Motor control device

Abstract

This motor control device includes a road surface reaction force characteristic setting unit that sets a road surface reaction force characteristic coefficient on the basis of vehicle information. The road surface reaction force characteristic setting unit includes: a vehicle behavior prediction unit that uses a vehicle behavior prediction model that includes, as an optimization coefficient, a candidate for the road surface reaction force characteristic coefficient, to predict a plurality of vehicle behaviors corresponding to the vehicle information and the optimization coefficient; a vehicle behavior determination unit that determines one appropriate vehicle behavior from among the plurality of vehicle behaviors predicted by the vehicle behavior prediction unit; and a road surface reaction force characteristic determination unit that determines, as the road surface reaction force characteristic coefficient, the optimization coefficient used to predict the vehicle behavior determined by the vehicle behavior determination unit.

Description

Motor control device

[0001] The present disclosure relates to a motor control device that controls the drive of an electric motor for steering angle control.

[0002] The following Patent Document 1 discloses a motor control device that includes a manual steering command value generation unit that generates a manual steering command value using steering torque, an integrated angle command value calculation unit that calculates an integrated angle command value by adding the manual steering command value to an automatic steering command value, and a control unit that controls the angle of an electric motor based on the integrated angle command value.

[0003] The manual steering command value calculation unit in Patent Document 1 calculates the manual steering command value using a reference EPS model. Specifically, the manual steering command value calculation unit calculates the manual steering command value based on an equation of motion that includes, as coefficients, a spring stiffness coefficient (spring constant) and a viscous damping coefficient for applying a virtual reaction force.

[0004] The motor control device of Patent Document 1 further includes a road surface reaction force characteristic setting unit that sets the spring stiffness coefficient and viscous damping coefficient used by the manual steering command value generating unit based on vehicle environment information, which is information related to the vehicle's driving environment.

[0005] In the motor control device described in Patent Document 1, in order to optimize the values ​​of the spring stiffness coefficient and viscous damping coefficient taking into account vehicle behavior, it is necessary to set the optimal spring stiffness coefficient and viscous damping coefficient by repeating test runs while changing the spring stiffness coefficient and viscous damping coefficient for each combination of driving conditions and steering conditions.

[0006] International Publication No. 2023 / 144895

[0007] An object of one embodiment of the present disclosure is to provide a motor control device that makes it easy to set a road reaction force characteristic coefficient that takes vehicle behavior into consideration.

[0008] One embodiment of the present disclosure provides a motor control device including: a manual steering command value generation unit that generates a manual steering command value based on an equation of motion obtained from a reference model of a steering device, the equation of motion including a road surface reaction force characteristic coefficient; an integrated angle command value calculation unit that calculates an integrated angle command value by adding the manual steering command value to an automatic steering command value; a control unit that controls an electric motor for steering angle control based on the integrated angle command value; and a road surface reaction force characteristic setting unit that sets the road surface reaction force characteristic coefficient based on vehicle information, wherein the road surface reaction force characteristic setting unit includes: a vehicle behavior prediction unit that predicts a plurality of vehicle behaviors according to the vehicle information and the optimization coefficients using a vehicle behavior prediction model that includes candidates for the road surface reaction force characteristic coefficients as optimization coefficients; a vehicle behavior determination unit that determines an appropriate vehicle behavior from the plurality of vehicle behaviors predicted by the vehicle behavior prediction unit; and a road surface reaction force characteristic coefficient determination unit that determines, as the road surface reaction force characteristic coefficient, the optimization coefficient used for predicting the vehicle behavior determined by the vehicle behavior determination unit.

[0009] This configuration makes it easy to set the road reaction force characteristic coefficient taking into account the vehicle behavior.

[0010] The above and other objects, features, and advantages of the present disclosure will become apparent from the following description of the embodiments with reference to the accompanying drawings.

[0011] FIG. 1 is a schematic diagram showing a general configuration of an electric power steering system to which a motor control device according to an embodiment of the present disclosure is applied. FIG. 2 is a schematic diagram showing a vehicle and a target trajectory. FIG. 3 is a block diagram showing the electrical configuration of a motor control ECU. FIG. 4 is a schematic diagram showing a torsion bar torque T ｔｂ Assist torque command value T ａｓｓｔ Fig. 5 is a schematic diagram showing an example of a reference EPS model used in a manual steering command value generating unit. Fig. 6 is a block diagram showing the configuration of an angle control unit. Fig. 7 is a block diagram showing the configuration of a road surface reaction force characteristic setting unit.

[0012] Description of Embodiments of the Present Disclosure One embodiment of the present disclosure provides a motor control device including: a manual steering command value generation unit that generates a manual steering command value based on an equation of motion obtained from a reference model of a steering device, the equation including a road surface reaction force characteristic coefficient; an integrated angle command value calculation unit that calculates an integrated angle command value by adding the manual steering command value to an automatic steering command value; a control unit that controls an electric motor for steering angle control based on the integrated angle command value; and a road surface reaction force characteristic setting unit that sets the road surface reaction force characteristic coefficient based on vehicle information, wherein the road surface reaction force characteristic setting unit includes: a vehicle behavior prediction unit that predicts a plurality of vehicle behaviors in accordance with the vehicle information and the optimization coefficients using a vehicle behavior prediction model that includes candidates for the road surface reaction force characteristic coefficients as optimization coefficients; a vehicle behavior determination unit that determines an appropriate vehicle behavior from the plurality of vehicle behaviors predicted by the vehicle behavior prediction unit; and a road surface reaction force characteristic coefficient determination unit that determines, as the road surface reaction force characteristic coefficient, the optimization coefficient used for predicting the vehicle behavior determined by the vehicle behavior determination unit.

[0013] This configuration makes it easy to set the road reaction force characteristic coefficient taking into account the vehicle behavior.

[0014] In one embodiment of the present disclosure, the vehicle behavior prediction model is composed of the reference model and a vehicle model.

[0015] In one embodiment of the present disclosure, the vehicle behavior prediction unit predicts the plurality of vehicle behaviors after imposing constraints on the predicted vehicle behaviors.

