vehicle 1.The vehicle status information includes vehicle speed, lateral acceleration, yaw rate, etc. This vehicle status information is obtained by sensor group 40. The driving environment information acquisition device 60 transmits the acquired driving environment information (ENV) to the control device 100.

[0029] The controller 100 is a microcomputer equipped with a processor and memory. The controller 100 is also referred to as an ECU (electronic control unit). Various processing operations by the controller 100 are achieved by the processor, which executes a control program stored in the memory. A functional configuration of the controller 100 according to the present embodiment is described in more detail below. 2. Functional configuration of the control device

[0030] Fig. Figure 3 is a block diagram representing a functional configuration of the control device 100 according to the present embodiment. The control device 100 comprises a driving control unit 110, a driver state determination unit 140, and a setting unit 150 for a variable limit value as functional blocks. These functional blocks are accessed by the processor of the control device 100, which executes a control program stored in memory. 2-1. Driving control unit 110

[0031] The driving control unit 110 controls the driving of vehicle 1. In particular, the driving control unit 110 includes a steering control unit 120, which controls the steering of vehicle 1, and a driving control unit 130, which controls the driving (acceleration and deceleration) of vehicle 1.

[0032] The steering control unit 120 performs power steering control, which assists the driver's steering operation. Specifically, the steering control unit 120 calculates an assistance torque based on the sensor-detected torque MT and the vehicle speed. The steering control unit 120 then controls the operation of the EPS device 20 in such a way that the assistance torque is generated. Consequently, the steering effort required by the driver is reduced.

[0033] Furthermore, the steering control unit 120 performs "automatic steering control," which automatically turns wheel 2. The EPS device 20 is also used in the automatic steering control. In particular, the steering control unit 120 determines a steering angle command value θ*, which is a target value for the steering angle θ of wheel 2. The steering control unit 120 also obtains the steering angle θ information from the steering angle sensor 43. Alternatively, the steering control unit 120 can calculate the steering angle θ from the steering wheel angle MA. The steering control unit 120 then controls the operation of the EPS device 20 such that the steering angle θ follows the steering angle command value θ.

[0034] For example, the steering control unit 120 executes the automatic steering control in such a way that the vehicle 1 follows the target path PT (see Fig. 1) For this purpose, the steering control unit 120 calculates the target path PT and a target path steering angle θPT based on the driving environment information ENV. The target path steering angle θPT is the steering angle θ required for the vehicle 1 to follow the target path PT. Various methods for calculating the target path PT and the target path steering angle θPT have been proposed. In the present embodiment, the method for calculating them is not specifically limited. The steering control unit 120 sets the steering angle command value θ* to the target path steering angle θPT and controls the operation of the EPS device 20 such that the steering angle θ follows the target path steering angle θPT. Consequently, the vehicle 1 drives to follow the target path PT. This is how automatic steering control is performed in automated driving control, LTA, and the like.

[0035] The vehicle control unit 130 performs "automatic vehicle control," which automatically controls the movement of vehicle 1. Automatic vehicle control includes acceleration and deceleration control. The vehicle control unit 130 performs automatic vehicle control by controlling the operation of the vehicle device 30 (the drive device and the brake device). For example, the vehicle control unit 130 performs automatic vehicle control such that vehicle 1 travels along the target path PT.

[0036] The driving control unit 110 can perform “automated driving control”, which includes both the automatic steering control and automatic driving control described above.

[0037] In addition, the vehicle control unit outputs various messages to the driver via the HMI unit 50 (output device). These messages include, for example, an alarm, a transition requirement (TD), etc. The transition requirement prompts the driver to start manual driving.

[0038] The driver can use the HMI unit 50 (input device) to turn on / off (activate / deactivate) the driving control (i.e., automatic steering control, automatic driving control and automated driving control) performed by the driving control unit 110. 2-2. Driver condition assessment unit 140

[0039] While the automatic steering control is being executed by the vehicle control unit 110, the driver can intervene in the steering of the vehicle 1. A driver state in which the driver intervenes in the steering of the vehicle 1 is hereinafter referred to as a "hands-on state". For example, a driver state in which the driver performs the steering operation, i.e., the driver actively operates the steering wheel, is the hands-on state. As another example, a driver state in which the driver holds the steering wheel 3 to counteract the automatic steering control by the vehicle control unit 110 is also the hands-on state.

