WO2005113269A1 - Semi-active chassis concept having a reduced number of sensor - Google Patents

Abstract

The invention relates to a semi-active chassis concept having a reduced number of sensors for use in vehicles, especially motor vehicles. Said chassis concept comprises level sensors (4, 5, 6), sensing a change in travel of a vehicle body (1) in the vertical direction and, depending thereon, providing first sensor signals (N_VL, N _VR, N_HM); acceleration sensors (7, 8), sensing accelerations of the vehicle body (1) in the vertical direction and, depending thereon, providing second sensor signals (A_Z_VL, A_Z_VR); and a control device (10) for providing actuation signals for adjustment devices. The invention also relates to a method for producing actuation signals for adjustment devices (2, 3) of a semi-active chassis concept.

Description

 Semi-active chassis system with a reduced number of sensors

The invention relates to a semi-active chassis system with a reduced number of sensors for vehicles, in particular motor vehicles, with level sensors, acceleration sensors, a control device and adjusting devices, and a method for generating control signals for adjusting devices of a semi-active chassis system. The invention further relates to a method for generating control signals for control devices of a semi-active chassis system, in particular a semi-active chassis system.

Semi-active chassis systems, for example for motor vehicles with air suspension, have adjustable shock absorbers as actuating devices. Most vehicle manufacturers use three body acceleration sensors in such systems. Is not discussed in detail the structure and operation of such semi-active suspension systems ^is' generally known, so that it.

German patent specification DE 198 55 310 C2 describes an active suspension system for vehicles in which several sensors are used to generate signals which are at least correlated with vertical body accelerations. The disadvantage of such an arrangement is, above all, the relatively high costs involved in installing nes each sensor and for the wiring of the sensors with connection and supply lines and the associated control units and brackets.

The object on which the present invention is based is to provide a more economical suspension system for vehicles and a method for generating control signals for actuating devices of the suspension system, in particular maintaining the same driving comfort.

This object is achieved according to the invention by a semi-active chassis system with the features of patent claim 1 and by a method with the features of patent claim 7.

Accordingly, it is provided:

A semi-active chassis system with a reduced number of sensors for vehicles, in particular motor vehicles, with level sensors which record a change in a path of a vehicle body between the latter and the axles in the vertical direction and provide dependent first sensor signals; Acceleration sensors that record accelerations of the vehicle body in the vertical direction and provide dependent second sensor signals; a control device for providing control signals for actuating devices, the control device ^■ a first device for signal processing and processing, ^■ a second device for calculating state variables of the vehicle body based on the first and second sensor signals, which has a sensor fusion algorithm with which at least one further sensor signal is compensated, and ^{■ contains} a third device for generating the control signals for the actuating devices; Actuators for adjusting the vehicle body in the vertical direction. (Claim 1)

A method for generating control signals for control devices of a semi-active chassis system, with the method steps: (a) processing sensor signals; (b) calculating spring and damper forces; (c) calculating lift and pitch accelerations of the vehicle body; (d) calculating roll acceleration of the vehicle body; (e) calculating superimposition of the lift and pitch acceleration of the vehicle body; (f) calculating state variables of the vehicle body; (g) evaluating a control algorithm or control algorithm using the calculated state variables; and (h) generating drive signals. (Claim 7)

Advantageous refinements and developments of the invention can be found in the subclaims and the descriptions with reference to the drawings.

The idea underlying the present invention is to reduce the number of vertical body acceleration sensors required. The number of body acceleration sensors not required in this way is without replacement and is only compensated for by an additional algorithm in the associated control unit. In a preferred embodiment, the chassis system according to the invention has two acceleration sensors above the front axle of the vehicle, which are arranged at a distance from one another. A third conventional acceleration sensor above the rear axle can advantageously also be omitted without replacement.

The signal of the missing third acceleration sensor is advantageously compensated for by means of a sensor fusion algorithm which is implemented in the control device. This sensor fusion algorithm of the second device contains the following functional blocks: a fourth device for calculating spring and damper forces, - a fifth device for calculating lifting and pitching accelerations of the vehicle body, a sixth device for calculating roll acceleration of the vehicle body, and a seventh device for calculating superimposition of lifting and pitching acceleration of the vehicle body - an eighth device for calculating state variables of the vehicle body

The roll acceleration of the vehicle and the superimposition of lift and pitch acceleration are typically calculated using the signals from the remaining acceleration sensors above the front axle.

