KR102814378B1 - Steering cotrol method and system of motor driven power steering system - Google Patents

Abstract

The present invention relates to a steering control method and system for an electric power steering system. In the present invention, a steering control method and system for an electric power steering system are introduced, characterized in that a virtual steering system model is secured in which a steering reaction mechanism is provided between a steering wheel and a rack bar, and the rack reaction mechanism is provided on the rack bar; a state equation having the momentum of the steering reaction mechanism and the rack reaction mechanism as state variables is derived for the virtual steering system model; a target steering torque applied to the steering reaction mechanism is calculated through numerical integration of the state equation; and a control amount of a steering motor is feedback-controlled so as to follow the target steering torque.

Description

{STEERING COTROL METHOD AND SYSTEM OF MOTOR DRIVEN POWER STEERING SYSTEM}

The present invention relates to a steering control method and system for an electric steering system, which improves the development efficiency of control technology by enabling prediction of steering performance by variously changing the characteristics of the steering system using a virtual steering system model.

In the case of the existing MDPS (MOTOR DRIVEN POWER STEERING), the primary target restoration torque is calculated using a tuning table for the steering angle and vehicle speed.

Then, the assist scale is calculated based on the driver's steering torque, and the final target restoration torque is calculated and determined by multiplying the first target restoration torque by that ratio.

However, the graph of the steering angle and target restoration torque for vehicle speed has a problem that it is achieved through actual vehicle tuning, and the target restoration torque calculated by the tuning table also causes differences in restoration performance depending on the friction distribution of the vehicle/parts even within the same vehicle model.

The matters described as background technology above are only intended to enhance understanding of the background of the present invention, and should not be taken as an acknowledgment that they correspond to prior art already known to those skilled in the art.

KR 10-2017-0019669 A

The present invention has been made to solve the above-mentioned problems, and provides a steering control method and system of an electric steering system that is easy to set a target steering torque and is robust to friction distribution of a vehicle/parts by applying a virtual steering system model capable of obtaining a reaction force input to a rack bar to obtain a target steering torque.

In order to achieve the above object, the present invention may be characterized by including a step in which a controller secures a virtual steering system model in which a steering reaction mechanism is provided between a steering wheel and a rack bar, and a rack reaction mechanism is provided on the rack bar; a step in which the controller derives a state equation having momentum of the steering reaction mechanism and the rack reaction mechanism as state variables for the virtual steering system model; a step in which the controller calculates a target steering torque applied to the steering reaction mechanism through numerical integration of the state equation; and a step in which the controller feedback-controls a control amount of a steering motor to follow the target steering torque.

The above virtual steering system model applies the steering angle velocity of the upper part of the torsion bar, which is the steering reaction force mechanism, and the rack force as input variables; the torsion bar stiffness, the torsion bar damper, the steering angle velocity of the lower part of the torsion bar, the pinion radius, the rack bar moving speed, the rack bar weight, and the rack spring stiffness and the rack damper, which are the rack reaction force mechanisms, as system characteristic parameters; and the target steering torque determined by the relationship between the input variables and the system characteristic parameters can be applied as an output variable.

For the above virtual steering system model, a state equation can be derived using a bond graph.

By setting the above torsion bar torsional displacement and rack bar momentum as state variables, a state equation for the torsion bar torsional displacement differential value and rack bar momentum differential value can be derived.

The state equation for the differential value of the above torsion bar torsional displacement is derived as in the following equation (1); and the state equation for the differential value of the above rack bar momentum can be derived as in the following equation (2).

..........................(Formula 1)

..(Formula 2)

W _in : Steering angle speed (top of torsion bar)

F _rack : Rack Force

K _t : Torsion bar stiffness

B _t : Torsion bar damper

W _C : Steering angle speed (torsion bar bottom)

R _p : pinion radius

V _r : Rack bar movement speed

M: Rack bar weight

K _rt : Rack spring stiffness

B _rt : Rack damper

X _r : Rack bar displacement

The above target steering torque can be calculated by the following equation (3).

