Method for suspension control of an active suspension system of a motor vehicle, motor vehicle, and system comprising a vehicle-external computing device

Abstract

The invention relates to a method for suspension control of an active suspension system of a motor vehicle, the method comprising the steps: - acquiring sensor data (SD) from at least one vehicle sensor and associated contextual data (KD) during a first journey of the motor vehicle (S1); - transmitting those sensor data (SD) that exceed a threshold value, together with the associated contextual data (KD), as a data set to a central electronic computing device (S2); - aggregating a plurality of data sets from journeys which have at least one position in common (S3); - grouping the aggregated data sets according to events derivable therefrom (S4); - determining at least one characteristic parameter (CK) for each group and / or for each event (S5); - during a further journey of the motor vehicle, determining whether an event is present for a position along this route (S6); - adjusting damping and / or suspension characteristics of the active suspension system on the basis of the characteristic parameter (CK) (S7). The invention also relates to a motor vehicle and to a system.

Description

[0001] 2024P01690WG

[0002] 1

[0003] Mercedes-Benz Group AG

[0004] Method for chassis control of an active chassis of a motor vehicle, motor vehicle and system with a vehicle-external computing device

[0005] The invention relates to a method for chassis control of an active chassis of a motor vehicle, in which chassis pre-control is carried out using historically aggregated data, according to claim 1. Furthermore, the invention relates to a motor vehicle with an active chassis according to claim 9, and to a system with a vehicle-external, central electronic computing device to which at least two motor vehicles are assigned, according to claim 10.

[0006] Active suspension systems, or vehicle suspensions, often have fast actuators for adjusting spring and / or damping. Ideally, the suspension would respond effectively to large potholes or bumps, such as speed bumps. However, this is often suboptimal today because the exact geometry of the unevenness is unknown. Sensors for geometrically measuring such unevenness, such as stereo cameras, lidars, and the like, are frequently unavailable in production vehicles and / or exhibit significant measurement inaccuracies, especially at higher speeds. Furthermore, potholes are not always fully visible, making it difficult to adjust the suspension accordingly.

[0007] For example, DE 102020 007 770 A1 discloses a method for determining road surface irregularity using a system of a motor vehicle, wherein the road surface irregularity is determined as a function of a detected suspension travel or the suspension travel speed.

[0008] The object of the present invention is to provide a method, a motor vehicle and a system by means of which an adaptation 2024P01690WG can be carried out in a particularly advantageous manner.

[0009] 2 of a chassis, in particular its damping and / or suspension, is enabled to adapt to current road conditions, in particular unevenness of the ground.

[0010] This problem is solved according to the invention by the subject matter of the independent claims. Advantageous embodiments with further developments of the invention are specified in the dependent claims, in the description and in the drawing.

[0011] A first aspect of the invention relates to a method for chassis control of an active chassis of a motor vehicle, in which chassis pre-control is carried out using historically aggregated data in the form of data sets.

[0012] The motor vehicle in question is, in particular, a passenger car or truck. The active suspension system includes, for example, an adjustable actuator at each wheel, which allows for adjustments to the spring stiffness and / or damping characteristics. Thus, the active suspension system enables changes to both the spring rate and the damping. The suspension control system is used to control and / or regulate the corresponding actuators. The suspension control system allows for adjustments to be made based on the expected setpoint curve. Since it is generally a control system, it can improve vehicle handling to ensure stability.

[0013] The method according to the invention comprises the following steps:

[0014] In a first step, sensor data from at least one vehicle sensor, which can detect unevenness in the ground, and associated context data, which at least include the respective position during a first journey of the vehicle, are acquired.

[0015] In a second step, the sensor data exceeding a threshold, along with the associated contextual data, is transmitted as a data set to a central, and therefore, in particular, external electronic computing unit. 2024P01690WG

[0016] 3

[0017] In a third step, several data sets from several journeys are aggregated, for example from the motor vehicle or another motor vehicle which is designed to record the sensor data, whereby data sets are aggregated that have at least one position in common.

[0018] In a fourth step, the aggregated data sets are grouped according to events that can be derived from them, such as driving over a pothole, which can be detected as an unevenness in the ground by at least one vehicle sensor.

[0019] In a fifth step, at least one characteristic parameter of the respective group and / or the respective event is determined, in particular for each chassis control system.

