A power steering system

Abstract

The invention relates to a power steering system (1) comprising a steering wheel (2) and a return function (REC) which automatically returns the steering wheel (2) to a given reference position (P0), for example a central position, for which it comprises a function (FK1) for calculating a return speed setpoint which calculates a steering wheel speed setpoint (V_sw_set) as a function of the difference between the effective instantaneous position of the steering wheel (P_sw) and the reference position (P0), i.e. the steering wheel position error (ΔP), and a function (FK2) for calculating a return assistance which defines a return assistance setpoint (T_rec) as a function of the difference between the effective speed of the steering wheel (V_sw) and the steering wheel speed setpoint (V_sw_set), i.e. the steering wheel speed error (ΔV), the system (1) comprising a dynamic saturation function (RDS) which defines at least one saturation threshold (SAT_high, SAT_low) which is set as a function of the value of the steering wheel speed setpoint (V_sw_set) and which is then applied to dampen the peaks calculated by the return assistance setpoint (T_rec) so as to limit the return assistance setpoint in an authorized domain (D2_T).

Description

Technical Field

[0001] The present invention relates to a power steering system, which includes a return function to the center of the steering wheel. Background Technology

[0002] After the driver turns the steering wheel and then typically releases it to turn, the return function uses an auxiliary motor in a manner known per se to return the steering wheel to a neutral center position, which corresponds to a straight trajectory.

[0003] Thus, it is known that a return function is used, which performs servo control on the rotation speed of the steering wheel.

[0004] This return function determines the steering wheel speed setpoint based on the deviation between the actual angular position of the steering wheel and its center position. This setpoint is designed to return the steering wheel to the center position, thereby adjusting the setpoint applied to the auxiliary motor to achieve the desired angular velocity of the steering wheel.

[0005] However, the inventors discovered that the implementation of this return function may have some drawbacks.

[0006] In practice, during certain maneuvers, the driver may begin to steer sharply, that is, begin to turn the steering wheel significantly in the first direction and then quickly reverse the direction, that is, quickly bring the steering wheel to the center position. This can happen, for example, when changing direction at an intersection (a situation known as a "street corner") or making a sharp turn (hairpin). When the driver brings the steering wheel to the center position, the driver may manually give the steering wheel an actual rotational speed that is greater than the steering wheel speed setpoint defined by the return function.

[0007] By braking the steering wheel to counteract manual operation, the return function tends to suppress return.

[0008] This repositioning inhibition behavior is the source of the driver's artificial driving sensation, which is inconsistent with the feeling the driver expects in this situation.

[0009] Similarly, when a driver turns the steering wheel in a direction that moves it away from its center position, that is, when the driver increases the steering angle of the steering system, the known return function sometimes produces viscous behavior that is opposite to the rotation of the steering wheel. And because the driver is giving the steering wheel a high rotational speed, this viscous behavior is more noticeable and therefore uncomfortable. Summary of the Invention

[0010] Therefore, the purpose of this invention is to overcome the above-mentioned disadvantages and to propose a new power steering system that ensures the steering wheel effectively returns to the center position without causing viscous interference during steering wheel operation.

[0011] The purpose of this invention is achieved through a power steering system comprising a steering wheel and a return function designed to automatically return the steering wheel to a reference position, such as the center position, when the steering wheel is in a position different from a given reference position. The return function includes a return speed setpoint calculation function that calculates a steering wheel speed setpoint based on a deviation between the instantaneous actual position of the steering wheel and the reference position, termed "steering wheel position error," followed by a return auxiliary calculation function that calculates the return speed setpoint based on the actual speed of the steering wheel and the direction... A setpoint called a "return assist setpoint" is defined by the deviation between the steering wheel speed setpoints, referred to as "steering wheel speed error," and is used to control the assist motor to converge the actual speed of the steering wheel to the steering wheel speed setpoint. The system is characterized by including a dynamic saturation function that defines at least one saturation threshold, which is adjusted based on the value of the steering wheel speed setpoint. The dynamic saturation function is then used to mitigate calculations performed by the return assist calculation function to include the return assist setpoint in a domain called an "authorized return assist domain," the range of which is related to the value of the steering wheel speed setpoint.

[0012] Advantageously, the dynamic saturation function according to the invention allows for the definition of a saturation threshold that changes in real time to adapt to the vehicle's maneuvering conditions, depending on whether the steering wheel is in a left-hand or right-hand position relative to its reference position, whether the driver steers the steering wheel away from their driving position or conversely steers it closer to its reference position, and finally depends on the speed at which the driver shifts the steering wheel to perform their maneuvering, particularly the rotational angular velocity.

[0013] In this way, the dynamically adjusted saturation threshold allows for the elimination of some components known as "bad components" at any time. These components may exist in the steering wheel speed error signal, at the input of the return assist calculation function, or in the return assist setpoint signal, at the output of the return assist calculation function, and when represented in the return assist setpoint, they correspond to artificial viscous components that distort the driver's feel and reduce driving comfort.

[0014] Advantageously, the dynamic saturation function according to the invention allows for the removal of these undesirable components, and only these undesirable components, so that in the return auxiliary setpoint finally applied to the auxiliary motor, only the "pure" return component adapted to the ongoing operation is retained.

[0015] By providing dynamic saturation, which allows for differentiation of different steering wheel handling situations at any time and quantifies in each situation the portion of the return assist setpoint corresponding to the desired "pure" return component and the portion corresponding to the undesirable viscous component caused by the return function, the dynamic saturation function can thus be directly calculated or extracted from the return assist setpoint calculated by the return assist calculation function, which retains only the "pure" useful return component. This invention refines the operation of the return function, thereby improving driving comfort because it provides the driver with a more faithful return feel to the handling situation, and is therefore more intuitive, regardless of the handling situation. Attached Figure Description

[0016] Other objects, features, and advantages of the invention will become more apparent when reading the following description and using the accompanying drawings, which are provided purely for illustrative and non-limiting purposes, wherein:

[0017] Figure 1 The block diagram illustrates a first example of a power steering system equipped with a return function and a dynamic saturation function according to the invention, wherein the dynamic saturation function operates at the output of the return assist calculation function to dampen the return assist setpoint previously calculated by the return assist calculation function.

[0018] Figure 2 The block diagram illustrates the dynamic calculation of the high and low saturation thresholds according to the present invention, applicable to... Figure 1 The first example.

[0019] Figure 3 Another variation of the dynamic calculation of high and low saturation thresholds according to the invention is shown, applicable to Figure 1 The first example.

[0020] Figure 4 , 5 Figures 6 and 7 illustrate the operating principle of the invention in a four-quadrant diagram, which represents the position of the steering wheel on the horizontal axis and the actual speed of the steering wheel on the vertical axis. The operating principle is based on the authorized domain operation saturation relative to the curves between zero and the steering wheel speed setpoint in the steering, slow counter-steering, and fast counter-steering situations.

[0021] Figure 7 The block diagram illustrates a second example of a power steering system according to the invention, equipped with a return function and a dynamic saturation function, wherein the dynamic saturation function acts at the input of the return assist calculation function to mitigate steering wheel speed errors, which are then taken into account to calculate the return assist setpoint. Detailed Implementation

[0022] The present invention relates to a power steering system 1 intended for equipping vehicles.

[0023] In a manner known per se, this power steering system 1 includes a steering wheel 2.

[0024] The steering wheel 2 is arranged such that the driver can change the position P_sw of the steering wheel 2 to manipulate the power steering system 1 in order to limit the direction of travel of the vehicle.

[0025] Of course, the steering wheel 2 can take any suitable form, such as the conventional form of a rotatable ring, or even the conventional form of a tiltable joystick.

[0026] It should be noted that, preferably, the steering wheel 2 is designed to be actuated by rotation by the driver, such that the angular position and angular velocity of the steering wheel 2 are preferably referenced.

[0027] However, the present invention is fully applicable to steering wheels 2 composed of linear control components that can translate rather than rotate. In this case, the linear position and linear speed of the steering wheel will be taken into account, and necessary modifications can be made without departing from the scope of the present invention.

[0028] The power steering system 1 also includes a steering mechanism, preferably a rack and pinion steering mechanism, in a manner known per se, the displacement of which is controlled by the steering wheel 2 and allows for changes in the steering wheel of the vehicle, or preferably the steering angle of several steering wheels, i.e., the yaw direction.