An advantageous arrangement provides that the sixth device and the seventh device are arranged parallel to one another and in each case parallel to the fourth and fifth device and each have a calculation stage for their have output signals from the signals of the first and second acceleration sensors.

In particular, the modal speeds as an example of state variables of the vehicle body are typically calculated using an integration level and a weighting level. It is possible for the integration stage to be arranged in the signal course before the weighting stage. It is advantageous if the calculation carried out in the weighting stage is carried out using the so-called weighted linear least squares method.

Advantageously, the software components of the sensor fusion algorithm, which forms the so-called observer or has the function of an observer, and the control algorithm, which forms the so-called control algorithm, are almost completely decoupled. The sensor fusion algorithm calculates the state variables of the vehicle body and transfers them directly to the control algorithm. For the damping force calculation of the sensor fusion algorithm, it is advantageous if a feedback is arranged between the output of the control algorithm and the block for calculating spring and damper forces. This feedback enables information about the actuation of the actuating devices, which are required for the damper force calculation.

The sensor fusion algorithm advantageously makes it possible to save an acceleration sensor with its installation components. The compensation of its signals by means of a software component is advantageously simple and, for example, requires only minor changes or additions to existing systems. For example, the remaining body acceleration sensors above the front axle at the front left and front right at their usual in- remain at the installation site. Likewise, the level sensors do not have to be placed elsewhere. In addition, due to the decoupling of the control algorithm and sensor fusion algorithm, only minor changes need to be made in the control device.

In a further embodiment, the second device has a Cayman filter.

In the method, method steps (d) and (e) are carried out with signals from the first and second acceleration sensors. In one embodiment, method step (f) has the following method sub-steps: (al) pseudo-integrating acceleration signals of method steps (c), (d) and (e) and (a2) weights of the pseudo-integrated signals according to their respective signal quality.

Process sub-step (a2) includes weighting using the weighted linear least squares method.

Method step (b) is additionally carried out with feedback control signals from method step (h).

The process steps (b) to (f) can be carried out in a further embodiment by means of a Cayman filter.

The invention is explained in more detail below with reference to the exemplary embodiments given in the schematic figures of the drawing. It shows:

Figure 1 is a simplified schematic representation of a vehicle with a semi-active chassis system according to the prior art. 2 shows a block diagram of a control device for the chassis system according to FIG. 1 according to the prior art; 3 shows a simplified schematic illustration of a vehicle with a semi-active chassis system according to the invention; and FIG. 4 shows a block diagram of a control device for the chassis system according to the invention according to FIG. 3. FIG. 5 shows a detailed illustration of the block diagram of FIG. 4.

In all figures of the drawing, the same or functionally identical parts and / or assemblies - unless otherwise stated - have been given the same reference numerals.

1 shows a simplified schematic illustration of a vehicle with a semi-active chassis system according to the prior art.

In Fig. 1, the reference numeral 1 designates a vehicle body which is shown in a highly simplified manner. Front wheels 26 and rear wheels 27 of the vehicle, as well as actuators front wheels 2 and actuators rear wheels 3 are shown in simplified form. The actuating devices are front wheels and rear wheels 2, 3, for example adjustable shock absorbers with springs and connect the vehicle body 1 to the respective wheels 26, 27 via axles (not shown). A first level sensor 4 at the front left and a second level sensor 5 at the front right each measure the distance between the Vehicle body 1 for the respective axle. At the rear of the vehicle, the distance of the vehicle body 1 to the axle or to the stabilizer of the rear wheels 27 is measured via a third level sensor 6. Furthermore, the vehicle body 1 is equipped on the front in the area of its front axle 25 in its connection area with a first acceleration sensor 7 at the front left, a second acceleration sensor 8 at the front right and at the rear with a third acceleration sensor 9 at the rear right.

A coordinate system is shown at the top left in FIG. 1, which shows the axes x (direction of travel), y and the vertical axis z. The movement of the vehicle body 1 is divided into three degrees of freedom in this simplified linear vehicle model:

A stroke movement ZZ in the direction of the z-axis, a so-called roll movement KAPPA as a swivel movement around the x-axis (represented by a semicircle with arrow head) and a so-called pitch movement THETA as a swivel movement around the y-axis (represented by a semicircle with arrow) ,

Information from the signals from the level sensors 4, 5, 6 and the acceleration sensors 7, 8, 9 is used for the calculations of the above-mentioned speeds in the three degrees of freedom of movement. This takes place in a control device 10, as shown in FIG. 2.

