JP5525357B2 - Servo control device - Google Patents

Description

  The present invention relates to a servo control device in which an integral term is optimized, and more particularly to a servo control device that assists in steering a vehicle.

  In general, a servo control device that controls to follow a target value by using the position, orientation, orientation, etc. of an object as a controlled variable incorporates integral control in order to compensate for a deviation that constantly occurs between the target value and the target value. .

  For example, in a vehicle steering assist device that applies a steering force to a steering system so that the vehicle travels along a target position, when a disturbance such as a cross slope (cant) of a road surface occurs, the vehicle is in a target lateral position in the lane. The offset travel may be performed, and disturbance compensation is performed by applying an integral term of the offset amount to the control system.

  However, the integral control has a characteristic that the phase is delayed with respect to the disturbance change. For this reason, for example, when the road surface cant is switched, there is a possibility of being steered in the reverse direction. In order to avoid this, it is necessary to reset the integral value at an appropriate timing, and Patent Document 1 proposes a technique for resetting the integral value in accordance with the lateral position of the vehicle.

Japanese Patent No. 4200196

  However, the technique disclosed in Patent Document 1 is inappropriate because the integral value is reset by a threshold value with respect to the lateral position of the lane, so that an erroneous reset may occur or a reset delay may occur depending on the disturbance condition. There is a risk that the timing of the vehicle will be over, resulting in a situation where the amount of overshoot of the vehicle increases.

  The present invention has been made in view of the above circumstances, and provides a servo control device that can appropriately set an integral term in servo control according to the state of disturbance and compensate for the influence of the disturbance while reducing overshoot. The purpose is to do.

  In order to achieve the above object, a servo control device according to the present invention is a servo control device that repeatedly executes a predetermined control process for a controlled object and controls the state of the controlled object to follow a target value based on the execution result. A disturbance amount acquisition unit that acquires a disturbance amount applied to the control target, and an integration control unit that integrates a deviation between the state quantity of the control target and the target value, and the integration control unit includes the disturbance control unit. The integral amount is set to zero at the timing when the sign of the disturbance amount acquired by the amount acquisition unit changes.

  According to the present invention, an integral term in servo control can be appropriately set according to the state of disturbance, and the influence of disturbance can be compensated while reducing overshoot.

Configuration diagram of a servo control device applied to a vehicle steering assist device Explanatory drawing showing the driving lane in the vehicle coordinate system Steering control system block diagram Flow chart of integral control processing Explanatory drawing showing changes in vehicle behavior when disturbance change is small Explanatory drawing showing changes in vehicle behavior when disturbance change is large

Embodiments of the present invention will be described below with reference to the drawings.
In the present embodiment, an example will be described in which the servo control device is applied to a steering assist device for maintaining the lane of a vehicle. The steering assist device achieves a control target by calculating a target steering torque for keeping the traveling path of the vehicle following the target course and turning the steered wheels based on the target steering torque. In FIG. 1, the left and right front wheels FL and FR of the vehicle 1 are steered via an electric power steering device (EPS) 2.

  The EPS 2 steering system is connected to a steering wheel 3 steered by a driver via a steering shaft 4 and is connected to front wheels FL and FR via left and right tie rods 5 that can extend and contract in the axial direction. Has been. The tie rod 5 is disposed on the left and right as a ball screw type linear actuator driven by a motor 6 provided in the EPS 2.

  The EPS 2 motor 6 is driven and controlled by an EPS control unit (EPS_ECU) 20. The EPS control unit 20 supports the steering operation of the driver by adding the driving torque of the motor 6 to the steering torque of the driver. The EPS control unit 20 is connected to the control unit (ECU) 30 via, for example, an in-vehicle network via a CAN (Controller Area Network) bus and the like, and the lane maintenance is performed by a control command transmitted from the control unit 30. For this purpose, the steering support is shared and controlled.

  In addition, a plurality of control units that share vehicle control such as an engine control unit, a transmission control unit, and a brake control unit are connected to the in-vehicle network to which the EPS control unit 20 and the control unit 30 are connected. Each control unit transmits / receives data to / from each other and exchanges various types of information.

  In the present embodiment, a stereo camera 7 is provided as a sensor for recognizing the traveling environment of the host vehicle in steering assistance. The stereo camera 7 is formed by unitizing two cameras arranged with a predetermined baseline length and an image processing engine that processes a pair of stereo images captured by the two cameras. The surrounding situation in a predetermined area ahead is photographed through the front window. Three-dimensional information about the environment outside the vehicle is obtained from the image captured by the stereo camera 7, and the travel lane of the vehicle 1 is recognized based on the three-dimensional information.

