VEHICLE TYPE IDENTIFICATION

Description

[0001] The invention lies in the field of the control of vehicle treatment systems.

[0002] The disclosure relates in particular to the image-based (e.g., camera-based) acquisition of a vehicle in the area of ​​a vehicle treatment system for the recognition of vehicle attributes (e.g., the vehicle type and / or the vehicle contour) by means of a (cloud-based) data processing service for the optimization of the treatment process in the vehicle treatment system. The invention relates to a computer-aided method for providing a (cloud-based) data processing service.

[0003] The disclosed methods pursue a common inventive concept, wherein the functionality is preferably distributed across two or more computer systems involved, in particular across at least one control unit on a vehicle treatment system and at least one server that are connected and interact with each other (e.g. via the Internet or another network).

[0004] The term "vehicle handling facilities" includes, in particular, vehicle washing facilities and vehicle care facilities. These facilities can be designed for the treatment of cars, trucks, trailers, rail vehicles, or other types of vehicles.

[0005] Typical examples of vehicle treatment systems are gantry car washes or conveyor car washes. These systems generally include several treatment elements, such as rotating brushes, polishing, or drying equipment, which allow for at least semi-automated vehicle treatment. The treatment typically involves one or more surface treatments of the vehicle (e.g., exterior washing, paint care, wheel cleaning, underbody cleaning, etc.). Depending on the system's design, the vehicle may remain stationary or move through the system during treatment.

[0006] The present disclosure is, in principle, applicable to all types of vehicles. Typically, a vehicle treatment plant is designed for one type of vehicle, e.g., passenger cars, trucks, or rail vehicles. Even among vehicles of the same type (e.g., passenger cars), there can be significant differences between the individual vehicles to be treated.

[0007] A particular challenge in automated vehicle treatment is adapting the treatment process to the different vehicle types, models, special equipment and different vehicle contours and / or the individual vehicle.

[0008] Prior art (CN 105 205 468 A) discloses a car wash system with an image acquisition device that serves to recognize the vehicle identification numbers (VINs) of the vehicles to be washed. The recognized VIN is compared with a database, and vehicle type information, e.g., size information, is retrieved to automatically select the washing program. From CN 110 936 925 A, a multi-axis robotic car wash system is known with a method in which the position and vehicle type of the vehicle to be washed are recorded using a camera and a position sensor.

[0009] German patent DE 20 2013 005 162 U1 discloses an access control system that performs a rights check based on a license plate captured by a remote reading device in order to grant access to a "device intended for vehicles". WO 2018 / 024336A1 discloses a system for the maintenance of non-user-driven vehicles (e.g., for a rental car station) in which a camera-based detection device automatically determines the cleaning needs of a vehicle.

[0010] During surface treatment (e.g., automated washing, polishing, or drying), suitable treatment tools (e.g., rotating brushes, polishing cloths, or the nozzle of a blower) are typically moved across the vehicle's surface. The treatment tool and the vehicle surface should be positioned at an appropriate distance from each other. Often, the treatment tool is moved in a guided motion relative to the vehicle (e.g., from front to back), and / or the vehicle is moved in a corresponding direction relative to the treatment tool. A combination of movement of the treatment tool (e.g., vertical movement) and movement of the vehicle (e.g., horizontal movement) can also be used to create a relative movement along a suitable path between the treatment tool and the vehicle.

[0011] In state-of-the-art technology, sensors on the vehicle cleaning system are used to guide the cleaning agents (e.g., a rotating brush) along the individual vehicle surface. Often, the user can also select a specific wash program.

[0012] The vehicle treatment system typically detects the individual vehicle section by section during the washing process and incrementally determines a suitable path for surface treatment. Prior art includes techniques, for example, where a suitable distance to the vehicle is continuously determined during treatment by measuring the back pressure of the treatment agent upon contact.

[0013] The object of the present invention is to optimize the treatment of vehicles in a vehicle treatment plant.

[0014] A first aspect of the disclosure is a computer-implemented method for controlling a vehicle treatment system. This first method is preferably implemented in a control system of the vehicle treatment system. The method can be implemented, in particular, by one or more appropriately designed software modules. Alternatively or additionally, one or more specially designed data processing systems can also be used.

[0015] The method comprises several steps, each of which can be performed individually or multiple times. The claimed methods are not limited to the sequence of steps.

[0016] First, an image signal is acquired by an image acquisition device. The image acquisition device preferably comprises one or more cameras. The cameras or similar means for capturing images of the surroundings are preferably directed at a specific area (e.g., the entrance area or the area in front of an operating terminal) of the vehicle treatment system. Depending on the design of the vehicle treatment system, it may be advantageous to capture the vehicle to be analyzed in a specific area or at a specific moment.

[0017] The image signal preferably comprises one or more digital images. The image signal can, in particular, comprise a sequence of images and / or a video. The image signal contains one or more representations of the vehicle to be detected. In particular, images from different perspectives can be combined and / or fused within the image signal.

