Control device, electric power steering apparatus, and control method

Abstract

To improve accuracy of lane keeping control while reducing a sense of discomfort felt by a steering person from a steering wheel.SOLUTION: A control device includes: a first assist control section which can execute lane keeping control and generates a first input value; a second assist control section which generates a second input value on the basis of a steering input value; and a disturbance sensitivity control section to which a command value calculated on the basis of the first input value multiplied by a first gain and the second input value multiplied by a second gain is input. When the first gain is K1, and the second gain is K2, the following relation is satisfied: K1=1-K2. The disturbance sensitivity control section includes a first model following control section which generates a correction value for correcting the command value, on the basis of output of a control target and a first nominal model.SELECTED DRAWING: Figure 4

Description

[Technical field]

[0001] The present invention relates to a control device, an electric power steering device, and a control method. [Background technology]

[0002] 2. Description of the Related Art Conventionally, an electric power steering system mounted on a vehicle is known (for example, see Patent Document 1). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2018-183046 A Summary of the Invention [Problem to be solved by the invention]

[0004] The electric power steering system as described above may include, for example, a control device capable of executing lane keeping control. The lane keeping control is a control that applies a driving force to a steering mechanism by a motor to keep the vehicle in the lane when the vehicle is about to deviate from the lane. When the control amount of the lane keeping control is increased, the steering wheel feels more uncomfortable to the steerer. On the other hand, when the control amount of the lane keeping control is reduced in order to reduce the steering wheel feeling of the steerer, it becomes difficult to sufficiently control the vehicle by the lane keeping control, and there is a problem that the accuracy of keeping the vehicle in the lane is reduced.

[0005] In consideration of the above circumstances, one of the objects of the present invention is to provide a control device and a control method that can improve the accuracy of lane keeping control while reducing the discomfort that the helmsman feels from the steering wheel, and an electric power steering device equipped with such a control device. [Means for solving the problem]

[0006] One aspect of the control device of the present invention is a control device for controlling a steering mechanism mounted on a vehicle, capable of executing lane keeping control for keeping the vehicle within a lane, and including a first assist control unit for generating a first input value, a second assist control unit for generating a second input value based on a steering input value input from a steering wheel of the vehicle, and a disturbance sensitivity control unit to which a command value calculated based on the first input value multiplied by a first gain and the second input value multiplied by a second gain is input. The first gain and the second gain are variable values. When the first gain is K1 and the second gain is K2, K1=1-K2 holds. The disturbance sensitivity control unit has a first model following control unit for generating a correction value for correcting the command value based on an output of a control object including a motor provided in the steering mechanism and a first nominal model based on a configuration of the control object. The first model following control unit is configured so that a transfer function of the controlled object is constrained to a transfer function of the first nominal model in a frequency band where a complementary sensitivity gain, which is a gain in a gain characteristic of a complementary sensitivity function with respect to a modeling error between the controlled object and the first nominal model, is approximately 1. The complementary sensitivity function includes a variable parameter that changes based on the second gain.

[0007] One aspect of an electric power steering device of the present invention includes the above control device and the steering mechanism.

[0008] One aspect of the control method of the present invention is a control method for controlling a steering mechanism mounted on a vehicle, the control method including: executing lane keeping control for keeping the vehicle within a lane, generating a first input value by the lane keeping control, generating a second input value based on a steering input value input from a steering wheel of the vehicle, calculating a command value based on the first input value multiplied by a first gain and the second input value multiplied by a second gain, executing model following control for generating a correction value for correcting the command value based on an output of a control object including a motor provided in the steering mechanism and a first nominal model based on a configuration of the control object, and constraining a transfer function of the control object to a transfer function of the first nominal model by the model following control in a frequency band where a complementary sensitivity gain, which is a gain in a gain characteristic of a complementary sensitivity function with respect to a modeling error between the control object and the first nominal model, is approximately 1. The first gain and the second gain are variable values. When the first gain is K1 and the second gain is K2, K1=1-K2 holds true. The complementary sensitivity function includes a variable parameter that changes based on the second gain. Effect of the Invention

[0009] According to one aspect of the present invention, it is possible to improve the accuracy of lane keeping control while reducing the discomfort felt by the driver from the steering wheel. [Brief description of the drawings]

[0010] [Figure 1] FIG. 1 is a diagram illustrating an electric power steering device according to a first embodiment. [Diagram 2] FIG. 2 is a block diagram showing the configuration of the control device in the first embodiment. [Diagram 3] FIG. 3 is a diagram illustrating a case where a vehicle equipped with the lane keeping system in the first embodiment travels along a lane. [Figure 4] FIG. 4 is a block diagram showing the configuration of the lane keeping system in the first embodiment. [Diagram 5]FIG. 5 is a block diagram showing the configuration of the imaging device and the first assist control unit in the first embodiment. [Figure 6] FIG. 6 is a diagram for explaining the lane keeping control in the first embodiment. [Figure 7] FIG. 7 is a block diagram showing the configuration of the lane keeping control unit in the first embodiment. [Figure 8] FIG. 8 is a block diagram showing the configuration of the lateral deviation control unit in the first embodiment. [Figure 9] FIG. 9 is a Bode diagram showing an example of the gain of the transfer function of the lateral deviation adjustment unit and the gain of the transfer function of the yaw rate adjustment unit in the first embodiment. [Figure 10] FIG. 10 is a block diagram showing the configuration of the yaw rate control unit in the first embodiment. [Figure 11] FIG. 11 is a block diagram showing the configuration of the second assist control unit in the first embodiment. [Figure 12] FIG. 12 is a graph showing an example of the relationship between the input torque and the assist torque in the first embodiment. [Figure 13] FIG. 13 is a block diagram showing the configuration of the vehicle stabilization control unit in the first embodiment. [Figure 14] FIG. 14 is a Bode plot showing the gain of the transfer function of the second nominal model in the first embodiment. [Figure 15] FIG. 15 is a block diagram showing a configuration of the second state feedback unit in the first embodiment. [Figure 16] FIG. 16 is a block diagram showing the configuration of the disturbance sensitivity control unit in the first embodiment. [Figure 17] FIG. 17 is a graph illustrating the gain characteristic of the complementary sensitivity function of the first model following control unit in the first embodiment, and the gain characteristic of the reciprocal of the modeling error between the transfer function of the first controlled object and the transfer function of the first nominal model. [Figure 18] FIG. 18 is a graph showing an example of the relationship between the steering angle and the self-aligning torque. [Figure 19] FIG. 19 is a block diagram showing the configuration of a lane keeping system in the second embodiment. [Figure 20] FIG. 20 is a block diagram showing the configuration of the yaw rate control unit in the third embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0011] First Embodiment An electric power steering device 1000 of this embodiment shown in FIG. 1 is mounted on a vehicle V. As shown in FIG. 1, the electric power steering device 1000 includes a steering mechanism 530 and a control device 100. The steering mechanism 530 includes a steering mechanism section 520 and an auxiliary mechanism section 540. The electric power steering device 1000 controls the auxiliary mechanism section 540 by the control device 100, thereby controlling a steering torque T generated in the steering mechanism section 520 when a driver of the vehicle V steers a steering wheel 521. h The assist torque is generated to assist the driver in operating the steering wheel 521. The assist torque reduces the burden on the driver when the driver operates the steering wheel 521. The driver of the vehicle V is a person who steers the steering wheel 521 of the vehicle V.

[0012] The steering mechanism 520 has a handle 521, a steering shaft 522, universal joints 523A, 523B, an input shaft 524a, an output shaft 524b, a rack and pinion mechanism 525, a rack shaft 526, left and right ball joints 552A, 552B, tie rods 527A, 527B, knuckles 528A, 528B, and left and right tires 529A, 529B. That is, the steering mechanism 530 has a handle 521, a steering shaft 522, universal joints 523A, 523B, an input shaft 524a, an output shaft 524b, a rack and pinion mechanism 525, a rack shaft 526, left and right ball joints 552A, 552B, tie rods 527A, 527B, knuckles 528A, 528B, and left and right tires 529A, 529B.

[0013] The steering shaft 522 is a shaft extending from the handle 521 steered by the driver. One end of the input shaft 524a is connected to the end of the steering shaft 522 opposite to the end connected to the handle 521 via universal joints 523A and 523B. Thus, the handle 521 is connected to the input shaft 524a via the universal joints 523A and 523B and the steering shaft 522. The output shaft 524b is connected to the input shaft 524a via a torsion bar 546 described later. More specifically, one end of the output shaft 524b is connected to the other end of the input shaft 524a via the torsion bar 546. The other end of the output shaft 524b is connected to the rack shaft 526 via a rack and pinion mechanism 525.

[0014] The input shaft 524a and the output shaft 524b are arranged coaxially. The input shaft 524a and the output shaft 524b are rotatable around the same central axis. The input shaft 524a and the output shaft 524b are rotatable relative to each other within a range in which a torsion bar 546 (described later) can be twisted.

[0015] The auxiliary mechanism 540 has a steering torque sensor 541, a steering angle sensor 542, a motor 543, a reduction gear mechanism 544, an inverter 545, and a torsion bar 546. That is, the steering mechanism 530 has the steering torque sensor 541, the steering angle sensor 542, the motor 543, the reduction gear mechanism 544, the inverter 545, and the torsion bar 546. The torsion bar 546 connects the input shaft 524a and the output shaft 524b. The torsion bar 546 is disposed coaxially with the input shaft 524a and the output shaft 524b. In the following description, a virtual axis passing through the common central axis of the input shaft 524a, the output shaft 524b, and the torsion bar 546 is referred to as a rotation axis R. The torsion bar 546 can twist around the rotation axis R.

[0016] The steering torque sensor 541 detects the amount of twist of the torsion bar 546 around the rotation axis R, thereby detecting the steering torque T hDetects the steering torque T h is a torsion bar torque generated in the torsion bar 546, and is a torsional moment about the rotation axis R. The steering angle sensor 542 detects a rotation angle θ of the input shaft 524a about the rotation axis R. a The rotation angle θ of the input shaft 524a can be detected. a is the steering angle θ of the steering wheel 521. h That is, the steering angle sensor 542 detects the rotation angle θ of the input shaft 524a. a By detecting the steering angle θ of the steering wheel 521, h Based on the steering torque sensor 541 and the steering angle sensor 542, the rotation angle θ of the output shaft 524b can be detected. b The rotation angle θ of the output shaft 524b can be detected. b is the steering angle θ s It is.

[0017] The inverter 545 converts DC power into three-phase AC power, which is a pseudo-sine wave of U-phase, V-phase, and W-phase, in accordance with a motor drive signal input from the control device 100, and supplies the power to the motor 543. The motor 543 is connected to the output shaft 524b via a reduction gear mechanism 544. The three-phase AC power is supplied to the motor 543 from the inverter 545. The motor 543 is, for example, an interior permanent magnet synchronous motor (IPMSM), a surface mounted permanent magnet synchronous motor (SPMSM), a switched reluctance motor (SRM), or the like. The motor 543 generates a steering torque T by receiving the three-phase AC power from the inverter 545. h The motor 543 transmits the generated auxiliary torque to the output shaft 524b via the reduction gear mechanism 544.

[0018] The control device 100 is a control device that controls a steering mechanism 530 mounted on a vehicle V. In this embodiment, the control device 100 can control a first control object 560. The first control object 560 is a control object including a motor 543 provided in the steering mechanism 530. In this embodiment, the first control object 560 has a steering mechanism part 520, a torsion bar 546, a motor 543, and a reduction mechanism 544. Since the first control object 560 includes an input shaft 524a and an output shaft 524b that can rotate relatively to each other via the torsion bar 546, the motion of the first control object 560 cannot be described only by a simple equation of motion of a one-inertia system. The first control object 560 changes between a one-inertia system and a two-inertia system depending on the strength with which the steerer grips the handle 521. The stronger the steerer grips the handle 521, the closer the first control object 560 becomes to a one-inertia system. The more loosely the helmsman grips the steering wheel 521, the closer the first controlled object 560 becomes to a two-inertia system. In this manner, the first controlled object 560 is configured as a two-inertia system.

[0019] The control device 100 is electrically connected to the inverter 545. The control device 100 generates a motor drive signal based on detection signals detected by the steering torque sensor 541, the steering angle sensor 542, and the vehicle speed sensor 300 mounted on the vehicle V, and outputs the motor drive signal to the inverter 545. The control device 100 controls the rotation of the motor 543 via the inverter 545 to control the first controlled object 560. More specifically, the control device 100 controls the switching operations of a plurality of switching elements included in the inverter 545. Specifically, the control device 100 generates a control signal for controlling the switching operations of each switching element and outputs the control signal to the inverter 545. Each switching element is, for example, a MOSFET (Metal-Oxide-Semiconductor Field-Effect Transistor). In the following description, the control signal for controlling the switching operation of each switching element is called a "gate control signal."

[0020] The control device 100 detects the steering torque T hThe control device 100 generates a torque command value based on the above, and controls the torque and rotation speed of the motor 543 by, for example, vector control. The vector control is a method of decomposing the current flowing through the motor 543 into a current component that contributes to torque generation and a current component that contributes to magnetic flux generation, and independently controlling each of the mutually orthogonal current components. The control device 100 is not limited to vector control, and may perform other closed-loop control. The rotation speed of the motor 543 is expressed, for example, by the number of rotations the rotor rotates in one minute [rpm (revolutions per minute)], the number of rotations the rotor rotates in one second [rps (revolutions per second)], or the like.

[0021] The control device 100 receives the steering torque T directly from the steering torque sensor 541. h Alternatively, the control device 100 may calculate the steering torque T h The control device 100 may directly receive the steering angle θ of the steering wheel 521 from the steering angle sensor 542. h Alternatively, the control device 100 may input the steering angle θ h The value of may be calculated.

[0022] Moreover, the control device 100 and the motor 543 are modularized and manufactured and sold as a motor module. The motor module includes the motor 543 and the control device 100, and is suitable for use in the electric power steering device 1000. Moreover, the control device 100 can be manufactured and sold as a control device for controlling the electric power steering device 1000, independent of the motor 543.

[0023] 2 shows a typical example of the configuration of the control device 100 in this embodiment. The control device 100 includes, for example, a power supply circuit 111, an angle sensor 112, an input circuit 113, a communication I / F 114, a drive circuit 115, a ROM 116, and a processor 200. The control device 100 can be realized as a printed circuit board (PCB) on which these electronic components are mounted.