[0016] In one embodiment of the present disclosure, the vehicle behavior determination unit is configured to determine the appropriate vehicle behavior using an evaluation function that includes, as variables, a plurality of pieces of information selected from the vehicle information and the optimization coefficients used in predicting the vehicle behavior by the vehicle behavior prediction unit.

[0017] In one embodiment of the present disclosure, the evaluation function is expressed as a sum of values ​​obtained by multiplying the squares of a plurality of pieces of information selected from the vehicle information and the optimization coefficients used in predicting the vehicle behavior by the vehicle behavior prediction unit by weights.

[0018] DETAILED DESCRIPTION OF EMBODIMENTS OF THE PRESENT DISCLOSURE Hereinafter, embodiments of the present disclosure will be described in detail with reference to the accompanying drawings.

[0019] FIG. 1 is a schematic diagram showing a general configuration of an electric power steering system to which a steering device according to an embodiment of the present disclosure is applied.

[0020] The electric power steering system 1 includes a steering wheel (handle) 2 as a steering member for steering the vehicle, a steering mechanism 4 that steers steered wheels 3 in conjunction with the rotation of the steering wheel 2, and a steering assist mechanism 5 that assists the driver in steering. The steering wheel 2 and the steering mechanism 4 are mechanically connected via a steering shaft 6 and an intermediate shaft 7.

[0021] The steering shaft 6 includes an input shaft 8 connected to the steering wheel 2 and an output shaft 9 connected to the intermediate shaft 7. The input shaft 8 and the output shaft 9 are connected via a torsion bar 10 so as to be capable of relative rotation.

[0022] A torque sensor 12 is disposed near the torsion bar 10. The torque sensor 12 detects the torsion bar torque T applied to the steering wheel 2 based on the amount of relative rotational displacement between the input shaft 8 and the output shaft 9. ｔｂ In this embodiment, the torsion bar torque T ｔｂ For example, the torque for steering left is detected as a positive value, and the torque for steering right is detected as a negative value. The larger the absolute value of the torque, the greater the torsion bar torque T ｔｂ The magnitude of is assumed to be large.

[0023] The steering mechanism 4 is made up of a rack-and-pinion mechanism including a pinion shaft 13 and a rack shaft 14 as a steering shaft. The steered wheels 3 are connected to each end of the rack shaft 14 via tie rods 15 and knuckle arms (not shown). The pinion shaft 13 is connected to the intermediate shaft 7. The pinion shaft 13 rotates in conjunction with the steering of the steering wheel 2. A pinion 16 is connected to the tip of the pinion shaft 13.

[0024] The rack shaft 14 extends linearly in the left-right direction of the vehicle. A rack 17 that meshes with the pinion 16 is formed in the axial middle of the rack shaft 14. The pinion 16 and the rack 17 convert the rotation of the pinion shaft 13 into axial movement of the rack shaft 14. By moving the rack shaft 14 in the axial direction, the steered wheels 3 can be steered.

[0025] When the steering wheel 2 is steered (rotated), this rotation is transmitted to the pinion shaft 13 via the steering shaft 6 and the intermediate shaft 7. The rotation of the pinion shaft 13 is then converted into axial movement of the rack shaft 14 by the pinion 16 and the rack 17. As a result, the steered wheels 3 are steered.

[0026] The steering assist mechanism 5 includes an electric motor 18 for generating a steering assist force (assist torque), and a reducer 19 for amplifying the output torque of the electric motor 18 and transmitting it to the steering mechanism 4. The reducer 19 is made up of a worm gear mechanism including a worm gear 20 and a worm wheel 21 that meshes with the worm gear 20. The reducer 19 is housed in a gear housing 22 that serves as a transmission mechanism housing.

[0027] In the following, the reduction ratio (gear ratio) of the reducer 19 is N Ｒ Reduction ratio N Ｒ is the worm wheel angle θ, which is the rotation angle of the worm wheel 21 ｗｗ The worm gear angle θ is the rotation angle of the worm gear 20 relative to the ｗｇ The ratio (θ ｗｇ / θ ｗｗ ) is defined as

[0028] The worm gear 20 is rotationally driven by the electric motor 18. The worm wheel 21 is connected to the output shaft 9 so as to be rotatable integrally therewith.

[0029] When the worm gear 20 is rotationally driven by the electric motor 18, the worm wheel 21 is rotationally driven, and motor torque is applied to the steering shaft 6, causing the steering shaft 6 (output shaft 9) to rotate. The rotation of the steering shaft 6 is then transmitted to the pinion shaft 13 via the intermediate shaft 7. The rotation of the pinion shaft 13 is converted into axial movement of the rack shaft 14, thereby turning the steered wheels 3. In other words, by rotating the worm gear 20 with the electric motor 18, steering assistance by the electric motor 18 and steering of the steered wheels 3 become possible. The electric motor 18 is provided with a rotation angle sensor 23 for detecting the rotation angle of the rotor of the electric motor 18.

[0030] The torque applied to the output shaft 9 (an example of a drive target of the electric motor 18) includes a motor torque by the electric motor 18 and a disturbance torque T ｌｃ Disturbance torque other than the motor torque T ｌｃ Torsion bar torque T ｔｂ , road reaction torque (road load torque) T ｒｌ , friction torque T ｆ etc. are included.

[0031] Torsion bar torque T ｔｂ is the torque applied to the output shaft 9 from the steering wheel 2 side by the force applied to the steering wheel 2 by the driver, the force generated by steering inertia, etc.

[0032] Road reaction torque T ｒｌ is the torque applied to the output shaft 9 from the steered wheels 3 side via the rack shaft 14 due to the self-aligning torque generated in the tire, the force generated by the suspension and tire-wheel alignment, the frictional force of the rack-and-pinion mechanism, etc.

[0033] The vehicle is equipped with a CCD (Charge Coupled Device) camera 25 that photographs the road ahead in the direction of travel of the vehicle, a GPS (Global Positioning System) 26 for detecting the vehicle's position, a radar 27 for detecting road shapes and obstacles, a map information memory 28 that stores map information, and a vehicle speed sensor 29.

[0034] The CCD camera 25, GPS 26, radar 27, map information memory 28, and vehicle speed sensor 29 are connected to a host ECU (Electronic Control Unit) 201 for performing driving assistance control. Based on the information obtained by the CCD camera 25, GPS 26, radar 27, and vehicle speed sensor 29 and map information, the host ECU 201 performs surrounding environment recognition, vehicle position estimation, route planning, etc., and determines control target values ​​for steering and drive actuators.

