Repair support system

Abstract

The invention relates to a repair support system comprising at least one computing unit (8) which, supported by artificial intelligence, is designed to automatically analyze at least one image depicting a vehicle (2) and to identify at least one defective component (4) of the vehicle (2) depicted in the at least one image, to propose at least one new component (12) for replacing the at least one defective component (4) and to request it from a source (10) for new components (12).

Description

[0001] The invention relates to a repair support system and a method for supporting the repair of a vehicle.

[0002] A method for optimizing the use of artificial intelligence in a manufacturing, repair or maintenance process is known from publication DE 10 2023 200 016 A1.

[0003] The publication DE 10 2014 226 554 A1 describes an identification and repair support device.

[0004] Publication US 2002 / 044104 A1 describes an augmented reality system for situation-specific support of an interaction between a user and a technical device.

[0005] Against this background, it was a task to provide suitable support for the repair of a vehicle.

[0006] This problem is solved by a repair support system and a method with the features of the independent claims. Embodiments of the repair support system and the method are described in the dependent claims and the description.

[0007] The repair support system according to the invention is designed for a vehicle typically designed as a motor vehicle and has at least one computing unit which, supported by artificial intelligence (AI), is designed to automatically analyze at least one image depicting the vehicle and to identify at least one defective component of the vehicle depicted in the at least one image, usually among other intact components of the vehicle, and to suggest at least one new component as a spare part for replacing the at least one defective component and to request and / or order this from a source for new components or spare parts.

[0008] The repair support system has at least one camera as a sensor, which is designed to capture at least one image, i.e., just one image or several successive images, e.g., from different perspectives, possibly in a video, of the vehicle and to provide it to at least one computing unit for automatically analyzing the at least one image.

[0009] Once at least one new component has been provided from the source, the at least one defective component is removed from the vehicle by a technician and / or a robot, and the at least one new component is installed and / or fitted into the vehicle by the technician and / or the robot. During the replacement process, at least one image of the vehicle can be captured by at least one camera and provided to at least one computing unit. The robot is then monitored and / or controlled by at least one computing unit using artificial intelligence during the component replacement, even in real time. Alternatively or additionally, the technician is monitored during the replacement process, taking into account at least one image captured during the process, and monitored by at least one computing unit, e.g., via a camera.guided, if necessary, via a human-machine interface of the repair support system.

[0010] The artificial intelligence is trained to automatically analyze and assess, during the exchange process, usually at least after the replacement of the at least one defective component with the at least one new component or spare part, at least one new image showing the vehicle with the at least one new component instead of the at least one component identified as defective and already removed from the vehicle, to determine whether the at least one new component is correctly installed in the vehicle.

[0011] The artificial intelligence is further trained to document the vehicle's condition using at least one image taken by at least one camera before and after the replacement of at least one defective component with at least one new component, and to compare the condition and / or any changes in the condition resulting from the replacement. It is possible that the at least one defective component is replaced by at least one identical new component. It is also possible that the artificial intelligence replaces the defective, technically outdated component with at least one technically advanced new component, whereby this at least one technically advanced component is automatically identified, suggested, and / or selected beforehand by the artificial intelligence.

[0012] The method according to the invention is designed to support the repair of a vehicle. In this method, at least one computing unit, supported by artificial intelligence, automatically analyzes at least one image depicting the vehicle and identifies at least one defective component of the vehicle among intact components, usually based on its appearance, provided that the at least one defective component is present and depicted in the at least one image. To replace the at least one component identified as defective, the at least one computing unit, supported by artificial intelligence, proposes at least one new component and requests and / or orders it from a source of new components.

[0013] After the component is replaced, at least one camera takes a new picture of the vehicle, focusing on the new component. At least one processing unit then analyzes and verifies whether the old, defective component was replaced correctly and whether the new component was correctly installed in the vehicle. If the processing unit determines that the new component was incorrectly installed after replacement, it suggests, with the support of artificial intelligence, how to install it correctly.

[0014] The system is designed so that at least one computing unit and its associated artificial intelligence can support multiple repairs of different vehicles that are identical in construction and / or of the same type. For each vehicle, at least one defective component is replaced with at least one new component during each repair. The artificial intelligence is trained and / or educated based on previously performed repairs to support future repairs involving component replacement.