[0029] The power steering system 1 also includes a return function denoted as REC for "return control", which is designed to automatically return the steering wheel 2 to the reference position P0, such as the center position P0, when the steering wheel 2 is in a position P_sw that is different from the given reference position (i.e., when P_sw ≠ P0).

[0030] Preferably, the reference position P0 of the steering wheel will correspond to the center position of the steering mechanism (and therefore the power steering system 1), which corresponds to the trajectory of the vehicle on a straight line.

[0031] In this case, the reference position P0 will conventionally be associated with the zero-angle position of the steering wheel 2, which defines the origin of the coordinate system. That is, P0 = 0, and the steering wheel moves to the left or right relative to the origin.

[0032] However, as a variation, a reference position P0 corresponding to a non-zero steering angle can be defined, that is, corresponding to the turning position of steering wheel 2. This is especially likely if the reference position P0 is adjusted by a third-party autonomous driving function, such as an automatic lane-keeping function (called "lane keeping"), when the autonomous driving function causes the vehicle to follow a curved trajectory corresponding to a non-zero reference position P0, rather than a straight trajectory corresponding to a zero reference position P0.

[0033] Furthermore, it should be noted that information representing the position of the steering wheel P_sw can be provided in any suitable manner, such as by a position sensor placed on the steering column carrying the steering wheel 2, or by a position sensor placed on any other part of the steering mechanism, the position of which represents the position P_sw of the steering wheel 2, especially if the mechanism is connected to the steering wheel 2 via a motion transmission element.

[0034] Therefore, in particular, for example, the position of the steering wheel P-sw can be evaluated based on the position information of the motor shaft of the auxiliary motor 3 that drives the rack and steering wheel 2.

[0035] If necessary, any suitable algorithm can be used to estimate the angular position of the steering wheel P_sw from data obtained from the power steering system 1 or data accessible to the power steering system 1.

[0036] The return function REC advantageously performs servo control of the steering wheel speed V_sw, which allows the steering wheel 2 to return to its reference position P0 by making the steering wheel 2 follow a defined speed curve defined by the steering wheel speed setpoint V_sw_set, which depends in particular on the position P_sw of the steering wheel at the moment of consideration.

[0037] Therefore, as Figure 1 or Figure 7 As shown, the return function REC includes a function for calculating the return speed setpoint, denoted as FK1. This function first calculates the steering wheel speed setpoint V_sw_set from the deviation between the instantaneous actual position P_sw of the steering wheel and the reference position (P0) (called "steering wheel position error" ΔP).

[0038] Therefore:

[0039] ΔP = P_sw - P0;

[0040] V_sw_set=FK1(ΔP)

[0041] More specifically,

[0042] V_sw_set=FK1(ΔP)=K1(abs(ΔP))*[sign(ΔP)*(-1)]

[0043] in:

[0044] "abs" represents the "absolute value" of a mathematical function.

[0045] "Sign" represents the mathematical function "sign". It returns +1 if the expression being tested is positive and -1 if the expression being tested is negative.

[0046] And K1 here represents the first law (or "function") of the predetermined conversion, which allows the return speed setpoint calculation function FK1 to associate the appropriate steering wheel speed setpoint V_sw_set with each steering wheel position error (absolute) value ΔP.

[0047] The gain applied by the first conversion law K1 is preferably adjusted according to the vehicle's longitudinal speed V_vehic, and more preferably according to a decreasing function of the vehicle speed V_vehic. Therefore, when the vehicle accelerates, the absolute value of the gain applied by the first conversion law K1 decreases.

[0048] The gain applied by the first conversion law K1 preferably varies according to the position of the steering wheel P_sw, and more specifically depends on the value of the steering wheel position error ΔP.

[0049] More preferably, according to the increasing function of the position error ΔP (from the position P_sw respectively), the absolute value of the gain applied by the first conversion law K1 will increase with the position error ΔP (from the position of the steering wheel P_sw respectively), that is, as the steering wheel 2 moves away from its reference position P0, the absolute value of the gain applied by the first conversion law K1 will increase.

[0050] The first transformation law K1 can take the form of a map, such as a three-dimensional map, which is preferably a map representing the gain that can be applied based on the vehicle speed V_vehic and the steering wheel position P_sw.

[0051] The map can be stored in the non-volatile memory of the power steering system 1.

[0052] In fact, if P0 = 0, then ΔP = P_sw.

[0053] For ease of description, the position of the steering wheel P_sw can be consistent with the steering wheel position error ΔP in the following text.

[0054] Furthermore, according to notation convention, the position of the steering wheel P_sw is considered positive when it rotates to the right relative to the reference position P0, and negative when it rotates to the left relative to the reference position P0. Of course, the opposite convention can be used without departing from the scope of this invention.

[0055] Similarly, it can be assumed that when the steering wheel moves from left to right (clockwise), the speed of the steering wheel, V_sw, is positive, while when the steering wheel moves from right to left (counterclockwise), the speed is V-sw.

[0056] Of course, in order to ensure that the steering wheel returns to the reference position P0, the steering wheel speed setting value V_sw_set should be opposite to the sign of the steering wheel position P_sw (especially the sign of the steering wheel position error ΔP).

[0057] This explains why in Figure 1 and 7 In the charts and formulas above, when calculating the steering wheel speed setpoint V_sw_set, the absolute value of the steering wheel position error ΔP and a sign correction factor of type [sign(ΔP)*(-1)] are strictly used by considering algebraic values ​​(with signs) of different magnitudes:

[0058] ΔP=P_sw-P0

[0059] V_sw_set=FK1(ΔP)=K1(abs(ΔP))*[sign(ΔP)*(-1)]

[0060] Of course, any other formula or model that is effective can be used without departing from the scope of the invention.

[0061] The return function (REC) also includes a return auxiliary calculation function FK2, which then defines a setpoint called the "return auxiliary setpoint" T_rec based on the deviation between the actual speed of the steering wheel V_sw and the steering wheel speed setpoint V_sw_set (called the "steering wheel speed error" ΔV). This setpoint is designed to control the auxiliary motor 3 to converge the actual speed of the steering wheel V_sw to the steering wheel speed setpoint V_sw_set.

[0062] ΔV = V_sw_set – V_sw

[0063] T_rec=FK2(ΔV)

[0064] More specifically, T_rec = FK2(ΔV) = K2(abs(ΔV)) * symbol(ΔV).

[0065] Here, K2 represents the second law (or "function") of the predetermined conversion, which allows the return assist calculation function FK2 to associate the appropriate return assist setpoint value T_rec with each steering wheel speed error (absolute) value ΔV.

[0066] Preferably, the return assist setpoint T_rec is uniform under force, especially torque, and the auxiliary motor 3 should apply the force to change the actual speed V_sw of the steering wheel and make the actual directional speed V_sw closer to the directional speed setpoint V_sw_set.

[0067] Therefore, the return auxiliary setpoint T_rec more preferably represents the power supply current that is desired to be applied to the auxiliary motor 3 so that the latter can transmit the desired return force (or torque).

[0068] The gain applied by the second conversion law K2 is preferably adjusted according to the longitudinal speed V_vehic of the vehicle.

[0069] The gain applied by the second conversion law Iaw K2 preferably varies depending on the value of the steering wheel speed error ΔV.

[0070] More preferably, the absolute value of the gain applied by the second conversion law K2 will increase with the increase of the steering wheel speed error ΔV, as an increasing function of the speed error ΔV. That is, the farther the actual steering wheel speed V_sw is from the steering wheel speed setpoint V_sw_set, the greater the absolute value of the gain applied by the second conversion law K2, thereby promoting rapid convergence.

[0071] The second transformation law K2 can take the form of a (second) map, such as a three-dimensional map, which is preferably a map representing the gain that can be applied according to the vehicle speed V_vehic and the steering wheel speed error ΔV.

[0072] The map can be stored in the non-volatile memory of the power steering system 1.

[0073] The auxiliary motor 3 is understood herein as motor 3, preferably a rotary motor, and more preferably an electric motor, such as a brushless motor, which can act on the steering wheel 2 via the steering mechanism when necessary to change the displacement speed V_sw 2 of the steering wheel 2.

[0074] If the power steering system 1 includes a steering mechanism mechanically connected to the steering wheel 2, for example when the steering rack is driven by a pinion fixed to the steering column driven by the steering wheel, then the motor 3 may advantageously correspond to an auxiliary motor 3 for assisting in manipulating the steering mechanism.