[0035] In this embodiment, there are two driving modes: a normal mode and a driving assistance mode. In the driving assistance mode, the host ECU 201 controls the automatic steering command value θ ＡＤ，ｃｍｄ In this embodiment, the driving assistance is a lane centering assist (LCA) that assists in steering so that the vehicle stays in the center of the driving lane. The driving assistance may be a lane keeping assist (LKA) that prevents the vehicle from deviating from the lane, or another type of driving assistance. ＡＤ，ｃｍｄ is the target value of the steering angle for making the vehicle travel along the target trajectory (target travel line).

[0036] In this embodiment, the automatic steering command value θ ＡＤ，ｃｍｄ is expressed as the amount of rotation (rotation angle) of the output shaft 9 from the neutral position, and the amount of rotation from the neutral position in the left steering direction is expressed as a positive value, and the amount of rotation from the neutral position in the right steering direction is expressed as a negative value. ＡＤ，ｃｍｄ is set based on, for example, the vehicle speed, the lateral deviation from a target trajectory (for example, the center line of the lane), and the yaw angle deviation of the vehicle from the target trajectory. ＡＤ，ｃｍｄ The process of setting the value is well known, so a detailed description will be omitted here.

[0037] The host ECU 201 also outputs a mode signal S indicating whether the driving mode is the normal mode or the driving assistance mode. ｍｏｄｅ , automatic steering command value θ ＡＤ，ｃｍｄ In this embodiment, the vehicle information output from the host ECU 201 is the yaw angle deviation ψ ｅ , yaw rate deviation dψ ｅ / dt, lateral deviation y ｅ , lateral deviation y ｅ The time derivative dy ｅ / dt, lateral deviation y ｅ The integral value of ∫y ｅ , the yaw angle ψ of the target trajectory ｒｅｆ , the yaw angle ψ of the target trajectory ｒｅｆ The differential value dψ ｒｅｆ / dt and vehicle speed V. In the following, the vehicle information output from the host ECU 201 will be collectively referred to as C ｉｎｆ It may be expressed as:

[0038] In this embodiment, the vehicle information used to predict the vehicle behavior includes the vehicle information C ｉｎｆ In addition, torsion bar torque T ｔｂ , actual steering angle θ c and actual steering angle θ c The time derivative dθ c / dt is also included. Actual steering angle θ c is the rotation angle of the output shaft 9. Vehicle information C used to predict vehicle behavior ｉｎｆ , T ｔｂ , θ cc , dθ c / dt is an example of "vehicle information" in this disclosure.

[0039] FIG. 2 is a schematic diagram showing a vehicle and a target trajectory.

[0040] In FIG. Ｂ and y Ｂ are the x-axis and y-axis of the vehicle coordinate system based on the vehicle 101. Ｂ is an axis that has its origin at the center of gravity of the vehicle 401 (hereinafter referred to as "vehicle center of gravity G") and extends in the front-rear direction of the vehicle. Ｂ is an axis extending in the lateral direction of the vehicle with the vehicle center of gravity G as its origin.Ｇ and y Ｇ are the x-axis and y-axis of the global coordinate system (world coordinate system), respectively.

[0041] Yaw angle deviation ψ ｅ is the yaw angle ψ of the vehicle 401 and the yaw angle ψ of the target trajectory 402 ｒｅｆ The angle deviation from (ψ-ψ ｒｅｆ ) The yaw rate deviation dψ ｅ / dt is the yaw angle deviation ψ ｅ The time derivative of (dψ / dt-dψ ｒｅｆ / dt).

[0042] Lateral deviation y ｅ is the distance from the target trajectory 402 to the vehicle center of gravity G. ｅ is the lateral position y of the vehicle 401 and the lateral position y of the target trajectory 402 ｒｅｆ Position deviation (y-y ｒｅｆ )

[0043] Mode signal S ｍｏｄｅ , automatic steering command value θ ＡＤ，ｃｍｄ and vehicle information C output from the host ECU 201 ｉｎｆ is given to the motor control ECU 202 via the in-vehicle network. ｔｂ The output signal of the rotation angle sensor 23 is input to the motor control ECU 202. The motor control ECU 202 controls the electric motor 18 based on these input signals and information provided by the host ECU 201.

[0044] FIG. 3 is a block diagram showing the electrical configuration of the motor control ECU 202.

[0045] The following mainly describes the operation when the driving mode is the driving assistance mode.

[0046] The motor control ECU 202 includes a microcomputer 40, a drive circuit (inverter circuit) 31 controlled by the microcomputer 40 to supply power to the electric motor 18, and a current (hereinafter, "motor current I") flowing through the electric motor 18. ｍ and a current detection circuit 32 for detecting the current.

[0047] The microcomputer 40 includes a CPU and memory (ROM, RAM, non-volatile memory, etc.), and functions as a plurality of functional processing sections by executing predetermined programs. The functional processing sections include a rotation angle calculation section 41, a reduction ratio division section 42, a road surface reaction force characteristic setting section 43, an assist torque command value setting section 44, a manual steering command value generation section 45, an integrated angle command value calculation section 46, an angle control section 47, a first weight multiplication section 48, a second weight multiplication section 49, an addition section 50, a torque control section (current control section) 51, and a weight setting section 52.

[0048] The rotation angle calculation unit 41 calculates the rotor rotation angle θ of the electric motor 18 based on the output signal of the rotation angle sensor 23. ｍ The reduction ratio division unit 42 calculates the rotor rotation angle θ ｍ Reduction ratio N Ｒ By dividing by this, the rotor rotation angle θ ｍ The rotation angle (actual steering angle) θ of the output shaft 9 ｃ In this embodiment, the actual steering angle θ ｃ is expressed as the amount of rotation (rotation angle) of the output shaft 9 from the neutral position, and the amount of rotation from the neutral position in the left steering direction is expressed as a positive value, and the amount of rotation from the neutral position in the right steering direction is expressed as a negative value.