[0018] The image acquisition device can be implemented as a separate component and / or as a software module. The camera or other sensor can be integrated into the image acquisition device or implemented as a separate component. Data from an existing camera on a vehicle treatment system can also be used.

[0019] A vehicle image message is generated from one or more image signals. This vehicle image message is then sent to a data processing service. The vehicle image message preferably includes one or more images of the detected vehicle within the vehicle handling facility. In addition to image information, the vehicle image message may also contain supplementary information, such as details about the vehicle handling facility, and / or processing instructions for a data processing service.

[0020] The data processing service according to the invention is preferably accessible via the Internet or another data network. One or more messages can be exchanged between a remote data processing service and one or more vehicle treatment systems via a suitable communication interface.

[0021] Various data and / or network protocols are available to the person skilled in the art for sending messages. The message can, in particular, be structured as a data packet. A message within the meaning of this disclosure is computer-readable information. The message preferably comprises one or more data structures with functional data relating to the vehicle, the vehicle handling system, an image acquisition device, and / or a data processing service. A message can be sent directly to the recipient or via an intermediary.

[0022] A particular advantage of the invention lies in the centralized (especially cloud-based) processing of image information from the vehicles to be treated. In a preferred embodiment, data from various vehicle treatment systems, which may be geographically distributed, can be processed by a central data processing service. Further developments and improvements to the data processing can be carried out at a central location, e.g., on a server accessible via the internet, and made directly available to a large number of systems.

[0023] Another advantage of sending vehicle image messages to a remote data processing service is the possibility of developing a cost-effective service based on the processing of the vehicle data. For example, an operator of a vehicle treatment facility could obtain additional functions as a service ("feature as a service") from the manufacturer of the facility or another provider.

[0024] The data processing service can generate a control parameter message based on the vehicle image message sent by a vehicle treatment system, which can be used for optimized control of the vehicle treatment system.

[0025] The control parameter message can be exchanged between a vehicle treatment system and a data processing service via the same or a different communication interface. The vehicle treatment system receives the control parameter message. The control parameter message preferably includes a vehicle attribute and / or a control instruction for the vehicle treatment system that can be derived from a vehicle attribute. The control parameter message may, in particular, have been generated for a specific vehicle treatment system.

[0026] Both sending and receiving data by the components disclosed herein can be implemented using various communication architectures. Communication can be based on a server-client or service architecture. Time- and / or event-driven distributed program flows can also be used. For the purposes of this disclosure, a received message can be either actively retrieved or passively received. The same applies to sending a message.

[0027] An important aspect of the disclosure lies in the distributed functionality across local components at the vehicle treatment system and remote components of a (cloud-based) data processing service. The data processing service can, in principle, be implemented at any location and preferably serve multiple vehicle treatment systems. Regarding the distribution of functional components between local and remote computer components, the disclosure includes several embodiments based on a common fundamental concept.

[0028] It is advantageous for plant-related data to be processed on the plant side. Complex image processing, which involves high computing capacity and / or ongoing development, is preferably carried out using a remote data processing service.

[0029] A particularly advantageous aspect is the provision of a central data intermediation service in the cloud, enabling vehicle treatment systems to communicate via defined interfaces and data formats. Data processing (e.g., image processing and extraction of vehicle attributes) can be adapted in the background and, if necessary, distributed across multiple (internal or external) services or components. In this way, the data processing systems can be further developed centrally in one location, while a large number of vehicle treatment systems can maintain operation.

[0030] The control parameter message is generated by the data processing service and enriched with data that can be used for optimized control of the vehicle treatment system. A control signal for controlling the vehicle treatment system can be generated based on the control parameter message.

[0031] To generate the control signal, data from the vehicle treatment system itself and / or data from a control parameter message of a remote data processing service can be used. The combination of locally available sensor data, for example, for locating the vehicle during the treatment process, and vehicle attributes determined by image processing performed in the cloud is particularly advantageous.

[0032] The control signal can be either a logical data set within a control program of the vehicle treatment system or a direct control signal for a specific treatment agent of the vehicle treatment system.

[0033] The second computer-implemented method according to the invention provides the data processing service for vehicle treatment systems. This second method also comprises several steps, each of which can be executed individually or multiple times. The method is not limited to the sequence described here.

[0034] The data processing service receives one or more vehicle image messages from one or more vehicle processing systems. In addition to the one or more vehicle images, the vehicle image messages may also contain further processing data and / or requests.

[0035] The purpose of processing the vehicle image messages in the remote data processing service is to determine and provide vehicle-related information for the optimized treatment of the vehicle in the vehicle treatment facility.

[0036] The vehicle image message comprises a vehicle image relating to a vehicle within the vehicle handling system. The vehicle image can depict the vehicle in whole or in part. The one or more vehicle images are processed using suitable image processing equipment. This processing takes place in several steps. These steps are performed by external processing equipment. According to the invention, a vehicle type is determined from a vehicle image in a first step. For this purpose, an external image processing service is used.