FIG. 2 shows a block diagram of a control device for the chassis system according to FIG. 1 according to the prior art. For example, the chassis system can be a so-called ADS air suspension system (ADS = Adaptive Damping System). The known control device 10 consists of a block for signal processing 11. The respective signals of the level sensors 4, 5, 6 and the acceleration sensors 7, 8, 9 serve as input signals, with A_Z_VL the signal of the first acceleration sensor 7 at the front left, A_Z_VR the signal of the second Acceleration sensor 8 front right, AZH the sig- nal of the third acceleration sensor 9 rear right, N_VL the signal of the first level sensor 4 front left, N_VR the signal of the second level sensor 5 front right and N_HM the signal of the third level sensor 6 rear center.

The signal processor 11 calculates the necessary values for the control algorithm 12 of the control device 10. The control algorithm 12 uses this to generate certain control signals from its output 24 for the front wheel and rear wheel control devices 2, 3, which are not shown here, in accordance with the respective driving situation of the vehicle.

3 shows a simplified schematic illustration of a vehicle with a semi-active chassis system according to the invention. The simplified representation of the vehicle body 1 and the other components corresponds to FIG. 1 except for the point marked by a dashed circle in the rear area of the vehicle body 1. At this point the third acceleration sensor 9 is omitted at the rear right. The front acceleration sensors 7 remain at the front left and 8 at the front right and retain their installation location. The level sensors 4, 5, 6 also remain in their original location.

The use of the information from the three level sensors 4, 5, 6 in conjunction with the two remaining acceleration sensors 7, 8 enables the third acceleration sensor 9 to be omitted without replacement at the rear right. The sensor loss is compensated for by means of a software component in the control device 10, as shown in FIG. 4.

FIG. 4 shows the block diagram of the control device 10 similar to FIG. 2. The control device 10 has the already existing device 12 for calculating the Control signals. Within the first device 11 for signal processing there is a second device 13 sensor fusion algorithm which compensates for the signals of the omitted acceleration sensor 9. The input signals of the signal processing 11 remain except for the signal from the omitted acceleration sensor 9 A_Z_HR.

FIG. 5 shows a detailed illustration of the block diagram of FIG. 4, which is used in particular for calculating the wheel-related, vertical assembly speeds above the wheels 26, 27, which in turn are input variables in the control algorithm 12. The input signals of the acceleration sensors 7, 8 and the level sensors 4, 5, 6, as described in FIGS. 2 and 4, are connected to the signal processing unit 11. The signal processor 11 passes on its output signals to a fourth device 14 for calculating spring and damper forces. Their output signals are processed in a fifth device 15 for calculating stroke and pitching accelerations. At the same time, the input signals A_Z__VL of the acceleration sensor 7 at the front left and A_Z_VR of the acceleration sensor 8 at the front right are used by a sixth device 16 for calculating roll acceleration and a seventh device 17 for calculating superimposition of stroke and pitching accelerations. The respective output signals of devices 16 and 17 and the output signals of device 15 are used as input signals in addition to the signals coming from device 15 for an eighth device 18 for calculating modal speeds.

The device 18 has an integration level 19 and a weighting level 20. The modal speeds calculated in device 18 are used as signals VZ as signal stroke speed, VJKAPPA as signal roll speed and V_THETA as signal pitch speed are forwarded to a ninth device 21 for calculating wheel-related body speeds. The device 21 has a tenth device 22 for signal postprocessing in its input stage. The wheel-related body speeds generated in the device 21 are used as input signals for the known and unchanged device 12 for calculating the control signals.

It is advantageous that the output 24 of the control algorithm generation 12 is connected via a feedback connection 23 to an input of the device 14 for calculating spring and damper forces. This is the only connection between the control algorithm 12 and the software component sensor fusion algorithm 13. In addition to this, the aforementioned functional parts are completely decoupled, so that the effort required to change the control unit software is advantageously minimal.

The second device 13 is shown in FIG. 4 with dash-dotted lines.

