DE102022204011A1 - Method for monitoring a mechanically uncoupled actuator system and mechanically uncoupled actuator system - Google Patents

Abstract

The invention relates to a method for monitoring a mechanically uncoupled actuator system (1), wherein the actuator system (1) has a feedback actuator module (3) and an actuator module (4), which are coupled to one another in terms of signals, the feedback actuator module (3) and the actuator module ( 4) are each monitored by means of an external diagnosis, with a passive external diagnosis and/or an active external diagnosis being carried out outside the module, with the passive external diagnosis using a respective model (13, 14) for the feedback actuator module (3) and the actuator module (4 ) a respective function is checked, and wherein as part of the active external diagnosis, a coupling between the feedback actuator module (3) and the actuator module (4) is temporarily canceled and the feedback actuator module (3) and / or the actuator module (4) are controlled with predetermined test parameters in order to check a respective function based on a model, and wherein a checking result (20) is provided. The invention further relates to a mechanically uncoupled actuator system (1).

Description

The invention relates to a method for monitoring a mechanically uncoupled actuator system and a mechanically uncoupled actuator system.

Steer-by-wire (SbW) steering systems, which can be used in large series in the automotive sector in the future, must have significantly higher reliability than conventional steering systems. This is because SbW steering systems have no mechanical connection between a steering wheel and a steering gear. A complete failure of the steering system would result in the vehicle no longer being steerable. A complete failure must therefore be prevented and the effects of errors must be recognized early on. Furthermore, automated driving assistance systems also require increased safety of the vehicle components.

In general, in mechanically uncoupled actuator systems, that is, in “by-wire” systems, it is no longer possible for a human operator (e.g. a driver or pilot) to directly access an actuator in a mechanically mediated manner.

From the DE 100 53 335 B4 a SbW steering system for vehicles is known, with a steering module that acts on the position of the steered wheels of the vehicle, with a steering rotation angle sensor that detects the position of the steered wheels, with a steering wheel module that transmits the effects of the road on the vehicle to the driver, with a Steering wheel rotation angle sensor that detects the driver's steering request, with a supply module for supplying the SbW steering system with electrical energy and with a fallback level, the steering module, steering rotation angle sensor, steering wheel module, steering wheel rotation angle sensor and supply module being connected to one another via signal lines and the steering module, steering rotation angle sensor, steering wheel module and steering wheel rotation angle sensor being connected to the supply module via electrical supply lines are supplied with electrical energy, wherein the steering module, steering rotation angle sensor and steering wheel rotation angle sensor are constructed redundantly, and wherein the signal lines and supply lines are present redundantly, wherein the steering wheel module is constructed redundantly, wherein the steering module and the steering wheel module have at least two computer modules, and wherein the function of the at least Two computer modules are each monitored by a monitoring module.

The invention is based on the object of improving a method for monitoring a mechanically uncoupled actuator system and a mechanically uncoupled actuator system.

The object is achieved according to the invention by a method with the features of patent claim 1 and a mechanically uncoupled actuator system with the features of patent claim 10. Advantageous embodiments of the invention result from the subclaims.

In particular, a method for monitoring a mechanically uncoupled actuator system is provided, wherein the actuator system has a feedback actuator module and an actuator module, which are coupled to one another in terms of signals, the feedback actuator module and the actuator module each being monitored by means of an external diagnosis, with a passive external diagnosis and /or an active external diagnosis is carried out, wherein as part of the passive external diagnosis a respective function is checked based on a respective model for the feedback actuator module and the actuator module, and whereby as part of the active external diagnosis a coupling between the feedback actuator module and the actuator module is temporarily canceled and that Feedback actuator module and / or the actuator module are controlled with predetermined test parameters in order to check a respective function based on a model, and a check result is provided.

Furthermore, in particular, a mechanically uncoupled actuator system is created, comprising a feedback actuator module and an actuator module, which are coupled to one another in terms of signals, the actuator system being set up to monitor the feedback actuator module and the actuator module in each case by means of an external diagnosis, and for this purpose a passive external diagnosis and in each case external to the module /or to carry out an active external diagnosis, and as part of the passive external diagnosis to check a respective function based on a respective model for the feedback actuator module and the actuator module, and as part of the active external diagnosis to temporarily cancel a coupling between the feedback actuator module and the actuator module and the feedback actuator module and/or or to control the actuator module with specified test parameters in order to check a respective function based on a model and to provide a verification result.