[0040] On the other hand, a driver state in which the driver does not intervene in the steering of vehicle 1 is subsequently referred to as a "hands-off state". One could say that the hands-off state is the driver state excluding the hands-on state.

[0041] Determining whether the driver state is the hands-on or hands-off state is important for automatic steering control. As an example, consider a case where the transition requirement described above is issued by the vehicle control unit 110. If the vehicle control unit 110 completes the automatic steering control at a time when the driver state has not yet become the hands-on state, it is possible that the vehicle 1 will leave its lane. It is preferred that the vehicle control unit 110 completes the automatic steering control after confirming that the driver state is the hands-on state.

[0042] The driver state determination unit 140 is provided for this purpose. The driver state determination unit 140 performs a "driver state determination processing" that determines whether the driver state is the hands-on state or the hands-off state. In particular, the driver state determination unit 140 compares the sensor-detected torque MT, which is detected by the torque sensor 42, with a "determination threshold TH". If an absolute value of the sensor-detected torque MT is greater than the determination threshold TH (|MT| > TH), the driver state determination unit 140 determines that the driver state is the hands-on state. Conversely, if the absolute value of the sensor-detected torque MT is equal to or less than the determination threshold TH (|MT| ≤ TH), the driver state determination unit 140 determines that the driver state is the hands-off state. 2-3. Unit of measurement 150 for a variable limit value

[0043] The setting unit 150 for a variable limit value defines the determination limit value TH, which is used in the driver state determination processing described above. When setting the determination limit value TH, a frictional force caused by parts such as gears must be taken into account. The reason for this is that a frictional torque caused by the frictional force may be included in the sensor-detected torque MT.

[0044] In the present embodiment, the focus is particularly on an “upper frictional force” that is upstream of the torque sensor 42. Specifically, the upper frictional force is a frictional force acting on the steering shaft 4 between the installation position (i.e., the first position) of the torque sensor 42 and the steering wheel 3 when the steering shaft 4 is rotated. For example, the upper frictional force is caused by components such as gears in the steering column 6. When the steering shaft 4 is rotated, the sensor-detected torque MT, detected by the torque sensor 42, includes a frictional torque caused by the upper frictional force. An absolute value of a component of the sensor-detected torque MT caused by the upper frictional force is hereinafter referred to as an “upper frictional torque TF”.

[0045] Consider a situation where the driver takes their hands off the steering wheel 3 (i.e., a hands-free state). When the driving control unit 110 executes the automatic steering control to turn the wheel 2, the steering shaft 4 and the steering wheel 3 rotate accordingly. In this case, the torque sensor 42 detects the upper frictional torque TF, caused by the upper frictional force, as the sensor-detected torque MT. That is, the upper frictional torque TF ≠ zero is detected as the sensor-detected torque MT, even though the driver takes their hands off the steering wheel 3.

[0046] To avoid incorrect determination of the driver's condition, the determination limit TH is set equal to or greater than the upper friction torque TF. However, the magnitude of the upper friction torque TF is not constant but varies. In particular, the magnitude of the upper friction torque TF varies according to the temperature environment and the age-related deterioration of parts.

[0047] With regard to the foregoing, according to the present embodiment, the variable limit setting unit 150 variably sets the determination limit TH according to the variation of the upper friction torque TF. In particular, the variable limit setting unit 150 repeatedly estimates the upper friction torque TF. Furthermore, the variable limit setting unit 150 sets the determination limit TH such that it is equal to or greater than the estimated value of the upper friction torque TF. Consequently, the accuracy of the driver state determination processing is improved. This contributes to an increase in confidence in the vehicle control system 10.

[0048] The following describes in more detail a setting process for a variable limit value by the setting unit 150 for a variable limit value according to the present embodiment. 3. Determination Processing for a Variable Limit 3.1 Inverse Input Friction Estimation Method

[0049] First, a method for estimating the upper friction torque TF in a freehand state, in which the driver removes their hands from the steering wheel 3, is described. This method is subsequently referred to as an "inverse input friction estimation method".