[0024] A vehicle speed sensor 300, a steering torque sensor 541, and a steering angle sensor 542 mounted on the vehicle V are communicatively connected to the processor 200. The vehicle speed is transmitted from the vehicle speed sensor 300 to the processor 200. The steering torque T h The processor 200 receives the steering angle θ h is sent.

[0025] The processor 200 is a semiconductor integrated circuit, and is also called a central processing unit (CPU) or a microprocessor. The processor 200 sequentially executes a computer program, which is stored in the ROM 116 and describes a group of instructions for controlling the driving of the motor, to realize a desired process. In addition to the processor 200, or instead of the processor 200, the control device 100 may have a field programmable gate array (FPGA) equipped with a CPU, a graphics processing unit (GPU), an application specific integrated circuit (ASIC), an application specific standard product (ASSP), or a combination of two or more circuits selected from these circuits. The processor 200 sets a current command value according to an actual current value and a rotation angle of the rotor of the motor 543, and generates a pulse width modulation (PWM) signal, and outputs the PWM signal to the drive circuit 115.

[0026] The power supply circuit 111 is connected to an external power supply (not shown). The power supply circuit 111 generates a DC voltage required for each component of the control device 100. The DC voltage generated in the power supply circuit 111 is, for example, 3V or 5V.

[0027] The angle sensor 112 detects the rotation angle of the rotor of the motor 543 and outputs it to the processor 200. The angle sensor 112 may be a resolver, a Hall element such as a Hall IC, or an MR sensor having a magnetic resistance element. The processor 200 can calculate the angular velocity ω [rad / s] of the motor 543 based on the electrical angle θm of the motor 543 obtained based on the angle sensor 112. Note that the control device 100 may include, instead of the angle sensor 112, a speed sensor capable of detecting the rotation angular velocity of the motor 543 and an acceleration sensor capable of detecting the rotation angular acceleration of the motor 543.

[0028] A motor current value detected by a current sensor (not shown) is input to the input circuit 113. In the following description, the motor current value detected by the current sensor (not shown) is referred to as an "actual current value." The input circuit 113 converts the level of the input actual current value to an input level of the processor 200 as necessary, and outputs the actual current value to the processor 200. A typical example of the input circuit 113 is an analog-to-digital conversion circuit.

[0029] The communication I / F 114 is, for example, an input / output interface for transmitting and receiving data in accordance with an in-vehicle control area network (CAN: Controller Area Network).

[0030] The driving circuit 115 is typically a gate driver or a pre-driver. The driving circuit 115 generates a gate control signal according to a PWM signal and provides the gate control signal to the gates of a plurality of switching elements of the inverter 545. For example, when the motor 543 to be driven is a motor that can be driven at a low voltage, the driving circuit 115 as a gate driver may not necessarily be required. In that case, the function of the gate driver in the driving circuit 115 may be implemented in the processor 200.

[0031] The ROM 116 is electrically connected to the processor 200. The ROM 116 is, for example, a writable memory, a rewritable memory, or a read-only memory. An example of the writable memory is a PROM (Programmable Read Only Memory). An example of the rewritable memory is a flash memory and an EEPROM (Electrically Erasable Programmable Read Only Memory). The ROM 116 stores a control program including a group of instructions for causing the processor 200 to control the motor drive. For example, the control program stored in the ROM 116 is temporarily deployed in a RAM (not shown) at boot time.

[0032] 3, a lane keeping system 1100 is configured by the control device 100 and the imaging device 400. The lane keeping system 1100 is a system for keeping a vehicle V driven by a human in the center of a lane L. When the vehicle V is about to deviate from the center of the lane L, the lane keeping system 1100 controls the steering mechanism 530 by the control device 100 based on an image of the lane L captured by the imaging device 400, to return the vehicle V to the center position of the lane L. The imaging device 400 is attached to, for example, the windshield FG of the vehicle V.

[0033] As shown in Fig. 4, the imaging device 400 has an imaging device main body 410 and an imaging device control unit 420. The imaging device main body 410 captures an image of a portion of a lane L located in front of a vehicle V. The imaging device main body 410 is, for example, a camera having a CCD (Charge Coupled Device) image sensor. The imaging device control unit 420 controls the imaging device main body 410. The imaging device control unit 420 is a semiconductor integrated circuit, similar to the processor 200. The imaging device control unit 420 outputs target coordinates x, y based on an image captured by the imaging device main body 410.

[0034] FIG. 4 shows an example of the functional blocks of the processor 200 of this embodiment. The processor 200, which is a computer, sequentially executes processes or tasks required for controlling the steering mechanism 530 using each functional block. Each functional block of the processor 200 shown in FIG. 4 may be implemented in the processor 200 as software such as firmware, may be implemented in the processor 200 as hardware, or may be implemented in the processor 200 as software and hardware. The processing of each functional block in the processor 200 is typically described in a computer program in software module units and stored in the ROM 116. However, when using an FPGA or the like, all or a part of these functional blocks may be implemented as a hardware accelerator. In addition, the control method of the control device 100 in this embodiment may be implemented by having a computer execute a desired operation by a program.

[0035] The processor 200 has a controller 201, a subtractor SU1, and a torque conversion unit 223. The controller 201 has a first assist control unit 210, a second assist control unit 220, a vehicle stabilization control unit 240, a disturbance sensitivity control unit 290, a first gain adjustment unit 610, a second gain adjustment unit 620, and an adder AD5. In other words, the control device 100 includes the first assist control unit 210, the second assist control unit 220, the vehicle stabilization control unit 240, the disturbance sensitivity control unit 290, the first gain adjustment unit 610, the second gain adjustment unit 620, and the adder AD5. In other words, the processor 200 of the control device 100 is implemented with functions corresponding to each of the first assist control unit 210, the second assist control unit 220, the vehicle stabilization control unit 240, the disturbance sensitivity control unit 290, the first gain adjustment unit 610, the second gain adjustment unit 620, and the adder AD1.

[0036] The first assist control unit 210 is capable of executing lane keeping control to keep the vehicle V within the lane L. As shown in FIG. 5, the first assist control unit 210 has a vehicle state calculation unit 211 and a lane keeping control unit 212. The vehicle state calculation unit 211 calculates a yaw rate target value γ , which is a target value of the yaw rate γ of the vehicle V, based on an input from the imaging device 400. r , and the lateral displacement y of the vehicle V of Calculate the yaw rate target value γ r is the yaw rate γ required for the vehicle V to stay within the lane L.

[0037] The yaw rate γ of the vehicle V is a parameter indicating a change in the yaw angle φ, which is a sway angle of the vehicle V in the left-right direction. In other words, the yaw rate γ is an angular velocity when the vehicle V sways in the left-right direction. In the example shown in FIG. 3, in a lane L having a curve that turns to the left, the vehicle V before turning the curve is shown by a solid line, and the vehicle V after turning the curve is shown by a two-dot chain line. The angle formed by a virtual line CL1 extending in the traveling direction of the vehicle V shown by a solid line and a virtual line CL2 extending in the traveling direction of the vehicle V shown by a two-dot chain line is the yaw angle φ changed when the vehicle V turns the curve. Note that the virtual lines CL1 and CL2 coincide with the optical axis of the imaging device main body 410, for example, when viewed from above in the vertical direction. The optical axis of the imaging device main body 410 passes through the center in the left-right direction of the area IR imaged by the imaging device main body 410.

[0038] FIG. 6 shows an xy coordinate system. The x-axis is an axis that extends in one horizontal direction. The y-axis is an axis that extends in one horizontal direction and perpendicular to the x-axis. FIG. 6 shows an example of a vehicle V traveling in the x-axis direction turning in the y-axis direction. The lateral deviation y of is the target route R t (x) is the deviation of the vehicle V in the y-axis direction from the target route R t (x) is the center of gravity V of vehicle V g The target route is located on the center line LCL of the lane L. The x-y coordinate position of the vehicle V shown in FIG. 6 is the center of gravity V g The x and y coordinates are the positions of the horizontal displacement y of is the target route R t (x) and center of gravity V g In addition, at the position of the vehicle V shown in FIG. g The x and y coordinates of are (0,0).

[0039] The vehicle state calculation unit 211 calculates the yaw rate target value γ based on the target coordinates x, y input from the imaging device 400. r and lateral displacement y of In this embodiment, the vehicle state calculation unit 211 calculates the coordinates (x k-1,y k-1 ) and coordinates (x k ,y k ) and coordinates (x k+1 ,y k+1 The three target coordinates (x, y) are input. The three target coordinates (x, y) are the target path R on the target center line LCL. t The coordinates are included in (x). k-1 ,y k-1 ) is the target path R at the current position of the vehicle V along the x-axis. t (x) is the coordinate on the k+1 ,y k+1 ) is the target path R at a position in the x-axis direction ahead of the current position of the vehicle V in the x-axis direction. t (x) is the coordinate on the k ,y k ) is the coordinate (x k-1 ,y k-1 )'s x-axis position and coordinate (x k+1 ,y k+1 ) at the x-axis position of the center between the x-axis position of t (x) is the coordinate on the target route R t (x) is the target y-axis position relative to the x-axis position. Target path R t (x) is expressed, for example, as in the following formula (1).

[0040]

number

[0041] The vehicle state calculation unit 211 calculates three coordinates (x k-1 ,y k-1 ),(x k ,yk ),(x k+1 ,y k+1 ) into the following equation (2) to calculate the values ​​of each term in equation (1). of is calculated.

[0042]

number

[0043] The vehicle state calculation unit 211 calculates the change rate C rp , curvature C p , and the coordinate x k-1 ,x k+1 Substituting this into the following equation (3), we obtain the coordinates (x k-1 ,y k-1 ) in the target route R t Radius of curvature R of (x) c (x k-1 ) is calculated.

[0044]

number

[0045] The vehicle state calculation unit 211 calculates the radius of curvature R obtained based on the above formula (3). c (x k-1 ) into the following equation (4) to obtain the yaw rate target value γ r Calculate.

[0046]

number

[0047] In this manner, the vehicle state calculation unit 211 calculates the yaw rate target value γ r and lateral displacement y of As shown in FIG. 5, the yaw rate target value γ r and lateral displacement y ofis input to the lane keeping control unit 212.

[0048] The lane keeping control unit 212 performs lane keeping control. The lane keeping control unit 212 moves the vehicle V along a target route R t The first command torque T required to run on (x) LKA In this embodiment, the first command torque T LKA is a "first input value" generated by the first assist control unit 210. As shown in Fig. 7, the lane keeping control unit 212 has a yaw rate control unit 213, a lateral deviation control unit 214, and an adder AD2. That is, the first assist control unit 210 has the yaw rate control unit 213, the lateral deviation control unit 214, and an adder AD2.

[0049] The lateral deviation control unit 214 determines the lateral deviation y of Based on the lateral displacement command torque T y In this embodiment, a lateral deviation command torque T y is a "lateral deviation command value" generated in the lateral deviation control unit 214. As shown in FIG. 8, the lateral deviation control unit 214 has a command torque generation unit 214a and a lateral deviation adjustment unit 214b. The command torque generation unit 214a calculates the lateral deviation y of Based on the lateral displacement command torque T ya In this embodiment, the command torque generator 214a generates a lateral deviation y of The lateral displacement y of The lateral displacement command torque T that converges to zero ya The command torque generating unit 214a performs PI control. The command torque generating unit 214a has a P gain unit 214c, an I gain unit 214d, an anti-windup control unit 214e, an integral value reset unit 214f, and an adder AD7.

[0050] The P gain section 214c and the I gain section 214d are of is input. Lateral displacement y of The gain K of the P gain unit 214c p The value multiplied by is input to adder AD7.of The gain K of the I gain unit 214d I The value multiplied by is input to the anti-windup control unit 214e. The anti-windup control unit 214e performs anti-windup control. In I control, if the integral value continues to stick to the set upper and lower limit values, it will exceed the target value and an overshoot will occur. In order to suppress this overshoot, the anti-windup control unit 214e performs anti-windup control. The output from the anti-windup control unit 214e is input to the integral value reset unit 214f.

[0051] The integral value reset unit 214f ​​receives the output from the anti-windup control unit 214e and a signal S indicating whether lane keeping control is being performed. LKA The signal S LKA When the signal S indicates that lane keeping control is being performed, the integral value reset unit 214f ​​functions as an integrator having a transfer function of 1 / s, where s is a Laplace transformer. LKA When the signal indicates that lane keeping control is not being executed, the integral value reset unit 214f ​​sets the output to zero. The output from the integral value reset unit 214f ​​is input to an adder AD7. The adder AD7 adds the output from the P gain unit 214c and the output from the integral value reset unit 214f ​​to obtain a lateral deviation command torque T ya and generates a lateral displacement command torque T ya to the lateral deviation adjustment unit 214b.

[0052] The lateral deviation adjustment unit 214b adjusts the lateral deviation command torque T ya is adjusted according to the frequency, and the lateral displacement command torque T y In order to switch the frequency band, the lateral deviation adjustment unit 214b outputs the lateral deviation command torque T ya The low frequency band is extracted from the lateral displacement command torque T y As shown in FIG. 9, the transfer function C yThe gain of (s) is high in a band with a low frequency f and is low in a band with a high frequency f. More specifically, the transfer function C y The gain of (s) is the transfer function C y (s) is lower than the gain. The second frequency band fb2 is a frequency band higher than the first frequency band fb1. The first frequency band fb1 is a frequency band equal to or lower than the frequency fc. The second frequency band fb2 is a frequency band higher than the frequency fc. The frequency fc is, for example, equal to or higher than 0.1 Hz and equal to or lower than 0.2 Hz. Note that the first frequency band fb1 and the second frequency band fb2 may be frequency bands provided separately.

[0053] The lateral deviation adjustment unit 214b is, for example, a low-pass filter with a cutoff frequency of the above-mentioned frequency fc. The lateral deviation adjustment unit 214b may be a phase lag compensation unit that performs phase lag compensation with a pole at the above-mentioned frequency fc. The lateral deviation adjustment unit 214b adjusts the lateral deviation command torque T ya The torque in the frequency band below fc is called the lateral displacement command torque T y In this way, the lateral deviation adjustment unit 214b outputs the lateral deviation command torque T ya The lateral deviation adjustment unit 214b adjusts the yaw rate target value γ before being converted into torque. r By adjusting according to the frequency f, the lateral displacement command torque T y may be adjusted according to the frequency f.