The method and the actuator system make it possible to monitor the feedback actuator module and the actuator module in an improved manner. In particular, in addition to the usual self-diagnosis, an external diagnosis is also possible light. This is achieved by carrying out a passive external diagnosis and/or an active external diagnosis external to the module. As part of the passive external diagnosis, a respective function is checked using a respective model for the feedback actuator module and the actuator module. For this purpose, relationships between input variables and output variables and thus the behavior of the modules are checked with the help of the respective model. As part of the active external diagnosis, a coupling between the feedback actuator module and the actuator module is temporarily canceled and the feedback actuator module and / or the actuator module is controlled with predetermined test parameters. This only ever happens if an active third-party diagnosis is possible, that is, if the actuator system is not currently being operated. This allows a specific function to be checked on a model-based basis. In particular, the aim of active third-party diagnostics is to check sensors, power electronics and actuators for correct functioning. In particular, parameters or parameter ranges can be implemented that do not occur in normal operation, such as edge or extreme areas. The verification result is provided, for example as an analog or digital signal, in particular as a digital data packet.

A mechanically uncoupled actuator system is in particular an actuator system in which there is no mechanical connection between a feedback actuator module, that is, a module for detecting a manual actuation, and an actuator module, that is, a module for executing the action derived from the manual actuation. but only a signaling connection exists (the actuator system can therefore also be referred to as a “by-wire” system). The manual actuation is therefore not implemented mechanically into the action, but rather is detected in particular by means of a sensor system and implemented via the signaling connection using an actuator system. In the same way, feedback from the actuator module to the feedback actuator module only takes place via signaling technology. A feedback actuator module in particular includes a controller for controlling or regulating feedback that is given to a person during a manual operation. An actuator module in particular includes a controller for controlling or regulating an action derived from the detected manual actuation. The modules include in particular data processing devices that carry out the control and/or regulation and in particular a self-diagnosis and in particular an external diagnosis.

The models can fundamentally be designed differently; for example, characteristics and/or characteristic maps can be used as the respective model. A machine learning method, for example a trained neural network, can also be used as a model. The neural network is then trained to estimate output variables for each queried input variable, which can be compared with queried output variables to check a function of the modules. The models represent the ideal behavior of the modules, especially of the modules' electric motors. It can be provided that the aging behavior of the modules, in particular the motors of the modules, is taken into account, for example by adapting the models depending on the operating time. This means that expected, but uncritical, aging of the modules or motors and/or other components in the models can be taken into account.

• - Eingangsgröße: Phasenspannung, Ausgangsgröße: Phasenstrom;
• - Eingangsgröße: Spannungsimpuls, Ausgangsgröße: Verlauf eines Stroms, hierüber können insbesondere Induktivitäten und Widerstände einzelner Phasen bestimmt werden;
• - Eingangsgröße: Drehungen des Motors, Ausgangsgröße: Verschiebung einer Zahnstange (Position);
• - Eingangsgröße: Zahnstangenposition, Ausgangsgröße: Reaktion eines Fahrzeugs (z.B. Querbeschleunigung, Gierrate etc.);
• - Eingangsgröße: Lastimpuls, Ausgangsgröße: Einbruch einer Spannung an einer Spannungsversorgung, hierüber kann insbesondere ein Widerstand eines Versorgungspfades bestimmt werden.

Input variables can be, for example, phase voltages and/or phase currents. Output variables can be, for example, phase currents and/or motor angles. In principle, input variables and output variables can be chosen arbitrarily, as long as they allow a model-based check of the function of the modules (especially the motors). With the combination of respective phase currents as input variables and a respective motor angle as output variable, it can be checked, for example, whether an estimated or predicted motor angle is set for given phase currents or not. For example, an electrical resistance of the motor can be determined and checked from a combination of respective phase voltages as input variables and respective phase currents as output variables. If electrical resistance increases over time, this may indicate that the motor is aging. Such aging is then detected, for example with the help of a threshold value comparison. Furthermore, the combinations of input variables and output variables listed below can also be used for model-based checking:

• - Input variable: phase voltage, output variable: phase current;
• - Input variable: voltage pulse, output variable: course of a current, this can be used to determine inductances and resistances of individual phases;
• - Input variable: rotations of the motor, output variable: displacement of a rack (position);
• - Input variable: rack position, output variable: reaction of a vehicle (e.g. lateral acceleration, yaw rate, etc.);
• - Input variable: load pulse, output variable: voltage drop on a power supply; this can be used in particular to determine the resistance of a supply path.

