Trajectory-based chassis control

Abstract

Method for controlling an active suspension of a vehicle depending on road height profile data which lie in a prediction of a trajectory of the vehicle, wherein the prediction of the trajectory is calculated by means of a dynamic mathematical model at least on the basis of steering angle data from a steering angle sensor of the vehicle, and wherein the prediction of the trajectory is assigned respective road height profile data acquired by at least one environment sensor, wherein at least one actuator of the active suspension is controlled on the basis of the prediction of the trajectory, and wherein the prediction of the trajectory is used to control the active suspension only if the trajectory lies in an area which has been classified as drivable on the basis of environment data acquired by the at least one environment sensor.

Description

[0001] The present invention relates to a method for controlling an active suspension system of a vehicle based on road elevation profile data contained in a prediction of the vehicle's trajectory. The present invention further relates to a vehicle with a control unit configured to carry out the presented method.

[0002] To prepare predictive chassis systems, such as a vehicle's active suspension systems, for a future road surface, the road elevation profile ahead of the vehicle is traditionally measured by environmental sensors and processed into an estimated road height profile signal. This signal is then provided to a predictive chassis control system, such as an active suspension system, which uses it to calculate a proactive control signal for the respective actuators. The time advantage gained through this predictive calculation using environmental sensors can, for example, increase driving comfort by precisely adjusting the actuators to the specific environment detected by the sensors.

[0003] To control a predictive steering system, the road height signal is needed where each wheel of the vehicle is traveling; that is, the road height signal is required for each vehicle trajectory. Traditionally, the road height signal is output based on a trajectory derived from currently known vehicle movement data, such as steering angle, vehicle speed, or acceleration. If the steering angle does not change, the trajectory remains valid for a short period within the section of road the vehicle is traveling. However, if the driver makes steering movements, the vehicle's trajectory changes dynamically, meaning that the road height information, based on data from past time windows, is now outside the current trajectory and can no longer be used.The deviation between the predicted trajectory and the vehicle's actual trajectory increases with the distance between the predicted trajectory and the vehicle. Dynamic steering movements by the driver, for example, can cause the predicted trajectory to deviate from the current lane and, consequently, from the area actually being driven on.

[0004] German patent application DE 10 2012 015 492 A1 discloses a method for operating a chassis for a vehicle in which a road height profile is determined and assigned to the respective predefined categories.

[0005] US patent application US 2010 / 0023211 A1 discloses a control system for a vehicle chassis, wherein a sensor measures a road surface ahead and a variable is calculated based on the sensor's measurement data, which is used to adjust the chassis actuators.

[0006] A method for operating an active vehicle suspension is disclosed in German patent application DE 10 2010 018 902 A1, in which a road elevation profile signal is analyzed for periodically recurring changes in elevation and the suspension actuators are adjusted accordingly, if necessary. Elevation profiles are determined in DE 10 2012 009 882 A1, DE 10 2012 004 198 A1, and DE 10 2013 016 974 A1. DE 101 33 117 A1 discloses a method for road classification.

[0007] Against this background, a method for controlling an active vehicle suspension is presented based on road height profile data contained in a trajectory prediction of the vehicle, wherein the trajectory prediction is calculated using a dynamic mathematical model based at least on steering angle data from a steering angle sensor of the vehicle, and wherein the trajectory prediction is assigned to the respective road height profile data acquired by at least one environmental sensor, wherein at least one actuator of the active suspension is controlled based on the prediction, and wherein the trajectory prediction is used to control the active suspension only if the trajectory lies in an area that has been classified as drivable based on environmental data acquired by the at least one environmental sensor.

[0008] Specific features are derived from the description and the dependent requirements.

[0009] The presented method is particularly useful for controlling a vehicle's chassis using environmental data acquired by an environmental sensor to calculate a future trajectory of the vehicle. This future trajectory is calculated using a mathematical model that processes at least the vehicle's steering angle data. Specifically, the method is designed to use only trajectories that lie within a realistically drivable area.

[0010] In the context of the present invention, environmental data refers to data obtained from a measurement of the vehicle's current surroundings using an environmental sensor, such as a camera. Environmental data includes, in particular, road height information.

[0011] In the context of the present invention, road height profile data refers to data that indicate the surface characteristics of a given environment. In particular, it is intended that road height profile data be derived from environmental data acquired by an environmental sensor.