[0020] In a sixth step, during a further journey of the motor vehicle and / or another motor vehicle along a route or track, it is determined whether an event has occurred for a position on this route.

[0021] Subsequently, in a seventh step, damping and / or suspension of the active chassis is adjusted based on at least one characteristic parameter for the event and depending on a current driving condition of the motor vehicle and / or the other motor vehicle, in particular before or until the position associated with the event is reached.

[0022] The term "acquisition" refers specifically to recording and / or storing data in a memory area, such as a ring buffer of the vehicle sensor and / or an associated control unit, like a suspension control unit. The sensor data is characterized in such a way that it is suitable for detecting road irregularities, such as potholes or similar, based on information derived from it. These road irregularities possess properties that would make a change in the damping and / or suspension of an active suspension system advantageous when driving over them. The sensor data can, for example, include the suspension travel and / or the suspension travel speed of a particular tire of the vehicle.

[0023] In addition to the sensor data, contextual data is also recorded, which includes, in particular, a position based on GPS data, for example. Furthermore, the 2024P01690WG

[0024] 4

[0025] Contextual data includes vehicle speed, lane selection, and similar information. In particular, this contextual data is suitable for deriving the respective event based on information extracted from it.

[0026] In the second step, only sensor data exceeding a threshold is transmitted to the central computing unit, which is typically configured as a server or backend for a vehicle fleet operator and / or manufacturer, or similar entity. This threshold can be configured so that, for example, only sensor data exceeding a certain threshold when driving over a pothole is transmitted, thus indicating an event such as driving over a pothole. This allows for a particularly compact database on the central, and especially vehicle-external, electronic computing unit, resulting in particularly efficient storage consumption.

[0027] To aggregate and collect data sets, data is gathered from multiple vehicles equipped with appropriate sensors. This allows for a degree of redundancy by recording positions along a route or track, or positions that appear during the initial journeys of the vehicles, with sufficient frequency. This ensures that derived events correspond with a high probability to actual events. The data can then be grouped based on these events.In this way, chassis control can be carried out using historically aggregated data. This involves checking whether, during a subsequent journey of the vehicle or another vehicle, data is available in the central electronic processing unit that has already detected an event at least at one location along the route and determined a corresponding parameter. This parameter is then checked against the current driving condition, specifically by comparing the context data associated with the historical sensor data with the current driving condition. If there is a sufficient similarity, the damping or suspension adjustment is suggested and, in particular, initiated. For this purpose, at least one parameter is transmitted from the central processing unit to the vehicle and / or the other vehicle during a journey along the route.In other words, the first step involves storing chassis suspension data, and the next step sends sensor values, or data exceeding a threshold, to a backend. Steps 3 and 4 in the backend are the aggregation and clustering of the data, respectively. In the fifth step, characteristic values ​​are derived from the data, such as a mean, maximum, slope, frequency, and / or similar parameters. In the sixth step, during a subsequent drive, the system checks whether data for this section is available in the backend. If so, the characteristic values, or at least one parameter or characteristic value, are used to define parameters for a forward chassis control system.The seventh step involves evaluating the suspension or damping control based on the events and assessing them together with the characteristic values, particularly within a feedback loop. This process is used continuously during a journey, both to acquire sensor data and to adjust the chassis.

[0028] One advantage of this method is, in particular, an increase in ride comfort and / or control accuracy. Furthermore, it protects vehicle components through particularly effective chassis control when traversing uneven road surfaces that put stress on them. In addition, it can create a more favorable perception, for example, in the area of ​​"digital luxury" for a vehicle manufacturer.

[0029] In an advantageous embodiment of the invention, the derivation of the respective event based on initial information from the context data is performed by the central computing unit. Additionally or alternatively, the grouping of events based on secondary information from the context data is also performed by the central computing unit. In other words, the respective data set is assigned or aggregated into events based on initial information from the context data. Additionally or alternatively, clustering is performed for each event, so that similar trends in sensor values ​​or sensor data can be grouped, for example, depending on the driving state and / or the location or position. This offers the advantage that the 2024P01690WG

[0030] 6

[0031] Aggregation and / or grouping can be carried out particularly efficiently and in a semantically meaningful way.