[0015] Additionally, it is possible for the artificial intelligence executed by the at least one computing unit to communicate with at least one vehicle device, such as at least one on-board computer, an electrical control unit (ECU), at least one sensor, at least one actuator, and / or the vehicle's electrical system, before, during, and after the component replacement, exchanging signals with information about the component and the replacement to be carried out or already carried out. Integrating the at least one computing unit into the vehicle or a vehicle system and enabling communication between the at least one computing unit and the vehicle allows for the optimization of the repair or repair process.

[0016] In this process, the artificial intelligence can suggest maintenance after a vehicle component has been replaced. This is possible based on data about components and their repairs, which is collected, provided to the AI, and used for training. This also applies to data on repairs that the AI ​​collects during repairs and / or while performing the process itself. Through the collected data and its continuous learning capability, the AI ​​is able to make and / or provide proactive maintenance suggestions, thereby preventing potential component problems before they occur.Artificial intelligence can be used to develop advanced diagnostic functions for detecting defective components, identifying more complex or less obvious problems and thus defective components that might otherwise be overlooked.

[0017] It is possible that an embodiment of the presented repair support system is configured to carry out an embodiment of the presented method, wherein the embodiment of the presented method is carried out with the embodiment of the presented repair support system.

[0018] The presented repair support system is AI-supported and equipped with at least one camera and at least one processing unit, which together form a repair support device and / or a generative AI unit or device for carrying out the procedure. It is possible that a sensor and an AI-supported processing unit combine to form a wearable unit or repair support device, which can be designed and / or described as AR smart AI glasses, i.e., glasses for augmented reality with artificial intelligence, i.e., as an optical sensor with artificial intelligence. Such repair support devices suitable for the procedure are known, for example, under the brand names "Humane AI Pin" or "Rabbit R1". A repair support device usable in the procedure is designed to perceive its environment, including the vehicle, with the camera and to respond to user input, e.g.,...The repair support device enables the technician to react to and perceive, capture, and / or understand the vehicle and its components using generative artificial intelligence, e.g., through multimodal LLMs (Large Language Models) or visual LLMs, with the user being assisted by the device. The repair support device can, for example, recognize what is in the environment, how to interact with it, and may also ask questions about the environment. In this implementation, the repair support device captures and / or recognizes vehicle components, distinguishing between intact and defective components.

[0019] The repair support system allows for precise documentation of the vehicle's condition before, during, and after replacement. This documentation of the vehicle's condition and repair can also be provided to the vehicle manufacturer in a quality-assured manner. This enables, for example, the efficient and accurate selection and installation of the necessary replacement component for the defective component through information exchange between the at least one processing unit and the manufacturer. The system can be used to replace electrical and / or mechanical components, i.e., purely electrical, electromechanical, or purely mechanical components without electronics. Any defect or damage to such a purely mechanical component can be identified using the at least one image depicting the component.For such components without electronics, the process also allows for image- and AI-supported documentation and / or feedback of each replacement. Alternatively or additionally, a defective non-electronic component or component without electronics identified by the technician and / or the repair support device can be included in the documentation and / or feedback within the image- and AI-supported process.

[0020] It is possible that the at least one optical sensor and / or the at least one computing unit is configured as an AI-supported device for detecting the vehicle and its components. This device, e.g., a repair support device, combines the at least one sensor and the at least one computing unit, with the artificial intelligence having been and / or being trained in advance with all possible steps for repairing and thus replacing the at least one component for all potential components of a class of vehicles or vehicle class. Furthermore, the at least one AI-supported computing unit is networked with the manufacturer's vehicle condition documentation.

[0021] Artificial intelligence can be supplemented with additional data and regularly trained, and / or it can train itself automatically, with automatic updates. In this process, the artificial intelligence can be trained with new techniques for diagnosing and / or detecting defective components, as well as techniques for repair.

[0022] The repair support system improves the efficiency and accuracy of repairs and ensures precise documentation of the vehicle's condition. Additionally, repair costs can be reduced by automating and shortening the time required for identifying and ordering spare parts through artificial intelligence. Furthermore, continuous learning and adaptation of the artificial intelligence to new situations and challenges can make the entire repair process more efficient.

[0023] The self-learning repair support system can identify each vehicle component as either defective or intact during the repair process. Once a defective component is identified, the correct new component or replacement part can be selected, taking into account the specific hardware and / or software combination of the defective component with other, typically intact, vehicle components. Additionally, the repair support system ensures error-free documentation and / or the return documentation of each replacement. The at least one AI-supported computing unit and the at least one device in the vehicle and / or a robot used for component replacement can exchange data and interact using AI and / or a multimodal language model (multimodal Large Language Model, LLM).