[0037] The vehicle type can be, in particular, a manufacturer-specific model or model variant. Alternatively or additionally, the vehicle type can also describe a vehicle category (e.g., station wagon, sedan, SUV). In a second step, (further) vehicle attributes can be determined based on the vehicle type, which can be used for optimized handling of the vehicle in the vehicle handling system. The vehicle attribute includes data about the vehicle's geometry.

[0038] The disclosure includes various embodiments for determining the vehicle attributes based on the vehicle image message.

[0039] The vehicle attribute can be derived from a visual representation of the vehicle. If necessary, additional information from other sources, such as vehicle databases, is used in addition to the visual representation of the vehicle.

[0040] The vehicle attribute can include, in particular, group-specific information and / or vehicle-specific information. Group-specific information would be, for example, the exterior contour or a dimension of a standard production vehicle of a specific vehicle type from a manufacturer. For instance, image processing of the vehicle image can be used to determine the manufacturer and model variant, and a specification of this vehicle model can then be retrieved from a database based on the model. Group-specific information can be obtained, in particular, from third-party databases and / or data collections specifically created for this purpose.

[0041] Alternatively or in addition to group-specific information, the vehicle attribute can also include vehicle-specific information. Examples of vehicle-specific information include the individual degree of soiling of the vehicle or a special body configuration. Determining vehicle attributes to generate a control parameter message can, in particular, involve a combination of group-specific and vehicle-specific information. It is especially advantageous to assign the depicted vehicle to a pre-known vehicle category (e.g., manufacturer and model variant) and to compare the actual depiction of the vehicle with a standard version. This allows for the identification of deviations between the individual vehicle and the standard or group-specific version of the vehicle category.Such deviations can also be transmitted as additional data for a vehicle attribute.

[0042] A particularly advantageous embodiment is the determination of a vehicle contour and / or a virtual vehicle model. A vehicle contour comprises geometric dimensions of the vehicle's outer skin. The vehicle contour can be represented, in particular, as a two- or three-dimensional curve or surface. The vehicle contour provides information about the surfaces of the vehicle to be treated in the vehicle treatment system.

[0043] A particular advantage of determining one or more vehicle attributes based on a vehicle image is that data about the vehicle geometry can be provided to the vehicle treatment system before and / or during a treatment program. During treatment (e.g., washing), the vehicle can often only be captured section by section or incrementally. It is therefore particularly advantageous to generate an image of the vehicle before the treatment process begins (e.g., in the entry area) and to derive the most complete possible geometric information about the vehicle from it.

[0044] It is particularly advantageous to provide vehicle attributes of an image-captured vehicle to the vehicle treatment system in real time or near real time. Preferably, the methods disclosed herein are designed to capture a vehicle with an image capture device and provide vehicle attributes for an imminent treatment program. This provision is achieved through automated processing of the image information. For example, the driver of a vehicle in the entry area of ​​the vehicle treatment system can select a treatment program that is optimized and / or adapted to the vehicle by the technology disclosed herein. Preferably, only a few seconds or minutes elapse between the image capture of the vehicle and the adaptation of the treatment program.

[0045] Another advantage of the present invention is that by processing one or more vehicle images, a virtual vehicle model that is as complete as possible can be determined.

[0046] If information about the geometry or other properties of the vehicle is provided before the start of a vehicle treatment program, the treatment program can be parameterized or adapted in such a way as to improve the results of the vehicle treatment or to optimize the efficiency of the treatment process.

[0047] The data processing service provides one or more control parameter messages for the vehicle's surface treatment in the vehicle treatment system. Depending on the embodiment, the control parameter message can contain different data. In a first embodiment, the control parameter message includes one or more of the determined vehicle attributes (e.g., a vehicle contour or a virtual vehicle model). The conversion of the vehicle attributes into control commands for the optimized treatment of the vehicle in the vehicle treatment system can be performed by the data processing service and / or by a component at the vehicle treatment system. Distributing the processing steps across these components can offer various advantages depending on the use case. For example, it may be advantageous to forward the vehicle attributes determined in the cloud directly or only after preprocessing to the vehicle treatment system.Especially when the conversion of vehicle attributes into concrete control commands depends on the type of system, it can be advantageous to perform these calculations within the vehicle treatment system itself. Alternatively or additionally, control information or instructions can also be transmitted from the data processing service to the vehicle treatment system via the control parameter message. This can be particularly beneficial if the calculation of a control instruction or control signal is computationally intensive or requires frequent adjustments. For example, additional information for adapting a treatment program can also be calculated in the cloud-based data processing service and sent to the vehicle treatment system via the control parameter message.

[0048] The control parameter message preferably comprises a combination of determined vehicle attributes and / or control instructions for the vehicle treatment system. The control parameter message is designed for vehicle treatment, particularly for vehicle surface treatment. The control parameter message may, in particular, include path and / or movement data for one or more treatment devices (e.g., a washing brush).