[0032] In a further advantageous embodiment of the invention, the sensor data includes the detection of acceleration, in particular of the motor vehicle, or of one of its components, such as suspension travel and / or velocity and / or spring force and / or suspension travel and / or suspension travel velocity and / or a control variable of a chassis component and / or quality and / or body movement and / or pitching behavior and / or roll behavior and / or component load. In other words, the at least one sensor unit is designed to detect corresponding spring forces, travel distances and / or travel speeds, in particular of all wheels.Additionally or alternatively, control variables from chassis actuators, such as pumps, valves, electrical and / or hydraulic actuators, as well as vehicle-specific state variables, such as performance or comfort, can be determined, particularly with regard to component stress and / or pitch and / or roll behavior and / or control. Performance, for example, can be considered as a function of the set control variable or a difference between the target control variable and the actual control variable. This offers the advantage that various sensor data can be used in diverse ways to determine road surface unevenness.

[0033] In a further advantageous embodiment of the invention, the context data and / or the current driving state comprise speed and / or tire pressure and / or direction of travel and / or lane and / or road type. In other words, the context data or the current driving state describe the state or condition of a component of the motor vehicle and / or the entire motor vehicle and / or an environment in which the motor vehicle is located, in particular the roadway and its type or lane. This offers the advantage that the event can be determined and thus precisely defined using the context data in combination with the sensor data. Furthermore, the precise definition of each event allows for the advantageous aggregation of identical events.

[0034] In a further advantageous embodiment of the invention, the at least one vehicle sensor is a camera and / or radar and / or lidar and / or ultrasonic sensor 2024P01690WG

[0035] 7. Additionally or alternatively, a classification of a road surface type or a surface irregularity, such as pothole, cobblestones, dirt track, sleeping policeman, drainage ditch, etc., can be determined or has been determined via semantic segmentation and / or a 3D survey of a surface, in particular the lanes, and thus by means of a height profile. In other words, a sensor is used as the at least one vehicle sensor whose sensor data includes an image and / or a 3D point cloud. If sensor data, such as images, is available, the aforementioned semantic segmentation can be carried out, in particular using a method such as computer vision, whereby corresponding objects or the road surface type and / or a surface irregularity can be recognized. A 3D survey can be carried out using point clouds, from which corresponding road surface types and in particular surface irregularities can also be derived.This offers the advantage that the process can be operated particularly efficiently and that events can be advantageously assigned to ground irregularities based on the data. Characterizing parameters (see step 5) can also be derived from the sensor data, especially from 3D point clouds.

[0036] In a further advantageous embodiment of the invention, a mean value and / or an extreme value and / or a ripple and / or a periodicity and / or a control quality and / or a control comfort is determined for at least one characteristic parameter for a driving condition. In other words, the characteristic parameter serves in particular as a key figure or as a multidimensional characteristic value and quantifies, in particular, the operating parameters that are advantageous for corresponding components of the vehicle, such as the chassis and / or wheels and / or a body, especially with regard to, for example, component load or mean values ​​derived from the events and the like. This results in the advantage that the damping or the suspension of the active chassis can be adjusted particularly advantageously based on the characteristic parameter.

[0037] In a further advantageous embodiment of the invention, the damping and / or suspension of the chassis is adjusted based on a similarity measure. In other words, a difference is calculated between current context data (i.e., in the current driving state) and context data aggregated from groups or events. This difference is then compared to a (second) threshold value that describes the similarity measure. If a certain level of similarity is present, the adjustment is made using at least one parameter from the aggregated data with a high degree of similarity. For example, adjustment can be made only if the same lane is driven at a similar speed and with comparable tire pressure. This offers the advantage of achieving a particularly high level of driving comfort without unnecessary adjustments.

[0038] In a further advantageous embodiment of the invention, the damping and / or suspension of the chassis is adjusted depending on a specific quality level. In other words, the event and the relevant parameter are evaluated, particularly within the vehicle, whereby, for example, a quality level for comfort, sportiness, and / or component protection is specified. This quality level is added to the at least one parameter. If a corresponding position or location is approached during a subsequent drive, the adjustment, and in particular the transmission, of the at least one parameter to the vehicle and / or the other vehicle only occurs if the quality level exceeds a predefined (third) threshold. Otherwise, the vehicle continues with the standard parameter settings. Thus, a feedback loop for the continuous evaluation and improvement of the process can be easily implemented.