[0050] Fig. Figure 4 is a conceptual diagram to illustrate the inverse input friction estimation method. In the case of the freehand state, the vehicle control unit 110 executes the automatic steering control to turn the wheel 2. Fig. Figure 4 shows a variation of the sensor-detected torque MT when wheel 2 is turned by the automatic steering control in hands-free mode. A vertical axis represents the sensor-detected torque MT, and a horizontal axis represents a steering parameter that represents the turning of wheel 2. In the Fig. In example 4, the steering wheel angle MA is used as the steering angle parameter. It is also possible to use the steering angle θ, the yaw rate, or the lateral acceleration instead of the steering wheel angle MA.

[0051] If wheel 2 is turned in the hands-free state, the steering wheel 3 is rotated accordingly, and the torque is generated in the opposite direction to that of the steering operation by the driver, as shown in Fig. Figure 4 illustrates this. Specifically, the sensor-detected torque MT assumes a negative value when the steering wheel angle MA increases, and a positive value when the steering wheel angle MA decreases. In both cases, the absolute value of the sensor-detected torque MT is nearly constant. This constant value corresponds to the upper frictional torque TF. One can say that the upper frictional torque TF is the absolute value of the sensor-detected torque MT at the time when the steering wheel 3 begins to rotate due to the automatic steering control in the hands-free state.

[0052] Furthermore, as in Fig. As shown in Figure 4, the hysteresis width HIS of the sensor-detected torque MT remains almost constant regardless of the steering wheel angle MA. Half of the hysteresis width HIS corresponds to the upper frictional torque TF.

[0053] The variable limit setting unit 150 estimates the upper friction torque TF based on the sensor-detected torque MT when the wheel 2 is turned by the automatic steering control. Specifically, the variable limit setting unit 150 estimates the absolute value of the sensor-detected torque MT at the time the steering wheel 3 begins to rotate due to the automatic steering control as the upper friction torque TF. As another example, the variable limit setting unit 150 estimates half the hysteresis width HIS of the sensor-detected torque MT due to the automatic steering control as the upper friction torque TF. Using the hysteresis width HIS makes it possible to eliminate the influence of noise and vibration components.

[0054] Fig. Figure 5 shows a situation in which the upper frictional torque TF increases or is elevated. As the upper frictional torque TF increases, the hysteresis width HIS also increases. The setting unit 150 for a variable limit value can obtain a final value for the upper frictional torque TF by repeatedly executing the estimation process. 3.2 Normal Input Friction Estimation Method

[0055] Next, a method for estimating the upper friction torque TF in a steering state where the driver is performing the steering operation is described. This method is subsequently referred to as a "normal input friction estimation method".

[0056] Fig. Figure 6 is a conceptual diagram to explain the normal input friction estimation method. A format of Fig. 6 is the same as in Fig. 4. As is generally known, the MA-MT property in the steering state is represented by a Lissajous signal waveform. Unlike the case of Fig. 4. The absolute value of the sensor-detected torque MT is not constant. The variation tendency of the sensor-detected torque MT also differs from that in the case of Fig. 4. In particular, the sensor-detected torque MT increases when the steering wheel angle MA increases. Conversely, the sensor-detected torque MT decreases when the steering wheel angle MA decreases.

[0057] The variable limit setting unit 150 estimates the upper friction torque TF based on the sensor-detected torque MT when wheel 2 is turned by the driver's steering operation. Specifically, the variable limit setting unit 150 estimates half the hysteresis width HIS of the sensor-detected torque MT at a steering angle neutral point (i.e., MA = zero) as the upper friction torque TF.

[0058] It should be noted that the upper frictional torque TF, estimated by the normal input friction estimation method, may include the influence of a lower frictional force. The lower frictional force is a frictional force on the side of wheel 2 of the torque sensor 42. For example, the lower frictional force is caused by gears in the impact mechanism 5. A lower frictional torque is an absolute value of a component of the sensor-detected torque MT caused by the lower frictional force and is pre-calculated. The variable limit setting unit 150 can estimate a value obtained by subtracting the lower frictional torque from half the hysteresis width HIS as the upper frictional torque TF. 3.3 Functional configuration example of the setting unit for a variable limit value

[0059] Fig. Figure 7 is a block diagram illustrating an example of a functional configuration of the setting unit 150 for a variable limit value. The setting unit 150 for a variable limit value includes a freehand determination unit 151, an estimation unit 152 for upper friction, and a limit value setting unit 153.