[0049] The methods described here achieve a common technical effect during vehicle treatment in the vehicle treatment system. Each method can be executed by one or more computers. It is particularly possible and advantageous for the first method, which controls the vehicle treatment system, to be executed on a control unit or computer located at the vehicle treatment system. The second method, which provides the data processing service, is preferably executed on one or more computers or servers accessible via a network (e.g., the Internet). Alternatively or additionally, one of the methods and / or parts of a method can be executed on a central control unit for locally connected vehicle treatment systems. This can be particularly advantageous for a large network of vehicle treatment systems (e.g., in a large car wash).

[0050] In addition to the methods described above, corresponding data processing systems and computer programs are also required. A system for providing a data processing service as described in the second method preferably comprises one or more servers. The system may, in particular, include a server with a data transmission module. The data transmission module may, in particular, receive vehicle images and transmit them to further processing modules of the service provider itself and / or third-party services. This may involve, in particular, preprocessing or further processing of the vehicle images. Data may also be transmitted between several modules and / or servers of the service provider itself and / or third-party providers to determine the vehicle attributes in one or more steps.The data processing system at the vehicle treatment facility can consist of a central control unit or several distributed computing units. Parts of the data processing can also be performed on an image acquisition device and / or a control unit for a treatment device.

[0051] Further advantageous embodiments are described below with reference to the drawings. Drawing description

[0052] The invention is illustrated in the drawings in an exemplary and schematic manner. They show: Figure 1: Interaction between a vehicle treatment system and a remote data processing service; Figure 2: The detection of a vehicle at a vehicle treatment system; Figure 3: The adaptation of a vehicle treatment to a determined vehicle attribute; Figure 4: The detection of a special body on a vehicle; Figure 5: The treatment of a vehicle with an undercut in the vehicle contour. Detailed description

[0053] Figure 1Figure 1 schematically illustrates the interaction between the first and second computer-implemented methods described here. A vehicle (60') is detected in the area of ​​a vehicle treatment system (10) by an image acquisition device (15). The image acquisition device (15) preferably comprises one or more cameras (16). The one or more cameras (16) are preferably directed towards an entry area or an area of ​​the vehicle treatment system (10) upstream of the treatment zone. Figure 1 The camera is only shown schematically as a component.

[0054] The vehicle treatment system preferably comprises a control device (20). The control device (20) can be either a component of the system control or a separate computing and / or communication unit. The control device (20) preferably comprises one or more communication modules. Preferably, the control device (20) communicates with components of the vehicle treatment system or sensors via one or more WUS systems. Communication with one or more data processing services preferably takes place via a communication interface (25). The communication interface (25) can, for example, be a cellular modem and / or a network connection.

[0055] The control device receives image information from the image acquisition device (15). The image signal (IM) preferably comprises one or more images of the vehicle (60). The image signal (IM) can include preprocessed information from an image acquisition device and / or raw data from one or more sensors (for example, a camera). The image signal (IM) can be transmitted, in particular, as data packets, a message, or a similar analog or digital signal.

[0056] A vehicle image message (VD) can be generated based on one or more image signals. The vehicle image message (VD) is preferably transmitted from the control device (20) to a data processing service (30) via a communication interface (25). The vehicle image message (VD) preferably comprises one or more vehicle images (65). The vehicle image (65) can, for example, include an image of the vehicle from a specific perspective and / or pre-processed image information from one or more image signals (IM). The fusion or combination of multiple camera images can be particularly advantageous.

[0057] The data processing service (30) can be implemented by one or more servers accessible via the Internet or another network. The functionality of the data processing service can be implemented in different embodiments. In Figure 1A particularly advantageous embodiment is shown.

[0058] In the embodiment shown, the data processing service (30) comprises a data transmission module (35). The data transmission module can be operated either as a software module or as a separate server. The data transmission module (35) can send and receive data, e.g., a vehicle image (65) or a determined vehicle attribute (61). Alternatively or additionally, the data transmission module (35) can also process data and / or generate requests to other systems or services.

[0059] According to the invention, the data processing service is implemented using an external image processing service (40). The image processing service (40) can, for example, receive and process image information to recognize, categorize, and / or evaluate objects. The image processing service (40) can, in particular, be an external service provided by a third party.

[0060] For example, preprocessing of image information can take place in the data transmission module (35) shown.

[0061] According to the invention, a type query (TR) is generated. The type query (TR) is generated by the data transmission module (35) and transmitted to an image processing service (40). The image processing service processes the type query (TR). The type query (TR) can, in particular, include one or more vehicle images (65). Depending on the type query (TR), the image processing service (40) returns a vehicle type message (VT). The vehicle type message (VT) is transmitted by the image processing service (40) to a data transmission module (35). The vehicle type message (VT) preferably includes data about a vehicle type. The vehicle type can, for example, be identified by a model designation of a specific manufacturer. The vehicle type can also consist of an abstract category of vehicles (e.g., station wagon).