[0054] As shown in FIG. 7, the yaw rate control unit 213 controls the yaw rate target value γ r Based on the yaw rate command torque T γ In this embodiment, the yaw rate command torque T γ is a "yaw rate command value" generated in the yaw rate control unit 213. The yaw rate control unit 213 receives a yaw rate target value γ r , the travelling speed V of vehicle V v , and the coordinate (x k+1,y k+1 ) in the target route R t Curvature C of (x) p 10, the yaw rate control unit 213 has a yaw rate linearization phase compensation unit 213a, a yaw rate adjustment unit 213b, a yaw rate phase adjustment unit 213c, and a torque conversion unit 213d.

[0055] The yaw rate linearization phase compensation unit 213a linearizes the steering torque T h In this embodiment, the yaw rate linearization phase compensation unit 213a linearly approximates the relationship between the input yaw rate target value γ r Phase lag compensation is performed for a given frequency f d The gain of the frequency band above the specified frequency f d is a parameter f set by the model following control in the vehicle stabilization control unit 240. n The parameter f n is the natural frequency f of the yaw rate γ γ This is a parameter that corresponds to the given frequency f d is, for example, 0.5 Hz. The natural frequency f γ The parameter f corresponds to n is, for example, 2 Hz or more and 4 Hz or less. r The target yaw rate value γ r1 to the yaw rate adjustment unit 213b.

[0056] In this embodiment, since the disturbance sensitivity control unit 290 can compensate for disturbances, the vehicle stabilization control unit 240 can perform model following control to control the characteristics of the vehicle V and the steering torque T h The relationship between the yaw rate and γ is T h = Gγ, where G is a proportionality constant determined by tuning. The range in which the vehicle stabilization control unit 240 can constrain the characteristics of the vehicle V to the above characteristics is determined by the input steering torque Th The frequency f of the yaw rate γ is the natural frequency f γ Therefore, the yaw rate linearization phase compensation unit 213a compensates for the natural frequency f γ The yaw rate target value γ in a frequency band equal to or higher than a predetermined frequency f that is lower than r By reducing the gain of T h Within the range satisfying = Gγ, the yaw rate target value γ r This allows the steering torque T h and the yaw rate γ can be used by linear approximation. In this embodiment, the yaw rate linearization phase compensation unit 213a performs phase delay compensation, thereby realizing the delay time that occurs from when the helmsman recognizes the predicted arrival point until when the helmsman actually steers the steering wheel 521 in the case where control is not performed by the first assist control unit 210. This reduces the discomfort felt by the helmsman when steering the steering wheel 521 even when assist control is performed by the first assist control unit 210.

[0057] The yaw rate adjustment unit 213b adjusts the yaw rate target value γ r1 is adjusted according to the frequency. As shown in FIG. 9, the transfer function C γ The gain of (s) is low in a band with a low frequency f and is high in a band with a high frequency f. More specifically, the transfer function C γ The gain of (s) is the transfer function C of the yaw rate adjustment unit 213b in the second frequency band fb2 that is higher than the first frequency band fb1. γ The gain of the yaw rate adjustment unit 213b is lower than the gain of the yaw rate linearization phase compensation unit 213a. The yaw rate adjustment unit 213b is, for example, a high-pass filter with a cutoff frequency set to the above-mentioned frequency fc. The yaw rate adjustment unit 213b may be a phase lead compensation unit that performs phase lead compensation with a zero point at the above-mentioned frequency fc. The yaw rate adjustment unit 213b adjusts the yaw rate target value γ r1 The value of the frequency band above fc is set as the yaw rate target value γ r2In this embodiment, the yaw rate adjustment unit 213b outputs the yaw rate target value γ r1 By adjusting according to the frequency f, the yaw rate command torque T γ is adjusted according to the frequency f.

[0058] In this embodiment, the transfer function of the yaw rate adjustment unit 213b is C γ (s), and the transfer function of the lateral deviation adjustment unit 214b is C y (s) and the constant that is approximately 1 is K γy Then, C γ (s)=K γy -C y (s) holds. In this specification, the constant K γy is approximately 1" means that the constant K γy In addition to the case where is strictly 1, there is a constant K γy The case where the transfer function C of the yaw rate adjustment unit 213b is 0.8 or more and 1.2 or less is also included. γ (s) and the transfer function C of the lateral deviation adjustment unit 214b y (s) and (s) are roughly complementary to each other.

[0059] For example, when the lateral deviation adjustment unit 214b is a phase lag compensation unit, the transfer function C y (s) is expressed, for example, by the following equation (5).

[0060]

number

[0061] Parameter K r , 0≦K r By adjusting the transfer function C of the lateral deviation adjustment unit 214b within the range of ≦(f1+f2) / f2, the weighting of the lateral deviation control and the yaw rate control can be adjusted. y (s) is expressed by the above formula (5), and K r When is 1, the transfer function C of the yaw rate adjustment unit 213b γ (s) is expressed by the following equation (6).

[0062]

number

[0063] As shown in FIG. 10, the yaw rate phase adjustment unit 213c receives the yaw rate target value γ r2 and the travelling speed V of vehicle V v and the coordinate (x k+1 ,y k+1 ) in the target route R t Curvature C of (x) p The yaw rate phase adjustment unit 213c receives the yaw rate target value γ r2 The phase of the yaw rate phase adjustment unit 213c is adjusted by the transfer function C γp (s) is the transfer function P of the second nominal model NM2 described later. nv Natural frequency f of yaw rate γ at (s) γ The parameter f corresponds to nIn this embodiment, the transfer function C of the yaw rate phase adjustment unit 213c can be converted into different parameters. γp (s) is the transfer function P of the second nominal model NM2 nv Natural frequency f of yaw rate γ at (s) γ The parameter f corresponds to n The transfer function C of the yaw rate phase adjustment unit 213c can be converted into different parameters. γp (s) is expressed, for example, by the following equation (7).

[0064]

number

[0065] As shown in FIG. 10, the yaw rate phase adjustment unit 213c uses the above transfer function C γp (s) determines the yaw rate target value γ r2 Adjust the yaw rate target value γ r3 The torque conversion unit 213d outputs the yaw rate target value γ r3 The yaw rate command torque T γ As described above, in this embodiment, the steering torque T h The relationship between the yaw rate and γ is h =Gγ, the gain of the torque conversion unit 213d is the proportionality constant G.

[0066] The yaw rate adjustment unit 213b adjusts the yaw rate target value γ r1 By adjusting according to the frequency f, the yaw rate command torque T γ The yaw rate adjustment unit 213b is provided after the torque conversion unit 213d, and adjusts the yaw rate command torque Tγ may be adjusted directly according to the frequency f.

[0067] As shown in FIG. 7, the yaw rate command torque T γ and the lateral deviation command torque T y are added together in an adder AD2 to obtain the first command torque T LKA As shown in FIG. 4, the first command torque T LKA is input to the first gain adjustment section 610.

[0068] The first gain adjustment unit 610 adjusts the first command torque T LKA The first gain adjuster 610 adjusts the gain of the first command torque T LKA is multiplied by the first gain K1 to obtain the first command torque T LKA1 The first gain K1 is a variable value. The first gain K1 is expressed as K1=1-K2 using the second gain K2 of the second gain adjustment unit 620. The first command torque T LKA1 is input to the adder AD5 and the vehicle stabilization control unit 240.

[0069] The subtractor SU1 receives the steering angle θ h The subtractor SU1 receives the steering angle θ h Based on the steering torque sensor 541 and the steering angle sensor 542, the steering angle θ s The result is output to the torque conversion unit 223.

[0070] The value output from the subtractor SU1 is input to the torque conversion unit 223. The torque conversion unit 223 multiplies the value output from the subtractor SU1 by a gain to obtain the steering torque T h The gain in the torque conversion unit 223 is the spring constant K tor Steering angle θ h and steering angle θ s The difference between the spring constant K of the torsion bar 546 torThe value multiplied by is the torsion bar torque, i.e., the steering torque T h The steering torque T h is input to the second gain adjustment section 620.

[0071] The second gain adjustment unit 620 adjusts the steering torque T h The second gain adjustment unit 620 adjusts the gain of the steering torque T h is multiplied by the second gain K2 to obtain the steering torque T h1 The second gain K2 is equal to or greater than 0 and equal to or less than 1. The second gain K2 is a function of the steering torque T h According to the above-mentioned relationship of K1=1-K2, when the second gain K2 is 1, the first gain K1 is 0, when the second gain K2 is 0.5, the first gain K1 is 0.5, and when the second gain K2 is 0, the first gain K1 is 1. The larger the value of the second gain K2, the larger the steering torque T h The influence of the first command torque T output from the first assist control unit 210 becomes larger as the value of the first gain K1 becomes larger. LKA The first gain K1 when the first assist control unit 210 executes the lane keeping control is larger than the first gain K1 when the first assist control unit 210 does not execute the lane keeping control. h1 is input to an adder AD5. The adder AD5 calculates the steering torque T h1 First command torque T LKA1 The result is output to the second assist control unit 220.

[0072] The second assist control unit 220 controls the second command torque T r In this embodiment, the steering input value is the steering torque T h In this embodiment, the second command torque T r is a "second input value" generated by the second assist control section 220. The second assist control section 220 generates the second command torque Tr and controls the torque of the motor 543, thereby controlling the reaction force transmitted to the steering wheel 521. As shown in FIG. 11, the second assist control unit 220 has a base assist control unit 221 and a phase compensator 222.

[0073] The base assist control unit 221 calculates the torsion bar torque generated in the torsion bar 546, i.e., the steering torque T h The input torque T is calculated based on ad Based on this, the self-aligning torque T generated in the tires 529A, 529B of the vehicle SAT An assist torque T that compensates for at least a part of ass In this embodiment, the base assist control unit 221 generates the steering torque T h The steering torque T h1 and the first command torque T LKA The first command torque T is multiplied by the first gain K1. LKA1 The input torque T is calculated by adding ad Based on this, the assist torque T ass Generate.

[0074] In this embodiment, the base assist control unit 221 controls the input torque T ad Assist Gain K ass Multiplying this by the assist torque T ass The input torque T ad and assist torque T ass In the graph of FIG. 12, the horizontal axis represents the input torque T ad and the vertical axis is the assist torque T ass and the input torque T ad Assist torque T ass The slope of the assist gain K ass The input torque T ad and assist torque T ass and assist gain K ass What is dT? ass / dT ad =K assAs shown in Fig. 12, the assist torque T ass For example, the input torque T ad As shown in FIG. 11, the assist torque T output from the base assist control unit 221 ass is input to the phase compensator 222.

[0075] The base assist control unit 221 is configured to control the input torque T ad and the travelling speed V of vehicle V v Based on the assist torque T ass The base assist control section 221 may generate, for example, an input torque T ad and the running speed V v and assist torque T ass The assist torque T ass may be determined.

[0076] The phase compensator 222 in this embodiment adjusts the gain within a range of possible steering frequencies when the helmsman operates the steering wheel 521, and compensates for the stiffness of the torsion bar 546. The possible range of steering frequencies is, for example, 5 Hz or less. When the steering frequency is 5 Hz or less, the phase compensator 222 adjusts the assist torque T ass For example, first-order phase compensation may be applied to the first-order phase compensation, which is represented by a transfer function C(s) of, for example, equation (8).

[0077]

number

[0078] The phase compensator 222 is a phase compensator for the assist torque T ass and the second command torque T rThat is, the second input value, the second command torque T r is the assist torque T ass For example, the phase compensator 222 is a stabilizing compensator, and the assist torque T ass A stable phase compensation can be applied to the phase compensator 222. The phase compensator 222 can have a second- or higher-order transfer function whose frequency characteristic is variable according to the gain of the transfer function C(s). The second- or higher-order transfer function is expressed using a responsiveness parameter and a damping parameter. The second- or higher-order transfer function C(s) can be expressed by, for example, equation (9). By making the order of the transfer function C(s) second-order, it is possible to provide damping to the characteristics of the transfer function C(s). By changing the damping, it is possible to adjust the phase characteristic.

[0079]

number

[0080] In equation (9), s is the Laplace transformer, ω3 is the frequency of the zero of the transfer function C(s), ω4 is the frequency of the pole of the transfer function C(s), ζ3 is the damping ratio of the zero, and ζ4 is the damping ratio of the pole. The pole frequency ω4 is lower than the zero frequency ω3.

[0081] As shown in FIG. 4, the vehicle stabilization control unit 240 receives the first command torque T LKA1 and the second command torque T output from the second assist control unit 220. r 13, the vehicle stabilization control unit 240 has a second model following control unit 241, a second state feedback unit 242, a subtractor SU5, and an adder AD6.

[0082] The subtractor SU5 receives the first command torque T LKA1 and the second command torque T r The command torque T rL The subtractor SU5 subtracts the input command torque T rLThe second correction torque T output from the second model following control unit 241 is f2 The output from the subtractor SU5 is input to an adder AD6 and a second model following control unit 241. The adder AD6 calculates a command torque T rL1 is output to the disturbance sensitivity control unit 290. In this embodiment, the command torque T rL1 is the first command torque T multiplied by the first gain K1. LKA1 and the second command torque T multiplied by the second gain K2 r This is the "command value" calculated based on the above.

[0083] In this embodiment, the second model following control unit 241 is a model following controller configured to perform model following control. The second model following control unit 241 controls the command torque T rL The second correction torque T f2 is generated based on the yaw rate γ and the second nominal model NM2. In this embodiment, the second correction torque T f2 is the command torque T rL The second nominal model NM2 is an internal model used as a model for constraining the second controlled object 580 when controlling the second controlled object 580. In this embodiment, the second controlled object 580 is the actual steering angle θ t This is a plant model that takes the actual steering angle θ as input and the yaw rate γ as output. t is the angle at which tires 529A, 529B are inclined in the left-right direction of the vehicle body with respect to the front-rear direction of the vehicle body of the vehicle V when viewed in the vertical direction. The second controlled object 580 shown in FIG. 13 is the vehicle V. The transfer function G θ γ (s) is determined by the characteristics of the vehicle V.

[0084] In this embodiment, the second model following control unit 241 calculates a second correction torque T f2 and generates a command torque TrL The second model following control unit 241 has a second inverse nominal model 243, a filter 244, a torque conversion unit 245, and a subtractor SU6.