[0060] The hands-free determination unit 151 performs "hands-free determination processing". In hands-free determination processing, the hands-free determination unit 151 determines whether or not the driver takes their hands off the steering wheel 3. In other words, the hands-free determination unit 151 determines whether or not the driver's state is the hands-free state.

[0061] For example, the freehand determination unit 151 performs the freehand determination processing based on the variation tendency of the sensor-detected torque MT when the wheel 2 is turned. In the case of the freehand state, as in Fig. As shown in Figure 4, the sensor-detected torque MT assumes a negative value when the steering wheel angle MA increases, and the sensor-detected torque MT assumes a positive value when the steering wheel angle MA decreases. On the other hand, in the case of Fig. 6. The sensor-detected torque MT increases when the steering wheel angle MA increases, and the sensor-detected torque MT decreases when the steering wheel angle MA decreases. It is therefore possible to determine whether the driver's state is hands-free or not, based on the variation tendency of the sensor-detected torque MT.

[0062] Fig. Figure 8 shows a modification example. In this modification example, the vehicle control system 10 also has a driver monitor 70. The driver monitor 70 includes a steering wheel touch sensor, a gap sensor, a camera, and the like. The hands-free determination unit 151 determines whether or not the driver takes their hands off the steering wheel 3, based on a measurement result from the driver monitor 70.

[0063] The upper friction estimation unit 152 performs an "upper friction estimation process." In the upper friction estimation process, the upper friction estimation unit 152 repeatedly estimates the upper friction torque TF based on the sensor-detected torque MT when the wheel 2 is turned. Specifically, when the driver removes their hands from the steering wheel 3 while performing automatic steering control, the upper friction estimation unit 152 estimates the upper friction torque TF using the inverse input friction estimation method (see Figure 152). Fig. 4 and Fig. 5) Therefore, it is possible to estimate the upper frictional torque TF with high accuracy.

[0064] When the driver performs the steering operation, the estimation unit 152 can estimate the upper friction torque TF using the normal input friction estimation method (see below). Fig. 6).

[0065] The limit setting unit 153 performs a "limit setting process". In this process, the limit setting unit 153 variably sets the limit value TH according to the estimated upper frictional torque TF. Specifically, the limit setting unit 153 sets the limit value TH to be equal to or greater than the estimated value of the upper frictional torque TF. For example, the limit value TH is expressed by the following equation (1). TH=TF+α≤LIM

[0066] In equation (1), parameter α is a tolerance with respect to an error and parameter LIM is an upper limit. If the limit value TH exceeds the upper limit LIM, the limit setting unit 153 can inform the driver of an anomaly via the HMI unit 50. 3-4. Example of a process for determining a variable limit value

[0067] Fig. Figure 9 is a flowchart that illustrates an example of the specification processing for a variable limit value by the specification unit 150 for a variable limit value. A processing flow that is in Fig. The process shown in 9 is executed repeatedly for each specific cycle.

[0068] In step S10, the freehand determination unit 151 performs the freehand determination processing. If the driver removes their hands from the steering wheel 3 (step S10; yes), the processing continues with step S20. Otherwise (step S10; no), the processing continues with step S40.

[0069] In step S20, the estimation unit 152 estimates the upper frictional torque TF for an upper friction using the inverse input friction estimation method (see below). Fig. 4 and Fig. 5). Processing then continues with step S30.

[0070] In step S30, the estimation unit 152 sets a flag FL to "1" for upper friction. Flag FL indicates the presence or absence of a history of the inverse input friction estimation. An initial value of the flag is "zero". Processing then continues with step S60.

[0071] In step S40, the estimation unit 152 for upper friction determines whether or not the FL flag is "zero". If the FL flag is "zero" (step S40; yes), processing continues with step S50. Conversely, if the FL flag is "1" (step S40; no), processing ends in the current cycle.

[0072] In step S50, the estimation unit 152 estimates the upper frictional torque TF for an upper friction using the normal input friction estimation method (see below). Fig. 6). Processing then continues with step S60.

[0073] In step S60, the limit setting unit 153 sets the determination limit TH according to the preceding equation (1). Processing then continues with step S70.