[0062] Advantageously, one or more vehicle attributes (61) are determined using one or more data processing services (30) based on one or more vehicle images. The vehicle attributes (61) can be derived directly or indirectly from the image representation of the vehicle. Depending on the embodiment of the invention, one or more vehicle attributes can be determined in one or more steps, either in combination or independently of one another. Figure 1 Figure 6 shows a particularly advantageous embodiment of the invention in which a two-stage processing process takes place. Based on the vehicle type determined by image processing (e.g., manufacturer X, model Y, variant Z), one or more group-typical vehicle attributes (61) can be determined. A group-typical vehicle attribute (61) is, for example, the vehicle contour as it appears in Figure 1This is schematically indicated by a dashed line. Such geometric information for a specific vehicle type can, for example, be queried from a database. Alternatively or additionally, vehicle attributes can also be derived directly from a vehicle image (65), e.g., by optical measurement.

[0063] In a particularly advantageous embodiment, a property query (FR) is generated and / or transmitted to a vehicle database (50). For example, the property query (FR) can include a previously determined vehicle type and / or a desired property (e.g., a dimension, a virtual vehicle model, a contour line, equipment variants).

[0064] Based on the property query (FR), a group-typical property of the recorded vehicle type can be determined from a set of previously known vehicle types and properties associated with these types.

[0065] One or more requested properties can be transmitted using one or more vehicle property messages (V). For example, the vehicle database (50) can report the requested vehicle property, in particular a specific vehicle attribute (61), back to the data transmission module (35) and / or make it available for collection.

[0066] Alternatively or in addition to the one in Figure 1 In addition to the indicated two-stage processing, one or more further investigation steps can be carried out to determine a vehicle attribute.

[0067] For example, a determined vehicle contour and / or a virtual vehicle model can be checked for specific properties relevant to vehicle handling. The processing of geometry information and / or vehicle condition information for identifying critical areas on the vehicle surface is particularly advantageous. A critical area of ​​a vehicle surface could be, for example, a protruding antenna (62) or an undercut (63) in the vehicle's geometry.

[0068] The control parameter message (CP) can include one or more vehicle attributes (61) as well as control instructions for the vehicle treatment system (10) that can be derived from them.

[0069] A control device (20) on a vehicle handling system (10) receives a control parameter message (CP). The control device (20) can be the same or a different control device that received the image signal (IM) or sent the vehicle image message (VD).

[0070] The control parameter message (CP) includes one or more vehicle attributes (61) that can be used to treat the vehicle (60') in question. A control signal (CS) for controlling the vehicle treatment system (10) is generated based on the control parameter message (CP).

[0071] Depending on the embodiment of the invention, the processing of the data contained in the control parameter message (CP) can be carried out differently. In a first embodiment, the control parameter message (CP) contains a vehicle attribute (61), in particular a vehicle contour, which is converted into a suitable treatment movement (C) in the control device (20). In a further embodiment, the control parameter message (CP) already includes control instructions that can be implemented by the vehicle treatment system (C). For example, the data processing service (30) can already include a treatment contour (61a) and / or a treatment movement derivable from one or more vehicle attributes (61).

[0072] The control signal (CS) is designed to control one or more treatment devices (11), in particular washing brushes. When controlling one or more treatment devices (11), one or more parameters for a rotational and / or translational movement of a treatment device (11) can be specified. Alternatively or additionally, a torque and / or force, a dosage value, a flow rate, or a treatment duration for a treatment device (11) can also be specified.

[0073] The present disclosure allows for the optimized control of a wide variety of treatment devices (11). For example, it is possible to optimize the control of a washing brush, a polishing device, a drying blower, an application device, or another known treatment device in the vehicle treatment systems already described above.

[0074] Figure 2Figure 1 shows a possible sequence of a method for controlling a vehicle treatment system (10). A vehicle (60) is preferably captured by an image acquisition device (15) in a specific area of ​​the vehicle treatment system (10). For example, this area can be located in an entry area (70) of the vehicle treatment system. Alternatively or additionally, the vehicle can also be captured in another area, in particular in a treatment area (71) of the vehicle treatment system (10). It is advantageous to capture the vehicle (60) in an area where the vehicle can be captured as completely as possible by the image acquisition device. It is also preferably the vehicle (60) to be captured at a time before the start of the treatment program. This is particularly advantageous if the vehicle attributes (61) derived from the image acquisition are available in real time for the immediately subsequent treatment process.It is intended to be used in near real-time. Depending on the quality of the communication link and the computing power of the data processing service, feedback of the relevant vehicle attributes can be provided quickly enough to allow the user of the vehicle treatment system to immediately improve the treatment of their vehicle.

[0075] In a particularly advantageous embodiment, a virtual vehicle model (61) is generated based on one or more vehicle images. A virtual vehicle model is a virtual representation of the vehicle in a machine-readable data structure. In particular, a virtual vehicle model can comprise a parameter set, a CAD model, a collection of dimensions, a 3D model, or a similar representation of a vehicle.

[0076] Based on a virtual vehicle model, the treatment of the vehicle (60') can be planned even before the start of a treatment program. For example, a path for a treatment agent (11) along a surface of the vehicle can be calculated in advance.