[0049] The road surface reaction force characteristic setting unit 43 sets the spring stiffness coefficient k and viscous damping coefficient c used by the manual steering command value generating unit 45. Hereinafter, the spring stiffness coefficient k and the viscous damping coefficient c may be collectively referred to as the road surface reaction force characteristic coefficient. The spring stiffness coefficient k and the viscous damping coefficient c are examples of the "road surface reaction force characteristic coefficient" in the present disclosure. The road surface reaction force characteristic setting unit 43 is an example of the "road surface reaction force characteristic setting unit" in the present disclosure. Details of the road surface reaction force characteristic setting unit 43 will be described later.

[0050] The assist torque command value setting unit 44 sets the assist torque command value T ａｓｓｔ The assist torque command value setting unit 44 sets the vehicle information C ｉｎｆand the torsion bar torque T detected by the torque sensor 12. ｔｂ Based on this, the assist torque command value T ａｓｓｔ Set the torsion bar torque T ｔｂ Assist torque command value T ａｓｓｔ An example of the setting is shown in FIG.

[0051] Assist torque command value T ａｓｓｔ is set to a positive value when the electric motor 18 is to generate a steering assist force for steering to the left, and is set to a negative value when the electric motor 18 is to generate a steering assist force for steering to the right. ａｓｓｔ is the torsion bar torque T ｔｂ The torsion bar torque T ｔｂ For negative values ​​of , it is taken to be negative.

[0052] Assist torque command value T ａｓｓｔ is the torsion bar torque T ｔｂ The absolute value of the y-axis is set to be larger as the absolute value of y-axis increases, and the absolute value of the y-axis is set to be smaller as the vehicle speed V increases.

[0053] The assist torque command value setting unit 44 sets the torsion bar torque T ｔｂ is multiplied by a preset constant to obtain the assist torque command value T ａｓｓｔ may be calculated.

[0054] The manual steering command value generating unit 45 basically generates a steering angle (more precisely, a rotation angle θ of the output shaft 9) corresponding to the steering wheel operation when the driver operates the steering wheel 2. ｃ ) is the manual steering command value θ ＭＤ，ｃｍｄ The manual steering command value generating unit 45 is provided to set the torsion bar torque T ｔｂ and the assist torque command value T set by the assist torque command value setting unit 44. ａｓｓｔ and the spring stiffness coefficient k and the viscous damping coefficient c set by the road surface reaction force characteristic setting unit 43 are used to calculate the manual steering command value θ ＭＤ，ｃｍｄThe manual steering command value generating unit 45 will be described in detail later.

[0055] The integrated angle command value calculation unit 46 calculates the automatic steering command value θ set by the host ECU 201. ＡＤ，ｃｍｄ , manual steering command value θ ＭＤ，ｃｍｄ The integrated angle command value θ ｉｎｔ，ｃｍｄ Calculate the following.

[0056] The angle control unit 47 calculates an integrated angle command value θ ｉｎｔ，ｃｍｄ Based on this, the integrated angle command value θ ｉｎｔ，ｃｍｄ The integrated motor torque command value T ｍｉｎｔ，ｃｍｄ The angle control unit 47 will be described in detail later.

[0057] The first weight multiplier 48 multiplies the assist torque command value T ａｓｓｔ The second weight multiplier 49 multiplies the integrated motor torque command value T ｍｉｎｔ，ｃｍｄ is multiplied by the second weight W2. The first weight W1 and the second weight W2 are set by the weight setting unit 52.

[0058] The adder 50 calculates the assist torque command value W1·T after the first weight multiplication (after the first weighting process). ａｓｓｔ and the integrated motor torque command value W2·T after multiplication by the second weight (after the second weighting process) ｍｉｎｔ，ｃｍｄ By adding these, the motor torque command value T ｍ，ｃｍｄ Calculate the following.

[0059] When the driving mode is the normal mode, the weight setting unit 52 sets the first weight W1 to 1 and the second weight W2 to 0. When the driving mode is the driving assistance mode, the weight setting unit 52 sets the first weight W1 to 0 and the second weight W2 to 1. Therefore, when the driving mode is the normal mode, the assist torque command value T ａｓｓｔ is the motor torque command value T ｍ，ｃｍｄ On the other hand, when the driving mode is the driving assistance mode, the integrated motor torque command value T ｍｉｎｔ，ｃｍｄ is the motor torque command value T ｍ，ｃｍｄ is given to the torque control unit 51 as

[0060] The torque control unit 51 controls the motor torque of the electric motor 18 to be equal to the motor torque command value T ｍ，ｃｍｄ As the torque control unit 51, for example, the torque control unit (47) shown in Fig. 2 and Fig. 9 of Patent Document 1 (WO 2023 / 144895) can be used. However, in Fig. 2 and Fig. 9 of WO 2023 / 144895, the motor torque command value T ｍ，ｃｍｄ Is T ＊ ｍ，ｉｎｔ In this case, the torque control unit 51 is configured to control the motor torque command value T ｍ，ｃｍｄ The torque control unit 51 calculates a current command value by dividing the motor current I ｍ Feedback control is performed so that the current command value approaches the reference value.

[0061] FIG. 5 is a block diagram showing the configuration of the angle control unit 47.

[0062] The angle control unit 47 calculates an integrated angle command value θ ｉｎｔ，ｃｍｄ Based on this, the integrated motor torque command value T ｍｉｎｔ，ｃｍｄ The angle control unit 47 includes a low-pass filter (LPF) 61, a feedback control unit 62, a feedforward control unit 63, a disturbance torque estimating unit 64, a torque adding unit 65, a disturbance torque compensating unit 66, a reduction ratio dividing unit 67, and a reduction ratio multiplying unit 68.

[0063] The reduction ratio multiplication unit 68 multiplies the motor torque command value T ｍ，ｃｍｄ The reduction ratio N of the reducer 19 Ｒ By multiplying by ｍ，ｃｍｄ is the output shaft torque command value N acting on the output shaft 9 (worm wheel 21). Ｒ ・T ｍ，ｃｍｄ Convert to.

[0064] The low-pass filter 61 calculates the integrated angle command value θ ｉｎｔ，ｃｍｄ The integrated angle command value θ after low-pass filtering is ｉｎｔＬ，ｃｍｄis given to the feedback control section 62 and the feedforward control section 63.