[0074] In step S70, the limit setting unit 153 compares the limit value TH with the upper limit LIM. If the limit value TH is equal to or less than the upper limit LIM (step S70; yes), processing ends in the current cycle. Conversely, if the limit value TH exceeds the upper limit LIM (step S70; no), processing continues with step S80.

[0075] In step S80, the control unit 150 performs error handling for a variable limit value. For example, the control unit 150 informs the driver of an anomaly via the HMI unit 50. Furthermore, the vehicle control unit 110 can inform the driver of a "completion of automatic steering control" by the HMI unit 50. In this case, the vehicle control unit 110 completes the automatic steering control after confirming that the hands-on state has persisted for a specific period of time.

[0076] According to the processing procedure described in Fig. As shown in Figure 9, the limit value TH changes, for example, as follows. First, the driver performs manual driving. The limit value TH is set based on the upper friction torque TF, which is estimated using the normal input friction estimation method. Then, the automated driving control is started by the driving control unit 110. After the automated driving control is started, the limit value TH is set based on the upper friction torque TF, which is estimated using the inverse input friction estimation method. This means that the accuracy of the limit value TH is improved. The limit value TH is then updated repeatedly. 4. Different examples of vehicle control

[0077] The driving control unit 110 executes driving control based on a result of the driver state determination processing by the driver state determination unit 140. Different examples of driving control are described below. 4-1. First example

[0078] Fig. 10 is a flowchart that represents a first example of vehicle control. A Fig. The processing sequence shown in Figure 10 is executed repeatedly for each specific cycle.

[0079] In step S100, the drive control unit 110 determines whether the drive control is on (activated). The driver can switch the drive control on / off using the HMI unit 50. If the drive control is off (step S100; no), processing ends. Conversely, if the drive control is on (step S100; yes), processing continues with step S110.

[0080] In step S110, the vehicle control unit 110 refers to the result of the driver state determination processing. If the driver state is the hands-on state, or if the hands-on state persists for a certain period of time (step S110; yes), the processing continues with step S120. Conversely, if the driver state is the hands-off state, or if the hands-off state persists for a certain period of time (step S110; no), the processing continues with step S130. In step S120, the vehicle control unit 110 performs the

[0081] Lane Tracking Assistance (LTA) control is activated. At step S130, the vehicle control unit 110 executes the automated driving control. In this way, as in the first example, the vehicle control unit 110 switches the driving control between Lane Tracking Assist and Automated Driving according to the driver's state. 4.2 Second example

[0082] Fig. 11 is a flowchart that presents a second example of vehicle control. One in Fig. The processing sequence shown in step 11 is executed repeatedly in each specific cycle.

[0083] In step S200, the driving control unit 110 executes the driving control, assuming the hands-on state is in place. Driving control with the assumed hands-on state is exemplified by the steering assist control during manual driving, the lane following assist (LTA) control, etc.

[0084] In step S210, the vehicle control unit 110 refers to the result of the driver state determination processing. If the driver state is the hands-off state, or if the hands-off state persists for a certain period of time (step S210; yes), the processing continues with step S220. Otherwise (step S210; no), the processing ends in the current cycle.

[0085] At step S220, the driving control unit 110 sends an alarm to the driver via the HMI unit 50. For example, the driving control unit 110 issues an alarm message such as "Hold steering wheel". Consequently, the driver's state is expected to return to the hands-on state. 4-3. Third example

[0086] Fig. Figure 12 is a flowchart that represents a third example of vehicle control. A processing sequence that is shown in Fig. The process shown in 12 is executed repeatedly for each specific cycle.

[0087] In step S300, the driving control unit 110 executes the automated driving control.

[0088] In step S310, the vehicle control unit 110 informs the driver about the transition requirement using the HMI unit 50. Processing then continues with step S320.

[0089] In step S320, the vehicle control unit 110 refers to the result of the driver state determination processing. If the driver state is the hands-on state, or if the hands-on state persists for a certain period of time (step S320; yes), the processing continues with step S330. Otherwise (step S320: no), the processing ends in the current cycle.