[0039] A second aspect of the invention relates to a motor vehicle which has an active chassis and is designed to carry out a method according to the first aspect.

[0040] Advantageous embodiments and further developments, as well as advantages of the first aspect of the invention, are to be regarded as advantageous embodiments and further developments, as well as advantages of the second aspect of the invention, and vice versa.

[0041] A third aspect of the invention relates to a system with a vehicle-external, central electronic computing unit, to which at least two motor vehicles according to the second aspect of the invention are assigned, and the system is further configured to operate the two motor vehicles by means of a method according to the first aspect of the invention. Advantageous embodiments and further developments as well as advantages of the first and second aspects of the invention are to be regarded as advantageous embodiments and further developments as well as advantages of the third aspect of the invention, and vice versa.

[0042] Further advantages, features, and details of the invention will become apparent from the following description of a preferred embodiment and from the drawing. The features and combinations of features mentioned above in the description, as well as those mentioned below in the figure description and / or shown in the figures alone, can be used not only in the combinations specified, but also in other combinations or individually, without departing from the scope of the invention.

[0043] It shows:

[0044] Fig. 1 shows a schematic flowchart for a method for controlling the chassis of an active chassis of a motor vehicle; and

[0045] Fig. 2 is a schematic diagram of a possible sequence of the procedure according to Fig. 1.

[0046] Fig. 1 shows a schematic flowchart for a method for controlling the active suspension of a motor vehicle, in which suspension pre-control is performed using historically or previously aggregated data or datasets. The method comprises the following steps S1 to S7:

[0047] In a first step S1, sensor data SD from at least one vehicle sensor is acquired, which can detect an unevenness in the ground on a roadway or road surface, and associated context data KD, which at least includes the respective position during a first journey of the motor vehicle.

[0048] In a second step S2, the sensor data SD exceeding a threshold and the associated context data KD are transmitted as 2024P01690WG.

[0049] 10

[0050] Data set to a central, especially vehicle-external, electronic computing device.

[0051] In a third step S3, several data sets of journeys are aggregated, which have at least one position in common.

[0052] In a fourth step S4, the aggregated data sets are grouped according to events E that can be derived from them. Groups or clusters C are formed in this process.

[0053] In a fifth step S5, at least one characteristic CK of the respective group and / or the respective event E is determined.

[0054] In a sixth step S6, during a further journey of the motor vehicle and / or another motor vehicle along a route or track which has at least one position or location in common with the first journey or the aggregated journeys of the data records, it is determined whether a corresponding event E exists for at least one position.

[0055] Subsequently, in the seventh step S7, damping and / or suspension of the active chassis is adjusted based on at least one parameter CK for the event E depending on a current driving state of the motor vehicle and / or the other motor vehicle until the corresponding position is reached and thus before, for example, the unevenness of the ground is driven over.

[0056] The invention is based on the understanding that active suspension systems, or rather the suspension systems of motor vehicles, particularly passenger cars, have fast actuators for adjusting the spring rate and / or damping. However, especially with larger potholes or bumps, such as speed bumps, which are detectable road irregularities, suspension control is not optimal, as the exact geometry of the irregularity is often unknown. Sensors for geometric measurement are often unavailable in production vehicles and / or exhibit significant measurement inaccuracies at higher speeds. Furthermore, potholes are often not fully visible. The method presented here uses aggregated historical, especially map-based, data to achieve effective pre-control of the suspension, thereby enabling a particularly high level of driving comfort through suspension control and circumventing the aforementioned problems.The method presented here involves collecting data from damping-relevant events during driving, storing and clustering it in the backend or the central computing facility, deriving characteristic curves from the data, and downloading these curves when driving on a known road to enable parameters for feedforward control.

[0057] In other words, step S1 involves recording sensor values, for example, in a ring buffer of length T. Possible sensor values ​​include accelerations ax, ay, and / or az and / or velocities vx, vy, and / or vz, each in vehicle coordinates. Other sensor values ​​include, for example, spring forces, travel, or travel speeds, particularly of all wheels, and / or control variables of chassis actuators such as pumps, valves, electrical and / or hydraulic actuators, as well as chassis-specific state variables such as quality and / or comfort, in particular body movement, pitch and / or roll behavior, and / or component load, and / or control parameters, which include set control variables such as currents, valve positions, torques, forces, damping values, and / or the control parameterization used.