A vehicle is in particular a motor vehicle. In principle, a vehicle can also be another land, rail, water, air or space vehicle.

In one embodiment it is provided that the feedback actuator module and the actuator module each monitor each other and carry out the respective external diagnosis. This means that existing modules can be used to carry out the monitoring. In particular, no powerful central computer is necessary or no computing power is required for such a central computer.

In one embodiment it is provided that monitoring is carried out additionally or alternatively using at least one external monitoring device. As a result, monitoring can be carried out additionally or alternatively using an additional device. In particular, this allows central monitoring and/or bundling of monitoring activities in one facility. In the case of a vehicle, the additional device can be, for example, a central computing device (central computer) of the vehicle. Furthermore, a backend server can also be such an additional device, so that remote diagnosis is also possible. For this purpose, the actuator system is connected to the backend server via a communication connection, for example a mobile phone network or the Internet.

In one embodiment it is provided that input variables and output variables of the feedback actuator module and/or the actuator module are collected and stored, with the external diagnosis being carried out based on the collected and stored variables. This allows the external diagnosis to be carried out in particular at a later time and/or at a different location. In particular, the external diagnosis can be carried out at a time when computing power is (cheaply) available.

In one embodiment it is provided that the passive external diagnosis is carried out continuously. This means that a current check result is continuously available. In this way, errors can be detected without a long delay.

Alternatively, in one embodiment it can be provided that the passive external diagnosis is carried out in predetermined situations and/or at predetermined times. In this way, the computing power required for third-party diagnosis can be saved and/or used in a targeted manner.

In one embodiment it is provided that the active external diagnosis is carried out in predetermined situations and/or at predetermined times. This means that the active third-party diagnosis can always be carried out in suitable situations and/or at suitable times. In the case of an SbW steering system, the active external diagnosis can, for example, always be carried out before the vehicle starts driving (e.g. after activating the ignition) or after the vehicle ends. Furthermore, a rest phase, for example when the vehicle is waiting at a traffic light, can also be used to carry out an active external diagnosis.

In one embodiment it is provided that parameters of at least one vehicle are retrieved and/or received and are taken into account in the passive external diagnosis and/or the active external diagnosis. In particular, this increases a basis for comparison and thus increases diagnostic coverage and robustness against misdiagnoses. For example, parameters of the vehicle in which the actuator system is installed and/or parameters can be retrieved directly from other vehicles. Furthermore, parameters of other vehicles can also be retrieved from a backend server. Parameters can include, for example, a speed and/or an acceleration of the vehicle, a yaw rate, a lateral acceleration, camera data and/or radar data, etc. The parameters can be used, for example, to identify driving maneuvers, from which ideal models can be generated using pattern recognition. In the example of a steering system, such an ideal model can, for example, contain a dependence of a lateral acceleration on a curve shape and a speed of the vehicle.

In one embodiment it is provided that the mechanically uncoupled actuator system is a steer-by-wire (SbW) steering system. The SbW steering system includes a steering wheel module or force feedback actuator (FFA) as a feedback actuator module, which detects a desired steering angle and provides a driver with a realistic hand torque. The SbW steering system also includes a steering module or Road Wheel Actuator (RWA) as an actuator module with a steering gear, which converts the driver's request into a wheel steering movement.

Further features for the design of the actuator system result from the description of embodiments of the method. The advantages of the actuator system are the same as in the embodiments of the method.

• 1 eine schematische Darstellung einer Ausführungsform des mechanisch ungekoppelten Aktuatorsystems am Beispiel eines Steer-by-Wire-Lenksystems;
• 2 ein schematisches Ablaufdiagramm einer Ausführungsform des Verfahrens zum Überwachen eines mechanisch ungekoppelten Aktuatorsystems.

The invention is explained in more detail below using preferred exemplary embodiments with reference to the figures. Show here:

• 1 a schematic representation of an embodiment of the mechanically uncoupled actuator system using the example of a steer-by-wire steering system;
• 2 a schematic flow diagram of an embodiment of the method for monitoring a mechanically uncoupled actuator system.