  The control unit 30 receives signals from the steering angle sensor 8, the wheel speed sensor 9, the yaw rate sensor 10, the lateral acceleration sensor 11, and other various sensors (not shown). The steering angle sensor 8 is attached to the steering shaft 4 and detects the steering angle of the steering wheel 3. The detected steering angle is transmitted to the control unit 30 via the EPS control unit 20. The wheel speed sensor 9 is attached to each of the left and right front wheels FL and FR and the left and right rear wheels RL and RR of the vehicle 1 and generates a pulse signal having a cycle corresponding to the wheel speed of each wheel. Output to 30. The yaw rate sensor 10 and the lateral acceleration sensor 11 are configured as MEMS (Micro Electro Mechanical Systems) sensors, detect the rotational speed around the vertical axis of the vehicle body and the lateral acceleration, and output them to the control unit 30.

  The control unit 30 recognizes the travel environment recognition information based on the image captured by the stereo camera 7 and the vehicle state quantity detected based on signals from the steering angle sensor 8, the wheel speed sensor 9, the yaw rate sensor 10, the lateral acceleration sensor 11, and the like. Based on this, a target steering torque for lane keeping steering assist control is calculated. This lane keeping steering assist control realizes a target steering torque calculated mainly by the control unit 30 in the servo control system by the EPS 2, EPS control unit 20, and control unit 30.

  That is, the control unit 30 estimates the own vehicle traveling path based on the recognition information of the vehicle traveling environment and the vehicle state quantity, and calculates a target steering torque for causing the own vehicle traveling path to follow the target course. The calculated target steering torque is transmitted from the control unit 30 to the EPS control unit 20, and the EPS control unit 20 achieves the control target by drivingly controlling the motor 6 based on the target steering torque.

  Specifically, in the vehicle coordinate system of FIG. 2 in which the center of gravity of the host vehicle 1 is the origin, the vehicle width direction is the X axis, and the vehicle body front side is the Z axis, the forward gazing point L (for example, about 1 second later). Then, the center position Xc of the traveling lane at the gazing point L is obtained, and the lane (the lane indicated by the broken line in FIG. 2) having the center position Xc as a locus is set as the target course (target lane) that the host vehicle 1 should follow. Then, the curvature K of the target course based on the curve radius of the traveling lane is obtained.

  Next, a travel locus of the host vehicle (a vehicle locus indicated by a two-dot chain line in FIG. 2) is estimated based on the state quantity of the vehicle, and the vehicle yaw angle θyaw formed by the host vehicle with respect to the target course, The estimated lateral position of the host vehicle (the host vehicle position with respect to the lane center position) Xe is obtained. The estimated lateral position Xe of the vehicle at the gazing point L can be estimated using vehicle specifications such as steering angle, vehicle speed, wheel base and steering gear ratio, vehicle-specific stability factors, and the vehicle speed and yaw rate can also be estimated. It can also be estimated using.

Then, as shown in the following equation (1), the steering torque based on the curvature K of the lane for feedforward control with respect to the turning of the curve, and the deviation (Xc−Xe) between the lane center position Xc and the estimated lateral position Xe Steering torque for converging to the target value in the closed loop, steering torque for converging the yaw angle θyaw to the target value in the closed loop, and sideways of the vehicle in the integral control for removing the deviation due to steady deviation and disturbance generated in the closed loop system The target steering torque Td is calculated by adding the steering torque based on the position Xv.
Td = Gκ · K + Gl · (Xc−Xe) + Gy · θyaw + Gi · ＋ Ga · Xvdt (1)
Where Gκ: Feed forward gain
Gl: Feedback gain for the lateral position of the vehicle
Gy: Feedback gain for the yaw angle of the vehicle
Gi: Vehicle side position integral gain
Ga: Vehicle lateral position addition width gain

  In the present embodiment, the curve radius (curvature), the lateral position of the vehicle, and the yaw angle are expressed by positive values on the left side of the central axis in the front-rear direction of the vehicle 1. The right side is represented by a negative value.