[0065] The feedback control unit 62 calculates the steering angle estimated value ^θ calculated by the disturbance torque estimating unit 64. ｃ is the integrated angle command value θ after low-pass filtering. ｉｎｔＬ，ｃｍｄ The feedback control unit 62 includes an angle deviation calculation unit 62A and a PD control unit 62B. The angle deviation calculation unit 62A calculates an integrated angle command value θ ｉｎｔＬ，ｃｍｄ and the estimated steering angle ^θ ｃ The angle deviation Δθ (= θ ｉｎｔＬ，ｃｍｄ -^θ ｃ ) is calculated. The angle deviation calculation unit 62A calculates the integrated angle command value θ ｉｎｔＬ，ｃｍｄ and the actual steering angle θ calculated by the reduction ratio division unit 42 (see FIG. 3). ｃ The angle deviation (θ ｉｎｔＬ，ｃｍｄ -θ ｃ ) may be calculated as the angle deviation Δθ.

[0066] The PD control unit 62B performs a PD calculation (proportional differential calculation) on the angle deviation Δθ calculated by the angle deviation calculation unit 62A, thereby obtaining a feedback control torque T ｆｂ The feedback control torque T ｆｂ is given to the torque adder 65.

[0067] The feedforward control unit 63 is provided to compensate for a delay in response due to the inertia of the electric power steering system 1, thereby improving the response of the control. The feedforward control unit 63 includes an angular acceleration calculation unit 63A and an inertia multiplication unit 63B. The angular acceleration calculation unit 63A calculates an integrated angle command value θ ｉｎｔＬ，ｃｍｄ By differentiating twice, the target angular acceleration d ２ θ ｉｎｔＬ，ｃｍｄ / dt ２ Calculate the following.

[0068] The inertia multiplication unit 63B multiplies the target angular acceleration d calculated by the angular acceleration calculation unit 63A by ２ θ ｉｎｔＬ，ｃｍｄ / dt ２ , the inertia J of the electric power steering system 1 ｓBy multiplying by ｆｆ (=J ｓ ・d ２ θ ｉｎｔＬ，ｃｍｄ / dt ２ ) is calculated. ｓ is obtained from, for example, a physical model (not shown) of the electric power steering system 1. The feedforward control torque T ｆｆ is given to the torque adder 65 as an inertia compensation value.

[0069] The torque adder 65 calculates the feedback control torque T ｆｂ The feedforward control torque T ｆｆ By adding ｆｂ +T ｆｆ ) is calculated.

[0070] The disturbance torque estimating unit 64 is provided to estimate a nonlinear torque (disturbance torque: torque other than motor torque) that occurs as a disturbance in the plant (the object to be controlled by the electric motor 18). Ｒ ・T ｍ，ｃｍｄ and the actual steering angle θ ｃ Based on this, the disturbance torque (disturbance load) T ｌｃ , steering angle θ ｃ and steering angle differential value (angular velocity) dθ ｃ / dt is estimated. ｌｃ and steering angle θ ｃ The estimated values ​​of ^T ｌｃ and ^θ ｃ It is expressed as:

[0071] As the disturbance torque estimating unit 64, for example, the disturbance torque estimating unit (64) shown in Fig. 6 and Fig. 8 of Patent Document 1 (WO 2023 / 144895) can be used. However, in Fig. 6 and Fig. 8 of WO 2023 / 144895, the actual steering angle is calculated based on θ ｃ rather than θ ｃ，ｉｎｔ is shown.

[0072] The disturbance torque estimated value ^T calculated by the disturbance torque estimator 64 ｌｃis given to the disturbance torque compensator 66 as a disturbance torque compensation value. The steering angle estimated value ^θ calculated by the disturbance torque estimator 64 ｃ is given to the angle deviation calculation unit 62A.

[0073] The disturbance torque compensator 66 calculates the basic torque command value (T ｆｂ +T ｆｆ ) to the estimated disturbance torque value ^T ｌｃ By subtracting ｓｉｎｔ，ｃｍｄ (=T ｆｂ +T ｆｆ -^T ｌｃ ) is calculated. As a result, the integrated steering torque command value T ｓｉｎｔ，ｃｍｄ (torque command value for the output shaft 9) is obtained.

[0074] Integrated steering torque command value T ｓｉｎｔ，ｃｍｄ is given to the reduction ratio division unit 67. The reduction ratio division unit 67 calculates the integrated steering torque command value T ｓｉｎｔ，ｃｍｄ Reduction ratio N Ｒ By dividing by , the integrated motor torque command value T ｍｉｎｔ，ｃｍｄ This integrated motor torque command value T ｍｉｎｔ，ｃｍｄ is provided to the second weight multiplier 49 (see FIG. 3).

[0075] The manual steering command value generating unit 45 will be described in detail. In this embodiment, the manual steering command value generating unit 45 uses a reference EPS model to generate a manual steering command value θ ＭＤ，ｃｍｄ Set.

[0076] FIG. 6 is a schematic diagram showing an example of a reference EPS model used in the manual steering command value generating unit 45.

[0077] This reference EPS model is a single inertia model that includes a lower column. The lower column is an example of a plant that is driven by the electric motor 18. The lower column corresponds to the output shaft 9 and the worm wheel 21. However, this model is just one example, and the reference EPS model may also be an inertia model that includes a configuration other than the above (for example, the rack shaft 14).

[0078] In FIG. ｐis the inertia of the lower column (hereinafter referred to as "column inertia"), and θ ｃｏｌ is the rotation angle of the lower column, and T ｔｂ is the torsion bar torque. ｔｂ , torque N acting on the output shaft 9 from the electric motor 18 Ｒ ・T ｍ and road reaction torque (virtual reaction force) T ｒｌ is given.

[0079] Road reaction torque T ｒｌ is expressed by the following equation (1) using a spring stiffness coefficient k, which is the stiffness coefficient of the virtual spring, and a viscous damping coefficient c, which is the viscous damping coefficient of the virtual damper.

[0080]

[0081] The equation of motion of the reference EPS model is expressed by the following equation (2).

[0082]

[0083] In formula (2), J ｐ ・d ２ θ ｃｏｌ / dt ２ is the inertia torque acting on the lower column.