[0075] However, if the power steering system is a system called "steer-by-wire" system, in which there is no mechanical link for motion transmission between the steering wheel 2 on one side and the steering mechanism (rack, tie rod, wheel) on the other side, then preferably there is a master motor for driving the steering mechanism (thereby changing the yaw direction of the wheel). This master motor will be activated to move the wheel at a speed V_sw_set corresponding to the target steering wheel speed setpoint. An auxiliary "feedback" motor specifically assigned to the steering wheel will restore the position and behavior of the steering mechanism on the steering wheel, as well as the feel provided by the road surface, and will act as a "follow-up" action on the steering wheel 2 as the steering mechanism and the master motor, and thus tend to impart a speed V_sw to the steering wheel 2, which will converge to the applicable steering wheel speed setpoint V_sw_set.

[0076] To better understand this invention, we will... Figure 4 , Figure 5 and Figure 6 The four quadrant diagrams shown illustrate different manipulation scenarios.

[0077] These graphs show the instantaneous position (or "steering wheel angle") of the steering wheel under consideration, P_sw, on the horizontal axis, or more specifically, the steering wheel position error ΔP, that is, the deviation between the actual steering wheel position P_sw and the reference position P0.

[0078] On the vertical axis, they represent the speed of the steering wheel, V_sw.

[0079] The origin of the coordinate system corresponds to the coordinate point (P0, 0), that is, the horizontal coordinate corresponds to the coordinate point of the reference position P0 (preferably zero), and the vertical coordinate corresponds to the coordinate point of zero steering wheel speed (V_sw = 0).

[0080] The curve L1 corresponding to the steering wheel speed setpoint V_sw_set is also represented in this coordinate system because it is generated by the application of the first transformation law K1, and more generally by the return speed setpoint calculation function FK1.

[0081] It should be noted that the curve L1 representing the return speed setpoint calculation function FK1 preferably has an S-shaped curvature, with a steep slope starting from the reference position P0, and then gradually decreasing in slope as the steering wheel 2 moves away from the reference position P0, i.e. as the position error ΔP increases (absolute value), to converge to the asymptote, preferably strictly monotonic.

[0082] However, depending on the vehicle and power steering system 1 construction, the shape of curve L1 (and therefore the corresponding function FK1) will remain adjustable.

[0083] It should also be noted that the speed setpoint curve L1 (more typically a four-quadrant diagram) is centered on and symmetrical about the origin (P0, 0) of the coordinate system, thereby managing in a similar manner the case where the steering wheel 2 rotates to the right relative to the reference position P0 (the half-plane corresponds to the positive abscissa) and the case where the steering wheel rotates to the left relative to the reference position P0 (the half-plane corresponds to the negative abscissa).

[0084] For the sake of simplicity, the following explanation will refer to the case where the steering wheel 2 is turned to the right, so the steering wheel position P_sw and the steering wheel position error ΔP are positive.

[0085] Of course, the operation of this invention when the steering wheel is turned to the left is deduced by analogy.

[0086] It should also be noted that, advantageously, as long as the return assist setpoint T_rec is derived from the second transformation law K2, or more generally from the return assist calculation function FK2, the representation of the steering wheel speed error ΔV on the vertical axis of the four quadrant diagrams is related to the representation that can be composed of the corresponding components of the return assist setpoint T_rec.

[0087] Furthermore, provided that the return-to-position auxiliary calculation function FK2 is preferably continuous and (strictly) monotonic, and therefore bijective, the function FK2 preferably has a form (S-shape) quite similar to that of the curve L1 on the graph, and will define a domain with the same properties as the curve L1 within the four-quadrant graph.

[0088] Therefore, for ease of description, reasoning can be performed on the same four-quadrant diagram and in the steering wheel speed error component (which is uniform in the torque component) at the return assist setpoint T_rec, which is the image of the steering wheel speed error obtained by the function FK2.

[0089] To simplify the representation, Figures 4 to 6 The curve in the graph represents the steering wheel speed V_sw on the vertical axis, thus allowing the steering wheel speed error ΔV to be shown, which is used as the input to the return-to-position auxiliary calculation function FK2.

[0090] Of course, the same reasoning can be applied with necessary modifications by considering the return auxiliary setpoint T_rec, which is output from the function FK2 and thus represents the considered steering wheel speed error ΔV, and the components of the return auxiliary setpoint T_rec on the vertical axis.

[0091] Figure 4 The diagram illustrates a steering situation where the driver moves the steering wheel 2 away from the reference position P0 to increase the steering angle of the electric power steering system 1; in this case, the steering wheel is moving to the right.

[0092] This situation is identified by the first operating point OP1 located in the northeast quadrant.

[0093] The northeast quadrant corresponds to the first domain D1, called the "steering domain" D1, where the steering wheel position error ΔP is positive and the steering wheel speed V_sw given by the driver is also positive.

[0094] Therefore, the amplitude of the absolute value of the steering wheel speed error ΔV from which the return assist setpoint T_rec is generated is greater than the amplitude of the steering wheel speed setpoint V_sw_set. Thus, taking into account the steering wheel position P_sw (and the steering wheel position error caused therefrom), the steering wheel speed setpoint V_sw_set is defined by the function FK1.

[0095] Graphically, the steering wheel speed setpoint V_sw_set corresponds to the distance on the vertical axis that separates curve L1 from the horizontal axis.

[0096] Therefore, we can conclude that:

[0097] |ΔV|>|V_sw_set|

[0098] To facilitate understanding of the principles of this invention, it can be considered that, in this turning situation, the steering wheel speed error corresponds to the accumulation of a component called the "viscous component" V_visc, which covers a portion of the speed error located in the steering domain (northeast quadrant), and the accumulation of a component called the "pure return component" V_pure, which extends to curve L1 in the southeast quadrant, i.e., to the steering wheel speed setpoint value V_sw_set (negative in this case), which applies to the position of the steering wheel P_sw, or more precisely, to the steering wheel position error ΔP over the time considered.

[0099] It should be noted that in the composition of the steering wheel speed error ΔV as defined in this way, the pure return component V_pure actually corresponds to the steering wheel speed setpoint V_sw_set calculated by the return speed setpoint calculation function FK1: V_pure = V_sw_set, while the viscous component V_visc corresponds to the "excess" relative to the steering wheel speed setpoint V_sw_set, and has the opposite sign to the steering wheel speed setpoint V_sw_set.

[0100] Therefore, the absolute value of the return assist setpoint T_rec obtained by applying the steering assist calculation function FK2 to the steering wheel speed error ΔV is greater than the absolute value of the return assist setpoint corresponding to a "pure" return action:

[0101] T_pure=FK2(ΔV=V_sw_set),

[0102] This can be obtained by applying the return assist calculation function FK2 to the steering wheel speed error, which is equal to the maximum value provided by the return speed setpoint calculation function FK1, i.e., equal to the steering wheel speed setpoint V_sw_set.

[0103] In fact:

[0104] If we set |ΔV|=|V_pure+V_visc|,

[0105] Where V_pure = V_sw_set, the pure homogeneous component, and V_visc, the excess viscous component energy relative to V_sw_set.

[0106] So ΔV|=|V_sw_set+V_visc|>|ΔV_sw_set|

[0107] Therefore, FK2(|ΔV|)=FK2(|V_sw_set+V_visc|)>FK2(|V_sw_set|)

[0108] That is, T_rec=FK2(|ΔV|)>FK2(|ΔV_sw_set|)

[0109] Therefore, there exists an excess value X such that:

[0110] T_rec=FK2(|ΔV_sw_set|)+X

[0111] In other words:

[0112] T_rec=T_pure+X

[0113] It is known that FK2(|ΔV_sw_set|) represents the maximum return force (maximum return torque) relative to the steering wheel position P_sw. This maximum return force corresponds to the pure return action denoted as T_pure. This means that the excess X corresponds to the undesirable viscous component generated in the return assist setpoint (in the application of function FK2) due to the presence of an excess velocity viscous component V_visc in the steering wheel speed error ΔV (at the input of the function FK2).

[0114] Therefore, the above expression can be represented in the following form:

[0115] T_rec = T_pure + T_visc

[0116] Where T_visc represents the undesirable viscosity component, which is caused by the presence of an excessive velocity component V_visc in the steering wheel speed error ΔV.