[0085] The second model following control unit 241 calculates that in a frequency band where the gain in the gain characteristic of the complementary sensitivity function for the modeling error between the second controlled object 580, which is a plant model, and the second nominal model NM2 is approximately 1, the transfer function of the second controlled object 580 is smaller than the transfer function P nv (s) is constructed to be bound by

[0086] In this specification, "the transfer function of a certain controlled object is constrained to the transfer function of a certain nominal model" means that the controlled object is controlled so that the transfer function of the certain controlled object appears to be the transfer function of a certain nominal model when, for example, the input / output relationship is viewed. In addition, in this specification, "a certain gain is approximately 1" includes, in addition to the case where the certain gain is 1, for example, the case where the certain gain is 0.8 or more and 1.2 or less. The numerical range is, for example, a range in which the gain of the substantial disturbance suppression characteristic can be adjusted to 1, taking into account the positive efficiency and the negative efficiency of the worm gear, when the reduction mechanism 544 connected to the motor 543 has a worm gear. Since the efficiency of the worm gear is about 0.8, it is necessary to adjust the gain to ±0.2 with respect to the target value 1.

[0087] The complementary sensitivity function for the modeling error between the second controlled object 580 and the second nominal model NM2 is the complementary sensitivity function of the inner loop configured by the second model following control unit 241. In this embodiment, the complementary sensitivity function in the second model following control unit 241 is equal to the transfer function Q3(s) of the filter 244. The transfer function Q3(s), which is the complementary sensitivity function, has a gain of approximately 0 dB, i.e., a gain of approximately 1 in the transfer function, in the frequency band that the filter 244 passes. In other words, in this embodiment, the second model following control unit 241 calculates the complementary sensitivity function G of the second controlled object 580 in the frequency band where the gain in the gain characteristic of the transfer function Q3(s) is approximately 1. θ γ (s) is the transfer function P of the second nominal model NM2 nv In this manner, the vehicle stabilization control unit 240 is configured to be constrained to the transfer function G θ γ (s) is converted into a transfer function P nv (s) is bound to

[0088] The second inverse nominal model 243 is an inverse model of the second nominal model NM2 used to constrain the second controlled object 580. The transfer function P nv (s) is expressed, for example, by the following equation (10). nv -1 (s) is expressed by the following equation (11).

[0089]

number

[0090]

number

[0091] Transfer function P of the second nominal model NM2 nv The gain in (s) is expressed, for example, as shown in FIG. 14. As shown in FIG. 14, the transfer function P nv The gain at (s) is the natural frequency f γ The parameter f corresponds to n The steady-state gain G θ γ (0), and the parameter f n As described above, the yaw rate linearization phase compensation unit 213a reduces the yaw rate target value γ r At a given frequency f d By reducing the gain in the above frequency bands, the transfer function P nv The gain at (s) is the steady-state gain G θ γ In the region where the value is (0), the second controlled object 580 can be controlled.

[0092] The output of the second controlled object 580, i.e., the yaw rate γ, is input to the second inverse nominal model 243. The second inverse nominal model 243 calculates the steering angle θ based on the above formula (11) and the input yaw rate γ. tp The steering angle θ tp is the actual steering angle θ input to the second nominal model NM2 when the output value of the second nominal model NM2 is the same as the output value of the second controlled object 580. t The steering angle θ tp is input to the torque conversion unit 245. The torque conversion unit 245 converts the steering angle θ tp Torque T p2 Convert it to and output it.

[0093] The subtractor SU6 subtracts the output of the subtractor SU5 from the output of the torque conversion unit 245 to obtain a differential torque T a2 The torque difference T output from the subtractor SU6 is generated. a2 is input to the filter 244. The filter 244 outputs the differential torque T a2 is filtered to obtain the second correction torque T f2 and output to the subtractor SU5.

[0094] The second model following control unit 241 constrains the second controlled object 580 to the second nominal model NM2, so that the natural frequency f γ The transfer function P of the second nominal model NM2 nv Parameter f in (s) n In this embodiment, the transfer function C γp Since (s) is expressed by the above equation (7), the transfer function P nv The denominator in (s) is the transfer function C γp (s) is cancelled by the numerator, and the transfer function C γp This can be substituted into the denominator of (s), which gives the parameter f n The value of the parameter f in equation (7) r Therefore, the transfer function C γp (s) gives the transfer function P of the second nominal model NM2. nv Natural frequency f of yaw rate γ at (s) γ The parameter f corresponds to n can be converted to a different value as desired. γ The parameter f corresponds to r For example, the curvature C p For example, the curvature C p When is small, the parameter f r Increase the curvature C p When is large, the parameter f r Reduce the parameter f r It is preferable that the frequency be about 2 Hz or more and 4 Hz or less.

[0095] The parameter f n If not constrained by, the natural frequency f of the yaw rate γ γ The natural frequency f of the yaw rate γ is, for example, about 1 Hz or more and 1.5 Hz or less. γ The parameter f r By constraining the yaw rate γ to γ Therefore, the resonance of the yaw rate γ can be suppressed, and the running of the vehicle V can be further stabilized.

[0096] The second state feedback unit 242 is configured to detect the yaw rate γ or the steering angle θ of the vehicle V. s Based on this, the state feedback torque T fb As shown in FIG. 15, the second state feedback unit 242 includes a yaw rate conversion unit 242a, a switch unit 242b, and a torque conversion unit 242c. The switch unit 242b outputs the state feedback torque T fb The yaw rate γ and steering angle θ of the vehicle V are s When there is a signal from the imaging device 400, the switch unit 242b outputs the yaw rate γ calculated based on the imaging device 400 to the torque conversion unit 242c. When there is no signal from the imaging device 400, the switch unit 242b switches between the steering angle θ s The yaw rate conversion unit 242a outputs the yaw rate γ calculated based on the input steering angle θ s The yaw rate conversion unit 242a converts the yaw rate γ into a yaw rate γ and outputs the converted yaw rate γ to the switch unit 242b. The transfer function of the yaw rate conversion unit 242a is G θ γ (s) / g tot It is expressed as G θ γ (s) is the actual steering angle θ t from the yaw rate γ and is the transfer function of the second controlled object 580.tot is the steering angle θ s From actual steering angle θ t The conversion gain is up to g tot is the steering angle θ s From actual steering angle θ t This parameter corresponds to the gear ratio up to

[0097] The torque conversion unit 242c converts the yaw rate γ input from the switch unit 242b into a state feedback torque T fb The transfer function C1(s) of the torque conversion section 242c is expressed by the following equation (12).

[0098]

number

[0099] As shown in FIG. 13, the state feedback torque T fb is input to the adder AD6 and added to the output from the subtracter SU5. The adder AD6 adds the state feedback torque T fb Adding this, the command torque T rL1 and outputs it to the disturbance sensitivity control section 290.

[0100] The transfer function of the second controlled object 580 described above, i.e., the transfer function G θ γ (s) is expressed by the following equation (13).

[0101]

number

[0102] The second state feedback unit 242 calculates a state feedback gain K d The state feedback torque T fb By feeding back, the damping ratio parameter ξ in the above formula (13) can be controlled. When the damping ratio parameter ξ is set to 1 by the second state feedback unit 242, the right-hand term in the above formula (13) can be approximated by a first-order lag transfer function as shown in the following formula (14).

[0103]

number

[0104] By approximating the transfer function of the second controlled object 580 as in the above equation (14), the second controlled object 580 can be approximated by the transfer function P nv In addition, since the second nominal model NM2 can be expressed by a first-order lag transfer function such as equation (10), even if the plant characteristics of the second controlled object 580 change due to a change in the number of passengers in the vehicle V, the natural frequency f of the yaw rate γ can be easily restricted by the second nominal model NM2. γ The desired parameter fr In this manner, in this embodiment, by combining the state feedback by the second state feedback unit 242 and the model following control by the second model following control unit 241, it is possible to realize characteristics of the vehicle V that are stable and easy to steer.

[0105] For example, the second controlled object 580 is simply constrained to the second nominal model NM2 by model following control to obtain the natural frequency f γ If the modeling error between the second controlled object 580 and the second nominal model NM2 is large, the robust stability is reduced, and the natural frequency f γ To address this issue, in the present embodiment, the state feedback gain K d is adjusted to obtain the transfer function G θ γ By first approximating (s), the transfer function G θ γ (s) is the transfer function P nv (s), and modeling errors can be reduced. As a result, the second model following control unit 241 performs model following control to obtain a stable and desirable natural frequency f γ The relationship between the modeling error and robust stability in the model following control will be described in detail in the model following control by the first model following control unit 230 described later.

[0106] As shown in FIG. 16, the disturbance sensitivity control unit 290 receives the command torque T rL1 is input to the disturbance sensitivity control section 290. The disturbance sensitivity control section 290 has a first model following control section 230, a first state feedback section 280, an adder AD3, and a subtractor SU2.

[0107] The first model following control unit 230 calculates the command torque T rL1 The first correction torque T f1 is generated based on the output of the first controlled object 560 including the motor 543 provided in the steering mechanism 530 and a first nominal model NM1 based on the configuration of the first controlled object 560. In this embodiment, the first correction torque T f1 is the command torque T rL1 When the command value is an angle, the correction value is an angle. f1 is the command torque T rL1 The first nominal model NM1 is a feedback torque that is fed back to the first control object 560. The first nominal model NM1 is an internal model used as a model that constrains the first control object 560 when controlling the first control object 560. The first nominal model NM1 will be described in detail later. The first model following control unit 230 is a model following controller configured to perform model following control. The specific configuration of the first model following control unit 230 will be described in detail later.

[0108] The subtractor SU2 subtracts the command torque T rL1 The first correction torque T output from the first model following control unit 230 is f1 The output from the subtractor SU2 is input to an adder AD3 and the first model following control unit 230. The adder AD3 calculates a command torque T rL2 The adder AD1 adds the steering torque T h and disturbance torque T d The command torque T rL3 The first controlled object 560 outputs the input to the command torque T rL3 The output is the steering angle θ s As shown in FIG. 13, the steering angle θ s is input to the conversion unit 570. The conversion unit 570 converts the steering angle θs Multiplying this by the conversion gain 1 / gtot gives the actual steering angle θ t The actual steering angle θ output from the conversion unit 570. t is input to the second controlled object 580. In addition, in FIG. 4, the first controlled object 560, the conversion unit 570, and the second controlled object 580 are collectively shown as a controlled object block 590.

[0109] Disturbance torque T d is the difference between the ideal output torque of the motor 543 and the actual output torque of the motor 543. d includes a disturbance torque applied from outside to the first controlled object 560. d is, for example, an excess torque caused by friction and rattle due to mechanical elements such as the motor 543 and the reduction gear mechanism 544, a torque ripple generated in the motor 543, and a self-aligning torque T SAT , and disturbance torque that may occur when traveling on an unpaved, bumpy road or gravel road.

[0110] As shown in FIG. 16, in this embodiment, the first model following control unit 230 controls the steering angle θ s Based on the first correction torque T f1 and generates a command torque T rL1 The first model following control unit 230 has a first inverse nominal model 231, a first filter 232a, a second filter 232b, an assist adjustment unit 270, a subtractor SU3, and an adder AD4. In this embodiment, the first filter 232a is a high-pass filter. The first filter 232a has a first cutoff frequency Cf1. The first cutoff frequency Cf1 is, for example, 2 Hz or more and 10 Hz or less. In this embodiment, the first cutoff frequency Cf1 is higher than 5 Hz and lower than 10 Hz.

[0111] In this embodiment, the second filter 232b is a low-pass filter. The second filter 232b has a second cutoff frequency Cf2 higher than the first cutoff frequency Cf1. The second cutoff frequency Cf2 is, for example, 3 Hz or more and 50 Hz or less. However, the upper limit of the second cutoff frequency Cf2 can be set in a range of about 140 Hz or more and 200 Hz or less. The order of the second filter 232b is third order or more. The second filter 232b may be composed of, for example, a plurality of low-pass filters. The first filter 232a and the second filter 232b are coupled in series.

[0112] The first model following control unit 230 calculates the transfer function P of the first controlled object 560 in a frequency band where the complementary sensitivity gain GT, which is the gain in the gain characteristic of the complementary sensitivity function T(s) for the modeling error between the first controlled object 560 and the first nominal model NM1, is approximately 1. s (s) is the transfer function P of the first nominal model NM1. sn (s) is constructed to be bound by

[0113] The complementary sensitivity function T(s) is a complementary sensitivity function of the inner loop configured by the first model following control unit 230. FIG. 17 shows a complementary sensitivity gain GT in the complementary sensitivity function T(s). The complementary sensitivity gain GT is a gain of the complementary sensitivity function T(s) as a transfer function, and is an absolute value of the complementary sensitivity function T(s). As shown in FIG. 17, the complementary sensitivity function T(s) has a gain of 0 dB, that is, a complementary sensitivity gain GT in the transfer function is 1 in a frequency band in which the frequency f is equal to or higher than the first cutoff frequency Cf1 and equal to or lower than the second cutoff frequency Cf2. In other words, the first cutoff frequency Cf1 is the lower limit of the frequency band in which the complementary sensitivity gain GT is 1. The second cutoff frequency Cf2 is the upper limit of the frequency band in which the complementary sensitivity gain GT is 1.

[0114] The transfer function P of the first controlled object 560 s (s) is the plant characteristic for which model following control is performed. s(s) is expressed, for example, by the following equation (15).

[0115]

number

[0116] The first inverse nominal model 231 is an inverse model of the first nominal model NM1 used to constrain the first controlled object 560. The transfer function Psn (s) is expressed, for example, by the following equation (16). The transfer function P sn -1 (s) is expressed, for example, by the following equation (17).

[0117]

number

[0118]

number

[0119] In equations (16) and (17), s is the Laplace transform, and J n is a parameter that represents the moment of inertia of the first nominal model NM1, and B n is a parameter that represents the viscous friction coefficient of the first nominal model NM1. Note that the transfer function P sn (s) and the transfer function P of the first inverse nominal model 231 sn -1 (s) is not limited to the examples shown in formulas (16) and (17), and is not particularly limited.