The 1 shows a schematic representation of an embodiment of the mechanically uncoupled actuator system 1 using the example of a steer-by-wire steering system 2.

The mechanically uncoupled actuator system 1 includes a feedback actuator module 3 and an actuator module 4, which are coupled to one another in terms of signals. In particular, the feedback actuator module 3 and the actuator module 4 are connected to one another via a (private) Controller Area Network (CAN) bus 5. In the example shown of the steer-by-wire steering system 2, the feedback actuator module 3 is a steering wheel module or Force Feedback Actuator (FFA) module and the actuator module 4 is a steering module or Road Wheel Actuator (RWA) module.

The actuator system 1 is set up to monitor the feedback actuator module 3 and the actuator module 4 each by means of an external diagnosis, and for this purpose to carry out a passive external diagnosis and/or an active external diagnosis external to the module.

As part of the passive external diagnosis, a respective function is checked using a respective model 13, 14 for the feedback actuator module 3 and the actuator module 4.

As part of the active external diagnosis, a coupling between the feedback actuator module 3 and the actuator module 4 is temporarily canceled and the feedback actuator module 3 and/or the actuator module 4 is controlled with predetermined test parameters in order to check a respective function based on a model.

A check result 20 is provided, for example by outputting a corresponding message on the CAN bus 5, which can then be evaluated by other devices in the vehicle and/or transmitted to a backend server 60.

It can be provided that the feedback actuator module 3 and the actuator module 4 monitor each other and carry out the respective external diagnosis. As part of the passive external diagnosis, the feedback actuator module 3 queries input variables 4-1 and output variables 4-2 of the actuator module 4 via the CAN bus 5 at the actuator module 4 and the actuator module 4 queries input variables 3-1 and output variables 3-2 of the feedback actuator module 3 via the CAN bus 5 at the feedback actuator module 3. For example, phase voltages of a respective electric motor can be queried as input variables 3-1, 4-1 and phase currents of the respective electric motor can be queried as output variables 3-2, 4-2. Alternatively or additionally, motor angles and/or other variables can also be queried. The feedback actuator module 3 checks a function of the actuator module 4 using a model 14, which represents an ideal behavior of the actuator module 4. In the example mentioned, it is checked in particular whether the phase currents match the given phase voltages or whether there are deviations (e.g. an increased electrical resistance), which indicate, for example, a defect or aging of the motor. The actuator module 4 checks a function of the feedback actuator module 3 using a model 13, which represents an ideal behavior of the feedback actuator module 4. In the example mentioned, it is also checked in particular whether the phase currents match the given phase voltages or whether there are deviations, for example indicate a defect or aging of the engine. The check result 20 (comprising, for example, information as to whether a deviation is above a predetermined threshold value or not) is then output.

The models 13, 14 can fundamentally be designed differently, for example characteristic curves and/or characteristic fields or machine learning methods can be used, for example a trained neural network which is trained to produce output data 3-1 for each queried input data 3-1, 4-1. 3, 4-3, which can be compared with the queried initial data 3-2, 4-2.

It can be provided that monitoring is carried out additionally or alternatively by means of at least one external monitoring device 30. This can be, for example, a central computer 51 of the vehicle 50. The external monitoring device 30 can also be one Be backend server 60, which is connected to vehicle 50 by means of a communication connection 61. Basically, the procedure is the same as in the embodiment in which the modules 3, 4 monitor each other.

It can be provided that input variables 3-1, 4-1 and output variables 3-2, 4-2 of the feedback actuator module 3 and/or the actuator module 4 are collected and stored, with the external diagnosis starting from the collected and stored variables 3-x , is carried out 4 times. For this purpose, the input variables 3-1, 4-1 and output variables 3-2, 4-2 are stored in a memory (not shown) of the feedback actuator module 3 and/or the actuator module 4 and/or the backend server 60 and then used at a later point in time of the respective model 13, 14 checked. This later point in time can be determined and/or fixed, for example, depending on the available computing power.

It can be provided that the passive external diagnosis is carried out continuously. This is done by mutual checking by the feedback actuator module 3 and/or the actuator module 4 or by the central computer 51 and/or the backend server 60, for which purpose the input variables 3-1, 4-1 and output variables 3-2, 4-2 are continuously displayed the other module 3, 4 or the central computer 51 and / or the backend server 60 are transmitted.