  As shown in the block diagram of FIG. 3, the steering control system using the target steering torque as an operation output includes a lane recognition unit 41, a curvature control unit 42, a lateral position control unit 43, an attitude control unit 44, and an integration control unit. Each control unit includes 45 and a steering system 46 using EPS2. Each control part 42-45 calculates the steering torque shown to the 1st term-the 4th term of the right-hand side of (1) type, and each steering torque is added together and it outputs to the steering system 46 by EPS2. Then, the vehicle motion generated by the steering system 46 is fed back to the control units 43 to 45.

  The lane recognition unit 41 is formed as a function of the stereo camera 7 or the control unit 30, and the curvature control unit 42, the lateral position control unit 43, the attitude control unit 44, and the integration control unit 45 are formed as functions of the control unit 30. Has been.

  Specifically, the lane recognition unit 41 detects, for example, a white line on a road on which the host vehicle travels from an image captured by the stereo camera 7 by edge detection or the like, and recognizes a travel lane of the host vehicle formed by the white line. To do. And the curvature K of the recognized traveling lane is calculated | required, the target lane position Xc in the gazing point L, the yaw angle (theta) yaw which the own vehicle with respect to a traveling lane, and the horizontal position Xv are calculated and output. The curvature control unit 42 multiplies the curvature K input from the lane recognition unit 41 by a predetermined gain Gκ, and calculates a target value for the steering torque of the feedforward control for turning along the curve.

  The lateral position control unit 43 and the attitude control unit 44 are both feedback control units for the target value. In the present embodiment, the lateral position control target value is set to the center position (target lane) of the travel lane, the yaw angle Using the target value as an angle with respect to the tangent line of the driving lane, a deviation (Xc−Xe) between the lane center position Xc and the estimated lateral position Xe is multiplied by a predetermined feedback gain Gl to calculate a steering torque for lateral position control of the vehicle. The steering torque for attitude control around the vertical axis of the vehicle is calculated by multiplying the yaw angle θyaw by a predetermined gain Gy.

  The integration control unit 45 integrates a value obtained by multiplying the lateral position Xv of the vehicle by a predetermined addition width gain Ga, and further performs an integration operation of multiplying the integration amount by a predetermined integration gain Gi, and according to the integration operation Calculate the steering torque. By the steering torque by the integral calculation, a deviation remaining in the lateral position control of the vehicle can be removed, and a deviation generated according to a disturbance amount due to a cross wind or a road surface cant can be compensated.

  The disturbance amount is estimated as a lateral force acting on the vehicle by a measured value by a sensor, a disturbance observer using a vehicle model, or the like. For example, by inputting the yaw rate by the yaw rate sensor 10 and the steering angle measured by the steering angle sensor 8 to the vehicle model and comparing the yaw rate calculated by the model, an external force acting on the host vehicle 1 due to a road surface cant or the like is obtained. Estimated as lateral force disturbance. The disturbance amount may be calculated by directly measuring the gradient of the road surface cant by using, for example, an inclination sensor.

  In the present embodiment, the curve radius (curvature), the lateral position of the vehicle, and the yaw angle are expressed by positive values on the left side of the central axis in the front-rear direction of the vehicle 1. Since the right side is represented by a negative value, the disturbance amount is also represented by a positive value on the left side of the central axis in the front-rear direction of the vehicle 1 and a negative value on the right side of the central axis in the front-rear direction of the vehicle 1. To do.

  In response to such disturbance to the vehicle, the steering control system according to the present embodiment performs integration control according to the amount of disturbance so as to smoothly and quickly converge to the target steering torque. Specifically, the integration value is reset at an appropriate timing with respect to the disturbance change, an appropriate integration amount according to the magnitude of the disturbance, an integration range restriction, and the like are performed.

  That is, in general, since the phase of the integration control is delayed with respect to the disturbance change, there is a possibility of steering in the reverse direction when the road surface cant is switched.To avoid this, the integration value is set at an appropriate timing. Need to reset. In the conventional integral control, the integral value is reset with reference to the lane curvature, or the integral value is reset according to the lateral position of the vehicle. In the former case, since the lane curvature and the disturbance direction do not necessarily match, there is a possibility that an erroneous reset may occur, and a situation where a vehicle offset occurs again due to this erroneous reset is assumed. In the latter case, since the reset timing is determined by the threshold value of the lane lateral position, there is a possibility that the timing becomes inappropriate depending on the disturbance situation, and as a result, the amount of overshoot of the vehicle may increase.

  For this reason, in this embodiment, the integral value is reset according to the state of the disturbance, and the integral value is reset when the sign of the lateral force disturbance estimated value changes. As a result, the reset timing can be set to an optimum timing according to the disturbance situation, and erroneous reset can be reduced and overshoot can be suppressed.