[0120] As shown in Fig. 16, the output of a first controlled object 560 is input to the first inverse nominal model 231. Specifically, the first inverse nominal model 231 is s The first inverse nominal model 231 is based on the above equation (17) and the input steering angle θ s Based on the torque T p That is, the first model following control unit 230 uses the first nominal model NM1 based on the output of the first controlled object 560 to output the torque T p Calculate the torque T p is equal to the value of the torque input to the first nominal model NM1 when the output value of the first nominal model NM1 is the same as the output value of the first controlled object 560.

[0121] The subtractor SU3 subtracts the output of the subtractor SU2 from the output of the first inverse nominal model 231 to obtain a differential torque T a That is, the subtractor SU3 generates the first correction torque T f1 After being fed back, the state compensation value V s The command torque T before feedback rL1 , torque T p Subtract the differential torque T from a Generates a differential torque T a For example, the disturbance torque T d The differential torque T output from the subtractor SU3 is a is input to the second filter 232b and subjected to low-pass filtering, and is then input to the first filter 232a and subjected to high-pass filtering. a The differential torque T filtered by the first filter 232a and the second filter 232b is input to an adder AD4. a In other words, the differential torque T filtered by the first filter 232a and the second filter 232b is in a state where frequency components lower than the first cutoff frequency Cf1 and frequency components higher than the second cutoff frequency Cf2 are removed. a is the frequency component T that is equal to or greater than the first cutoff frequency Cf1 and equal to or less than the second cutoff frequency Cf2. aM It is.

[0122] The assist adjustment unit 270 generates a compensation value for friction and disturbances to adjust the differential torque T a In this embodiment, the assist adjustment unit 270 adjusts the differential torque T a Among the frequency components T aM The assist adjustment section 270 is connected in parallel to the first filter 232a. The assist adjustment section 270 has a friction compensation value calculation section 250, a disturbance compensation value calculation section 260, and a subtractor SU4.

[0123] The subtractor SU4 subtracts the output value from the first filter 232a from the output value from the second filter 232b. Here, the output value from the second filter 232b is the differential torque T a The output value from the first filter 232a is a value obtained by removing frequency components higher than the second cutoff frequency Cf2 from the differential torque T a The value output from the subtractor SU4 is the torque difference T a The frequency component T aL The output of the subtractor SU4 is input to a friction compensation value calculation unit 250 and a disturbance compensation value calculation unit 260. aL The friction force and self-aligning torque T SAT , disturbance torque caused by backlash of first controlled object 560, and torque ripple occurring in first controlled object 560.

[0124] The friction compensation value calculation unit 250 calculates a friction compensation value V f The differential torque T a As described above, the value from the subtractor SU4 input to the friction compensation value calculation unit 250 is calculated based on the differential torque T a The frequency component T aL Therefore, in this embodiment, the friction compensation value calculation unit 250 calculates the differential torque T a Based on the frequency component lower than the first cutoff frequency Cf1, the friction compensation value V f Calculate.

[0125] The friction compensation value calculation unit 250 has a limiter 252 and a gain adjuster 253. The limiter 252 limits the output value from the subtractor SU4. When the input value exceeds an upper or lower threshold, the limiter 252 clips the input value to the upper or lower threshold. The gain adjuster 253 multiplies the output value from the limiter 252 by a gain K3. The friction compensation value calculation unit 250 calculates the differential torque T a The friction compensation value V f The threshold value of the limiter 252 and the value of the gain K3 are determined in advance based on the friction force actually generated in the first controlled object 560, for example.

[0126] The friction compensation value V output from the friction compensation value calculation unit 250 f is the differential torque T a Frequency component T aL Generally, since a moderate amount of friction is required for the first controlled object 560, the friction compensation value calculation unit 250 sets the friction compensation value V to a value smaller than the friction force actually generated in the first controlled object 560. f This makes it possible to realize highly accurate friction compensation while leaving an appropriate friction force on the first controlled object 560. The friction compensation value V f The targets of friction compensation by this include, for example, friction of the motor 543, friction of the reduction mechanism 544, and a difference between the left and right friction of the reduction mechanism 544.

[0127] Here, the differential torque T a Frequency component T aL In addition to the frictional force component, the self-aligning torque T SAT , the disturbance torque caused by the backlash occurring in the first controlled object 560, and the torque ripple occurring in the first controlled object 560 are also included. aL The friction compensation value V obtained by processing the above with the limiter 252 and the gain adjuster 253 is fThe self-aligning torque T generated in the first controlled object 560 is SAT , a disturbance torque caused by backlash occurring in the first controlled object 560, and a compensation value for compensating for at least a part of the torque ripple occurring in the first controlled object 560.

[0128] The vehicle V equipped with the electric power steering device 1000 can travel according to a travel mode having an automatic travel mode and a manual travel mode. In this case, the gain K3 of the gain adjuster 253 may be switched according to the travel mode. The greater the gain K3 of the gain adjuster 253, the greater the degree of reduction in friction.

[0129] The disturbance compensation value calculation unit 260 calculates the self-aligning torque T SAT A disturbance compensation value V that compensates for at least a part of d In this embodiment, the disturbance compensation value V d includes a compensation value for compensating for at least a part of the frictional force generated in the first controlled object 560, the disturbance torque caused by the backlash generated in the first controlled object 560, and the torque ripple generated in the first controlled object 560. The disturbance compensation value calculation unit 260 calculates the torque T p and command torque T rL1 The differential torque T a Based on the disturbance compensation value V d That is, the disturbance compensation value calculation unit 260 calculates the torque T p and command torque T rL1 The differential torque T a Based on the disturbance compensation value V d As described above, the value from the subtractor SU4 input to the disturbance compensation value calculation unit 260 is the differential torque T a 1. In this embodiment, the torque difference T a Based on the frequency components lower than the first cutoff frequency Cf1, the disturbance compensation value Vd Calculate.

[0130] The disturbance compensation value calculation unit 260 has a limiter 262 and a gain adjuster 263. The limiter 262 limits the output value from the subtractor SU4. When the input value exceeds the upper or lower threshold, the limiter 262 clips the input value to the upper or lower threshold. The threshold of the limiter 262 is different from the threshold of the limiter 252, for example. The gain adjuster 263 multiplies the output value from the limiter 262 by a gain K4. The transfer function P s (s) is the transfer function P of the first nominal model NM1. sn The maximum value of gain K4 of gain adjuster 263 is determined under the condition bound by (s). The value of gain K4 is different from the value of gain K3, for example. The value of gain K4 is, for example, not less than 0.1 and not more than 0.8 in the manual driving mode. The value of gain K4 is, for example, approximately 1 in the automatic driving mode. The value of gain K4 being approximately 1 includes the value of gain K4 being not less than 0.8 and not more than 1. Gain K4 of gain adjuster 263 may be switched depending on the driving mode of vehicle V.

[0131] Disturbance compensation value V d is the differential torque T a Frequency component T aL The external disturbance compensation value calculation unit 260 calculates, for example, a value that compensates for at least a part of the self-aligning torque component T SAT The disturbance compensation value V is about half of the value d The self-aligning torque T actually generated in the first controlled object 560 is calculated as follows: SAT is experimentally obtained in advance for each frequency. The threshold value of the limiter 262 of the external disturbance compensation value calculation unit 260 and the value of the gain K4 are set to a predetermined value in the manual driving mode. SAT The disturbance compensation value V is set to a value between 0.1 and 0.8 times the magnitude of dThe threshold value of the limiter 262 of the external disturbance compensation value calculation unit 260 and the value of the gain K4 are adjusted to a value that is calculated based on the self-aligning torque T SAT The disturbance compensation value V is set to approximately 1 times the magnitude of the d The disturbance compensation value V calculated by the disturbance compensation value calculation unit 260 is adjusted to a value calculated by the disturbance compensation value V d is the friction compensation value V calculated by the friction compensation value calculation unit 250. f is a different value.

[0132] Here, the differential torque T a Frequency component T aL The self-aligning torque T SAT In addition, the frequency component T aL The disturbance compensation value V obtained by processing the above by the limiter 262 and the gain adjuster 263 is d also includes a compensation value for compensating for at least a part of the frictional force occurring in the first controlled object 560, the disturbance torque caused by the backlash occurring in the first controlled object 560, and the torque ripple occurring in the first controlled object 560.

[0133] The first correction torque T used for the model following control in the first model following control unit 230 f1In order to apply friction compensation and disturbance compensation performed in the assist adjustment unit 270 to the above, it is necessary to pay attention to the stability conditions of the model following control. This condition is that the gain in the gain characteristic of the transfer function of the assist adjustment unit 270, which is constrained to a characteristic considering stability, does not exceed 1 according to the small gain theorem described later. This is derived from the design conditions of the second filter 232b. In this embodiment, the values ​​of the gains K3 and K4 in the gain adjusters 253 and 263 are set to a maximum of 1, and a subtractor SU4 is provided in front of the limiters 252 and 262 to apply subtraction processing so that the gain in the gain characteristic becomes 1 under this condition. In other words, the assist adjustment unit 270 behaves as a low-pass filter having a transfer function of 1-Q1(s). Q1(s) is the transfer function of the first filter 232a, which is a high-pass filter. The assist adjustment unit 270 performs low-pass filter processing having a transfer function of 1-Q1(s) on the torque output from the second filter 232b, and adjusts and outputs the processed value in each of the friction compensation value calculation unit 250 and the disturbance compensation value calculation unit 260.

[0134] For example, the values ​​of the gains K3 and K4 described above are determined so that the friction force components are 100% compensated for in both the manual driving mode and the automatic driving mode. Here, the friction compensation value calculation unit 250 compensates only for the friction force components, whereas the external disturbance compensation value calculation unit 260 compensates for all disturbances including the friction force components. In the automatic driving mode, for example, it is preferable to compensate for almost all disturbances, so the gain K4 of the external disturbance compensation value calculation unit 260 is approximately 1. In this case, the friction force components are also almost entirely compensated for by the compensation by the external disturbance compensation value calculation unit 260, so the gain K3 of the friction compensation value calculation unit 250 is zero. On the other hand, in the passive driving mode, the self-aligning torque T SATIt is preferable to compensate 100% of the frictional force components while leaving a portion of the tire reaction force including the above to make it easier for the helmsman to drive the vehicle V. Therefore, in the manual driving mode, it is preferable to set the gain K4 of the external disturbance compensation value calculation unit 260 to be smaller than 1, and to have the frictional force components not compensated for by the external disturbance compensation value calculation unit 260 compensated for by the friction compensation value calculation unit 250. For example, when the gain K4 of the external disturbance compensation value calculation unit 260 is 0.8 in the manual driving mode, the gain K3 of the friction compensation value calculation unit 250 is set to 0.2, and the 20% of the frictional force components not compensated for by the external disturbance compensation value calculation unit 260 are compensated for by the friction compensation value calculation unit 250.

[0135] The adder AD4 adds the output value from the assist adjustment unit 270 to the output value from the first filter 232a. aM Friction compensation value V f and disturbance compensation value V d The adder AD4 outputs the frequency component T aM and the friction compensation value V f and disturbance compensation value V d The first correction torque T f1 The first correction torque T output from the adder AD4 f1 is the input of the first controlled object 560, i.e., the command torque T rL1 In this manner, in the present embodiment, the first model following control unit 230 outputs the differential torque T a , i.e., frequency component T aM For the friction compensation value V f and disturbance compensation value V d The first correction torque T f1 Generate.

[0136] The first state feedback unit 280 calculates the apparent transfer function of the first controlled object 560 based on the output of the first controlled object 560 as a transfer function P sn To get closer to (s), the state compensation value V sThe command torque T rL1 The apparent transfer function of the first controlled object 560 is, for example, the transfer function of a portion located inside the feedback loop created by the first model following control unit 230 when the portion is regarded as one portion. Specifically, in this embodiment, the apparent transfer function of the first controlled object 560 is the transfer function of the entire portion between the subtractor SU2 and the output of the first controlled object 560, and is the transfer function of the portion combining the first state feedback unit 280 and the first controlled object 560. In this embodiment, the first state feedback unit 280 calculates the first correction torque T f1 After being corrected by and before being input to the first controlled object 560, the command torque T rL1 For the state compensation value V s Provide feedback.

[0137] State compensation value V s includes a compensation value that compensates for at least a part of the inertial force generated in the first controlled object 560, the viscous force generated in the first controlled object 560, and the frictional force generated in the first controlled object 560. More specifically, the state compensation value V s includes a compensation value that compensates for at least a part of the inertial force generated in the motor 543, the viscous force generated in the motor 543, and the frictional force generated in the motor 543. In this embodiment, the state compensation value V s is a compensation value including the inertial force generated in the motor 543, the viscous force generated in the motor 543, and the frictional force generated in the motor 543.

[0138] The first state feedback unit 280 includes an inertia compensator 281, a viscosity compensator 282, and a friction compensator 283. The inertia compensator 281 calculates a steering angle θ s The viscous compensator 282 calculates a compensation value for compensating for at least a part of the inertial force generated in the motor 543 based on the steering angle θ s The friction compensator 283 calculates a compensation value for compensating for at least a part of the viscous force generated in the motor 543 based on the steering angle θ sBased on this, a compensation value for compensating for at least a part of the frictional force generated in the motor 543 is calculated. In this embodiment, the state compensation value V s is composed of a compensation value calculated by an inertia compensator 281, a compensation value calculated by a viscosity compensator 282, and a compensation value calculated by a friction compensator 283. The compensation value calculated by the inertia compensator 281, the compensation value calculated by the viscosity compensator 282, and the compensation value calculated by the friction compensator 283 are output to an adder AD3 to calculate a first correction torque T f1 The command torque T after correction by rL1 is added to.

[0139] Next, the control by the first model following control unit 230 will be described in more detail. The first model following control unit 230 controls the first controlled object 560 using an inverse model of the first nominal model NM1 held as an internal model, that is, a first inverse nominal model 231. In this embodiment, the feedback loop created by the first model following control unit 230 makes it possible to compensate for torque ripples that depend on the angular velocity of the motor 543.

[0140] The first model following control unit 230 is similar in configuration to a conventional disturbance estimator (disturbance observer), but has different intended actions and effects. A conventional disturbance estimator estimates disturbance torque by making an inverse plant model, which is an internal model, a model close to the first control target 560, and reduces the influence of disturbance by adjusting the disturbance torque in advance.