..................(Formula 3)

T _{q_ref} : Target steering torque

K _t : Torsion bar stiffness

q ₃ : Torsion bar torsional displacement

B _t : Torsion bar damper

: Torsion bar torsional displacement differential

An assist scale is determined according to the steering torque, and by multiplying the assist scale by the target steering torque, it is possible to change the target steering torque according to the assist scale.

Here, 0 < AssistScale ≤ 1.

By changing at least one of the above system characteristic parameters, it may be possible to change the target steering torque.

The configuration of the present invention may be characterized by including: a steering system model storage unit storing a virtual steering system model in which a steering reaction mechanism is provided between a steering wheel and a rack bar, and the rack reaction mechanism is provided on the rack bar; a target steering torque calculation unit deriving a state equation having momentum of the steering reaction mechanism and the rack reaction mechanism as state variables for the virtual steering system model, and calculating a target steering torque applied to the steering reaction mechanism through numerical integration of the state equation; and a feedback controller that feedback-controls a control amount of a steering motor to follow the target steering torque.

Through the above-described problem solving means, the present invention enables prediction of steering performance by calculating target steering torque based on a virtual steering system model (VM), thereby increasing the efficiency of development of steering control technology. In particular, since a rack reaction force mechanism is connected to the steering system model, it is possible to calculate the target steering torque based on a state in which rack reaction force is input to the rack bar. Accordingly, it is possible to easily set the target steering torque for steering restoration, thereby providing an advantage of robustness against friction distribution of the vehicle/parts, and has the effect of improving the torque build-up of the steering torque.

Figure 1 is a drawing illustrating the configuration of a steering control system according to the present invention.
Figure 2 is a diagram showing the flow of a steering control process according to the present invention.
Figure 3 is a drawing showing the experimental results showing differences in restorability as the rack spring stiffness changes in the present invention.
Figure 4 is a drawing showing the experimental results showing differences in restorability as the rack damper in the present invention changes.
Figures 5 and 6 are drawings showing experimental results for explaining the correlation between torsion bar stiffness and rack spring stiffness in the present invention.

A preferred embodiment of the present invention is described in detail with reference to the attached drawings as follows.

An electric steering system applicable to the present invention may be a steering system that generates steering force or assists steering force using an electric motor, and may be an MDPS (Motor Driven Power Steering) system or an SBW (Steer-by-wire) system.

Meanwhile, the present invention is a steering control method that can be applied to the above-described steering system to set a target steering torque (T _{q_ref} ), comprising: a step of generating a virtual steering system model (VM); a step of deriving a state equation for the virtual steering system model (VM); a step of generating a target steering torque using the state equation; and a step of performing feedback control to track the target steering torque (T _{q_ref} ).

Referring to FIGS. 1 and 2, the present invention will be examined in detail. First, in the virtual steering system model setting step, a controller (CLR) secures a virtual steering system model (VM) in which a steering reaction mechanism (3) is provided between a steering wheel (1) and a rack bar (5), and a rack reaction mechanism (7) is provided on the rack bar (5).

Here, the steering reaction mechanism (3) may be a torsion bar, and may have a structure in which a torsion bar is connected between the steering wheel (1) and the rack bar (5), and further, the rack reaction mechanism (7) may have a structure in which a rack spring and a rack damper are connected to the rack bar (5), so that a virtual steering system model (VM) can be set.

And, in the state equation derivation step, the controller (CLR) derives a state equation with the momentum of the steering reaction mechanism (3) and the rack reaction mechanism (7) as state variables for the virtual steering system model (VM).

Next, in the target steering torque calculation step, the controller (CLR) calculates the target steering torque (T _{q_ref} ) applied to the steering reaction mechanism (3) through numerical integration of the state equation.

And, in the feedback control stage, the controller (CLR) feedback-controls the control amount of the steering motor to follow the target steering torque (T _{q_ref} ).