  In general, in order to reduce the delay in integral control, it is sufficient to increase the gain of the integral term. However, if the gain is uniquely increased, overshoot occurs and controllability deteriorates. For this reason, in the present embodiment, the integration addition value is adjusted according to the magnitude of the estimated disturbance amount and the integration range is limited. As a result, it is possible to cope with the delay of the integration control while suppressing the overshoot and improve the followability to the change in disturbance, and it is possible to perform appropriate compensation according to the situation of the change in disturbance.

  Specifically, the above integral control is realized by software processing executed by the control unit 30. Hereinafter, the integral control of the steering control system in the present embodiment will be described with reference to the flowchart of FIG.

  The flowchart in FIG. 4 is an integration control process that is repeatedly executed in the control unit 30 at a predetermined time period. In this integration control process, first, in the first step S1, environmental information such as the curvature and width of the traveling lane in which the host vehicle is traveling, vehicle information such as the lateral position of the host vehicle with respect to the target lane, and the yaw angle are acquired. .

  Next, it progresses to step S2, calculates the integral value of the lateral position of the own vehicle, and estimates lateral force disturbance at step S3. In step S4, it is determined whether or not the sign of the estimated disturbance value has changed from the + side to the-side or from the-side to the + side. As a result, when the sign of the disturbance estimated value has not changed, the process proceeds from step S4 to step S5, and it is checked whether the disturbance estimated value is less than a predetermined threshold value.

  If the estimated disturbance value is equal to or greater than the threshold value, the integral addition width is changed according to the magnitude of the disturbance in step S6 to correct the integral value. In the present embodiment, the change in the integral addition width is to change the addition width gain Ga to be multiplied by the lateral position Xv of the vehicle to a gain Ga1 having a relatively large value. This gain Ga1 (Ga1 <Ga ) Is used to increase the integral addition width when a disturbance greater than or equal to the threshold is input. In step S7, the integration value is limited according to the magnitude of the disturbance. In the present embodiment, the integration value is limited by changing the integration range (integration time) Iw to a relatively wide range (long time) Iw1 so that the integration value does not become too small for a large disturbance input. To ensure responsiveness.

  On the other hand, if the estimated disturbance value is less than the threshold value in step S5, the integral addition width is changed and the integral value is limited corresponding to the case where the disturbance is small in steps S8 and S9. That is, in step S8, the addition width gain Ga multiplied by the lateral position Xv of the vehicle is changed to a relatively small gain Ga2, and correction is performed to reduce the integral addition width when the disturbance becomes smaller than the threshold value. . In step S9, the integral range (integration time) Iw is changed to a relatively narrow range (short time) Iw2, thereby limiting the integral value so as not to become too large for a small disturbance input, and overshooting. Avoid the occurrence of

  After correcting and limiting the integral value according to the disturbance in steps S6 and S7 or steps S8 and S9, the process proceeds to step S11, and the corrected and limited integral value is multiplied by the integral gain Gi. Then, in step S12, the integration control target value is output and the process is exited.

  Thereafter, when the direction of the disturbance changes and the estimated value of the disturbance changes in step S4, the process proceeds from step S4 to step S10. In step S10, the integral value is reset to 0 corresponding to the timing at which the disturbance estimated value changes sign, and the process exits through steps S11 and S12. In this case, the integration control target value is also reset to 0 by resetting the integration value.

  The change in the vehicle behavior due to the reset of the integral value will be described with reference to FIGS. FIG. 5 shows an example of a behavior change when the disturbance change is small, and FIG. 6 shows an example of a behavior change when the disturbance change is large. In FIG. 5, the amount of disturbance gradually starts changing from time T1, and time T2 (T2 > T1) shows an example in which the direction of the disturbance is switched, and FIG. 6 shows an example in which the disturbance change is changed stepwise. 5 and 6, the horizontal axis represents time, and the vertical axis represents the target steering torque integrated value, the lateral position in the lane of the vehicle, and the lateral force disturbance amount from the top.

  When the disturbance change is small, in the conventional method of resetting by the lateral position in the lane, as shown by the broken line in FIG. 5, when the reset threshold is small, the disturbance amount starts to decrease, and the timing at which the lateral position exceeds the threshold The integral value is reset at (point A in the figure). At this reset timing, since the reset is performed at an earlier time than the disturbance switching point (time T2), the vehicle is offset in the + direction again and is integrated again in the same direction. As a result, overshoot of the vehicle due to integral control occurs, and the wobbling behavior increases.