[0090] At step S330, the drive control unit 110 completes the automated driving control. At this time, the drive control unit 110 can issue a confirmation message such as "Switching to manual driving" via the HMI unit 50. 5. Effects

[0091] According to the present embodiment, as described above, the vehicle control system performs the driving state determination processing based on the comparison between the sensor-detected torque MT and the determination limit TH. Furthermore, the vehicle control system 10 variably sets the determination limit TH according to a variation of the upper friction torque TF. In particular, the vehicle control system 10 repeatedly estimates the upper friction torque TF and sets the determination limit TH to be equal to or greater than the estimated value of the upper friction torque TF. Consequently, the accuracy of the driver state determination processing is improved.

[0092] As a first comparative example, we consider the technology disclosed in the aforementioned patent document 1 (Japanese unexamined patent application with publication number JP-2008-2090679). According to this first comparative example, the upper friction torque TF is estimated in advance through testing and assigned a fixed value. However, the actual upper friction torque TF varies, and thus the assigned value of the upper friction torque TF deviates from the actual value. An increase in this deviation causes a decrease in the accuracy of the driver state determination processing.

[0093] As a second comparative example, consider a case in which the determination limit TH is set to a sufficiently large or sufficiently small value, anticipating variation in the upper friction torque TF. In this case, the set value of the upper friction torque TF also deviates from the actual value, thus reducing the accuracy of the driver state determination processing.

[0094] If the determination threshold TH is set to an unnecessarily small value, the following problem arises. For example, the driver performs the steering operation in response to the transition requirement. The automatic steering control completes when it is determined that the driver state is the hands-on state. However, if the determination threshold TH is too small, it is likely that the driver state will be incorrectly determined to be the hands-on state, even though the actual driver state is still the hands-off state. If the automatic steering control completes regardless of the erroneous determination, vehicle 1 may leave its lane.

[0095] If the threshold value TH is set to an unnecessarily large value, the following problem arises. For example, the driver executes the steering operation in response to the transitional requirement. The automatic steering control continues until a point is reached when the driver's state is determined to be the hands-on state. As the threshold value TH increases, this point is further delayed, and thus the driver is likely to experience a rather strange sensation in the steering operation. It is also conceivable that the driver executes the steering operation quickly to avoid a collision, but the vehicle's avoidance behavior is delayed. In this way, if the threshold value TH is unnecessarily large, controllability with respect to steering deteriorates.

[0096] According to the present embodiment, the accuracy of the driver state determination processing is improved. Accordingly, erroneous determination of the driver state is suppressed. Furthermore, the decrease in controllability with regard to steering is suppressed. This leads to an increase in confidence in the vehicle control system 10. 6. Estimation of driver steering torque

[0097] The control device 100 according to the present embodiment can have a function for estimating a driver steering torque MTD. The driver steering torque MTD, which is a torque exerted by the driver to rotate the steering wheel 3, indicates the strength of the steering operation by the driver. The driver state determination unit 140 described above can perform the driver state determination processing using the estimated driver steering torque MTD instead of the sensor-detected torque MT.

[0098] Fig. Figure 13 is a block diagram representing a functional configuration of the control device 100. The control device 100 includes a driver steering torque estimation unit 160 and a parameter estimation unit 170 in addition to the preceding functional configuration. These functional blocks are accessed by the processor of the control device 100, which executes a control program stored in memory. The driver steering torque estimation unit 160 performs

[0099] “Driver steering torque estimation processing” is used to estimate the driver steering torque MTD. A method for estimating the driver steering torque MTD is not particularly restricted. For example, the driver steering torque estimation unit 160 uses a model of a steering system to estimate the driver steering torque MTD. In this case, the driver steering torque estimation unit 160 can estimate the driver steering torque MT based on the sensor-detected torque MT, a motor rotation angle δ, and a motor rotation angular velocity dδ / dt. The motor rotation angle δ, which is a rotation angle of a rotor of the electric motor of the EPS device 20, is detected by a rotation angle sensor (not shown). The motor rotation angular velocity dδ / dt is obtained by differentiating the motor rotation angle δ.

[0100] The steering system model used in driver steering torque estimation processing may include a mechanical parameter that varies according to an external environment. A typical example of a mechanical parameter that varies with the external environment is mechanical friction, which varies with ambient air temperature. If this mechanical parameter is fixed to a nominal value—that is, if a nominal model is used regardless of variation in the external environment—the estimation accuracy for driver steering torque (MTD) will be reduced.