[0084] The manual steering command value generating unit 45 is T ｔｂ The torsion bar torque T detected by the torque sensor 12 ｔｂ Substituting, T ｍ The assist torque command value T ａｓｓｔ By substituting the above and solving the differential equation (2), the rotation angle θ of the lower column is obtained. ｃｏｌ Then, the manual steering command value generating unit 45 calculates the obtained rotation angle θ of the lower column. ｃｏｌ The manual steering command value θ ＭＤ，ｃｍｄ Generate it as:

[0085] The equation of motion in equation (2) is T ｍ T ａｓｓｔ and θ ｃｏｌ θ ＭＤ，ｃｍｄ is equivalent to the equation of motion in which

[0086] Next, the road surface reaction force characteristic setting unit 43 will be described in detail. The road surface reaction force characteristic setting unit 43 receives the vehicle information C ｉｎｆ and the automatic steering command value θ ＡＤ，ｃｍｄ and the actual steering angle θ ｃ and torsion bar torque T ｔｂ The spring stiffness coefficient k and the viscous damping coefficient c are set based on the above.

[0087] FIG. 7 is a block diagram showing the configuration of the road surface reaction force characteristic setting unit 43.

[0088] The road surface reaction force characteristic setting unit 43 includes a vehicle behavior prediction unit 71 , a vehicle behavior determination unit 72 , and a road surface reaction force characteristic coefficient determination unit 73 .

[0089] The vehicle behavior prediction unit 71 calculates the candidates for the spring stiffness coefficient k and the viscous damping coefficient c as optimization coefficients Z ａ，１ and Z ａ，２ More specifically, the vehicle behavior is predicted using a vehicle behavior prediction model including an optimization coefficient Z ａ，１ , Z ａ，２ By changing the vehicle information C ｉｎｆ , T ｔｂ , θ ｐ , dθ ｐ / dt and optimization coefficient Z ａ，１ , Z ａ，２ In the following, we will predict multiple vehicle behaviors according to Z ａ，１ the first optimization coefficient Z ａ，１ That said, Z ａ，２ the second optimization coefficient Z ａ，２ This is sometimes the case.

[0090] The vehicle behavior prediction model is a physical model that is composed of a reference model of the steering device (see FIG. 6) and a vehicle model.

[0091] The vehicle state x, the optimization coefficient u, and the model disturbance w are expressed by the following equations (3), (4), and (5), respectively. The discretization method of equation (5) is described using the Euler method, but other discretization methods (e.g., Runge-Kutta method) may also be used. Note that in equation (4), T ｍis used, but T ｍ =T ａｓｓｔ is.

[0092]

[0093] The vehicle behavior prediction model is expressed by the following equation (6).

[0094]

[0095] In the formula (6), the definitions of the symbols other than those already explained are as follows:

[0096] l ｆ : Distance between the center of gravity of the vehicle and the front wheel axle l ｒ : Distance between the center of gravity of the vehicle and the rear wheel axle C ｆ : Front wheel cornering rigidity C ｒ : cornering stiffness of rear wheels I: yaw moment of inertia of vehicle m: vehicle mass N: steering gear ratio When the above equation (6) is discretized using the period (sampling time) for calculating the vehicle behavior, the following equation (7) is obtained.

[0097]

[0098] x in formula (7) ｉ＋１ is the vehicle behavior predicted by the vehicle behavior prediction model. ０ Then, for i=1, x １ = Ax ０ +Bu ０ +Golden Week ｏ At i=2, x ２ = Ax １ +Bu １ +Golden Week １ This becomes:

[0099] In this embodiment, the manual steering command value θ ＭＤ，ｃｍｄ (Motor torque command value T ｍ，ｃｍｄ ) is calculated in a period of T0 [sec], the vehicle behavior x ｉThe period for calculating and setting (updating) the road reaction force coefficients k and c (hereinafter referred to as the "first period T1") may be A·T0 [sec], where A is a natural number. For example, if T0 is 0.01 [sec] and A is 10, T1 becomes 0.1 [sec]. Note that if A=1, T1=T0.

[0100] The vehicle behavior prediction unit 71 predicts one or more vehicle behaviors at one or more time points within a period (prediction horizon) up to N×T1 seconds into the future, based on the above-mentioned equation (7) for each first period T1, where N is a natural number equal to or greater than 1. The following description will mainly focus on the case where N is equal to or greater than 2.

[0101] The vehicle behavior prediction unit 71 calculates the optimization coefficient Z ａ，１ , Z ａ，２ A plurality of time-series vehicle behavior data series are generated, each having a different time-series data series (hereinafter referred to as "time-series Z data series"), which are a combination of the above. The time-series Z data series (time-series Z data group) is composed of N steps of time-series Z data. The time-series vehicle behavior data series (time-series vehicle behavior data group) is composed of N steps of time-series vehicle behavior data. The number of types of time-series Z data series is M, where M is a natural number equal to or greater than 2. Therefore, the vehicle behavior prediction unit 71 generates M types of time-series vehicle behavior data series, each consisting of N steps of time-series vehicle behavior data, for each first period T1.

[0102] The first time series Z data (Z of the first step) of the M types of time series Z data series ａ，１ , Z in the first step ａ，２ ) is set randomly, for example. Here, the first optimization coefficient Z ａ，１ The first reference value Z ａ，１，ｒｅｆ and the second optimization coefficient Z ａ，２ The second reference value Z ａ，２，ｒｅｆ It is assumed that the values ​​of Z in the first step for each of the M types of time series Z data series are set in advance. ａ，１ is the first reference value Z ａ，１，ｒｅｆ Similarly, the Zａ，２ is the second reference value Z ａ，２，ｒｅｆ may be set by increasing or decreasing the value by a predetermined value.

[0103] The first optimization coefficient Z for each time series Z data series after the second step ａ，１ and the second optimization coefficient Z ａ，２ is the first optimization coefficient Z of the first step in the time series Z data series. ａ，１ and the second optimization coefficient Z ａ，２ Specifically, the first optimization coefficient Z ａ，１ is the first optimization coefficient Z ａ，１ The second optimization coefficient Z may be set by gradually increasing, gradually decreasing, or not changing. ａ，２ is the second optimization coefficient Z in the first step ａ，２ The value may be set by gradually increasing, gradually decreasing, not changing, or the like.