The functional principle of the sensor fusion algorithm is shown below using the calculation processes of the individual devices as shown in FIG. 5. The vertical speed signals of the vehicle body 1 above the rear axle are reconstructed from four sensor equations evaluated in each time step. Two of these are based on the acceleration sensors 7, 8 of the vehicle body 1 and two on a force and torque balance. There are therefore four equations for calculating three modal speeds (vertical stroke, roll, pitch). The linear system of equations is thus over-determined. The over-determination can be in resolve each time step using the so-called "Weighted Linear Least Squares Method" - that is, using the weighted, smallest error squares method.

In device 15, the stroke and pitching accelerations are calculated on a force and moment balance basis. Device 15 determines the stroke and pitch acceleration on the basis of four strut forces of the vehicle and a fixed vehicle mass or pitch inertia.

In addition to the two equations for lifting and pitching accelerations (see device 15), the device 16 determines a further equation for determining the modal accelerations of the vehicle body 1. Based on the signals A_Z_JVL from the acceleration sensor 7 at the front left and the signal A_Z_VR from the acceleration sensor 8 at the front right with the difference of which the roll acceleration of the vehicle is calculated by means of the distance between the two acceleration sensors 7, 8 remaining in the vehicle. Thus, despite the omission of the third acceleration sensor 9, this comfort-relevant variable continues to be directly available as a measurement signal for the control algorithm 12 at the rear right.

Furthermore, the two signals A_Z_VL of the acceleration sensor 7 at the front left and A_Z_VR of the acceleration sensor 8 at the front right are used for the calculation in device 17 for a variable which describes the superimposition of lifting and pitching accelerations of the vehicle body 1 over the front axle 25. There is therefore another equation for the subsequent determination of the three accelerations in the degrees of freedom of the vehicle body 1.

Exemplary calculation procedures for the above four equations are given below. The four equations are first summarized in a matrix:

-*Vi?,/>r )

- * Vi?, /> R)

A X = B

The variables mean: z stroke acceleration k roll acceleration θ pitch acceleration a _z , vi body acceleration front left a _z , vr body acceleration front right FFB.VI FpB. r PpB, hi # F _FB , hr strut forces front left / right, rear left / right m total mass of the vehicle

I ₀ moment of inertia of the vehicle s distance of body acceleration sensors (y direction) l _v distance of vehicle center of gravity to strut linkage points VA (x direction) l _h distance of vehicle center of gravity to strut linkage points HA (x direction)

In the weighting stage 20, the so-called weighted linear least squares method is used in order to best solve the overdetermined linear system of equations in terms of the quality of the signals used at each time step. With a weighting matrix W, the following equation results for the modal accelerations x:

x (t) = (A ^T WA) -1 A ^τ W b (t) with

* 0 0 0 0

W = w, 0 0, where the Wi is the weightings for the 0 0 w ₂ 0 0 0 0 w,

Lift and pitch accelerations are.

In integration stage 19, a so-called pseudo-integration takes place according to the equation:

In order to convert from the modal acceleration signals to modal speeds, a pseudo integration (Pl) is necessary. This filters out high-frequency as well as DC components from the acceleration signals and realizes an integration in the body's natural frequency range due to clever amplification-phase relationships.

Following the calculation of the lifting, rolling and pitching speeds, these are converted into vertical body speeds above the individual wheels using the geometric relationships. The following equation applies to the wheel-related, vertical body speeds v_vl, v_vr, v_hl, v_hr (vl = front left, vr = front right, hl = rear left, hr = rear right):

The device 12 control algorithm generation receives the wheel-related, vertical assembly speeds as input signals from device 13 with sensor fusion algorithm. This means that only the sensor fusion algorithm has been added to the previous series version. Only information about the actuation of the actuating devices front wheels and rear wheels 2, 3 via, for example, valves is required in the form of the feedback connection 23 for the damper force calculation in block 14 calculation of spring and damper forces.

Test results have already shown that the semi-active chassis system with only a first and second acceleration sensor 7 and 8 enables identical driving comfort as with a chassis system of the prior art with a third acceleration sensor 9. Due to the replacement of the third acceleration sensor 9 on the rear right without replacement Vehicle body 1 also advantageously dispenses with the wiring from this acceleration sensor 9 to the control unit and the control unit connection / processing of the sensor information which is omitted. The sensor fusion algorithm 13 is implemented in the existing control device 10 in the software area.