  Further, as shown by the one-dot chain line in FIG. 5, when the lateral position threshold is large, it is reset at a timing (point B) after the disturbance switching point (time T2), and the integral value is the target course. The situation occurs in the opposite direction. As a result, the amount of vehicle overshoot increases.

  On the other hand, in the integration reset according to the disturbance change in the present embodiment, as shown by the solid line in FIG. 5, the integration value is reset when the disturbance is switched (timing at time T2). As a result, the lateral position of the vehicle can be quickly converged to the target position while minimizing the amount of overshoot in the lateral position.

  On the other hand, when the disturbance change is large, as shown by the alternate long and short dash line in FIG. 6, in the conventional method of resetting according to the deviation of the horizontal position, the horizontal position is the threshold value even though the disturbance is changed. Overshoot is inevitable because there is a delay until it reaches. In this case, as indicated by a broken line in FIG. 6, when there is no integral reset, a large phase delay occurs, and as a result, the amount of overshoot of the vehicle increases.

  In contrast, in the integration reset according to the disturbance change in the present embodiment, as shown by the solid line in FIG. 6, even if the disturbance change is fast, the integral value can be reset quickly. As a result, it is possible to quickly converge the vehicle lateral position to the target position while minimizing the overshoot amount of the lateral position.

  As described above, in this embodiment, the integral value in servo control can be reset at an optimal timing according to the change in the amount of disturbance, and the integral control can be optimized according to the amount of the amount of disturbance. Thus, it is possible to effectively compensate for the influence of disturbance while reducing overshoot.

  In particular, when this servo control device is applied to a vehicle steering assist device, resetting the integral value when the direction of disturbance acting on the vehicle changes can reduce false resets and reset them at the optimal timing. The vehicle behavior can be stabilized by suppressing the overshoot.

  In addition, the integral addition value is adjusted according to the magnitude of the disturbance and the integration range is limited. Therefore, the integral control gain is unambiguously increased to suppress overshoot without degrading controllability. It is possible to cope with the delay of the integration control and improve the followability to the disturbance change, and it is possible to perform appropriate compensation according to the situation of the disturbance change.

1 Vehicle 20 EPS Control Unit 30 Control Unit 30
41 Lane recognition unit 42 Curvature control unit 43 Lateral position control unit 44 Posture control unit 45 Integration control unit 46 Steering system Ga addition width gain Td Target steering torque

Claims (8)

A servo control device that repeatedly executes a predetermined control process for a controlled object and controls the state of the controlled object to follow a target value based on the execution result,
A disturbance amount acquisition unit for acquiring a disturbance amount applied to the control target;
An integral control unit that integrates a deviation between the state quantity of the control target and the target value;
The integral control unit
A servo control device characterized in that the integral amount is set to zero at a timing when the sign of the disturbance amount acquired by the disturbance amount acquisition unit changes. The servo control device is a steering assist device that applies a steering torque to the steering system so that the vehicle travels along the target position.
The disturbance amount acquisition unit acquires a disturbance amount in a lateral direction with respect to the vehicle such as a crossing gradient and wind of the traveling road,
2. The servo control device according to claim 1, wherein the integration control unit integrates a deviation between the target position and a lateral position of the vehicle, and sets the integration amount to zero at a timing when the sign of the disturbance amount changes. . The integral control unit
More disturbance quantity obtained by the disturbance acquiring unit is large, the servo control device according to claim 1, wherein the widening the range of integration of the product fraction calculation. The integral control unit
More disturbance quantity obtained by the disturbance acquiring unit is small, the servo control device according to claim 1, wherein narrowing the integration range of the integration operation. The integral control unit
More disturbance quantity obtained by the disturbance acquiring unit is large, the servo control device according to claim 1, wherein increasing the added width of the product fraction calculation. The integral control unit
More disturbance quantity obtained by the disturbance acquiring unit is small, the servo control device according to claim 1, wherein reducing the added width of the product fraction calculation.   3. The servo control device according to claim 2, wherein the disturbance amount acquisition unit estimates the amount of disturbance based on a comparison between a yaw rate sensor value and a model-calculated yaw rate estimated value obtained by inputting a steering angle.   3. The servo control apparatus according to claim 2, wherein the disturbance amount acquisition unit estimates the amount of disturbance using a disturbance observer using information on a traveling path and a vehicle model.

Images