That is, according to the above configuration, by calculating the target steering torque (T _{q_ref} ) based on the virtual steering system model (VM), the steering performance can be predicted, thereby increasing the efficiency of the development of steering control technology. In particular, since the rack reaction force mechanism (7) is connected to the above steering system model (VM), the target steering torque (T _{q_ref} ) can be calculated based on the state in which the rack reaction force is applied to the rack bar (5), so that the target steering torque (T _{q_ref} ) for steering restoration can be easily set, which not only provides the advantage of being robust to the friction distribution of the vehicle/parts, but also improves the torque build-up of the steering torque.

In addition, referring to FIG. 1, a virtual steering system model (VM) according to the present invention is explained. The virtual steering system model (VM) applies the steering angle velocity (W _in ) of the upper end of the torsion bar and the rack force (F _rack ) as input variables.

In addition, the torsion bar stiffness (K _t ), the torsion bar damper (B _t ), the steering angle speed (W _C ) of the lower end of the torsion bar, the pinion radius (R _p ), the rack bar moving speed (V _r ), the rack bar weight (M), and the rack spring stiffness (K _rt ) and the rack damper (B _rt ) of the rack reaction mechanism (7) are applied as system characteristic parameters.

Additionally, the target steering torque (T _{q_ref} ) determined by the relationship between the above input variables and system characteristic parameters can be applied as an output variable.

That is, a virtual steering system model (VM) can be set up with two input variables, eight parameters, and one output variable, and in particular, by applying the rack spring stiffness (K _rt ) and rack damper (B _rt ) as system characteristic parameters, the restoration performance according to the restoration of the steering wheel is improved.

Meanwhile, a state equation can be derived for the above virtual steering system model (VM) using a bond graph, and the bond graph can be expressed as follows.

W _in : Steering angle speed (top of torsion bar)

F _rack : Rack Force

K _t : Torsion bar stiffness

B _t : Torsion bar damper

W _C : Steering angle speed (torsion bar bottom)

R _p : pinion radius

V _r : Rack bar movement speed

M: Rack bar weight

K _rt : Rack spring stiffness

B _rt : Rack damper

And, by using the above bond graph, the state equation is derived, and by setting the torsion bar torsional displacement and the rack bar momentum as state variables, the state equation for the torsion bar torsional displacement differential value and the rack bar momentum differential value can be derived.

The state equation for the differential value of the above torsion bar's torsional displacement can be derived as shown in the following equation (1).

..........................(Formula 1)

And, the state equation for the above rack bar momentum derivative can be derived as shown in the following equation (2).

..(Formula 2)

q ₃ : Torsion bar torsional displacement

P ₉ : Rack bar momentum

X _r : Rack bar travel displacement

: Rack bar momentum derivative

In addition, as explained above, the present invention calculates the target steering torque (T _{q_ref} ) applied to the torsion bar through numerical integration of the state equation, and the target steering torque (T _{q_ref} ) can be calculated by the following equation (3).

..................(Formula 3)

T _{q_ref} : Target steering torque

K _t : Torsion bar stiffness

q ₃ : Torsion bar torsional displacement

B _t : Torsion bar damper

: Torsion bar torsional displacement differential

Meanwhile, the present invention can be configured such that an assist scale is determined according to a steering torque measured by a torque sensor, and the target steering torque (T _{q_ref} ) calculated by the above formula (3) is multiplied by the assist scale, so that the target steering torque (T _{q_ref} ) can be changed according to the assist scale.

Here, 0 < AssistScale ≤ 1.

That is, as the driver's will to steer increases, the assist scale decreases, which reduces the target steering torque (T _{q_ref} ).

In addition, the present invention can be configured to allow a change in target steering torque (T _{q_ref} ) by changing at least one of the above system characteristic parameters.

For example, Fig. 3 shows the experimental results showing that a difference in restorability occurs as the rack spring stiffness (K _rt ) changes in the present invention. When the steering wheel is restored toward the neutral steering angle, it can be confirmed that when the rack spring stiffness (K _rt ) increases, the restoration speed increases and the residual angle decreases.

In addition, Fig. 4 shows the experimental results showing that a difference in restorability occurs as the rack damper (B _rt ) changes in the present invention. When the steering wheel is restored toward the neutral steering angle, it can be confirmed that the overshoot phenomenon is reduced as the rack damper (B _rt ) increases.