[0117] In other words, due to the presence of a velocity viscous component V_visc, which adds to the steering wheel speed error ΔV at the moment under consideration, the "original" application of the return assist calculation function FK2 (without the saturation purpose of this invention) will result in defining a return assist setpoint T_rec that is too high. In addition to the pure return component T_pure, which is useful and normally represents the return action intuitively expected by the driver, it also contains an undesirable viscous component T_visc, which tends to be opposite to the driver's steering wheel operation, thus giving a heavy feeling.

[0118] For ease of description, this will be... Figure 4 The same chart is used to describe the steering situation on the one hand, relative to the steering wheel speed V_sw and the steering wheel speed setpoint V_sw_set, and on the other hand, relative to the return assist setpoint (return torque setpoint) T_rec.

[0119] Therefore, in Figure 4 For more specific information regarding steering wheel speed V_sw in the aforementioned chart, please refer to:

[0120] The actual speed of the steering wheel, V_sw;

[0121] The steering wheel speed setpoint V_sw_set is defined by the speed setpoint calculation function FK1;

[0122] Turning domain D1 corresponds to the northeast quadrant for right turns and the southwest quadrant for left turns;

[0123] The second domain, D2, called the "pure return domain," is contained between the horizontal axis (V_sw = 0) and the curve L1, which represents the steering wheel speed setpoint V_sw_set defined by the function FK1.

[0124] Steering wheel speed error ΔV = V_sw_set – V_sw,

[0125] Where ΔV = V_pure + V_visc, V_pure = V_sw_set corresponds to a portion (component) of ΔV contained in the pure return domain D2, while V_visc corresponds to the excess portion (component) ΔV contained in the turning domain D1.

[0126] Similarly, regarding auxiliary setpoints, please refer to:

[0127] T_rec = FK2(ΔV), the return auxiliary setpoint generated by the return auxiliary calculation function FK2 (here in row mode, without saturation);

[0128] Thus, T_rec = T_pure + T_vis.

[0129] Where T_pure=FK2(V_sw_set) corresponds to the part (component) of T_rec that represents the pure return action, and is graphically contained in the pure return domain D2_T, while T_visc corresponds to the excess part (component) of T_rec, is contained in the steering domain (viscous domain) D1_T, and represents the undesirable additional action, that is, braking the driver's steering wheel operation.

[0130] D1_T is the first viscous turning region, which corresponds to the northeast (turning right) and southwest (turning left) quadrants, and contains an excess return torque opposite to the turning direction;

[0131] D2_T is the second pure eurygesis domain, which extends between the horizontal axis and the image curve of curve L1 through the eurygesis auxiliary calculation function FK2, that is, preferably between zero and FK2 (V_sw_set).

[0132] From the graphic perspective, in the above reference... Figure 4 In the detailed steering scenario, the absolute value of the steering wheel speed error ΔV exceeds the value of the return speed setpoint V_sw_set. Therefore, the resulting return auxiliary setpoint T_rec exceeds the pure return domain D2_T.

[0133] However, if the pure return component Tpure (which is velocity V_pure) corresponds to the intuitive feeling of the steering wheel 2 during the steering process, the viscous component T_visc (which is velocity V_visc) will produce an artificial heaviness in the steering wheel 2. This heaviness is more intense when the steering speed V_sw provided by the driver to the steering wheel is high (absolute value) at the same given position P_sw of the steering wheel (and therefore at the same position error ΔP).

[0134] Therefore, the purpose of the saturation proposed in this invention is to eliminate such undesirable viscous components T_visc, V_visc, so that the return of the steering wheel 2 in the case of steering is limited to a single pure return component T_pure, V_pure, which corresponds to a given steering wheel position P_sw.

[0135] Figure 5 The diagram illustrates a slow reverse steering situation, in which the driver manipulates the steering wheel 2 to bring it closer to its reference position P0 (from right to left) to reduce the steering angle of the power steering system 1 relative to the reference position P0. In this case, the steering wheel speed V_sw given by the driver is less in absolute value than the steering wheel speed setpoint V_sw_set and has the same sign as the steering wheel speed setpoint V_sw_set.

[0136] In practice, this slow reverse steering situation can particularly involve the following scenario: after the driver turns the steering wheel 2 to steer, he releases the steering wheel 2 and allows it to return freely to the reference position P0.

[0137] Figure 5 This slow reverse turn corresponds to the second operating point OP2, which is located in the pure return domain D2, in the southeast quadrant.

[0138] This time, the steering wheel speed error ΔV and the resulting return auxiliary setpoint T_rec=FK2(ΔV) are completely contained in the pure return region D2 or D2_T, so only the pure return component is included.

[0139] The function of this pure return component is to accelerate the return of the steering wheel. Therefore, taking into account the position of the steering wheel 2, and thus the distance ΔP that separates the steering wheel 2 from the reference position P0, the return can be made to have a natural consistency at the moment of consideration.

[0140] In this case, corresponding to the normal operation of the return function REC, the return auxiliary setpoint T_rec does not contain any bad components.

[0141] Therefore, it is not necessary to change the return assist setpoint T_rec, nor is it necessary to change the steering wheel speed error ΔV, which constitutes the return assist setpoint T_rec, through the return assist calculation function FK2, in order to obtain an intuitive and situation-appropriate return feel.

[0142] Therefore, in practice, the saturation characteristic of this invention will not actively alter the calculation performed by the return-to-position auxiliary calculation function FK2 under these circumstances.

[0143] Figure 6 A rapid reverse steering scenario is illustrated, in which the driver manipulates the steering wheel 2 to bring it closer to its reference position P0 (from right to left here) to reduce the steering angle of the power steering system 1 relative to the reference position P0. However, this time the absolute value of the steering wheel speed V_sw is higher than the steering wheel speed setpoint V_sw_set, and has the same sign as the steering wheel speed setpoint V_sw_set.

[0144] This rapid reverse steering corresponds to the third operating point, denoted as OP3, which is located in the third domain, here in the southeast quadrant, called the "damped domain" D3. This damped domain extends beyond the speed setpoint curve L1 relative to the horizontal axis and is opposite to the pure return domain D2 relative to curve L1.

[0145] In this case, the steering wheel speed error ΔV lies entirely within the damping domain D3 and corresponds to the damping component V_damp.

[0146] Therefore, the resulting return assist setpoint T_rec = FK2(ΔV) lies entirely within a damping domain D3_T, which extends beyond the image of curve L1 in the southeast quadrant via the return assist calculation function FK2. This corresponds to the positive damping component T_damp, which tends to decrease the steering wheel speed V_sw in absolute value, thus aligning it with the steering wheel speed setpoint V_sw_set.

[0147] Therefore, the damping component T_damp generates a resisting torque that resists the driver's desired return motion to brake the steering wheel 2.

[0148] Therefore, the object of the present invention is to remove this undesirable damping component T_damp again, or, in an equivalent manner, to remove the speed component V_damp, the root cause of which is the presence of this damping component T_damp in the return assist torque setpoint T_rec, thereby allowing the driver to naturally return the steering wheel to its desired high speed P_sw.

[0149] In fact, Figure 6 In this rapid reverse steering scenario shown, it can be assumed that since the driver returns the steering wheel 2 to its reference position P0 faster than the return function REC automatically does, no return assistance is required.

[0150] Specifically, the saturation unique to this invention allows, to some extent, the retrace function REC to be suppressed and a retrace auxiliary setpoint T-rec to be generated, which is zero, instead of generating the retrace auxiliary setpoint T_rec = T_damp by rigidly applying the retrace function REC and applying it to the auxiliary motor 3. This retrace auxiliary setpoint will correspond to the damping component T_damp that hinders manual operation by the driver.

[0151] Of course, with different signs, it is found symmetrically relative to the origin (P0, 0) of the coordinate system. When considering the case where the steering wheel 2 turns to the left relative to its reference position P0, that is, if the northwest quadrant and the southwest quadrant in the figure are taken into account, corresponding to the case of negative abscissa, the same domains D1, D2, D3 and the same principle apply.