[0104] The time-series Z data constituting the M types of time-series Z data series may be set in advance or may be set for each first period T1.

[0105] A method for generating time-series vehicle event data in any one time-series vehicle event data series will now be briefly described.

[0106] The vehicle behavior prediction unit 71 predicts the initial state x ０ Get w ０ Then, the vehicle behavior prediction unit 71 obtains u ０ (Z in the first step ａ，１ , Z in the first step ａ，２ ) to obtain the time-series vehicle behavior data (x １ = Ax ０ +Bu ０ +Gw ０ ) to generate the

[0107] Next, the vehicle behavior prediction unit 71 calculates w ０ Wow １ Set it as w １ (=w ０ ), x１ and u １ (Z in the second step ａ，１ , Z in the second step ａ，２ ) to obtain the time-series vehicle behavior data (x ２ = Ax １ +Bu １ +Gw １ ) is generated. In the same manner, time-series vehicle behavior data from the third step onwards is generated. Even when generating time-series vehicle behavior data from the third step onwards, the w obtained when generating the time-series vehicle behavior data in the first step is used. ０ But w ｉ is set as

[0108] For each first period T1, the vehicle behavior determination unit 72 determines an appropriate one time-series vehicle behavior data series from among the M types of time-series vehicle behavior data series generated by the vehicle behavior prediction unit 71. Specifically, the vehicle behavior determination unit 72 determines the appropriate one time-series vehicle behavior data series using, for example, an evaluation function J(m) of the following equation (8):

[0109]

[0110] In equation (8), θ ｐ is the actual steering angle θ ｃ and the automatic steering command value θ ＡＤ，ｃｍｄ The angle deviation (θ ｃ -θ ＡＤ，ｃｍｄ In formula (8), Z ａ，１，ｒｅｆ is the first optimization coefficient Z ａ，１ is a reference value and is set in advance. ａ，２，ｒｅｆ is the second optimization coefficient Z ａ，２ This is a reference value that is set in advance.

[0111] In formula (8), Q １ ~Q ７ , R １ and R ２ are weights for the corresponding parameters and are used to adjust the vehicle behavior. １ ~Q ７ , R １ , R ２The value of is preset from the viewpoint of how to set an appropriate vehicle behavior.

[0112] More specifically, the vehicle behavior determination unit 72 selects, as an appropriate vehicle behavior data series, one vehicle behavior data series that minimizes the evaluation function J(m) from among the M types of time-series vehicle behavior data series generated by the vehicle behavior prediction unit 71. Then, the vehicle behavior determination unit 72 determines, as an appropriate vehicle behavior, the vehicle behavior data that is first in chronological order from among the N pieces of time-series vehicle behavior data included in the selected time-series vehicle behavior data series. The reason for this is that the predicted data closest to the current time in chronological order is considered to have higher prediction accuracy than later predicted data.

[0113] The road reaction force characteristic coefficient determination unit 73 determines the optimization coefficient Z ａ，１ and Z ａ，２ are determined as the spring stiffness coefficient k and the viscous damping coefficient c, respectively.

[0114] The spring stiffness coefficient k and the viscous damping coefficient c determined by the road surface reaction force characteristic coefficient determination unit 73 are provided to the manual steering command value generation unit 45. As a result, the spring stiffness coefficient k and the viscous damping coefficient c used in the manual steering command value generation unit 45 are updated.

[0115] The entire optimization process from predicting vehicle behavior to determining the optimal road reaction force coefficient using an evaluation function may use an algorithm that can more efficiently search for an optimal solution, which is used in optimal control such as LQR control or model predictive control.

[0116] In addition, when N=1, the vehicle behavior prediction unit 71 calculates the optimization coefficient Z ａ，１ , Z ａ，２ A plurality of vehicle behavior data sets with different combinations of the optimization coefficient Z ａ，１ , Z ａ，２ Assuming that the number of types of combinations is M (M is a natural number of 2 or more), the vehicle behavior prediction unit 71 generates M types of vehicle behavior data for each first period T1.

[0117] The vehicle behavior determination unit 72 determines, as an appropriate vehicle behavior, one vehicle behavior data that minimizes the evaluation function J(m) of the equation (8) from among the M types of vehicle behavior data generated by the vehicle behavior prediction unit 71. The road surface reaction force characteristic coefficient determination unit 73 uses the optimization coefficient Z ａ，１ and Z ａ，２ are determined as the spring stiffness coefficient k and the viscous damping coefficient c, respectively.

[0118] In this embodiment, when the driving mode is the manual driving mode, the assist torque command value T ａｓｓｔ When the driving mode is the driving assistance mode, the electric motor 18 is controlled based only on the integrated motor torque command value T ｍｉｎｔ，ｃｍｄ That is, the electric motor 18 is controlled based on the assist torque command value T ａｓｓｔ a control mode in which the electric motor 18 is controlled based only on the integrated motor torque command value T ｍｉｎｔ，ｃｍｄ The control mode in which the electric motor 18 is controlled can be switched based on the above.

[0119] In addition, in this embodiment, the spring stiffness coefficient k and the viscous damping coefficient c for obtaining appropriate vehicle behavior can be automatically calculated, which makes it easy to set the road reaction force coefficient in consideration of the vehicle behavior in the driving assistance mode.

[0120] This eliminates the need to repeat test runs while changing the spring stiffness coefficient k and the viscous damping coefficient c for each combination of driving conditions and steering conditions in order to optimize the values ​​of the load spring stiffness coefficient and the viscous damping coefficient taking into account the vehicle behavior.

[0121] Although the embodiments of the present disclosure have been described above, the present disclosure can also be embodied in other forms.

[0122] The vehicle behavior prediction unit 71 may predict a plurality of vehicle behaviors by imposing constraints on the predicted vehicle behaviors. For example, the lateral deviation y ｅ A limit may be set on the absolute value of the lateral deviation y of the predicted vehicle behavior. ｅ Absolute value of |yｅ | is limited to D (D>0) or less, the vehicle behavior prediction unit 71 calculates the lateral deviation y ｅ Absolute value of |y ｅ Vehicle behaviors for which | is equal to or less than D are excluded, and the remaining vehicle behaviors are set as the vehicle behavior predicted for the first period T1. In this case, the road reaction force coefficient is set while observing the constraints on lateral deviation, so that the system can function as a lane keeping assist (LKA) that prevents the vehicle from leaving its lane.