[0012] By taking into account dynamic steering movements of the vehicle when calculating the prediction of the vehicle's trajectory, spontaneous steering maneuvers of a driver can be considered, which, in combination with a selective choice of corresponding environmental data, enables a precise adjustment of actuators of the vehicle's chassis depending on exactly the environmental or road height profile data that correspond to an actually driven section of the road.

[0013] It is conceivable that the presented method operates on the basis of at least two different forecast horizons, whereby a first near or short forecast horizon is calculated on the basis of steering angle data, in particular current steering angle data, and a second further or more distant forecast horizon is calculated on the basis of classified environmental data.

[0014] In particular, it is planned to combine a forecast horizon based on steering angle data with a forecast horizon based on environmental data and to calculate a trajectory solely based on steering angle data in a specific range of environmental data.

[0015] In the context of the present invention, a forecast horizon is understood to be a spatial distance of a boundary of a currently calculated forecast of a respective trajectory of a respective vehicle.

[0016] In another possible embodiment of the presented method, it is envisaged that the environmental data be used to recognize, i.e., classify, a lane currently being traveled by the vehicle.

[0017] Using environmental data, such as road elevation profile data, it is possible to identify a currently occupied lane and, if necessary, distinguish it from non-drivable areas, such as a shoulder. Once the relevant information about drivable areas is available in the respective environmental data, calculations for controlling the vehicle's active suspension can be limited to those environmental areas known to be drivable.

[0018] Furthermore, by selecting environmental data depending on the currently driven lane, the influence of environmental data on a respective prognosis to be calculated or on the settings of the chassis, based on data that was determined in undriven areas, can be avoided.

[0019] In another possible embodiment of the presented method, it is envisaged that a surface profile is classified based on currently recorded environmental data, in particular environmental data of a lane currently being traveled by the vehicle.

[0020] To narrow down the prediction of the vehicle's trajectory to likely traversed areas and thus increase the prediction's validity, it is conceivable to classify currently acquired environmental data. This would allow for the identification of drivable areas, such as a smooth road, and their differentiation from impassable areas, such as a rough shoulder. By using the knowledge about upcoming drivable areas generated through this classification of environmental data, potentially relevant environmental data can be further selected and refined, thereby preventing actuator malfunctions caused by environmental data from impassable areas.

[0021] In another possible embodiment of the presented method, it is envisaged that the classification of the environmental data, in particular the environmental data of the surface profile of the lane currently being traveled by the vehicle, and knowledge of the currently being traveled lane are used to assess a prediction of a respective trajectory by means of a validity indicator with regard to its suitability for controlling the active chassis.

[0022] Once the relevant environmental data has been classified and information is available, for example, in the form of a likely lane, it is possible to limit the vehicle trajectory prediction, calculated based on the steering angle changes detected by the steering angle sensor, to those areas of the environmental data that are actually drivable. Accordingly, only those road elevation profile data points that are classified as lying on the current lane and calculated to be within the vehicle trajectory prediction are used to control the active suspension.

[0023] It is conceivable that, to assess the suitability of a calculated trajectory prediction for a given vehicle, a validity index could be calculated that indicates the position of the calculated trajectory prediction within a recognized lane or within a drivable area. The validity index could, for example, take a value between zero and one, where zero represents a position of the trajectory prediction far from drivable areas and one represents a position directly within a given lane.

[0024] A possible algorithm for executing the presented procedure could, for example, proceed as described below: a) Detecting a currently occupied lane in environmental data captured by an environmental sensor. b) Classifying paved and unpaved road sections based on environmental data. c) Estimation of a future trajectory of the vehicle based on information about the currently traveled lane and knowledge of paved and unpaved sections in the environmental data via the results of steps a) and b) d) Assignment of road height information acquired by means of an environmental sensor to the vehicle trajectory calculated in step c). e) Limiting the trajectory prediction based on a current steering angle of the vehicle to areas that can actually be driven on.