[0141] In the present embodiment, the control by the first model following control unit 230 is carried out by a feedback loop to obtain a transfer function P s (s) is the transfer function P of the first nominal model NM1 that is included as an internal model. sn For example, if the first nominal model NM1 is defined so that there is no torque ripple, the transfer function P s(s) is constrained to a characteristic without torque ripple, and as a result, torque ripple can be reduced by applying torque ripple compensation. In addition, by making the first nominal model NM1 a low inertia model and constraining the first controlled object 560 with the first nominal model NM1, the first controlled object 560 can be treated as a low inertia model. In addition, by making the first nominal model NM1 a low viscosity model and constraining the first controlled object 560 with the first nominal model NM1, the first controlled object 560 can also be treated as a low viscosity model. By executing model following control by the first model following control unit 230, in addition to compensation for torque ripple of the motor 543, for example, loss torque compensation or motor inertia compensation is performed. In the above-mentioned equations (16) and (17), J n and B. n By appropriately setting s (s) can be given the desired frequency characteristics.

[0142] The transfer function P of the first controlled object 560 s (s) and the transfer function P of the first nominal model NM1 sn When the modeling error with respect to (s) is Δ(s), the transfer function P s (s) is expressed, for example, by the following equation (18).

[0143]

number

[0144] The transfer function P s (s), that is, the plant characteristic in the model following control of the first model following control unit 230 is not limited to the above example of equation (18) and may be expressed in any way. s (s) may be expressed as a two-inertia system or as an equation derived using a higher order approximation.

[0145] The transfer function P of the first controlled object 560s The gain characteristic of (s) has peaks at, for example, two frequency values. The modeling error Δ(s) appears, for example, near the higher frequency of the two peaks in the gain characteristic of the first controlled object 560. Therefore, as shown in FIG. 17, the inverse 1 / Δ(s) of the modeling error Δ(s) has a bottom in a relatively high frequency region. In FIG. 17, the modeling error Δ(s) is shown as an absolute value. When the modeling error Δ(s) becomes large, the transfer function P s (s) and the transfer function P of the first nominal model NM1 sn As a result, the deviation from the first nominal model NM1 becomes large, and the control of the first controlled object 560 by the first model following control unit 230 using the first nominal model NM1 becomes unstable. Therefore, in a region where the modeling error Δ(s) is relatively small, the first controlled object 560 can be stably and suitably controlled by setting the gain of the complementary sensitivity function T(s) to 1 and constraining the first controlled object 560 to the first nominal model NM1. The frequency characteristic of the modeling error Δ(s) is expressed by the transfer function P sn J in (s) n and B. n The frequency band in which the gain of the complementary sensitivity function T(s) is 1 can be adjusted by adjusting the first cutoff frequency Cf1 and the second cutoff frequency Cf2. This allows the gain of the complementary sensitivity function T(s) to be adjusted to 1 in a frequency band in which the modeling error Δ(s) is small.

[0146] 17, 1 / Δ(s) is relatively high in a frequency band equal to or lower than the second cutoff frequency Cf2, and drops sharply in a frequency band higher than the second cutoff frequency Cf2. The model following control that constrains the first controlled object 560 to the first nominal model NM1 can be stably performed, for example, in a range in which 1 / Δ(s) is greater than 1, that is, in a range in which 1 / Δ(s) is greater than 0 dB. Therefore, as shown in FIG. 17, by adjusting 1 / Δ(s) to be greater than 1 in a frequency band in which the gain of the complementary sensitivity function T(s) is 1, when the gain of the complementary sensitivity function T(s) is 1, the first controlled object 560 can be constrained to the first nominal model NM1 and controlled stably and suitably.

[0147] For example, in order to widen the frequency band in which the first controlled object 560 can be stably and suitably controlled by constraining it to the first nominal model NM1, the second cutoff frequency Cf2 may be increased within a range in which 1 / Δ(s) is not equal to or less than 1, that is, within a frequency band lower than the frequency at which the curve showing 1 / Δ(s) intersects with the horizontal axis in FIG. 17. However, if the second cutoff frequency Cf2 is increased too much, the gain of the complementary sensitivity function T(s) may remain relatively high in a frequency band higher than the second cutoff frequency Cf2, even though 1 / Δ(s) is lowered, which may cause unstable control. In contrast, in this embodiment, the order of the second filter 232b, which is a low-pass filter, is set to third or higher, so that the gain of the complementary sensitivity function T(s) can be sharply reduced in a frequency region higher than the second cutoff frequency Cf2. As a result, even if the second cutoff frequency Cf2 is set relatively high, the gain of the complementary sensitivity function T(s) can be immediately reduced in a frequency band higher than the second cutoff frequency Cf2, thereby preventing the control of the first controlled object 560 from becoming unstable.

[0148] The robust stability of the first model following control section 230 is guaranteed when the small gain theorem shown in the following equation (19) holds between the complementary sensitivity function T(s) and the modeling error Δ(s).

[0149]

number

[0150] As described above, in order to perform model following control using the first nominal model NM1 in the first model following control section 230, it is sufficient that the complementary sensitivity gain GT of the complementary sensitivity function T(s) is approximately 1, but in consideration of robust stability, it is necessary to satisfy the above formula (19). As can be seen from this, it is not possible to make the complementary sensitivity gain GT approximately 1 in all frequency bands and to satisfy formula (19), and therefore suppression of disturbances and the like by the first model following control section 230 and robust stability are not compatible.

[0151] As shown in Fig. 17, even in a low frequency region FA1 where the frequency f is lower than the first cutoff frequency Cf1, the complementary sensitivity gain GT of the complementary sensitivity function T(s) is smaller than 1. In the region where the complementary sensitivity gain GT of the complementary sensitivity function T(s) is smaller than 1, the command torque T rL1 As described above, in the high frequency region FA2 where the frequency is higher than the second cutoff frequency Cf2, the complementary sensitivity gain GT of the complementary sensitivity function T(s) is significantly lowered to control the first control target 560. f1 is hardly fed back to the input of the first controlled object 560. On the other hand, in the low frequency region FA1, the complementary sensitivity gain GT of the complementary sensitivity function T(s) is set to a certain magnitude, and the first correction torque T f1 is fed back to the input of first controlled object 560. In the low frequency region FA1, the compensation value generated in the above-described assist adjustment unit 270 is fed back to the input of first controlled object 560 in accordance with the complementary sensitivity gain GT of the complementary sensitivity function T(s).

[0152] The control device 100 performs torque control in the first assist control unit 210, the second assist control unit 220, and the vehicle stabilization control unit 240 for torque signals with low frequencies lower than the first cutoff frequency Cf1, and performs control to make the angular velocity substantially zero for disturbances with high frequencies higher than the second cutoff frequency Cf2, thereby realizing stabilization of steering so that the steering wheel 521 is not moved. To achieve this objective, the control device 100 uses the first assist control unit 210, the second assist control unit 220, and the vehicle stabilization control unit 240 to lower the high-frequency gain of the torque control, and uses the first model following control unit 230 to calculate the transfer function P s The reason for performing the latter process is to prevent the first controlled object 560 from reacting to a disturbance when the disturbance is input to the first controlled object 560.

[0153] Steering angle θ h The transfer function of the portion from after the subtractor SU1 to which the output of the disturbance sensitivity control unit 290 is input to before the adder AD1 to which the output of the disturbance sensitivity control unit 290 is input is defined as C2(s), and the disturbance torque T d (s), that is, the first correction torque T f1 If the ratio of these is the variable parameter m, the disturbance torque T d (s) to steering torque T h The transfer characteristics up to (s) are expressed by the following equation (20) using the complementary sensitivity function T(s), and the complementary sensitivity function T(s) is expressed by the following equation (21).

[0154]

number

[0155]

number

[0156] From the above formula (20), it can be seen that if the complementary sensitivity function T(s) is 1, the disturbance is completely suppressed, and if the complementary sensitivity function T(s) is smaller than 1, at least a part of the disturbance is transmitted. Therefore, by separating the disturbance to be transmitted and the disturbance to be suppressed in the frequency domain and frequency shaping the complementary sensitivity function T(s) accordingly, it is possible to design the disturbance sensitivity for each frequency domain. The variable parameter m in the above formulas (20) and (21) is the disturbance torque T d First correction torque T f1 The compensation rate is equal to T f1 =mT d For example, if the variable parameter m is 0.5, the first correction torque T f1 The disturbance torque T d is compensated by 50%. In this embodiment, when assist control is not executed by the first assist control section 210, the variable parameter m is equal to or greater than 0.1 and equal to or less than 0.8. When assist control is executed by the first assist control section 210, the variable parameter m is approximately 1. The variable parameter m being approximately 1 includes variable parameter m being equal to or greater than 0.8 and equal to or less than 1.

[0157] The transfer function P of the first controlled object 560 s If (s) is the above equation (18), then C2(s) = K tor +K ass Therefore, the steady-state gain T(0) of the complementary sensitivity function T(s) at s=0 is expressed by the following equation (22) from the above equation (21). tor The steering angle θ s The self-aligning torque T generated in the tires 529A, 529B of the vehicle V with respect to SAT The slope of K SAT and the input torque T ad Assist torque T ass The slope of K ass The steady-state gain is T(0), and the disturbance input to the first controlled object 560, i.e., the disturbance torque T d First correction torque T f1When the variable parameter, which is the ratio of these, is m, the following equation (22) holds.

[0158]

number

[0159] When the steering wheel 521 is in the center position or near the center position, and the vehicle V is traveling at a speed V v In the steady state, when the assist gain K ass is close to zero. Also, the spring constant K tor is the self-aligning torque gain K SAT Therefore, when the steering wheel 521 is at or near the center position and the vehicle V is traveling at a speed V v In the steady state, when the disturbance torque T d First correction torque T for (s) f1 , that is, a value close to the variable parameter m. The steady-state gain T(0) of the complementary sensitivity function T(s) expressed by the above formula (22) is 0.1 or more. In this embodiment, the steady-state gain T(0) is 0.3 or more. The steady-state gain T(0) is, for example, 0.5 or less. The steady-state gain T(0) is the minimum value of the complementary sensitivity gain GT in the low-frequency region FA1.

[0160] Conventionally, for example, the complementary sensitivity function T(s) in the low-frequency region FA1 had a waveform as shown by the two-dot chain line in Fig. 17. In this embodiment, by performing model following control by the above-mentioned first model following control section 230, the steady-state gain T(0) of the complementary sensitivity function T(s) can be made as shown in the above formula (22), and the complementary sensitivity function T(s) can be shaped into a waveform as shown by the solid line in Fig. 17 with respect to frequency f.

[0161] In this embodiment, the low-frequency portion TFA1, which is a portion of the low-frequency region FA1 in the waveform of the complementary sensitivity gain GT of the complementary sensitivity function T(s) shown in Fig. 17, is generated by the first filter 232a, the assist adjustment unit 270, and the adder AD4 in the first model following control unit 230. More specifically, the frequency component T aM and the friction compensation value V output from the assist adjustment unit 270 which functions as a low-pass filter. f and disturbance compensation value V d The waveform of the low-frequency portion TFA1 is created by adding together these. The value at which the frequency f becomes zero in the low-frequency portion TFA1 is the value at which s = 0 in the complementary sensitivity function T(s), and is the steady-state gain T(0) expressed by the above-mentioned equation (22).

[0162] As shown in FIG. 17, the first cutoff frequency Cf1 is yaw The first resonant frequency f yaw is the natural frequency f of the yaw rate γ of the vehicle V γ In other words, the lower limit of the frequency band where the complementary sensitivity gain GT is 1 is the natural frequency f γ The first resonant frequency f is equal to yaw The natural frequency of the yaw rate γ is higher than γ is calculated by the second model following control unit 241 as a parameter f r If not constrained to , the first resonant frequency f yaw is, for example, about 1 Hz or more and 1.5 Hz or less. In this embodiment, as described above, the natural frequency f γ The desired parameter f r Therefore, the first resonance frequency f yaw is approximately 2Hz or more and 4Hz or less.

[0163] The second resonant frequency f st is the resonance frequency of the steering mechanism 530. More specifically, the second resonance frequency f st is the inertia of the entire steering mechanism 530 and the self-aligning torque T SATThe second resonant frequency f is the resonant frequency of the mode formed by the spring element that generates the second resonant frequency f st is the first resonant frequency f yaw The second resonance frequency f is higher than the lower limit of the frequency band in which the complementary sensitivity gain GT is 1, that is, the first cutoff frequency Cf1. st The second resonance frequency f is, for example, about 3 Hz or more and 5 Hz or less. st The complementary sensitivity gain GT at the first resonant frequency f yaw The complementary sensitivity gain GT at the second resonant frequency f st The complementary sensitivity gain GT at the second resonant frequency f st The complementary sensitivity gain GT in is 0.8 or more.

[0164] In this embodiment, the high frequency portion TFA2, which is a portion of the high frequency region FA2 in the waveform of the complementary sensitivity gain GT of the complementary sensitivity function T(s) shown in FIG.

[0165] A low-frequency portion TFA1, which is a portion of the low-frequency region FA1 of the waveform of the complementary sensitivity gain GT of the complementary sensitivity function T(s), may be created by a phase-lead compensator. Specifically, the first filter 232a may be the phase-lead compensator. In this case, the complementary sensitivity function T(s) is a function including the transfer function of the second filter 232b as a low-pass filter and the transfer function of the phase-lead compensator. Even in this case, the complementary sensitivity gain GT of the complementary sensitivity function T(s) can be suitably adjusted to the above-mentioned numerical range, and the waveform of the complementary sensitivity gain GT as shown in FIG. 17 can be realized.

[0166] When a low-frequency portion TFA1, which is a portion of the low-frequency region FA1 of the waveform of the complementary sensitivity gain GT of the complementary sensitivity function T(s), is generated by a phase lead compensator, the transfer function F(s) of the phase lead compensator is expressed by the following equation (23).

[0167]

number

[0168] The variable parameter m is the steady-state gain F(0) in the above equation (23). That is, m=ω a / ω b Therefore, ω a =ω b When this happens, the self-aligning torque T SAT Including disturbance torque T d Specifically, ω a =ω b When the disturbance torque T d When the first assist control unit 210 performs lane keeping control, ω a =ω b or ω a ≒ω b Let m be 1 or a value close to 1, and the disturbance torque T d When the assist control by the first assist control section 210 is not performed and only the assist control by the second assist control section 220 is performed, the disturbance torque T d If the ratio of is n, then m=1-n holds. This gives the variable parameter m, and the parameter ω a ,ω b can be calculated.