Alternatively, it can be provided that the passive external diagnosis is carried out in predetermined situations and/or at predetermined times. For example, passive external diagnosis can be carried out at regular intervals. Alternatively or additionally, the passive third-party diagnosis can also be carried out whenever free computing power is available, for example because other processes require less computing power.

It can be provided that the active external diagnosis is carried out in predetermined situations and/or at predetermined times. For example, it can be provided that the active external diagnosis is always carried out after starting the vehicle 50 before starting the journey. Furthermore, the checking can also take place after parking before switching off the vehicle 50 or while the vehicle 50 is at a standstill at a red traffic light phase. For example, with modules 3, 4 decoupled from one another, different test parameters can then be run through, for example different torques can be requested from the steering wheel module (FFA) and/or the steering module (RWA) and the relationship between the requested torques, the phase currents and the phase voltages can be determined using the models 13 , 14 be checked. In particular, extreme values that do not occur during normal operation can be implemented.

It can be provided here that parameters of at least one vehicle 50 are called up and/or received and are taken into account in the passive external diagnosis and/or the active external diagnosis. This is done in particular via the CAN bus 5 in a vehicle control (not shown) of the vehicle 50 and/or via a communication interface (not shown) set up for this purpose, for example a V2X interface, in at least one other vehicle. This means that swarm data from a large number of vehicles can also be taken into account. In particular, for example, typical courses and/or relationships and/or patterns between the input variables 3-1, 4-1 and output variables 3-2, 4-2 can be recognized and for checking the input variables 3-1, 4-1 and output variables 3-2 , 4-2 of the feedback actuator module 3 and/or the actuator module 4 can be used. For this purpose, for example, a deviation from typical courses and/or relationships and/or patterns averaged over the swarm data can be determined and compared with threshold values in order to determine whether the feedback actuator module 3 and/or the actuator module 4 are working properly.

The 2 shows a schematic flow diagram of an embodiment of the method for monitoring a mechanically uncoupled actuator system. This is based on the exemplary SbW steering system of a vehicle. The method is carried out in particular by means of an actuator system according to one of the embodiments described above.

A measure 100 checks whether a situation and/or a time specified for an active external diagnosis exists. This is the case, for example, if the vehicle has just been started, is waiting at a red light and/or has been parked.

If it was determined in measure 100 that a specified situation and/or a specified time for an active third-party diagnosis does not exist, the process continues with measure 200. Otherwise, proceed to action 300.

In measure 200, a passive external diagnosis is carried out outside the module. Measure 200 includes measures 201a to 201d and measures 202a to 202d. It is For example, it is provided that the feedback actuator module and the actuator module each monitor each other and carry out the respective external diagnosis.

In measure 201a, the feedback actuator module queries input variables and output variables of the actuator module, for example phase voltages and associated phase currents.

In measure 201b, the phase voltages are supplied as input data to a model of the actuator module, in particular a motor model. Using the model, phase currents are estimated in addition to the phase voltages. The estimated phase currents are compared with the queried phase currents and a deviation is determined. To determine the deviation, a suitable measure is chosen, for example a mean square error of values over time.

In measure 201c, the determined deviation is compared with a specified threshold value. A comparison result is provided in measure 201d as a verification result, in particular in the form of an analog or digital signal, for example as a digital data packet.

The measures 202a to 202d are measures that the actuator module carries out in an analogous manner to check the feedback actuator module.

Measure 300 includes measures 301 and 302a to 302e as well as 303a to 303e. In measure 301, the feedback actuator module and the actuator module are decoupled in terms of signals. A torque applied to the feedback actuator module then no longer causes a torque to be applied to the actuator module and vice versa, so that the modules can be controlled separately and independently of one another.

In measure 302a, the feedback actuator module controls the actuator module with predetermined test parameters. In particular, a motor of the actuator module is controlled with the specified test parameters. The specified test parameters (specified motor torque, specified phase voltages, etc.) can in particular also include values that do not occur during normal operation of the modules.

In measure 302b, input variables and output variables of the actuator module are queried during control. In particular, phase voltages are queried as input variables and phase currents of the motor of the actuator module are queried as output variables.