[0025] Furthermore, the present invention relates to a vehicle with a control unit configured to control an active suspension of a vehicle depending on road height profile data which lie within a prediction of a trajectory of the vehicle, wherein the control unit is further configured to calculate the prediction by means of a dynamic mathematical model at least on the basis of steering angle data from a steering angle sensor of the vehicle and to assign the prediction of the trajectory to respective road height profile data detected by an environment sensor, wherein the prediction of the trajectory is used to control the active suspension only if the trajectory lies in an area which can be classified as drivable based on the environment data.

[0026] Further advantages and embodiments of the invention will become apparent from the description and the accompanying drawings.

[0027] It is understood that the features mentioned above and those to be explained below can be used not only in the combinations specified, but also in other combinations or on their own, without leaving the scope of the present invention.

[0028] The invention is schematically illustrated with reference to embodiments in the drawings and is described schematically and in detail with reference to the drawings. Fig. Figure 1 shows a schematic sequence of a possible design of the presented procedure.

[0029] In Fig. Figure 1 represents a vehicle 1 with an environmental sensor 3 in the form of a camera. The vehicle 1 also includes actuators 5 for adjusting an active chassis of the vehicle 1 and a control unit 7 configured to detect steering movements of the vehicle 1.

[0030] In order to adjust the actuators 5 to height differences in a lane 9, the environment sensor 3 detects an environment located in front of the vehicle 1, i.e. in particular the lane 9 and a shoulder 11.

[0031] Based on environmental data acquired by means of the environment sensor 3, lane 9 can be classified, for example by using a mathematical classifier that compares the environmental data with predefined patterns and recognizes lane 9 as belonging to a category of "drivable" due to its surface structure, which differs from that of the shoulder 11.

[0032] Once lane 9 has been detected, environmental data corresponding to shoulder 11 can be excluded from further analysis, so that further calculations are only performed on environmental data corresponding to lane 9.

[0033] Using the current steering angle of vehicle 1, which the control unit 7 detects, for example, via a steering angle sensor, a trajectory 13 of vehicle 1 can be predicted, i.e., calculated for waypoints to be traveled in the future. To assess the validity of trajectory 13, it is only used to control the actuators 5 if the trajectory lies within lane 9. For this purpose, calculations to determine the prediction of trajectory 13 can, for example, be limited to those environmental data points that lie within lane 9.

Claims

[1] Method for controlling an active suspension of a vehicle depending on road height profile data which lie in a prediction of a trajectory of the vehicle, wherein the prediction of the trajectory is calculated by means of a dynamic mathematical model at least on the basis of steering angle data from a steering angle sensor of the vehicle, and wherein the prediction of the trajectory is assigned to the road height profile data which are detected by at least one environment sensor, wherein at least one actuator of the active suspension is controlled on the basis of the prediction of the trajectory, and wherein the prediction of the trajectory is used to control the active suspension only if the trajectory lies in an area which has been classified as drivable on the basis of environment data detected by the at least one environment sensor. [2] Method according to one of the preceding claims, wherein the environmental data are used to detect a lane currently being traveled by the vehicle. [3] Method according to claim 2, wherein a surface profile of the lane currently traveled by the vehicle is classified using currently recorded environmental data. [4] Method according to claim 3, wherein the surface profile is classified at least into categories from the following list of categories: “fixed” and “unfixed”. [5] Method according to claim 3 or 4, wherein the classification of the surface profile of the lane currently traveled by the vehicle and knowledge of the lane currently traveled by the vehicle are used to assess a prediction of a respective trajectory by means of a validity indicator with regard to its suitability for controlling the active chassis. [6] Method according to claim 5, wherein road height information acquired by means of the environment sensor and / or a road height sensor is assigned to the trajectory prediction. [7] Method according to one of the preceding claims, wherein a camera sensor providing image data is selected as the environmental sensor. [8] Vehicle (1) with a control unit (7) configured to control an active suspension (5) of the vehicle (1) depending on road profile height data which lie within a prediction of a trajectory (13) of the vehicle (1), wherein the control unit (7) is further configured to calculate the prediction of the trajectory (13) using a dynamic mathematical model at least on the basis of steering angle data from a steering angle sensor of the vehicle (1) and to assign the prediction of the trajectory (13) to respective road height profile data acquired by an environment sensor (3), wherein the prediction of the trajectory (13) is used to control the active suspension (5) only if the trajectory (13) lies in an area (9) which can be classified as drivable on the basis of environment data acquired by the environment sensor (3).

Images