[0058] For vehicles equipped with environmental sensors such as cameras, radar, lidar, and / or ultrasonic sensors, the road surface type can be classified using semantic segmentation, for example, pothole, cobblestones, dirt track, sleeping policeman, drainage ditch, or similar features. Furthermore, a 3D survey of the surface, particularly of lanes, can be performed, allowing for the determination of an elevation profile.

[0059] In the second step, S2, a decision is made based on a threshold value. For example, whenever at least one of the aforementioned quantities exceeds a first threshold value, such as a spring force greater than 2000 N, the data in the ring buffer is written to a database, specifically the central electronic computing unit, which, together with contextual information KD, such as location, 2024P01690WG

[0060] 12. In particular, additional direction or navigation data, such as lane order, road type, etc., is provided. Otherwise, the data will be discarded.

[0061] In the third step S3, data records are aggregated in a database of the central electronic computing facility based on the context information or context data KD, so that data, for example, from lanes with potholes driven in the same direction, can be assigned to an event E.

[0062] In the fourth step, S4, each event E is clustered or grouped so that similar sensor readings, depending on the driving state (e.g., speed) and location (position), are assigned to a cluster C. Advantageously, the data in each cluster can be summarized, for example, by removing outliers and mapping the remaining readings over time, for example, with mean and variance.

[0063] In the fifth step, S5, characteristic parameters, or at least one parameter CK, and / or vectors of parameters can be determined and stored for each event E and each cluster C from the recorded sensor variables, state variables, and controlled variables, such as maximum values, minimum values, mean values, sign, frequency, minimum / maximum gradient or slope, ripple, peak-to-peak, period, periodicity, control quality, and / or control comfort. The at least one parameter CK is added to a database for the event E or cluster C. This parameter CK can, in turn, depend on other variables, such as speed, tire pressure, and the like. In other words, the at least one parameter can also be implemented as a multidimensional characteristic map.

[0064] In the sixth step, S6, if a vehicle (or motor vehicle) and / or another motor vehicle travels the route or path where at least one position is stored in the historical data, the system checks where an event E exists in the database along this route. If so, a similarity measure AM is calculated between current context data and the context data KD aggregated in groups or clusters CR. Whenever the similarity measure exceeds a threshold, an event E is selected from the group or cluster CE, and the feedforward control is parameterized with at least one characteristic value CK. For example, data from the same lane, at similar speeds, and with comparable air pressure are always used.

[0065] Finally, in the seventh step, S7, the event E is controlled using a parameter CK via an evaluation within the vehicle. This evaluation can be achieved, for example, by specifying a quality measure for comfort, sportiness (also dependent on the selected driving program), and / or component protection. This quality measure is thus added to at least one parameter CK. If another vehicle approaches event E, or if the vehicle repeats the journey, the parameter CK is only sent to the vehicle if the quality exceeds a certain threshold. Otherwise, the vehicle operates with the standard parameter settings. In other words, a feedback loop for the continuous evaluation and improvement of driving behavior is introduced.

[0066] Fig. 2 shows a flowchart of a possible implementation of the procedure, where box B1 indicates the start of the procedure or the start of the journey of the motor vehicle on a route.

[0067] Box B2 checks whether an entry, particularly one containing an event E, exists in a database of the central computing unit. This query can be performed continuously. If an entry exists, Box B3, which essentially corresponds to process step S6, determines whether the at least one characteristic value CK or standard parameter SP determined in step S5 should be used to perform feedforward control VS. Subsequently, Box B4 adjusts and / or disables the damping and / or suspension of the active chassis.

[0068] Figure 2 illustrates how a database containing data from a fleet of vehicles is used to record location-specific profiles of sensor values. For example, road irregularities such as potholes are summarized and stored on a map along with their location and direction. From these sensor values ​​(SG), characteristic values ​​(CK) are derived for parameterizing a controller, such as a feedforward control system for a chassis control unit. As a vehicle travels a route, the system continuously checks whether there are entries in the database along that route. If so, the controller is parameterized accordingly or, alternatively, operated with the default values.

[0069] In summary, this method communicates the lack of geometric knowledge about unevenness through historically aggregated values ​​and key figures, thus leading to an improvement in ride comfort.