[0077] In a particularly advantageous embodiment, a vehicle attribute (61) is linked to data from the vehicle treatment system (10). For example, sensor data from one or more sensors (18) in a control device (20) can be linked to information from one or more control parameter messages (CP).

[0078] One possible embodiment of a vehicle treatment system comprises one or more localization means (17). The localization means (17) may include one or more sensors (18). For example, the vehicle treatment system may include one or more light barriers, ultrasonic sensors, cameras, or other means for detecting the position of a vehicle (60') within the vehicle treatment system (10). It is particularly advantageous to align one or more vehicle attributes (61), in particular a virtual vehicle model, with the actual position of the vehicle (60'). In this way, a pre-calculated treatment method, in particular a treatment path, can be applied to the actual vehicle.

[0079] Figure 3shows schematically how, based on a vehicle attribute (61), e.g. a vehicle contour or a virtual vehicle model, a treatment program of the vehicle can be prepared and / or adapted and / or carried out with one or more treatment agents (11).

[0080] In one possible embodiment, a treatment contour (61a) is determined based on one or more vehicle attributes (61). Such a determination step can be carried out by the vehicle treatment system and / or by a data processing service. For example, a treatment movement (C) along a part of the vehicle surface can be planned for one or more treatment means (11).

[0081] In the Figure 3The illustrated embodiment shows, by way of example, the movement of a brush roller with a horizontal axis of rotation in a vehicle wash system. Alternatively or in addition to this treatment movement, treatment processes of other treatment agents (11) can be optimized for the vehicle (60) to be treated.

[0082] Figure 4 Figure 6 shows another interesting application of the technology disclosed herein, in which image processing is used to determine one or more special states or special equipment (62) of a vehicle (60). Alternatively or additionally to the vehicle attributes (61) already mentioned, a deviation of a vehicle (60) captured with a vehicle image from a group-typical state of the vehicle can be detected. This can be applied in particular to the case where a vehicle includes a non-standard attachment such as a spoiler or a roof rack.

[0083] In a particularly advantageous embodiment, the disclosed technology can detect a hazardous condition that is critical for treatment in the vehicle treatment system (10). Such a hazardous condition can be caused, for example, by an open window or a roof structure.

[0084] Particularly advantageous is a method for providing a data processing service in which the actual vehicle in a vehicle image (65) is compared with an expected vehicle based on a vehicle type and, in particular, deviations are identified.

[0085] Figure 5Figure 6 shows another interesting application of the invention. In a particular embodiment, critical areas of a vehicle geometry can be identified. For example, undercuts (63) on a vehicle and / or on a virtual vehicle model (61) can be identified. This is particularly advantageous for the advance planning and / or the adaptation of a treatment movement (C) to the geometry of the vehicle.

[0086] Adapting the vehicle treatment to the individual and / or group-specific vehicle geometry is particularly advantageous when certain vehicle characteristics cannot be detected, or can only be detected poorly, by the sensors typically used in a vehicle treatment system. This can be the case, in particular, because there is usually relative movement between the vehicle treatment system (10) and the vehicle (60) being treated. Depending on the design of the vehicle treatment system (10), the vehicle may, for example, be towed through a car wash and / or parts of the vehicle treatment system may move along the vehicle. Depending on the arrangement of the sensors for detecting the vehicle, determining the vehicle geometry using previously known methods may be complex or only possible to a limited extent.

[0087] In addition to the embodiments shown in the drawings, the disclosure includes further advantageous embodiments. The features of these embodiments can, in particular, be interchanged, omitted, replaced, or supplemented.

[0088] Particularly advantageous is a computer-implemented method according to claim 1, wherein the control signal (CS) is configured to adapt a predetermined treatment program of the vehicle treatment system (10) to the detected vehicle (60), in particular to one or more vehicle attributes (61) of the vehicle (60), for example the contour and / or special equipment of the vehicle.

[0089] For example, the control signal (CS) can be used to adjust one or more parameters for a predefined treatment program. Alternatively or additionally, the control signal (CS) can also include the treatment program itself.

[0090] Preferably, the treatment program is adapted to the vehicle contour by means of the control signal (CS). Specifying the vehicle contour and / or a path adapted to it is particularly advantageous.

[0091] Another advantageous embodiment is a computer-implemented method according to claim 1, wherein the control signal (CS) is configured to accelerate and / or decelerate and / or intensify and / or suspend the treatment of the vehicle (60), in particular in certain areas with respect to the vehicle surface.

[0092] By capturing images of the vehicle, critical areas of the vehicle surface can be identified.

[0093] It is particularly advantageous to adapt the treatment in the critical areas in order to achieve an optimal treatment result.

[0094] Another embodiment comprises a computer-implemented method according to claim 1, wherein the control parameter message (CP) includes data relating to a type deviation (62) of the vehicle (60).

[0095] A type deviation can be, in particular, a special condition and / or special equipment compared to a standard model of a determined vehicle type.

[0096] In In an advantageous embodiment of the invention, a virtual vehicle model is generated from the data obtained through image processing. This vehicle model can be generated by a data processing service and / or on the part of the vehicle treatment system.