[0169] In the case of sharing control in which the assist control by the first assist control unit 210 and the assist control by the second assist control unit 220 are performed simultaneously, the steering torque T h For parameter ω a In this case, the parameter ω a is expressed by the following equation (24).

[0170]

number

[0171] The variable parameter m is changed based on the second gain K2. The variable parameter m is expressed as m=m * ×K2+K1. When the assist control by the first assist control unit 210 is not performed, the second gain K2 is 1. When the second gain K2 is 1, the first gain K1 is zero, and the variable parameter m is a constant m * , that is, 0.1 or more and 0.8 or less. Also, from the relationship K1=1-K2, 1-m * m ** Then, the variable parameter m is m=1-m ** ×K2. From this formula, it can be seen that the larger the second gain K2, the smaller the variable parameter m, and vice versa. In other words, the variable parameter m and the second gain K2 have a complementary relationship. In this embodiment, the variable parameter m is equal to the gain K4 described above.

[0172] As can be seen from the above formula (22), the larger the variable parameter m, the larger the steady-state gain T(0) of the complementary sensitivity function T(s). In this embodiment, the steady-state gain T(0) of the complementary sensitivity function T(s) when the first assist control unit 210 executes assist control, i.e., lane keeping control, is larger than the steady-state gain T(0) of the complementary sensitivity function T(s) when the first assist control unit 210 does not execute lane keeping control.

[0173] In the control method for controlling the steering mechanism 530 mounted on the vehicle V, which is performed by the control device 100 described above, the lane keeping control is performed by adjusting the yaw rate target value γ r and the lateral displacement y of the vehicle V of and calculating the yaw rate target value γ r Based on the yaw rate command torque T γ and generating a lateral displacement y of Based on the lateral displacement command torque T y and a yaw rate command torque T γ is adjusted in accordance with the frequency f, and the lateral deviation command torque T y and adjusting the frequency f according to the frequency f.

[0174] In addition, the control method for controlling the steering mechanism 530 mounted on the vehicle V, which is performed by the above-mentioned control device 100, is to control the vehicle V to be within the lane L by controlling the first command torque T LKA and generating a steering torque T h Based on the second command torque T r and generating a first command torque T LKA and the second command torque T multiplied by the second gain K2 r Based on the command torque T rL1 and the command torque T rL1 The first correction torque T f1 based on the output of a first controlled object 560 including a motor 543 provided in a steering mechanism 530 and a first nominal model NM1 based on the configuration of the first controlled object 560; and by the model following control, a transfer function P s(s) is the transfer function P ns (s)

[0175] According to this embodiment, the control device 100 includes a first assist control unit 210 capable of performing lane keeping control to keep the vehicle V within the lane L. The first assist control unit 210 determines a yaw rate target value γ r and the lateral displacement y of the vehicle V of and a vehicle state calculation unit 211 that calculates a yaw rate target value γ r Based on the yaw rate command torque T γ and a yaw rate control unit 213 that generates a lateral deviation y of Based on the lateral displacement command torque T y The yaw rate control unit 213 has a lateral deviation control unit 214 that generates a yaw rate command torque T γ The yaw rate adjustment unit 213b adjusts the lateral deviation command torque T y The lateral deviation adjustment unit 214b adjusts the transfer function C γ The gain of (s) is the transfer function C of the yaw rate adjustment unit 213b in the second frequency band fb2 that is higher than the first frequency band fb1. γ The transfer function C of the lateral deviation adjustment unit 214b in the second frequency band fb2 is lower than the gain of y The gain of (s) is the transfer function C y Lower than the gain of (s).

[0176] Here, the yaw rate γ is adjusted by turning the steering wheel 521 to tilt the vehicle V in the direction of travel, whereas the lateral deviation y of is adjusted by how long the tilted state of the traveling direction is maintained. Therefore, when adjusting the yaw rate γ, the frequency at which the steering wheel 521 is steered is adjusted by the lateral deviation y ofWhen adjusting the frequency, the frequency is likely to be higher than the frequency at which the steering wheel 521 is steered. Therefore, in the first frequency band fb1, which is relatively low, the transfer function C γ The gain of (s) is lowered, and the transfer function C y By lowering the gain of (s), the yaw rate γ and lateral displacement y of Specifically, in the second frequency band fb2, which is relatively high, the yaw rate command torque T γ In the first frequency band fb1, which is relatively low, the influence of the lateral deviation y of The lateral displacement command torque T for adjusting y In this way, the frequency band that strengthens the control of the yaw rate γ and the lateral deviation y of By controlling the frequency band that strengthens the control of the yaw rate γ and the lateral deviation y of This allows the control device 100 to accurately control both the yaw rate γ and the lateral deviation y of By providing these two feedback means, each parameter can be adjusted with high precision.

[0177] According to this embodiment, the transfer function of the yaw rate adjustment unit 213b is set to C γ (s), and the transfer function of the lateral deviation adjustment unit 214b is C y (s) and the constant that is approximately 1 is K γy Then, C γ (s)=K γy -C y Therefore, the transfer function C γ (s) and the transfer function C of the lateral deviation adjustment unit 214b y (s) can be made to have a mutually complementary relationship. γ The frequency band where the effect of is large and the lateral displacement command torque T yTherefore, it is possible to more appropriately separate the frequency bands in which the influence of the yaw rate γ and the lateral deviation y of Therefore, the accuracy of keeping the vehicle V within the lane L in the lane keeping control can be improved.

[0178] According to the present embodiment, the control device 100 detects the actual steering angle θ of the tires 529A, 529B of the vehicle V. t The plant model with the input and the yaw rate γ as the output, that is, the transfer function G θ γ (s) is expressed as the transfer function P of the second nominal model NM2 by model following control. nv The vehicle stabilization control unit 240 has a second model following control unit 241 that performs model following control. The second model following control unit 241 controls the transfer function G θ γ (s) is the transfer function P of the second nominal model NM2 nv (s). The transfer function P of the second nominal model NM2 nv (s) is the natural frequency f of the yaw rate γ γ The parameter f corresponds to n The yaw rate control unit 213 includes a yaw rate target value γ r The transfer function of the yaw rate phase adjustment unit 213c is the transfer function P nv Natural frequency f of yaw rate γ at (s) γ The parameter f corresponds to n Therefore, the characteristic of the vehicle V can be constrained to the second nominal model NM2 by the model following control by the second model following control unit 241, and the natural frequency f γThe parameter f corresponds to n Let us assume that the desired value, i.e., the parameter f r This can further improve the running stability of the vehicle V.

[0179] According to the present embodiment, the control device 100 detects the steering torque T h Based on this, a second command torque T is input as a second input value to the vehicle stabilization control unit 240. r The first assist control section 210 is provided with a second assist control section 220 that generates a first command torque T LKA The second assist control section 220 generates a first command torque T LKA1 and the steering torque T multiplied by the second gain K2. h1 The first gain K1 and the second gain K2 are variable values. When the first gain is K1 and the second gain is K2, K1=1-K2 holds. Therefore, by adjusting the value of the second gain K2, the first command torque T LKA1 As a result, in a sharing control in which the assist control by the first assist control section 210 and the assist control by the second assist control section 220 are performed simultaneously, the ratio of the steering torque T h The amount of assist control by the first assist control section 210 can be adjusted according to the magnitude of the assist force.

[0180] According to the present embodiment, the first command torque T LKA1 is also input to the vehicle stabilization control unit 240. Therefore, an input similar to an input that is mechanically transmitted to the steering mechanism 530 when the helmsman steers the steering wheel 521 can also be input from the first assist control unit 210. This makes it possible to make the assist control by the first assist control unit 210 closer to the case where the helmsman steers the steering wheel 521. Therefore, when the assist control is performed by the first assist control unit 210, it is possible to reduce the sense of discomfort felt by the helmsman.

[0181] Conventionally, lane keeping control needs to function within a small torque range for safety reasons, which requires adjustment of the assist function for lane keeping control, and this adjustment takes time. The small torque range for safety is a steering torque T that can be easily corrected by the driver's will even if an abnormal output is output by lane keeping control. h In addition, in the past, when the torque for performing the lane keeping control, i.e., the control amount of the lane keeping control, was increased, a steering wheel felt uncomfortable, and it was difficult to increase the control amount of the lane keeping control, and there were cases where the accuracy of the lane keeping control could not be sufficiently improved.

[0182] In response to the above problem, according to the present embodiment, the control device 100 calculates a first command torque T LKA1 and a second command torque T as a second input value multiplied by a second gain K2. r The command torque T is calculated based on rL1 The disturbance sensitivity control unit 290 receives the command torque T rL1 The first correction torque T f1 The first model following control unit 230 generates a transfer function P of the first controlled object 560 in a frequency band where a complementary sensitivity gain GT, which is a gain in the gain characteristic of the complementary sensitivity function T(s) with respect to the modeling error Δ(s) between the first controlled object 560 and the first nominal model NM1, is approximately 1. s (s) is the transfer function P of the first nominal model NM1. ns The complementary sensitivity function T(s) is configured to be constrained to (s). The complementary sensitivity function T(s) includes a variable parameter m that varies based on the second gain K2.

[0183] Therefore, the self-aligning torque TSAT Disturbances such as the above can be compensated for. As a result, when lane keeping control is executed by the first assist control unit 210, lane keeping control can be executed with high accuracy even if the control amount by the first assist control unit 210 is reduced. Therefore, the accuracy of lane keeping control can be improved while reducing the control amount by the first assist control unit 210 to reduce the discomfort felt by the steering wheel 521 by the steerer. Furthermore, since disturbances can be compensated for by model following control, the above-mentioned adjustment of the assist function for lane keeping control is not required even if the control amount by the first assist control unit 210 is reduced. Furthermore, the complementary sensitivity function T(s) in the first model following control unit 230 includes a variable parameter m that changes based on the second gain K2. Therefore, the proportion of disturbances compensated for by the first model following control unit 230 is adjusted according to the change in the second gain K2. As a result, for example, when the second gain K2 becomes smaller and the lane keeping control by the first assist control unit 210 becomes stronger, the variable parameter m can be changed to increase the compensation proportion of the disturbance. Therefore, the compensation ratio for the disturbance can be adjusted according to the ratio between the control by the first assist control unit 210 and the control by the second assist control unit 220. This makes it possible to further improve the accuracy of the lane keeping control while further reducing the sense of discomfort felt by the steering wheel 521 by the driver.

[0184] Moreover, according to this embodiment, the first gain K1 when the first assist control unit 210 executes the lane keeping control is larger than the first gain K1 when the first assist control unit 210 does not execute the lane keeping control. The steady-state gain T(0) of the complementary sensitivity function T(s) when the first assist control unit 210 executes the lane keeping control is larger than the steady-state gain T(0) of the complementary sensitivity function T(s) when the first assist control unit 210 does not execute the lane keeping control. Therefore, when the first assist control unit 210 executes the lane keeping control, the proportion of disturbance compensated by the first model following control unit 230 can be made larger than when the lane keeping control is not executed. This makes it possible to further improve the accuracy of the lane keeping control while reducing the amount of control by the first assist control unit 210.

[0185] Moreover, according to this embodiment, the larger the variable parameter m, the larger the steady-state gain T(0) of the complementary sensitivity function T(s). The variable parameter m and the second gain K2 are in a mutually complementary relationship. Therefore, the larger the ratio of lane keeping control by the first assist control unit 210 to the control by the second assist control unit 220, the larger the compensation ratio of disturbance by the disturbance sensitivity control unit 290 can be continuously increased. This makes it possible to more suitably improve the accuracy of lane keeping control while suitably reducing the control amount by the first assist control unit 210 according to the ratio of control by the first assist control unit 210. Also, the assist control by the first assist control unit 210, i.e., the lane keeping control and the assist control by the second assist control unit 220, can be linearly and continuously adjusted. This makes it possible to suppress the switching characteristics of the control from becoming nonlinear in sharing control in which the assist control by the first assist control unit 210 and the assist control by the second assist control unit 220 are simultaneously performed, thereby reducing the discomfort of the steering person.

[0186] According to the present embodiment, the second assist control section 220 controls the torsion bar torque generated in the torsion bar 546, i.e., the steering torque T h The input torque T is calculated based on ad Based on this, the self-aligning torque T generated in the tires 529A, 529B of the vehicle V is SAT An assist torque T that compensates for at least a part of ass The second command torque T r is the assist torque T ass The variable parameter m is calculated based on the first correction torque T f1 The spring constant of the torsion bar 546 is K tor and the steering angle θ s The self-aligning torque T generated in the tires 529A, 529B of the vehicle V with respect to SAT The slope of K SAT and the input torque T ad Assist torque T assThe slope of K ass When the steady-state gain T(0) of the complementary sensitivity function T(s) is T(0) and the variable parameter is m, the above formula (22) holds. By being able to express the steady-state gain T(0) of the complementary sensitivity function T(s) by formula (22) including the variable parameter m, it becomes easier to adjust the disturbance compensation ratio in accordance with changes in the second gain K2.

[0187] In this embodiment, the yaw rate command value generated in the yaw rate control unit 213 and the lateral deviation command value generated in the lateral deviation control unit 214 are not limited to torque and may be an angle. When the yaw rate command value and the lateral deviation command value are angles, the first input value output from the first assist control unit 210 may be a torque or an angle. When the first input value output from the first assist control unit 210 is an angle, the input value input to the second assist control unit 220 is also an angle. In this case, the input value input to the vehicle stabilization control unit 240 is also an angle.

[0188] <Second embodiment> As shown in Fig. 19, the control device 100A of the lane keeping system 1100A of this embodiment is different from that of the first embodiment in the configuration of the controller 201A of the processor 200A. h The second assist control section 220A converts the input value into the steering torque T h The second command torque T r The second gain adjustment unit 620A adjusts the second command torque T r The second gain adjustment unit 620A is similar to the second gain adjustment unit 620 of the first embodiment, except that the second command torque T r is multiplied by the second gain K2 to obtain the second command torque T r1 The second command torque T r1to the vehicle stabilization control unit 240A.

[0189] In this embodiment, the first command torque T LKA1 is input only to the vehicle stabilization control unit 240A. LKA1 and the second command torque T r1 The command torque obtained by adding together the above is input. The other configurations of the vehicle stabilization control unit 240A are similar to the other configurations of the vehicle stabilization control unit 240 in the first embodiment.