[0152] To eliminate potentially unwanted components T_visc, V_visc, T_damp, and V_damp while preserving the sync effect, such as... Figure 1As shown, the power steering system 1 according to the present invention includes a dynamic saturation function, denoted as "return dynamic saturation" RDS, which defines at least one saturation threshold SAT_high and SAT_low. The dynamic saturation function RDS is adjusted according to the value of the steering wheel speed setpoint V_sw_set (at the moment under consideration). The dynamic saturation function RDS is applicable to the calculation performed by the return assist calculation function FK2, thereby including the return assist setpoint T_rec in a domain called the "authorized return assist domain".

[0153] For your convenience, please note the following:

[0154] The subfunction THRESH_RDS of the dynamic saturation function RDS is called the "threshold calculation subfunction". This function calculates the saturation threshold based on the steering wheel speed setpoint V_sw_set and SAT_high and SAT_low.

[0155] The subfunction SAT_RDS of the dynamic saturation function RDS is called the "saturation subfunction". It performs saturation operation by applying saturation thresholds SAT_high and SAT_low defined by THRESH_RDS to the signal it processes, in order to suppress the signal.

[0156] In this case, the authorized return auxiliary field corresponds to the reference mentioned above. Figures 4 to 6 The pure return domain D2_T is associated with the mentioned return auxiliary setpoint T_rec (return torque setpoint).

[0157] By extension, the premise of the authorized return auxiliary domain D2_T of the return auxiliary calculation function FK2 will be the authorized steering wheel speed setpoint domain corresponding to the pure return domain D2 associated with the steering wheel speed error ΔV.

[0158] Advantageously, the evolved saturation proposed according to the invention allows for the calculation and as close as possible to the saturation thresholds SAT_high and SAT_low, based on the exact steering wheel 2's operation at the considered instant, thereby adapting the return assist setpoint T_rec so that the return assist setpoint T_rec can be applied to the assist motor 3, from which any possible undesirable components T_damp and T_visc have been eliminated, and only said undesirable components are eliminated.

[0159] In fact, if the condition for applying the return assist calculation function FK2 is that, without the dynamic saturation function RDS, the application of function FK2 will generate a return assist setpoint T_rec that exceeds the authorized return assist domain D2_T, then during the process of the return assist calculation function FK2, the dynamic saturation function RDS intervenes to apply the considered saturation thresholds SAT_high and SAT_low. These thresholds replace values ​​considered inappropriate (e.g., excessive steering wheel speed error ΔV, or excessive return assist setpoint T_rec) in order to ultimately limit the return assist setpoint T_rec, thereby bringing the return assist setpoint T_rec back to the authorized return assist domain D2_T. This return assist setpoint T_rec will then be effectively applied to the auxiliary motor 3.

[0160] However, if the condition for applying the return auxiliary calculation function FK2 is that even without the dynamic saturation function RDS, the return auxiliary setpoint T_rec is still located in the authorized return auxiliary domain of D2_T, then the value used or obtained by the return auxiliary calculation function FK2 is compatible with the considered saturation threshold SAT_high or SAT_low. In this case, the dynamic saturation function RDS will not be modified in the process of the return auxiliary calculation function FK2, so that the original return auxiliary setpoint T_rec is applied to the auxiliary motor 3 as is.

[0161] Considering that the steering wheel speed setpoint V_sw_set, representing the return requirement, is used to define at least one saturation threshold SAT_high, SAT_low, it is advantageous to allow the saturation threshold to define the limits (boundaries) of the pure return domain D2, D2_T, outside which the generated return components correspond to the bad components T_visc, T_damp.

[0162] Preferably, the dynamic saturation function RDS, and more particularly the threshold calculation sub-function THRESH_RDS, defines a high saturation threshold SAT_high and a low saturation threshold SAT_low. The dynamic saturation function RDS, and more particularly the sub-function THRESH_RDS, is adjusted on the one hand according to the vehicle speed, and on the other hand according to the value of the steering wheel setpoint V_sw_set.

[0163] Preferably, the subfunction THRESH_RDS adjusts the calculation of the threshold THRESH_RDS based on the sign of the position of the steering wheel P_sw relative to the reference position P0 within the considered time period.

[0164] The fact that two limits are jointly defined and applied—namely, (algebraically) the lower limit formed by the low saturation threshold SAT_low and the upper limit formed by the high saturation threshold SAT_high—advantageously allows for the definition of an authorized domain between these two limits, which here corresponds to the pure return domains D2, D2_T extending from the low saturation threshold SAT_low to the high saturation threshold SAT_high, and excludes adjacent domains, in this case the (viscous) turning domains D1, D1_T and the damping domains D3, D3_T, and performs top and bottom clipping.

[0165] Therefore, it is possible to control the steering situation ( Figure 4 It can also control rapid steering situations. Figure 6 ).

[0166] Considering the sign of the steering wheel position P_sw, we can distinguish between steering wheel turning right and steering wheel turning left, and adjust the saturation thresholds SAT_high and SAT_low accordingly.

[0167] It should also be noted that the amplitudes of the high saturation threshold SAT_high and the low saturation threshold SAT_low should ideally be unequal (that is, their absolute values ​​should not be equal) so as not to be symmetrical about zero.

[0168] More specifically, preferably, and according to features that can be self-contained in the invention, one of the saturation thresholds, SAT_high and SAT_low, is set to zero, while the other saturation threshold corresponds to a non-zero value whose sign depends on the sign of the steering wheel speed setpoint V_sw_set.

[0169] More preferably, the absolute value of the other non-zero saturation threshold depends on the value of the steering wheel speed setpoint V_sw_set.

[0170] More specifically, the non-zero saturation threshold preferably corresponds to the aforementioned velocity setpoint curve L1, or to the image of the curve L1 of the second conversion law K2 used by the return auxiliary calculation function FK2.

[0171] The zero saturation threshold allows for the prevention in practice of a return assist setpoint T_rec that would have the same sign as the steering wheel position P_sw, meaning that the return assist setpoint T_rec would tend to move the steering wheel away from its reference position by doing so in the opposite direction to the return the driver expects.

[0172] Therefore, the zero saturation threshold allows the damping components V_damp and T_damp to be eliminated (cancelled), which would otherwise tend to result in the manual return of the steering wheel during rapid reverse steering.

[0173] In fact, the choice of zero threshold depends on the sign of the steering wheel position P_sw (relative to the reference position P0).

[0174] Here, given the notation conventions used, the zero saturation threshold algebraically corresponds to the high threshold SAT_high of the positive steering wheel position P_sw, which is when the steering wheel is turned to the right (east half of the four-quadrant diagram), but conversely, it corresponds to the low threshold of the negative steering wheel position (west half of the diagram), which is when the steering wheel is turned to the left.

[0175] The sign and value depend on another non-zero threshold of the steering wheel speed setpoint V_sw_set, which advantageously allows the magnitude (in absolute value) of the return auxiliary setpoint T_rec that is ultimately applied to the auxiliary motor 3 to be below a maximum value that corresponds to the maximum authorized pure return component T_pure that takes into account the position of the steering wheel P_sw over the time period under consideration.

[0176] Therefore, a non-zero saturation threshold allows the removal (cutting) of undesirable viscous components T_visc, V_visc, which exceed the pure homing components T_pure, V_pure in the case of turning.

[0177] Therefore, the two chosen saturation thresholds, SAT_high and SAT_low, allow for efficient handling of all identified problem situations. Figure 4 steering and Figure 6 Quick manual reverse steering) without interfering with normal slow reverse steering ( Figure 4 ).

[0178] According to the preferred features that can constitute the invention itself, in particular, regardless of how the return function REC sets the return auxiliary setpoint T_rec, the saturation thresholds (s) SAT_high and SAT_low defined by the dynamic saturation function RDS, in particular by the threshold calculation sub-function THRESH_RDS, are independent of the actual speed V_sw of the steering wheel 2.

[0179] Therefore, the authorized domains D2 and D2_T, defined and applied by the dynamic saturation function RDS, are related to the steering wheel speed setpoint V_sw_set and thus vary depending on the steering wheel speed setpoint V_sw_set. However, they remain independent of the actual steering wheel speed V_sw and therefore depend on the actual steering wheel speed V_sw and do not change. The actual steering wheel speed V_sw does not participate in determining the saturation thresholds SAT_high and SAT_low.

[0180] Therefore, unlike the return assist setpoint T_rec, which is defined by the "raw" calculation of the return assist calculation function FK2 from the steering wheel speed error ΔV, the authorized domains D2 and D2_T are not affected by the actual steering wheel speed V_sw, as they include (and experience) the effects of the actual steering wheel speed V_sw, saturation thresholds SAT_high and SAT_low.