[0097] The virtual vehicle model allows for the planning and / or simulation of vehicle treatment on a virtual model. The planned treatment can then be transferred to the treatment of the real vehicle. This allows for the inclusion of information about the vehicle that is typically unavailable during treatment by the vehicle treatment system. The treatment can then be compared with the treatment simulated on the virtual vehicle model, particularly during the process.

[0098] Advantageously, at least one vehicle attribute (61), in particular a vehicle contour or a virtual vehicle model, is related to the real vehicle (60'). For example, the virtual vehicle model can be localized with respect to a localization device (17), the vehicle (60), or the vehicle treatment system (10). By localizing the virtual model, the treatment devices (11) of the vehicle treatment system can be controlled based on data from the virtual model. This data can alternatively or additionally be combined with data from other sensors of the vehicle treatment system (e.g., contact pressure, light barriers, etc.).

[0099] Advantageously, the system is designed so that the vehicle image processing occurs (almost) in real time. For an optimal user experience, the customer's vehicle is captured before the treatment begins, for example, while the user selects a program at a control terminal. At the time of camera capture, the vehicle should ideally be as unobstructed as possible and fully captured. The vehicle images are processed online by the data processing service. The treatment program can then be automatically adapted to the vehicle immediately after it is captured.

[0100] According to the invention, the method for providing a data processing service comprises the following steps: generating a type query (TR) to determine a vehicle type in a vehicle image (65) and obtaining a vehicle type signal (VT) from an external image processing service (40).

[0101] In another embodiment, the method comprises the following steps: generating a property query (FR) to query a property of a given vehicle type, in particular from a vehicle database (50), and / or obtaining a vehicle property message (VF) from a vehicle database (50).

[0102] The disclosure also includes a method for providing a data processing service, wherein, in order to determine a vehicle attribute (61), the vehicle image (65) is compared with a group-typical vehicle of the assignable vehicle type, in particular to determine a type deviation (62) of the vehicle (60).

[0103] According to the invention, data is sent to an external image processing service. The image processing service is a trained system, for example, an artificial neural network or a support vector machine. Such a service is trained to recognize the vehicle type (vehicle manufacturer, model, variant).

[0104] Another embodiment is a method in which a determined vehicle attribute (61) includes vehicle-specific data, in particular regarding a degree of soiling, special equipment (62) and / or a type deviation compared to group-typical vehicles. The vehicle-specific data can be determined alternatively or additionally to group-typical data.

[0105] Depending on the embodiment of the method, the control parameter message (CP) includes at least one of the following elements: a determined vehicle attribute (61), in particular a virtual vehicle model, a vehicle contour, a vehicle shape, a vehicle equipment and / or a vehicle type; a type deviation (62) of the vehicle (60), in particular a special condition, special equipment; a soiling condition; a control instruction for the vehicle treatment system (10) derivable from a vehicle attribute (61) and / or a type deviation (62) and / or a soiling condition; one or more parameters for a treatment movement adapted to a vehicle attribute (61), in particular to a virtual vehicle model; a warning signal in relation to a hazardous situation, in particular in relation to a non-treatable special equipment (62). Reference symbol list

[0106] 10 Vehicle treatment facility, vehicle wash facility Vehicle treatment facility, vehicle washing facility 11 Treatment products, washing brush Treatment means, washing brush 15 Image capture device Imaging apparatus 16 camera Camera 17 Localization tools Localisation means 18 sensor sensor 20 Control device Control device 25 Communication interface Communication interface 30 Data processing service Data processing service 35 Data transmission module Data transmission module 40 Image processing service Image processing service 50 Vehicle database Vehicle database 60 vehicle Vehicle 61 Vehicle attribute, virtual vehicle model, vehicle contour, vehicle shape, vehicle equipment Vehicle attributes, virtual vehicle model, vehicle contour, vehicle shape, vehicle equipment 61a Treatment contour Treatment contour 62 Type deviation, special condition, special equipment, antenna Type deviation, special condition, special equipment, antenna 63 Undercut Undercut 65 Vehicle image Vehicle depiction 70 Entrance area Entry area 71 Treatment area Treatment area IN THE Image signal Image signal VD Vehicle image message Vehicle depiction message TR Type request Type request VT Vehicle type message Vehicle type message FR Property request Feature request VF Vehicle property message Vehicle feature message CP Control parameter message Control parameter message CS Control signal Control signal C Treatment movement, control effect Treatment motion, control effect

Claims

1. Computer-implemented method for controlling a vehicle treatment facility (10) comprising the following steps: - obtaining an image signal (IM) from at least one imaging apparatus (16) that captures a vehicle (60) in the area of the vehicle treatment facility (10), wherein the image signal comprises raw data from a camera; - sending a vehicle depiction message (VD) to a data processing service (30), wherein the vehicle depiction message (VD) comprises a vehicle depiction (65) relating to the vehicle in the area of the vehicle treatment facility; - the data processing service (30) generates a type request (TR) to determine a vehicle type in the vehicle depiction (65) and obtains a vehicle type signal (VT) from an external image processing service (40); wherein the data processing service (30) sends data to the external image processing service (40), which is a trained system and is trained to recognize the vehicle type; - the data processing service (30) provides a control parameter message (CP) for the surface treatment of the vehicle in the vehicle treatment facility; - receiving a control parameter message (CP) comprising, in particular, at least one of the following vehicle attributes (61) of the vehicle (60): a vehicle contour, a vehicle form; - generating a control signal (CS) for controlling the vehicle treatment facility (10) based on the control parameter message (CP).