Although the present invention has been described above on the basis of a preferred exemplary embodiment, it is not restricted to this but can be modified in a variety of ways. For example, it is conceivable that the invention can be used not only in a chassis system with an air spring, but also in chassis systems with other components, for example in active chassis. It is also conceivable that the remaining acceleration sensors 7, 8 are installed or attached to the vehicle body 1 at different installation locations.

It is also possible to implement the second device with a Cayman filter design.

Claims

Daimler Chrysler AG patent claims Semi-active chassis system with a reduced number of sensors for vehicles, in particular motor vehicles, with level sensors (4, 5, 6) which record a change in the path of a vehicle body (1) between it and the axles in the vertical direction and first sensor signals (N_VL, N_VR, N_HM) ready; Acceleration sensors (7, 8), which record accelerations of the vehicle body (1) in the vertical direction and provide dependent second sensor signals (A_Z_VL, A_Z_VR); A control device (10) for providing control signals for actuators (2, 3), the control device (10) ^■ a first device (11) for signal processing and processing, ■ a second device (13) for calculating state variables of the vehicle body ( 1), based on the first and second sensor signals, which has a sensor fusion algorithm with which at least one further sensor signal is compensated and ^{■ contains} a third device (12) for generating the control signals for the actuating devices (2, 3); Setting devices (2, 3) for adjusting the vehicle body (1) in the vertical direction. 2. Suspension system according to claim 1, characterized in that a first and a second acceleration sensor (7, 8) above the front axle (25) of the vehicle are arranged at a distance from one another. 3. Suspension system according to claim 1 or 2, characterized in that the sensor fusion algorithm of the second device (13) of the control device (10) has the following devices: - a fourth device (14) for calculating spring and damper forces, a fifth device ( 15) for calculating the lifting and pitching accelerations of the vehicle body (1), a sixth device (16) for calculating the rolling acceleration of the vehicle body (1), - a seventh device (17) for calculating the superimposition of the lifting and pitching acceleration of the vehicle body ( 1), - an eighth device (18) for calculating state variables of the vehicle body (1). 4. Suspension system according to claim 3, characterized in that the sixth device (16) and the seventh device (17) parallel to each other and in each case parallel to the first, fourth and fifth device (11, 14, 15) are arranged and each have a calculation stage for their output signals from the signals of the first (7) and second (8) acceleration sensors. 5. Suspension system according to claim 3 or 4, characterized in that the eighth device (18) includes an integration stage (19) for integrating acceleration signals and a weighting stage (20) for improving the quality of the calculated state variables. 6. Chassis system according to one of claims 3 to 5, characterized in that a feedback connection (23) is provided, which between the output (24) of the third device (12) for generating control algorithms for the control signals of the actuating devices and the fourth device (14) a feedback connection (23) is arranged to calculate spring and damper forces. 7. A method for generating control signals for actuating devices (2, 3) of a semi-active chassis system, in particular a semi-active chassis system according to one of claims 1 to 10, with the method steps: (a) conditioning sensor signals; (b) calculating spring and damper forces; (c) calculating lifting and pitching accelerations of the vehicle body (1); (d) calculating roll acceleration of the vehicle body (1); (e) calculating the superimposition of the lifting and pitching acceleration of the vehicle body (1); (f) calculating state variables (V_Z, V KAPPA, V THETA) of the vehicle body (1); (g) evaluating a control algorithm or control algorithm using the state variables from method step (f); and (h) generating drive signals. 8. The method according to claim 7, characterized in that the method steps (d) and (e) with signals (A_Z_VL, A_Z_VR) of the first and second acceleration sensors (7, 8) are carried out. 9. The method according to claim 7 or 8, characterized in that the method step (f) has the following method substeps: (al) pseudo-integrating acceleration signals, which are obtained in the method steps (c), (d) and (e), and (a2) Weights of the signals integrated in this way to improve the quality of the calculated state variables. 10. The method according to claim 9, characterized in that the partial process step (a2) includes weighting by means of the so-called weighted linear least squares method. 11. The method according to any one of claims 7 to 10, characterized in that method step (b) is additionally carried out with feedback control signals which are obtained in method step (h).