[0123] In the above embodiment, the vehicle behavior prediction unit 71 calculates the optimization coefficient Z ａ，１ , Z ａ，２ However, the vehicle behavior prediction unit 71 generates a plurality of time-series vehicle behavior data series with different "time-series data series of combinations of the optimization coefficient Z ａ，１ , Z ａ，２ In this case, the optimization coefficient Z ａ，１ , Z ａ，２ Although the combination of the optimization coefficient Z ａ，１ , Z ａ，２ The combination does not change.

[0124] In this case, the vehicle behavior determination unit 72 selects, as an appropriate time-series vehicle behavior data series, one vehicle behavior data series that has the smallest evaluation function J(m) from among the multiple types of time-series vehicle behavior data series generated by the vehicle behavior prediction unit 71. Then, the vehicle behavior determination unit 72 determines any vehicle behavior data in the selected time-series vehicle behavior data series as an appropriate vehicle behavior. Therefore, in this case, the optimization coefficient Z ａ，１ and Z ａ，２ are determined as the spring stiffness coefficient k and the viscous damping coefficient c, respectively.

[0125] In the above-described embodiment, the manual steering command value generating unit 45 uses the assist torque command value T ａｓｓｔUsing the manual steering command value θ ＭＤ，ｃｍｄ The road surface reaction force characteristic setting unit 43 generates the assist torque command value T ａｓｓｔ The road reaction force characteristics are set using the formula (4).

[0126] However, the manual steering command value generating unit 45 does not ａｓｓｔ is set to 0, and the manual steering command value θ ＭＤ，ｃｍｄ In this case, the road surface reaction force characteristic setting unit 43 may generate the assist torque command value T ａｓｓｔ may be set to 0 to set the road reaction force characteristics. In this case, the above-mentioned formula (4) is replaced by the following formula (9), and the above-mentioned formula (6) is replaced by the following formula (10). In formula (10), the matrix representing G is different from that in formula (6).

[0127]

[0128]

[0129] In the above embodiment, the evaluation function J is calculated based on the vehicle information ψ related to the vehicle state x. ｅ , dψ ｅ / dt,y ｅ , dy ｅ / dt, θ ｐ (=θ ｃ -θ ＡＤ，ｃｍｄ ), dθ ｐ / dt, ∫y ｅ And, (Z ａ，１ -Z ａ，１，ｒｅｆ ), (Z ａ，２ -Z ａ，１，ｒｅｆ ) as variables. The evaluation function J may include two or more pieces of information arbitrarily selected from these pieces of information as variables. In addition, θ defined by symbols other than x and u in equation (4) ＡＤ，ｃｍｄ may be set to a constant value throughout the prediction interval, or may be arbitrarily set as time series data if predictable.

[0130] In the above-described embodiment, the angle control unit 47 (see FIG. 5) includes the feedforward control unit 63, but the feedforward control unit 63 may be omitted. In this case, the feedback control torque T ｆｂ is the basic target torque.

[0131] In the above-described embodiment, an example in which the present disclosure is applied to a column-type EPS has been shown, but the present disclosure can also be applied to EPSs other than column-type EPSs. The present disclosure can also be applied to steer-by-wire systems.

[0132] Although the embodiments of the present disclosure have been described in detail, these are merely specific examples used to clarify the technical content of the present disclosure, and the present disclosure should not be construed as being limited to these specific examples, and the scope of the present disclosure is limited only by the appended claims.

[0133] 1...electric power steering device, 3...steered wheels, 4...steering mechanism, 18...electric motor, 43...road surface reaction force characteristic setting section, 44...assist torque command value setting section, 45...manual steering command value generation section, 46...integrated angle command value calculation section, 47...angle control section, 48...first weight multiplication section, 49...second weight multiplication section, 50...addition section, 51...torque control section, 52...weight setting section, 71...vehicle behavior prediction section, 72...vehicle behavior determination section, 73...road surface reaction force characteristic coefficient determination section

Claims

1. A motor control device comprising: a manual steering command value generation unit that generates a manual steering command value based on an equation of motion obtained from a reference model of a steering device, the equation including a road surface reaction force characteristic coefficient; an integrated angle command value calculation unit that calculates an integrated angle command value by adding the manual steering command value to an automatic steering command value; a control unit that controls an electric motor for steering angle control based on the integrated angle command value; and a road surface reaction force characteristic setting unit that sets the road surface reaction force characteristic coefficient based on vehicle information, wherein the road surface reaction force characteristic setting unit includes: a vehicle behavior prediction unit that predicts multiple vehicle behaviors according to the vehicle information and the optimization coefficients using a vehicle behavior prediction model that includes candidates for the road surface reaction force characteristic coefficients as optimization coefficients; a vehicle behavior determination unit that determines an appropriate vehicle behavior from the multiple vehicle behaviors predicted by the vehicle behavior prediction unit; and a road surface reaction force characteristic coefficient determination unit that determines the optimization coefficient used for predicting the vehicle behavior determined by the vehicle behavior determination unit as the road surface reaction force characteristic coefficient.

2. The motor control device according to claim 1, wherein the vehicle behavior prediction model is composed of the reference model and a vehicle model.

3. The motor control device according to claim 2, wherein the vehicle behavior prediction unit predicts the plurality of vehicle behaviors after imposing constraints on the predicted vehicle behaviors.

4. A motor control device as described in any one of claims 1 to 3, wherein the vehicle behavior determination unit is configured to determine the appropriate vehicle behavior using an evaluation function that includes, as variables, multiple pieces of information selected from the vehicle information and the optimization coefficients used in predicting vehicle behavior by the vehicle behavior prediction unit.

5. A motor control device as described in claim 4, wherein the evaluation function is expressed as the sum of values ​​obtained by multiplying the squares of multiple pieces of information selected from the vehicle information and the optimization coefficients used in predicting vehicle behavior by the vehicle behavior prediction unit by weights.

Images