[0101] The parameter estimation unit 170 is designed to suppress the decrease in estimation accuracy for the driver steering torque (MTD). Parameter estimation unit 170 performs "parameter estimation processing," which estimates the mechanical parameter according to the external environment. The estimated mechanical parameter is subsequently referred to as an "estimated parameter F." Parameter estimation unit 70 outputs the estimated parameter F to driver steering torque estimation unit 160. Driver steering torque estimation unit 160 uses the model based on the estimated parameter F to perform driver steering torque estimation processing. Consequently, the decrease in estimation accuracy for driver steering torque (MTD) is suppressed.

[0102] It should be noted that the parameter estimation unit 170 performs the parameter estimation processing for each specific cycle. The estimated parameter F is updated each time the parameter estimation processing is performed. The parameter estimation unit 170 can store a history of the estimated parameter F, starting from its initial value, in the memory of the control device 100.

[0103] Examples of parameter estimation processing using the parameter estimation unit 170 are described below. 6-1. First example

[0104] Fig. Figure 14 is a conceptual program for illustrating a first example of parameter estimation processing. In this first example, a mechanical parameter (for example, mechanical friction) is considered, which varies according to the ambient air temperature T. In this case, the estimated parameter is expressed as a function of the ambient air temperature T. The function, i.e., a correspondence relationship between the ambient air temperature T and the estimated parameter F, is given by a record or a formula. The parameter estimation unit 170 receives information about the ambient air temperature T, which is acquired by an ambient air temperature sensor (not shown). The parameter estimation unit 170 then obtains the estimated parameter F based on the ambient air temperature T and the function described above.

[0105] If the ambient air temperature sensor fails, the parameter estimation unit 170 can stop parameter estimation processing. In this case, the parameter estimation unit 170 outputs the estimated parameter obtained immediately before the ambient air temperature sensor failed. Since the ambient air temperature does not change rapidly, using the estimated parameter F obtained immediately before the ambient air temperature sensor failed can suppress the decrease in estimation accuracy for the driver steering torque (MTD).

[0106] If the fault or failure of the ambient air temperature sensor persists for a certain period of time or longer, the ambient air temperature may change during that period. In this case, the parameter estimation unit 170 can gradually adjust the estimated parameter F (output) back to its initial value. 6-2. Second example

[0107] Fig. Figure 15 is a conceptual diagram for deriving a second example of parameter estimation processing. In the second example, the parameter estimation unit 170 includes an estimation model 171 for a sensor-detected torque.

[0108] The estimation model 171 for a sensor-detected torque is a model for estimating the sensor-detected torque MT, which is detected by the torque sensor 42. The estimation model 171 for a sensor-detected torque is obtained by modifying the model used by the driver steering torque estimation unit 160 to estimate the driver steering torque MTD.

[0109] The parameter estimation unit 170 estimates the sensor-detected torque MT while setting a model parameter of the estimation model 171 to different values ​​for a sensor-detected torque. The parameter estimation unit 170 compares each estimated value of the sensor-detected torque MT with an actual value of the sensor-detected torque MT. Then, the parameter estimation unit 170 outputs a model parameter that most closely approximates the estimated value to the actual value, as the estimated parameter F.

[0110] Fig. Figure 16 is a flowchart to illustrate the second example of parameter estimation processing. A processing flow that is described in Fig. The process shown in 16 is executed repeatedly for each specific cycle.

[0111] In step S400, the parameter estimation unit 170 determines whether or not the steering operation is performed by the driver. For example, the parameter estimation unit 170 compares an absolute value of the driver steering torque MTD, estimated by the driver steering torque estimation unit 160, with a predetermined limit value. If the absolute value of the driver steering torque MTD is greater than the predetermined limit value, the parameter estimation unit 170 determines that the steering operation is performed by the driver. As another example, the parameter estimation unit 170 can determine whether or not the steering operation is performed by the driver based on a measurement result from the driver monitor 70 (see Figure 160). Fig. 8).

[0112] If the steering operation is performed by the driver (step S400; yes), the estimation accuracy for the sensor-detected torque MT may decrease. In this case, the parameter estimation unit 170 therefore outputs a final value of the estimated parameter F without executing the parameter estimation processing (step S410).