[0181] Advantageously, this allows for the decorrelation of a saturation threshold defined from (only) the steering wheel speed setpoint V_sw_set, that is, from the maximum return requirement, which may be useful at the considered position P_sw of steering wheel 2. On the other hand, the steering wheel speed error ΔV and the return assist setpoint T_rec both depend on the steering wheel speed error ΔV, and therefore both depend on the actual speed behavior of the steering wheel V_sw.

[0182] In this way, the adjustment of the saturation thresholds SAT_high and SAT_low becomes sensitive to the instantaneous return-to-center requirement, which is represented by the position of the steering wheel P_sw, and more specifically by the steering wheel speed setpoint V_sw_set. This steering wheel speed setpoint V_sw_set is associated with the steering wheel position P_sw through the return-to-center speed setpoint calculation function FK1, but it is not sensitive to the behavior of the steering wheel 2, which does not depend entirely on the return-to-center requirement, but also on the driver's intention and posture, represented here by the actual instantaneous speed V_sw of the steering wheel.

[0183] Therefore, the present invention can distinguish between pure return behavior and undesirable components caused by certain manual operations of the driver outside the limitations of such pure return behavior, and thus effectively, and preferably only, eliminate these undesirable components.

[0184] Based on one possibility of realization, and as Figure 7 As shown, the dynamic saturation function RDS acts on the input of the return assist calculation function FK2, thereby reducing the steering wheel speed error ΔV. This error is considered when calculating the return assist setpoint T_rec.

[0185] Formally, if SAT_RDS is represented as a subfunction of dynamic saturation defined by the dynamic saturation function RDS (that is, it applies the saturation threshold to the value under consideration), then in this configuration, we can obtain:

[0186] T_rec = FK2(SAT_RDS(ΔV))

[0187] Therefore, "preventative" wave cancellation is performed upstream of function FK2, which allows the steering wheel speed error ΔV to saturate so that the steering wheel speed error value ΔV, which is the input to the return-to-position auxiliary calculation function FK2, can be applied. This ensures that the image of function FK2 will be completely contained within the authorized return-to-position auxiliary domain D2_T, that is:

[0188]

[0189] therefore,

[0190] Therefore, it will be ensured that the wave suppression function includes the steering wheel speed error ΔV within the pure return domain D2, which is the premise for the authorized return auxiliary domain D2_T, constructed by the return auxiliary calculation function FK2 (as mentioned above). Figures 4 to 6 (as stated):

[0191]

[0192] Therefore, saturation thresholds SAT_high and SAT_low are defined uniformly at steering wheel speed, such that the saturation thresholds correspond to the limits (boundaries) of the pure return domain D2.

[0193] Therefore, preferably, the dynamic saturation function RDS may suppress the steering wheel speed error ΔV such that the steering wheel speed error ΔV is included between a first saturation threshold that is substantially equal to zero and preferably equal to zero and a second saturation threshold that is substantially equal to and preferably equal to the steering wheel speed setpoint V_sw_set.

[0194] The phrase "substantially equal to" indicates that there may be a predetermined tolerance range relative to the target values ​​(0 and V_sw_set, respectively), which is preferably equal to or less than + / -10%, and more preferably equal to or less than + / -5% of the steering wheel speed setting value V_sw_set.

[0195] Therefore, in Figures 4 to 6 In the eastern half-plane, corresponding to the case where the steering wheel 2 turns to the right of the reference position P0, the following settings can be made:

[0196] SAT_high = 0 (possibly + / - 10% * V_sw_set, or even + / - 5% * V_sw_set)

[0197] SAT_low = V_sw_set (possibly + / - 10% * V_sw_set, or even + / - 5% * V_sw_set)

[0198] Conversely, in the western half-plane, corresponding to a turn to the left of the reference position P0, due to the change in sign, we can obtain:

[0199] SAT_high = V_sw_set (possibly + / - 10% * V_sw_set, or even + / - 5% * V_sw_set)

[0200] SAT_low = 0 (possibly + / - 10% * V_sw_set, or even + / - 5% * V_sw_set)

[0201] The saturation thresholds SAT_high and SAT_low, which are uniform at steering wheel speeds, can be calculated using any appropriate subfunction THRESH_RDS.

[0202] An example of the threshold calculation sub-function THRESH_RDS is as follows: Figure 7 As shown.

[0203] In this example, the threshold calculation sub-function THRESH_RDS compares the zero and the steering wheel speed setpoint V_sw_set emitted by the return speed setpoint calculation function FK1, as shown below:

[0204] SAT_high = MAX[V_sw_set; 0]

[0205] SAT_low = MIN[V_sw_set; 0]

[0206] Of course, as mentioned above, the tolerance range of + / -10%*V_sw_set or even + / -5%*V_sw_set can be applied to these thresholds.

[0207] It should be noted that such a subfunction THRESH_RDS is particularly simple and quick to implement.

[0208] According to another possibility of realization, in terms of the final result it allows to obtain, it is substantially equivalent to the former, and as... Figure 1 As shown, the dynamic saturation function RDS acts on the output of the return auxiliary calculation function FK2 in order to eliminate the return auxiliary setpoint T_rec calculated by the return auxiliary calculation function FK2.

[0209] This leads to the conclusion that:

[0210] T_rec = SAT_RDS(FK2(ΔV))

[0211] Therefore, "therapeutic" damping is performed, which allows the return auxiliary setpoint T_rec to be saturated downstream of the function FK2, so that the output of the return auxiliary calculation function FK2 can be corrected if necessary, so that the resulting return auxiliary setpoint T_rec is indeed contained only in the authorized return auxiliary domain D2_T.

[0212] More specifically, see reference Figure 1 The return assist calculation function FK2 can calculate the original return assist setpoint T_rec_basic based on any value of the steering wheel speed error ΔV (which may or may not have been saturated according to the function RDS of the invention), while the dynamic saturation function RDS, especially the suppression sub-function SAT_RDS, then applies the saturation thresholds SAT_high and SAT_low to the original return assist setpoint T_rec_basic to obtain the appropriate return assist final setpoint T_rec.

[0213] Here, uniform saturation thresholds SAT_high and SAT_low are defined under the return torque (or force) such that the saturation thresholds correspond to the limits (boundaries) of the authorized return auxiliary domain D2_T.

[0214] Therefore, preferably and as Figure 1 , 2 As shown in Figure 3, the dynamic saturation function RDS is used to set the return auxiliary setpoint T_rec such that the return auxiliary setpoint T_rec is included between a first saturation threshold and a second saturation threshold. The first saturation threshold is substantially equal to and preferably equal to zero, and the second saturation threshold is substantially equal to and preferably equal to the image of the steering wheel speed error ΔV obtained by the return auxiliary calculation function FK2. The steering wheel speed error ΔV is equal to the steering wheel setpoint value V_sw_set.

[0215] The phrase “substantially equal to” here indicates that there may be a predetermined tolerance range relative to the target values ​​(here, 0 = FK2(0) and FK2(V_sw_set) respectively), which is preferably equal to or less than + / -10%, and more preferably equal to or less than + / -5% of the image value of the steering wheel speed setting value FK2(V_sw_set).

[0216] Therefore, in Figures 4 to 6 In the eastern half-plane, corresponding to the case where the steering wheel 2 turns to the right of the reference position P0, the following settings can be made:

[0217] SAT_high = FK2(0) = 0 (possibly + / - 10% * FK2(V_sw_set), or even + / - 5% * FK2(V_sw_set))

[0218] SAT_low = FK2(V_sw_set) (possibly + / - 10% * FK2(V_sw_set), or even + / - 5% * FK2(V_sw_set))

[0219] Conversely, in the western half-plane, corresponding to a turn to the left of the reference position P0, due to the change in sign, we can obtain:

[0220] SAT_high = FK2(V_sw_set) (possibly + / - 10% * FK2(V_sw_set), or even + / - 5% * FK2(V_sw_set))

[0221] SAT_low = FK2(0) = 0 (possibly + / - 10% * FK2(V_sw_set), or even + / - 5% * FK2(V_sw_set))

[0222] It should be noted that, for simplicity and consistency, the same second conversion law K2 is advantageously used, and more generally the same return-to-center auxiliary calculation function FK2 is used to calculate the return-to-center auxiliary setpoint T_rec from the steering wheel speed error ΔV, and to calculate the saturation thresholds SAT_high and SAT_low applicable to the wave-damping return-to-center auxiliary setpoint at the output of function FK2.