[0070] Particularly advantageous is the aggregation of events E, groups or clusters C, and at least one characteristic value CK across a fleet of vehicles, for example, by creating a card reader in a backend or the central electronic computing unit. Analogous to traffic jam signals, for example, the data of the events E, the clusters C, and the at least one state variable CK are fed into the vehicle or the vehicle in front, and thus into the vehicles, as needed or depending on the route.

[0071] Advantageously, at least one parameter CK can be supplemented by further parameters CK determined from on-board environmental sensors, such as cameras, radar, lidar, and / or ultrasound. This allows for the definition of a variety of surface types during the application of the method, for example, potholes, cobblestones, dirt roads, sleeping policemen, drainage ditches, potholes with lift (i.e., force primarily applied upwards in the z-direction and longitudinally backwards), or reciprocal excitation (i.e., force also applied laterally, such as connections or lateral body movements). For each surface type, a parameter set for optimized control or feedforward control can then be defined during development.

[0072] The chassis control, or chassis feedforward control, can be further improved by setting up a simulation that simulates all clusters C from the events E and, for example, defines further feedforward control parameters in such a way that an optimization criterion, such as maximum comfort and / or maximum component protection, is optimized.

[0073] Based on the method presented here, a motor vehicle capable of performing the corresponding procedure, or of being operated using the method presented here, will also be introduced. Furthermore, a system comprising the central computing unit and at least two associated motor vehicles will be presented, whereby the system can control the motor vehicles according to the method presented here.

[0074] Thus, possibilities for map-based chassis control based on historically aggregated data are presented here.

Claims

2024P01690WG 16 Mercedes-Benz Group AG Patent claims 1. Method for chassis control of an active chassis of a motor vehicle, in which chassis pre-control is carried out using historically aggregated data, comprising the steps: Acquisition of sensor data (SD) from at least one vehicle sensor capable of detecting a road surface irregularity, and associated contextual data (KD) which includes at least the respective position during an initial journey of the motor vehicle; (S1) Transmitting the sensor data (SD) that exceeds a threshold and the associated context data (KD) as a data set to a central electronic computing facility; (S2) - Aggregate multiple journey datasets that share at least one position; (S3) - Grouping the aggregated data sets according to events that can be derived from them; (S4) Determine at least one characteristic parameter (CK) of the respective group and / or event; (S5) During a further journey of the motor vehicle and / or another motor vehicle along a route, determine whether an event has occurred for a position on this route (S6): - Adjusting the damping and / or suspension of the active chassis based on at least one characteristic parameter (CK) for the event and depending on the current driving state of the vehicle and / or the other vehicle until the position associated with the event is reached. (S7) 2. Method according to claim 1, characterized in that 2024P01690WG 17. The derivation of the respective event based on initial information from the context data (KD) by the central computing facility and / or the grouping of the events based on second information from the context data (KD) by the central computing facility takes place.

3. Method according to claim 1 or 2, characterized in that the sensor data (SD) includes acceleration, velocity, spring force, spring travel, spring travel velocity, control variable of a chassis component, quality, body movement, pitching behavior, roll behavior and / or component load.

4. Method according to one of the preceding claims, characterized in that the context data (KD) and / or the current driving condition includes a speed, a tire pressure, a direction of travel, a lane and / or a road type.

5. Method according to one of the preceding claims, characterized in that the at least one vehicle sensor is designed as a camera, radar, lidar and / or ultrasonic sensor and / or a classification of a road type is carried out via semantic segmentation and / or a 3D measurement.

6. Method according to one of the preceding claims, characterized in that for the at least one characteristic parameter (CK) a mean value, an extreme value, a ripple, a periodicity, a control quality and / or a control comfort is determined for a driving condition.

7. Method according to one of the preceding claims, characterized in that the adjustment of the damping and / or the suspension of the chassis is carried out depending on a similarity measure. 2024P01690WG 18 8. Method according to one of the preceding claims, characterized in that the adjustment of the damping and / or the suspension of the chassis is made dependent on a quality.

9. Motor vehicle comprising an active chassis, designed to perform a method according to one of claims 1 to 9.

10. System comprising a vehicle-external, central electronic computing device, to which at least two motor vehicles according to claim 9 are assigned, and which is configured to operate the two motor vehicles by means of a method according to one of claims 1 to 8.

Images