That is, the rack spring and rack damper (B _rt ) applied to the above virtual steering system model (VM) are system characteristic parameters that have a great influence on the tuning scalability related to the restoring performance and torque build-up, and it is difficult to replace them by changing other system characteristic parameters.

This can be confirmed through the experimental results shown in FIGS. 5 and 6. Referring to FIG. 5, when the rack spring stiffness (K _rt ) is 0, it can be confirmed that even if the torsion bar spring stiffness is increased, there is no change in the hysteresis curve, and thus there is no change in the torque build-up.

However, as shown in Fig. 6, when the rack spring stiffness (K _rt ) is increased, it can be confirmed that although the influence on torque build-up increases as the torsion bar spring stiffness increases, there is a limit to the degree of change.

In this way, the present invention enables calculation of target steering torque (T _{q_ref} ) based on a state in which rack reaction force is input to the rack bar (5) by connecting a rack reaction mechanism (7) to a virtual steering system model (VM), thereby easily setting target steering torque (T _{q_ref} ) for steering restoration, thereby providing the advantage of robustness to friction distribution of the vehicle/parts, and improving torque build-up of steering torque.

Meanwhile, the steering control system of the electric steering system according to the present invention may be configured to include a steering system model storage unit (10), a target steering torque calculation unit (20), and a feedback controller (30).

Referring to Fig. 1, first, a virtual steering model (VM) is stored in the steering system model storage unit (10) in which a steering reaction mechanism (3) is provided between a steering wheel (1) and a rack bar (5), and a rack reaction mechanism (7) is provided on the rack bar (5).

And, in the target steering torque calculation unit (20), a state equation having the momentum of the steering reaction mechanism (3) and the rack reaction mechanism (7) as state variables is derived for the virtual steering system model (VM), and the target steering torque (T _{q_ref} ) applied to the steering reaction mechanism (3) is calculated through numerical integration of the state equation.

Additionally, the feedback controller (30) feedback-controls the control amount of the steering motor to follow the target steering torque (T _{q_ref} ).

Meanwhile, Fig. 2 is a drawing showing a steering control flow using the steering control device of Fig. 1. Referring to the drawing, a virtual steering system model (VM) consisting of a steering wheel (1), a torsion bar, a rack bar (5), a rack spring, and a rack damper is set and stored (S10).

Next, a bond graph for the virtual steering system model (VM) is generated (S20), and a state equation is derived using the bond graph. The state equation is derived by setting the torsion bar torsional displacement (q ₃ ) and the rack bar momentum (P ₉ ) as state variables (S30).

Then, the target steering torque (T _{q_ref} ) applied to the torsion bar is calculated through numerical integration of the above state equation (S40).

After this, the control amount of the steering motor is feedback controlled to follow the target steering torque (T _{q_ref} ) (S50).

As described above, the present invention calculates the target steering torque (T _{q_ref} ) based on a virtual steering system model (VM), thereby enabling prediction of steering performance, thereby increasing the efficiency of development of steering control technology. In particular, since the rack reaction force mechanism (7) is connected to the steering system model (VM), the target steering torque (T _{q_ref} ) can be calculated based on the state in which the rack reaction force is applied to the rack bar (5), thereby easily setting the target steering torque (T _{q_ref} ) for steering restoration, thereby providing the advantage of robustness against friction distribution of the vehicle/parts, and improving torque build-up of the steering torque.

In addition, when applied to an SBW system in which the mechanical connection structure between the steering wheel (1) and the steering gear box is disconnected, it is possible to generate steering reaction force and steering feel similar to that of a mechanical steering system.

Meanwhile, although the present invention has been described in detail only with respect to the above-mentioned specific examples, it will be apparent to those skilled in the art that various modifications and variations are possible within the technical scope of the present invention, and it is natural that such modifications and variations fall within the scope of the appended claims.