[0223] Therefore, the dynamic saturation function RDS eliminates the steering wheel speed error ΔV at the input of the return assist calculation function FK2 and / or eliminates the return assist setpoint T_rec at the output of the return assist function FK2, and if necessary, eliminates the original return assist setpoint T_rec_basic. The dynamic saturation function RDS will have the following effects: it retains only the useful component V_pure and only T_pure corresponding to the pure return requirement, and eliminates any undesirable components that will cause viscosity or damping in the return (V_visc, V_damp at steering wheel speed and / or T_visc, T_damp at the return assist setpoint, respectively, where the torque is uniform).

[0224] Therefore, as described above, a saturation threshold will be defined based on the return requirement, and the saturation threshold can be extracted from the return assist setpoint T_rec and, if necessary, removed from the return assist setpoint T_rec of any possible undesirable viscous component T_visc, which will be caused by the actual behavior at the steering wheel speed V_sw.

[0225] When the saturation thresholds SAT_high and SAT_low are applied to the return auxiliary setpoints T_rec and T_rec_basic at the output of the return auxiliary calculation function FK2, according to... Figure 3 The first variant shown, where the threshold calculation subfunction THRESH_RDS can, for example, compare the image value of the steering wheel speed setpoint V_sw_set with zero via the return auxiliary calculation function FK2, which is issued by the return speed setpoint calculation function FK1, as shown below (with the tolerances specified above if necessary):

[0226] SAT_high=MAX[FK2(V_sw_set);0]

[0227] SAT_low=MIN[FK2(V_sw_set);0]

[0228] According to the second variant, they are equal in purpose, such as Figure 2 As shown, the saturation threshold can be obtained using the following formula:

[0229] High saturation threshold:

[0230] SAT_high=[sign(P_sw)≠1]*FK2(V_sw_set),

[0231] Low saturation threshold:

[0232] SAT_low=[sign(P_sw)≠1]*FK2(V_sw_set)

[0233] in

[0234] P_sw represents the position of the steering wheel.

[0235] V_sw_set represents the steering wheel speed setpoint.

[0236] FK2 represents the back-position auxiliary calculation function.

[0237] If the sign of the expression being tested is positive, the function "sign" returns "1"; if the sign of the expression being tested is negative, it returns "-1".

[0238] The inequality symbol “≠” is a Boolean function that returns “1” if the two members of the equation being tested are different, and “0” if the two members of the equation being tested are equal.

[0239] It should be noted that, according to Figure 3 In the second variation, if the sign of the position P_sw of the steering wheel relative to the reference position P0 is positive, then the Boolean function 1≠1 returns zero because the values ​​are equal.

[0240] SAT_high = [sign(P_sw)≠1] * FK2(V_sw_set) = 0 * FK2(V_sw_set) = 0. Regardless of the threshold calculation function variable THRESH_RDS used, there is a zero saturation threshold, while the other non-zero saturation threshold is directly related to the steering wheel speed setpoint. As mentioned above, this makes the saturation of the steering wheel speed setpoint V_sw_set dependent, while making the saturation independent of the actual steering wheel speed V_sw.

[0241] Now refer to Figure 3 , 4The illustrative examples in section 5 illustrate the operation of the return function and the associated dynamic saturation.

[0242] For convenience and to facilitate understanding of the invention, we will now illustrate the saturation operation in the three cases studied above by giving virtual values ​​of the position and steering wheel speed, expressed in arbitrary units.

[0243] In addition, the dynamic saturation function RDS will be considered to act on the output of the return-to-base auxiliary calculation function FK2 to reference the original setpoint T_rec_basic for wave suppression and return-to-base auxiliary. Figure 1 The arrangements will be explained.

[0244] Of course, similar operations can be compared. Figure 7 The arrangement shown is described with reference to the suppression of steering wheel speed error at the input of the return auxiliary calculation function FK2.

[0245] In corresponding Figure 4 In the case of steering, set at OP1:

[0246] AP = P_sw > 0

[0247] V_sw = 3

[0248] V_sw_set=FK1(ΔP=-5

[0249] We can obtain:

[0250] ΔV=V_sw_set–V_sw=(-5)-(3)=-8

[0251] T_rec_basic=FK2(-8)

[0252] Moreover, as long as the steering wheel is turned to the right,

[0253] Sign(P_sw) = +1 because P_sw > 0

[0254] therefore

[0255] SAT_high=[sign(P_sw)≠1]*FK2(V_sw_set)=0*FK2(V_sw_set)=0 and

[0256] SAT_low=[sign(P_sw)≠-1]*FK2(V_sw_set)=1*FK2(V_sw_set)=FK2(-5)

[0257] In this case, from an algebraic perspective:

[0258] FK2(-8) < FK2(-5), that is, T_rec_basic < SAT_low,

[0259] More generally, T_rec_basic < SAT_low < SAT_high.

[0260] By applying the above saturation thresholds SAT_high = 0, SAT_low = FK2(V_sw_set) = FK2(-5) to the return assist original setpoint T_rec_basic = FK2(-8), the saturation function SAT_RDS thus returns FK2(-5), corresponding to the low saturation threshold.

[0261] Finally, it follows that:

[0262] T_rec = SAT_RDS(T_rec_basic) = SAT_low = FK2(V_sw_set) = FK2(-5), instead of FK2(-8).

[0263] Therefore, the return assist setpoint T_rec corresponds to the pure return component T_pure, which, due to saturation, corresponds to the result obtained by subtracting the viscous component T_visc = T_basic_rec - T_pure = T_basic_rec - FK2(V_sw_set) = FK2(ΔV) - FK2(V_sw_set) from the return assist original setpoint T_basic_rec.

[0264] Therefore, in the case of a turn, for a given steering wheel position P_sw (and, if necessary, for a given longitudinal vehicle speed V_vehic), regardless of the speed V_sw at which the driver drives the steering wheel 2, there is surely a return torque T_rec that is opposite to the movement away from the steering wheel, but whose intensity is limited and constant, that is, its intensity depends only on the distance of the steering wheel from its reference position P0 and not on the steering wheel drive speed V_sw.

[0265] In the case of Figure 5 the corresponding slow turn, for the same steering wheel position P_sw as before (and, if necessary, at the same longitudinal speed V_vehic as previously), and thus for the same steering wheel speed setpoint V_sw_set, at the operating point OP2 it is observed that:

[0266] V_sw = -2

[0267] V_sw_set = FK1(ΔP) = -5 (unchanged)

[0268] Therefore:

[0269] ΔV=V_sw_set–V_sw=(-5)-(-2)=-3

[0270] Therefore, T_rec_basic = FK2(-3).

[0271] The steering wheel is always turned to the right, and the saturation threshold remains unchanged: SAT_high = 0, SAT_low = FK2(-5).

[0272] Because FK2(-5) <FK2(-3)<0,

[0273] In other words, SAT_low <T_rec_basic<SAT_high,

[0274] Then, the suppression subfunction SAT_RDS will return FK2(-3), that is, without changing its basic setpoint T_rec_basic, which corresponds to the expected (pure) homing component.

[0275] Therefore, we obtain T_rec = T_rec_basic = FK2(ΔV) = FK2(-3).

[0276] In Figure 6 In the corresponding rapid reverse steering situation, the steering wheel is always kept in the same position P_sw as before. At the operation point OP3, the following is observed:

[0277] V_sw=-7

[0278] V_sw_set=FK1(ΔP)=-5(unchanged)

[0279] Therefore, we get:

[0280] ΔV=V_sw_set–V_sw=(-5)-(-7)=+2

[0281] Therefore, T_rec_basic = FK2(+2).

[0282] Similarly, the saturation thresholds are the same as in the previous example: SAT_high = FK2(0) = 0, SAT_low = FK2(-5).

[0283] Because FK2(+2)>FK2(0)=0,

[0284] In other words, T_rec_basic > SAT_high > SAT_low.

[0285] The suppression subfunction SAT_RDS then returns a value equal to the high saturation threshold SAT_high = 0, which is equivalent to eliminating the damping component T_damp that tends towards manual braking and repositioning.