In measure 302c, the phase voltages are supplied as input data to a model of the actuator module, in particular a motor model. Using the model, phase currents are estimated in addition to the phase voltages. The estimated phase currents are compared with the queried phase currents and a deviation is determined. To determine the deviation, a suitable measure is chosen, for example a mean square error of values over time.

In measure 302d, the determined deviation is compared with a predetermined threshold value. A comparison result is provided in measure 302e as a verification result, in particular in the form of an analog or digital signal, for example as a digital data packet.

Measures 303a to 303e are measures that the actuator module carries out in an analogous manner to check the feedback actuator module.

In a measure 304, the signaling connection between the feedback actuator module and the actuator module is restored.

You then continue with measure 100 again.

Further embodiments of the method have already been described above with reference to embodiments of the actuator system. The measures are always carried out in an analogous manner.

Reference symbol list

1

Actuator system

2

Steer-by-wire steering system

3

Feedback actuator module

3-1

Input variables

3-2

Output variables

4

Actuator module

4-1

Input variables

4-2

Output variables

5

CAN bus

13

Model (feedback actuator module)

14

Model (actuator module)

20

Verification result

30

external monitoring device

50

vehicle

51

Central computer

60

Backend server

61

Communication link

100

measure

200-202

Measures

300-304

Measures

QUOTES INCLUDED IN THE DESCRIPTION

This list of documents listed by the applicant was generated automatically and is included solely for the better information of the reader. The list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions.

Cited patent literature

• DE 10053335 B4 [0004]

Claims (10)

Method for monitoring a mechanically uncoupled actuator system (1), the actuator system (1) having a feedback actuator module (3) and an actuator module (4), which are coupled to one another in terms of signals, wherein the feedback actuator module (3) and the actuator module (4) are each monitored by means of an external diagnosis, whereby a passive external diagnosis and/or an active external diagnosis is carried out externally to the module, wherein as part of the passive external diagnosis, a respective function is checked using a respective model (13, 14) for the feedback actuator module (3) and the actuator module (4), and wherein as part of the active external diagnosis, a coupling between the feedback actuator module (3) and the actuator module (4) is temporarily canceled and the feedback actuator module (3) and / or the actuator module (4) are controlled with predetermined test parameters in order to check a respective function based on a model , and wherein a verification result (20) is provided. Procedure according to Claim 1 , characterized in that the feedback actuator module (3) and the actuator module (4) each monitor each other and carry out the respective external diagnosis. Procedure according to Claim 1 or 2 , characterized in that the monitoring is carried out additionally or alternatively by means of at least one external monitoring device (30). Method according to one of the preceding claims, characterized in that input variables (3-1,4-1) and output variables (3-2,4-2) of the feedback actuator module (3) and/or the actuator module (4) are collected and stored, whereby the external diagnosis is carried out based on the collected and stored variables (3-x,4-x). Method according to one of the preceding claims, characterized in that the passive external diagnosis is carried out continuously. Procedure according to one of the Claims 1 until 4 , characterized in that the passive external diagnosis is carried out in predetermined situations and/or at predetermined times. Method according to one of the preceding claims, characterized in that the active external diagnosis is carried out in predetermined situations and/or at predetermined times. Method according to one of the preceding claims, characterized in that parameters of at least one vehicle (50) are retrieved and/or received and are taken into account in the passive external diagnosis and/or the active external diagnosis. Method according to one of the preceding claims, characterized in that the mechanically uncoupled actuator system (1) is a steer-by-wire steering system (2). Mechanically uncoupled actuator system (1), comprising a feedback actuator module (3) and an actuator module (4), which are coupled to one another in terms of signaling, wherein the actuator system (1) is set up to monitor the feedback actuator module (3) and the actuator module (4) each by means of an external diagnosis, and for this purpose to carry out a passive external diagnosis and/or an active external diagnosis external to the module, and as part of the passive external diagnosis, to check a respective function using a respective model (13, 14) for the feedback actuator module (3) and the actuator module (4), and As part of the active external diagnosis, temporarily cancel a coupling between the feedback actuator module (3) and the actuator module (4) and control the feedback actuator module (3) and / or the actuator module (4) with predetermined test parameters in order to check a respective function based on a model, and to provide a verification result (20).

Images