1 : Steering wheel
3: Steering reaction mechanism
5: Rack bar
7: Rack reaction mechanism
10: Steering system model storage
20: Target steering torque calculation unit
30: Feedback controller
CLR : Controller
VM: Virtual steering system model

Claims (9)

A step for a controller to secure a virtual steering system model in which a steering reaction mechanism is provided between a steering wheel and a rack bar, and a rack reaction mechanism is provided on the rack bar;
A step in which the controller derives a state equation having the momentum of the steering reaction mechanism and the rack reaction mechanism as state variables for the virtual steering system model;
A step in which the controller calculates a target steering torque applied to the steering reaction mechanism through numerical integration of the state equation; and
A controller comprises a step of feedback-controlling a control amount of a steering motor to follow the target steering torque;
The above virtual steering system model is,
The steering angle velocity and rack force at the top of the torsion bar, which is the above steering reaction mechanism, are applied as input variables;
The above torsion bar stiffness, torsion bar damper, steering angle speed at the bottom of the torsion bar, pinion radius, rack bar moving speed, rack bar weight, and rack spring stiffness and rack damper as rack reaction mechanism are applied as system characteristic parameters;
A steering control method for an electric steering system, characterized in that a target steering torque determined by the relationship between the above input variables and system characteristic parameters is applied as an output variable. delete In claim 1,
A steering control method for an electric steering system, characterized by deriving a state equation using a bond graph for the above virtual steering system model. In claim 1,
A steering control method of an electric steering system, characterized in that the torsion bar torsional displacement and the rack bar momentum are set as state variables and a state equation for the torsion bar torsional displacement differential value and the rack bar momentum differential value is derived. In claim 4,
The state equation for the differential value of the above torsion bar torsional displacement is derived as shown in the following equation (1);
A steering control method of an electric steering system, characterized in that the state equation for the above rack bar momentum derivative is derived as in the following equation (2).
..........................(Formula 1)
..(Formula 2)
W _in : Steering angle speed (top of torsion bar)
F _rack : Rack Force
K _t : Torsion bar stiffness
B _t : Torsion bar damper
W _C : Steering angle speed (torsion bar bottom)
R _p : pinion radius
V _r : Rack bar movement speed
M: Rack bar weight
K _rt : Rack spring stiffness
B _rt : Rack damper
X _r : Rack bar displacement In claim 1,
A steering control method of an electric steering system, characterized in that the target steering torque is calculated by the following equation (3).
..................(Formula 3)
T _{q_ref} : Target steering torque
K _t : Torsion bar stiffness
q ₃ : Torsion bar torsional displacement
B _t : Torsion bar damper
: Torsion bar torsional displacement differential In claim 1,
A steering control method of an electric steering system, characterized in that an assist scale is determined according to a steering torque, and the target steering torque is changed according to the assist scale by multiplying the assist scale by the target steering torque.
Here, 0 < AssistScale ≤ 1. In claim 1,
A steering control method for an electric power steering system, characterized in that a target steering torque can be changed by changing at least one of the above system characteristic parameters. A steering system model storage unit in which a virtual steering system model in which a steering reaction mechanism is provided between a steering wheel and a rack bar and a rack reaction mechanism is provided on the rack bar is stored;
A target steering torque calculation unit that derives a state equation having the momentum of the steering reaction mechanism and the rack reaction mechanism as state variables for the virtual steering system model, and calculates a target steering torque applied to the steering reaction mechanism through numerical integration of the state equation; and
A feedback controller for feedback-controlling the control amount of the steering motor to follow the target steering torque;
The above virtual steering system model is,
The steering angle velocity and rack force at the top of the torsion bar, which is the above steering reaction mechanism, are applied as input variables;
The above torsion bar stiffness, torsion bar damper, steering angle speed at the bottom of the torsion bar, pinion radius, rack bar moving speed, rack bar weight, and rack spring stiffness and rack damper as rack reaction mechanism are applied as system characteristic parameters;
A steering control system for an electric steering system, characterized in that a target steering torque determined by the relationship between the above input variables and system characteristic parameters is applied as an output variable.

Images