[0286] T_rec = SAT_RDS(T_rec_basic) = SAT_high = FK2(0) = 0, instead of FK2(+2).

[0287] It should be noted that by choosing FK2(ΔV=0)=0, and therefore SAT_high=0 and T_rec=0 here, this is only equivalent to considering that if the driver has manually actuated the steering wheel 2 at a speed sufficient to reach (or even exceed) the “ideal” steering wheel speed setpoint V_sw_set as defined by function FK1, that is, when V_sw=V_sw_set, and therefore when ΔV=V_sw_set-V_sw=0, then it is useless for the auxiliary motor 3 to deliver a specific component to the return position.

[0288] Therefore, this choice allows for the absence of return assist when the driver does not require it, not to mention that it does not impede steering wheel movement when the driver performs the return to the reference position P0 at a speed faster than the speed nominally defined by the return speed setpoint calculation function FK1.

[0289] Of course, the present invention relates to a motor vehicle equipped with a power steering system 1 according to the present invention.

[0290] Of course, the present invention relates to a method for saturating the return assist setpoint, and more generally to a steering wheel return method according to any feature of the present invention.

[0291] Therefore, the present invention particularly relates to a steering wheel return method. When the steering wheel is in a position P_sw different from a given reference position P0, the method allows the steering wheel of the power steering system 1 to automatically return to the reference position P0, such as the center position. In this process, firstly, the steering wheel speed setpoint V_sw_set is calculated based on the deviation between the instantaneous actual position of the steering wheel and the reference position, referred to as the "steering wheel position error" ΔP. Then, a setpoint called the "return auxiliary setpoint" T_rec is calculated based on the deviation between the actual speed V_sw of the steering wheel and the steering wheel speed setpoint V_sw_set, referred to as the "steering wheel speed error" ΔV. The setpoint is used to control the auxiliary motor 3 to converge the actual speed P_sw of the steering wheel to the steering wheel speed setpoint P_sw_set. The method is characterized by including a dynamic saturation step, in which at least one saturation threshold SAT_high and SAT_low are defined, the saturation thresholds being adjusted according to the value of the steering wheel speed setpoint V_sw_set, and then applied to the calculation of the return auxiliary setpoint so that the return auxiliary setpoint T_rec is included in a domain called the "authorized return auxiliary domain" D2_T, the range of which is related to the value of the steering wheel speed setpoint V_sw_set.

[0292] Finally, the present invention relates to a computer-readable data medium, such as a hard disk, flash memory, USB key, optical disk, electronic card, or equivalent type, and contains code elements of a computer program that, when read by a computer, allow the implementation of any function and / or sub-function of the method according to the invention or the method thereof.

[0293] Of course, the present invention is by no means limited to the above-described variations, and those skilled in the art are particularly free to separate or combine any of the above features, or replace them with equivalents.

[0294] Specifically, the present invention relates to the use of a dynamic saturation function RDS in a power steering system 1 including a steering wheel return function REC, wherein the saturation thresholds SAT_high and SAT_low are defined by the steering wheel speed setpoint V_sw_set defined by the return function REC, but are independent of the actual speed V_sw of the steering wheel 2, so as to include the return assist setpoint T_rec applied by the return function REC to the assist motor 3 in the authorized return assist domain D2_T, the limits of which are set by the saturation thresholds SAT_high and SAT_low.

Claims

1. A power steering system (1) including a steering wheel (2) and a return function (REC) designed to automatically return the steering wheel (2) to the reference position (P0) when the steering wheel is in a position (P_sw) different from a given reference position (P0), the return function (REC) for this purpose comprising: The return speed setpoint calculation function (FK1) calculates the steering wheel speed setpoint (V_sw_set) based on the deviation between the instantaneous actual position (P_sw) of the steering wheel and the reference position (P0), referred to as "steering wheel position error" (ΔP); and the return assist calculation function (FK2) defines a setpoint called "return assist setpoint" (T_rec) based on the deviation between the actual speed (V_sw) of the steering wheel and the steering wheel speed setpoint (V_sw_set), referred to as "steering wheel speed error" (ΔV), which is used to control the assist motor (3) to adjust the steering wheel speed setpoint. The actual speed of the steering wheel (V_sw) converges to the steering wheel speed setpoint (V_sw_set). The system (1) is characterized by including a dynamic saturation function (RDS) that defines at least one saturation threshold (SAT_high, SAT_low), which is adjusted according to the value of the steering wheel speed setpoint (V_sw_set). The dynamic saturation function is then used to mitigate calculations performed by the return assist calculation function (FK2) to include the return assist setpoint (T_rec) in a domain called the "authorized return assist domain" (D2_T), the range of which is related to the value of the steering wheel speed setpoint (V_sw_set).

2. The power steering system according to claim 1, characterized in that, The Dynamic Saturation Function (RDS) defines a high saturation threshold (SAT_high) and a low saturation threshold (SAT_low). The Dynamic Saturation Function (RDS) is adjusted based on the vehicle speed (V_vehic) and the value of the steering wheel setpoint (V_sw_set).

3. The power steering system according to claim 2, characterized in that, One of the saturation thresholds, the high saturation threshold (SAT_high) and the low saturation threshold (SAT_low), is set to zero, while the other saturation threshold corresponds to a non-zero value, the sign of which depends on the sign of the steering wheel speed setpoint (V_sw_set).

4. The power steering system according to any one of the preceding claims, characterized in that, The dynamic saturation function (RDS) is applied to the input of the return-to-center auxiliary calculation function (FK2) to reduce the steering wheel speed error (ΔV). The steering wheel speed error is taken into account when calculating the return-to-center auxiliary setpoint (T_rec).

5. The power steering system according to claim 4, characterized in that, The Dynamic Saturation Function (RDS) dampens the steering wheel speed error (ΔV) such that the steering wheel speed error (ΔV) is included between a first saturation threshold that is substantially equal to zero and a second saturation threshold that is substantially equal to the value of the steering wheel speed setpoint (V_sw_set).

6. The power steering system according to any one of claims 1 to 3, characterized in that, The dynamic saturation function (RDS) is applied to the output of the return auxiliary calculation function (FK2), thereby eliminating the return auxiliary setpoint (T_rec, T_rec_basic) calculated by the return auxiliary calculation function (FK2).

7. The power steering system according to claim 6, characterized in that, The Dynamic Saturation Function (RDS) dampens the return assist setpoint (T_rec) such that the return assist setpoint (T_rec) is included between a first saturation threshold and a second saturation threshold, the first saturation threshold being substantially equal to zero, and the second saturation threshold being substantially equal to the image of the steering wheel speed error (ΔV) obtained by the return assist calculation function (FK2), the steering wheel speed error being equal to the value of the steering wheel setpoint (V_sw_set).

8. The power steering system according to claim 1, characterized in that, The saturation thresholds (SAT_high, SAT_low) defined by the dynamic saturation function (RDS) are independent of the actual speed (V_sw) of the steering wheel (2).

9. A motor vehicle equipped with a power steering system (1) according to any one of claims 1 to 8.

10. A steering wheel return method that allows the steering wheel (2) of a power steering system (1) to automatically return to the reference position (P0) when the steering wheel (2) is in a position (P_sw) different from a given reference position (P0), wherein, in the method, a steering wheel speed setpoint (V_sw_set) is first calculated based on the deviation between the instantaneous actual position (P_sw) of the steering wheel and the reference position (P0) called "steering wheel position error" (ΔP), and then a return assist setpoint (T_rec) is defined based on the deviation between the actual speed (V_sw) of the steering wheel and the steering wheel speed setpoint (V_sw_set) called "steering wheel speed error" (ΔV). A setpoint, the setpoint being used to control the auxiliary motor (3) to converge the actual speed (V_sw) of the steering wheel to the steering wheel speed setpoint (V_sw_set), the method being characterized in that it includes a dynamic saturation step in which at least one saturation threshold (SAT_high, SAT_low) is defined, the dynamic saturation step being adjusted according to the value of the steering wheel speed setpoint (V_sw_set) and then applied to the calculation of the return auxiliary setpoint so that the return auxiliary setpoint (T_rec) is included in a domain called the "authorized return auxiliary domain" (D2_T), the range of which is related to the value of the steering wheel speed setpoint (V_sw_set).

11. A data medium that is computer-readable and contains code elements of a computer program, wherein when the computer reads the medium, the code elements allow implementation of the method according to claim 10.

Images