[0190] In this embodiment, the vehicle stabilization control unit 240A receives a first command torque T LKA1 and the second input value, the second command torque T r1 and are input independently of each other. Therefore, the output from the first assist control section 210 adjusted by the first gain K1 and the output from the second assist control section 220A adjusted by the second gain K2 can be input to the vehicle stabilization control section 240A regardless of the output of the other. This makes it easy to adjust the control amount by each assist control section.

[0191] In this specification, "the first input value and the second input value are input to the vehicle stabilization control unit independently of each other" means, for example, that the first input value and the second input value input to the vehicle stabilization control unit are calculated without depending on the other input value.

[0192] Other configurations of the control device 100A are similar to the other configurations of the control device 100 in the first embodiment. Other configurations of the lane keeping system 1100A are similar to the other configurations of the lane keeping system 1100 in the first embodiment.

[0193] In this embodiment, the first input value output from the first assist control unit 210 may be an angle. In this case, the first input value may be input to the vehicle stabilization control unit 240A as an angle, or may be converted to a torque and then input to the vehicle stabilization control unit 240A. When the first input value is input to the vehicle stabilization control unit 240A as an angle, the second input value output from the second assist control unit 220A and input to the vehicle stabilization control unit 240A is also an angle.

[0194] <Third embodiment> As shown in Fig. 20, in the control device 100B of this embodiment, the yaw rate control unit 1213 does not have a yaw rate phase adjustment unit 213c, unlike the yaw rate control unit 213 in the first embodiment. r2 is input to the torque conversion unit 213d, and the yaw rate command torque T γ The other configurations of the yaw rate control section 1213 are similar to the other configurations of the yaw rate control section 213 in the first embodiment.

[0195] In this embodiment, the transfer function P nv (s) is expressed, for example, by the following equation (25).

[0196]

number

[0197] In this embodiment, the second model following control unit 241 calculates the transfer function G θ γ (s) is the natural frequency f γ The desired parameter f r By performing model following control to constrain the second control target 580 to a second nominal model NM2 having the following function, the transfer function G θ γ Natural frequency f of yaw rate γ at (s) γ That is, in this embodiment, the second model following control unit 241 can set the transfer function P nv Natural frequency f of yaw rate γ at (s) γ The parameter f corresponds to r By adjusting the transfer function G θ γ The natural frequency fγ of the yaw rate γ in (s) is adjusted. This makes it possible to further improve the running stability of the vehicle V.

[0198] Other configurations of the control device 100B are similar to those of the control device 100 in the first embodiment. The configuration of the control device 100B in this embodiment is also applicable to the control device 100A in the second embodiment.

[0199] The present invention is not limited to the above-described embodiment, and other configurations and methods may be adopted within the scope of the technical idea of ​​the present invention. The first frequency band is not particularly limited. The second frequency band is not particularly limited as long as it is a frequency band higher than the first frequency band. The transfer function of the yaw rate adjustment unit and the transfer function of the lateral deviation adjustment unit do not have to be complementary to each other. The variable parameter may be any parameter that changes based on the second gain. When lane keeping control is executed, the variable parameter and the second gain do not have to be complementary to each other.

[0200] The present technology can be configured as follows. [1] A control device for controlling a steering mechanism mounted on a vehicle, the control device being capable of executing lane keeping control for keeping the vehicle within a lane, the control device comprising: a first assist control unit that generates a first input value; a second assist control unit that generates a second input value based on a steering input value input from a steering wheel of the vehicle; and a disturbance sensitivity control unit to which a command value calculated based on the first input value multiplied by a first gain and the second input value multiplied by a second gain is input, the first gain and the second gain are variable values, and when the first gain is K1 and the second gain is K2, K1=1-K2 is satisfied, and the disturbance sensitivity control unit the control unit has a first model following control unit that generates a correction value for correcting the command value based on an output of a control object including a motor provided in the steering mechanism and a first nominal model based on a configuration of the control object, the first model following control unit is configured such that a transfer function of the control object is constrained to a transfer function of the first nominal model in a frequency band where a complementary sensitivity gain, which is a gain in a gain characteristic of a complementary sensitivity function with respect to a modeling error between the control object and the first nominal model, is approximately 1, and the complementary sensitivity function includes a variable parameter that changes based on the second gain. [2] The control device described in [1], wherein the first gain when the first assist control unit executes the lane keeping control is greater than the first gain when the first assist control unit does not execute the lane keeping control, and the steady-state gain of the complementary sensitivity function when the first assist control unit executes the lane keeping control is greater than the steady-state gain of the complementary sensitivity function when the first assist control unit does not execute the lane keeping control. [3] The control device according to [1] or [2], wherein the larger the variable parameter, the larger the steady-state gain of the complementary sensitivity function, and the variable parameter and the second gain have a mutually complementary relationship. [4] The steering mechanism has an input shaft to which a steering wheel steered by a driver is connected, an output shaft connected to the input shaft via a torsion bar, and the motor connected to the output shaft, and the second assist control unit generates an assist torque that compensates for at least a part of a self-aligning torque generated in a tire of the vehicle based on an input torque calculated based on a torsion bar torque generated in the torsion bar, the second input value is calculated based on the assist torque, the variable parameter is a ratio of the correction value to a disturbance input to the controlled object, and a spring constant of the torsion bar is K tor The gradient of the self-aligning torque generated in the tires of the vehicle relative to the steering angle, which is the rotation angle of the output shaft, is K SAT The gradient of the assist torque with respect to the input torque is K ass The control device according to any one of [1] to [3], wherein the following formula is satisfied when a steady-state gain of the complementary sensitivity function is T(0) and the variable parameter is m.

number

[10] An electric power steering device comprising the control device described in any one of [1] to [9] and the steering mechanism.

[11] A control method for controlling a steering mechanism mounted on a vehicle, the method comprising: executing lane keeping control for keeping the vehicle within a lane; generating a first input value by the lane keeping control; generating a second input value based on a steering input value input from a steering wheel of the vehicle; calculating a command value based on the first input value multiplied by a first gain and the second input value multiplied by a second gain; and generating a correction value for correcting the command value based on an output of a control object including a motor provided in the steering mechanism and a first nominal model based on a configuration of the control object. and constraining a transfer function of the controlled object to a transfer function of the first nominal model by the model following control in a frequency band where a complementary sensitivity gain, which is a gain in a gain characteristic of a complementary sensitivity function with respect to a modeling error between the controlled object and the first nominal model, is approximately 1, wherein the first gain and the second gain are variable values, K1=1-K2 holds when the first gain is K1 and the second gain is K2, and the complementary sensitivity function includes a variable parameter that changes based on the second gain.

[0201] The configurations and methods described in this specification can be combined as appropriate within the scope of not being mutually inconsistent. [Explanation of symbols]

[0202] 100, 100A, 100B... control device, 210... first assist control unit, 211... vehicle state calculation unit, 213, 1213... yaw rate control unit, 213b... yaw rate adjustment unit, 213c... yaw rate phase adjustment unit, 214... lateral deviation control unit, 214b... lateral deviation adjustment unit, 220, 220A... second assist control unit, 230... first model following control unit, 240, 240A... vehicle stabilization control unit, 241... second model following control unit, 290... disturbance sensitivity control unit, 400... imaging device, 521... hand 524a...input shaft, 524b...output shaft, 529A, 529B...tires, 530...steering mechanism, 543...motor, 546...torsion bar, 560...first controlled object (controlled object), 1000...electric power steering device, fb1...first frequency band, fb2...second frequency band, GT...complementary sensitivity gain, K1...first gain, K2...second gain, L...traffic lane, m...variable parameter, NM1...first nominal model, NM2...second nominal model, T(0)...steady-state gain, T(s)...complementary sensitivity function, T ad …input torque, T ass … Assist torque, T f1 …First correction torque (correction value), T LKA …First command torque (first input value), T r …Second command torque (second input value), T rL1 …Command torque (command value), T SAT …self-aligning torque, V…vehicle, y of ...lateral deviation, γ...yaw rate, γ r ,γ r1 ,γ r2 ,γ r3 …Yaw rate target value, θ s …Steering angle, θ t …Actual steering angle

Claims

1. A control device for controlling a steering mechanism mounted on a vehicle, a first assist control unit capable of executing lane keeping control to keep the vehicle within a lane and generating a first input value; a second assist control unit that generates a second input value based on a steering input value input from a steering wheel of the vehicle; a disturbance sensitivity control unit to which a command value calculated based on the first input value multiplied by a first gain and the second input value multiplied by a second gain is input; Equipped with the first gain and the second gain are variable values; The first gain is K １ and the second gain is K ２ Then, K １ = 1 - K ２ holds true, the disturbance sensitivity control unit has a first model following control unit that generates a correction value for correcting the command value based on an output of a control object including a motor provided in the steering mechanism and a first nominal model based on a configuration of the control object, the first model following control unit is configured such that a transfer function of the control object is constrained to a transfer function of the first nominal model in a frequency band in which a complementary sensitivity gain, which is a gain in a gain characteristic of a complementary sensitivity function with respect to a modeling error between the control object and the first nominal model, is approximately 1; The control device, wherein the complementary sensitivity function includes a variable parameter that varies based on the second gain.

2. The first gain in a case where the first assist control unit executes the lane keeping control is larger than the first gain in a case where the first assist control unit does not execute the lane keeping control, 2. The control device according to claim 1, wherein a steady-state gain of the complementary sensitivity function when the first assist control unit executes the lane keeping control is greater than a steady-state gain of the complementary sensitivity function when the first assist control unit does not execute the lane keeping control.

3. the larger the variable parameter, the larger the steady-state gain of the complementary sensitivity function; The control device according to claim 1 , wherein the variable parameter and the second gain have a complementary relationship with each other.

4. The steering mechanism includes an input shaft to which a steering wheel operated by a driver is connected, an output shaft connected to the input shaft via a torsion bar, and the motor connected to the output shaft, The second assist control unit generates an assist torque that compensates for at least a portion of a self-aligning torque generated in a tire of the vehicle, based on an input torque calculated based on a torsion bar torque generated in the torsion bar, The second input value is calculated based on the assist torque, the variable parameter is a ratio of the correction value to a disturbance input to the controlled object, The spring constant of the torsion bar is K ｔｏｒ The gradient of the self-aligning torque generated in the tire of the vehicle relative to the steering angle, which is the rotation angle of the output shaft, is K ＳＡＴ The gradient of the assist torque with respect to the input torque is K ａｓｓ 2. The control device according to claim 1, wherein the following formula is established when a steady-state gain of the complementary sensitivity function is T(0) and the variable parameter is m. [0010]

5. a vehicle stabilization control unit that constrains a transfer function of a plant model, which has an actual steering angle of a tire of the vehicle as an input and a yaw rate of the vehicle as an output, to a transfer function of a second nominal model by model following control; The vehicle stabilization control unit has a second model following control unit that performs model following control, the second model following control unit is configured such that a transfer function of the plant model is constrained to a transfer function of the second nominal model in a frequency band in which a gain in a gain characteristic of a complementary sensitivity function with respect to a modeling error between the plant model and the second nominal model is approximately 1; The control device according to claim 1 , wherein the first input value and the second input value are input to the vehicle stabilization control unit independently of each other.

6. The first assist control unit has a yaw rate control unit that generates a yaw rate command value based on a yaw rate target value, the transfer function of the second nominal model includes a parameter corresponding to a natural frequency of the yaw rate, the yaw rate control unit has a yaw rate phase adjustment unit that adjusts the phase of the yaw rate target value, 6. The control device according to claim 5, wherein the transfer function of the yaw rate phase adjustment unit includes a transfer function capable of converting a parameter corresponding to a natural frequency of the yaw rate in the transfer function of the second nominal model into a different value.

7. the transfer function of the second nominal model includes a parameter corresponding to a natural frequency of the yaw rate, 6. The control device according to claim 5, wherein the second model following control unit adjusts a parameter corresponding to a natural frequency of the yaw rate in a transfer function of the plant model by adjusting the parameter corresponding to a natural frequency of the yaw rate in a transfer function of the second nominal model.

8. The first assist control unit is a vehicle state calculation unit that calculates a yaw rate target value that is a target value of a yaw rate of the vehicle and a lateral deviation of the vehicle based on an input from an imaging device; a yaw rate control unit that generates a yaw rate command value based on the yaw rate target value; a lateral deviation control unit that generates a lateral deviation command value based on the lateral deviation; having The yaw rate control unit has a yaw rate adjustment unit that adjusts the yaw rate command value in accordance with a frequency, the lateral deviation control unit has a lateral deviation adjustment unit that adjusts the lateral deviation command value in accordance with a frequency, a gain of a transfer function of the yaw rate adjustment unit in a first frequency band is lower than a gain of a transfer function of the yaw rate adjustment unit in a second frequency band higher than the first frequency band; The control device according to claim 1 , wherein a gain of the transfer function of the lateral deviation adjustment unit in the second frequency band is lower than a gain of the transfer function of the lateral deviation adjustment unit in the first frequency band.

9. The transfer function of the yaw rate adjustment unit is C γ (s), and the transfer function of the lateral deviation adjustment unit is C ｙ (s) and a constant that is approximately 1 is K γｙ Then, C γ (s)=K γｙ -C ｙ The control device of claim 8 , wherein (s) holds true.

10. A control device according to any one of claims 1 to 9; The steering mechanism; An electric power steering device comprising:

11. A control method for controlling a steering mechanism mounted on a vehicle, comprising: Executing lane keeping control to keep the vehicle within a lane; generating a first input value by the lane keeping control; generating a second input value based on a steering input value input from a steering wheel of the vehicle; calculating a command value based on the first input value multiplied by a first gain and the second input value multiplied by a second gain; executing a model following control to generate a correction value for correcting the command value based on an output of a control object including a motor provided in the steering mechanism and a first nominal model based on a configuration of the control object; constraining a transfer function of the controlled object to a transfer function of the first nominal model in a frequency band in which a complementary sensitivity gain, which is a gain in a gain characteristic of a complementary sensitivity function with respect to a modeling error between the controlled object and the first nominal model, is approximately 1 by the model following control; Including, the first gain and the second gain are variable values; The first gain is K １ and the second gain is K ２ Then, K １ = 1 - K ２ holds true, The complementary sensitivity function includes a variable parameter that varies based on the second gain.

Images