[0113] On the other hand, if the steering operation is not performed by the driver (step S400; no), the parameter estimation unit 170 performs the parameter estimation processing. The parameter estimation processing includes the following steps S420 to S440.

[0114] In step S420, the parameter estimation unit 170 defines the model parameter of the estimation model 171 for a sensor-detected moment. Specifically, the parameter estimation unit 170 defines the model parameter in N patterns (where N is an integer equal to or greater than 2). For example, the parameter estimation unit 170 defines N types of a model parameter that follow a normal distribution with a median µ. The median µ is, for example, the last value of the estimated parameter F. As another example, the median µ can be extrapolated to the last value and one value before the last value of the estimated parameter F. The N types of model parameters can include a nominal value.

[0115] In step S430, the parameter estimation unit 170 estimates the sensor-detected torque MT using the sensor-detected torque estimation model 171 based on each of the N types of model parameters. For example, an input to the sensor-detected torque estimation model 171 is the motor rotation angle δ. A motor target command torque or a target current value can be used instead of the motor rotation angle δ.

[0116] In step S440, the parameter estimator 170 compares each of the estimated values ​​of the sensor-detected torque MT with an actual value of the sensor-detected torque MT. The parameter estimator 170 extracts a model parameter that can be used to obtain the estimated value closest to the actual value, and designates this as the optimal parameter. For example, the parameter estimator 170 calculates a difference between the actual value and the estimated value and extracts a model parameter that minimizes the difference as the optimal parameter. The parameter estimator 170 then outputs the extracted optimal parameter as the estimated parameter F. 6-3. Third example

[0117] As described above, the estimation unit 142 estimates the upper frictional torque TF for an upper friction (see above). Fig. 4 to Fig.9) The parameter estimation unit 170 can estimate a friction term (the estimated parameter F) of the steering system model based on the upper friction torque TF, which is estimated by the estimation unit 152 for an upper friction.

Claims

[1] Vehicle control system (10) installed in a vehicle (1), the vehicle (1) comprising: a wheel (2); a steering wheel (3); and a steering mechanism (5) coupled to the steering wheel (3) and configured to turn the wheel (2) according to a rotation operation of the steering wheel (3), wherein the vehicle control system (10) comprises: a torque sensor (42) configured to detect a torque applied to a first position of the steering shaft (4) as a sensor-detected torque (MT); and a control device (100) configured to perform automatic steering control that turns the wheel (2), where: an upper friction torque (TF) is an absolute value of the sensor-detected torque (MT) caused by a friction force acting on the steering shaft (4) between the first position and the steering wheel (3) when the steering shaft (4) is rotated; a hands-on state is a driver state in which a driver of the vehicle (1) intervenes in the steering of the vehicle (1); a hands-off state is a driver state other than the hands-on state; and characterized by that the steering mechanism (5) is coupled to the steering wheel (3) by a steering shaft (4), and in that the control device (100) is further configured to perform the following: an upper friction estimation processing that repeatedly estimates the upper friction torque (TF) based on the sensor-detected torque (MT) when the wheel (2) is turned; a limit setting processing that variably sets a determination limit value (TH) to be equal to or greater than an estimated value of the upper friction torque (TF); and a driver state determination processing that determines that the driver state is the hands-on state when the absolute value of the sensor detected torque (MT) is greater than the determination threshold (TH), and determines that the driver state is the hands-off state when the absolute value of the sensor detected torque (MT) is equal to or less than the determination threshold (TH). [2] Vehicle control system (10) according to claim 1, characterized by that the control device (100) determines whether or not the driver takes his hands off the steering wheel (3), and when the driver takes his hands off the steering wheel (3) during execution of the automatic steering control, the control device (100) executes the upper friction estimation processing. [3] Vehicle control system (10) according to claim 2, characterized by in the upper friction estimation processing, the control device (100) estimates the absolute value of the sensor-detected torque (MT) at a time when the steering wheel (3) starts to be rotated due to the automatic steering control as the upper friction torque (TF). [4] Vehicle control system (10) according to claim 2, characterized by in the upper friction estimation processing, the control device (100) estimates a half of a hysteresis width (HIS) of the sensor-detected torque (MT) due to the automatic steering control as the upper friction torque (TF).

Images