2. Computer-implemented method according to claim 1, wherein the data processing service (30) uses the model to retrieve a specification of this vehicle model from a database (of a third-party provider).

3. Computer-implemented method according to claim 1 or 2, wherein the manufacturer and model variant of the vehicle are determined by the external image processing service (40) and by image processing of the vehicle depiction (65).

4. Computer-implemented method according to one of the preceding claims, wherein the computer-implemented method is executed in a system control of the vehicle treatment facility (10) and wherein the vehicle treatment facility is a gantry car wash or a car wash tunnel and comprises rotating brushes.

5. Computer-implemented method according to one of the preceding claims, wherein the control signal (CS) is designed to adapt a predetermined treatment program of the vehicle treatment facility (10) to the detected vehicle (60), in particular to one or more vehicle attributes (61) of the vehicle (60), for example the contour and / or special equipment of the vehicle.

6. Computer-implemented method according to one of the preceding claims, wherein the control parameter message (CP) further comprises at least one of the following vehicle attributes (61) of the vehicle (60): a vehicle model, a vehicle type, vehicle equipment, and / or a vehicle condition.

7. Computer-implemented method according to one of the preceding claims, wherein the vehicle (60) is detected by the imaging apparatus (15) before the start of a treatment process of this vehicle (60), in particular in an entry area (70) of the vehicle treatment facility (10).

8. Computer-implemented method according to one of the preceding claims, wherein the control signal (CS) is designed to specify a treatment motion (C), in particular a movement vector, a path or a contact force, of at least one treatment means (11) and / or to adapt a treatment motion (C) to a vehicle attribute (61), in particular before the start of the treatment program and / or before the vehicle (60') enters the treatment area (71) of the vehicle treatment facility (10).

9. Computer-implemented method according to one of the preceding claims, wherein the control parameter message (CP) and / or the control signal (CS) is designed to specify the vehicle contour (61) for the treatment of the vehicle (60) in the vehicle treatment facility (10), in particular before the start of the treatment process.

10. Computer-implemented method according to one of the preceding claims, wherein a vehicle attribute (61), in particular a virtual vehicle model (61), is generated and / or derived from a control parameter message (CP).

11. Computer-implemented method according to one of the preceding claims, wherein the control parameter message (CP), in particular a vehicle attribute (61) contained therein, is related to data from at least one sensor (18) of the vehicle treatment facility (10), wherein the sensor (18) is preferably designed to detect the vehicle (60') in the vehicle treatment facility during a treatment process.

12. Computer-implemented method according to one of the preceding claims, wherein the control signal (CS) can be adapted to the control parameter message (CP), in particular in real time, to activate a treatment program of the vehicle treatment facility (10) immediately after the vehicle (60) has been detected.

13. Computer-implemented method for providing a data processing service (30) for vehicle treatment facilities (10), comprising the following steps: - receiving a vehicle depiction message (VD) comprising a vehicle depiction (65) relating to a vehicle (60) in the area of the vehicle treatment facility (10); - determining at least one vehicle attribute (61) of the imaged vehicle (60) by processing the vehicle depiction (65); - generating a type request (TR) to determine the vehicle type in the vehicle depiction (65), wherein the data processing service (30) sends data to an external image processing service (40), which is a trained system and is trained to recognize the vehicle type; - obtaining a vehicle type signal (VT) from the external image processing service (40); - providing a control parameter message (CP) for vehicle treatment of the vehicle (60) in the vehicle treatment facility (10).

14. Computer-implemented method according to claim 13, wherein, in order to determine the vehicle attribute (61) of the vehicle (60), in particular a vehicle contour or a virtual vehicle model, the vehicle type is determined in a first intermediate step and, in a second intermediate step, the vehicle attribute (61) is determined on the basis of the vehicle type in a vehicle data base (50).

15. Computer-implemented method according to one of the preceding claims, wherein the vehicle attribute (61), in particular a vehicle type message (VT), comprises group-typical data, in particular manufacturer-related and / or model-typical data, and / or a general vehicle type of the vehicle (60).

16. Computer-implemented method according to one of the preceding claims, wherein, in order to determine a vehicle attribute (61), the vehicle depiction (65) is assigned to a previously known vehicle category.

17. Data processing system comprising means for executing the method according to one of claims 1 to 12 and / or the method according to one of claims 13 to 16.

18. Computer program comprising instructions which, when executed by a computer, cause the computer to execute a method according to one of claims 1 to 12 and / or a method according to one of claims 13 to 16.