[0024] 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.

[0025] The invention is schematically illustrated with reference to embodiments in the drawing and is described schematically and in detail with reference to the drawing. Fig. Figure 1 shows a schematic representation of an embodiment of the repair support system according to the invention when carrying out an embodiment of the method according to the invention.

[0026] The in Fig. Figure 1, a schematically represented embodiment of the repair support system according to the invention, comprises a camera 6 as a sensor, a computing unit 8 for performing artificial intelligence, and a source 10. The camera 6 and the computing unit 8 can form a single device and be designed and / or designated as a repair support device. In the embodiment of the method according to the invention, at least four successive steps S1, S2, S3, and S4 are performed, which are based on Fig. 1 is illustrated.

[0027] A first step S1 of the embodiment of the method is carried out, for example, in a workshop where the camera 6 and the processing unit 8 are located. Here, the method is performed for a vehicle 2 designed as a motor vehicle, whereby the processing unit 8 activates a repair order for the vehicle 2 because a fault in the vehicle 2 has already been detected, for example, during a journey, e.g., by a control unit and / or the driver of the vehicle 2, and / or because an inspection or examination of the vehicle 2 was already scheduled.

[0028] In the first step S1, components 4 of the vehicle 2 are examined by the camera 6, which exchanges data with the AI-supported computing unit 8. Each visually identifiable component 4 currently installed in the vehicle 2 is analyzed, and the computing unit 8 can also exchange data with the vehicle 2's control unit. Here, the camera 6 and the computing unit 8 sensorily detect a component 4, in this case a purely mechanical component 4, and identify it as defective based on its appearance. The replacement of the defective component 4 is then planned and / or suggested as part of a repair. The camera 6 and the computing unit 8 are designed as a repair support device to distinguish between defective and intact components 4 of the vehicle 2 during an examination and analysis of its components 4, supported by artificial intelligence.

[0029] Such an analysis and detection of a defective component 4 can be carried out fully automatically by the computing unit 8 and the camera 6, whereby the camera 6 is moved relative to the vehicle 2 by a robot controlled by the computing unit 8, examining and / or inspecting its components 4. The camera 6 captures images of the components 4, which are then evaluated by the computing unit 8 using artificial intelligence. The artificial intelligence can then identify the component 4 depicted in an image as defective.

[0030] The first step S1 is followed (arrow 20) by a second step S2 of the process. The second step S2 comprises a proposal for a suitable new component 12 as a replacement for the component 4 identified as defective, and furthermore, an order for this new component 12 from a source 10, in this case a distributor and / or warehouse, for such components 12. The artificial intelligence proposes the new component 12 with regard to its hardware and software version. It is possible that the artificial intelligence selects the new component 12 from a selection of other new components 12 or from among several usable components 12. As soon as the new component 12 intended for replacement and / or required has been identified and / or selected, the artificial intelligence initiates an ordering process.The new component 12, as a replacement for the defective component 4, is ordered by the artificial intelligence from source 10, whereby it is requested directly from source 10, e.g. the warehouse, and / or ordered online from source 10 through communication between the artificial intelligence and source 10.

[0031] In a third step (S3) following the second step S2 (arrow 22), the new component 12 is delivered and thus ready for installation in vehicle 2. During the repair of the defective component 4, the robot and / or the technician can be supported by artificial intelligence, for example, via a human-machine interface of the repair support system. Such support includes removing the defective component 4 from vehicle 2 and installing the new component 12 in vehicle 2. The installation of the new component 12 is recorded by camera 6 and analyzed, typically checked, and / or monitored by the artificial intelligence.If an error is detected and / or identified by the artificial intelligence during installation, the technician and / or robot can be provided with a suggestion for correcting the error and for proper installation. In the third step, S3, after the installation of the new component 12, the artificial intelligence documents and / or updates the status of the vehicle 2, noting the new component 12 in the documentation of the vehicle 2 manufacturer, and ensuring that all changes to the vehicle 2 are properly documented.

[0032] Following this third step S3 of the procedure, the artificial intelligence of the repair support system can subsequently be trained in a fourth step S4 (arrow 24) during training 14. It is possible for the artificial intelligence to be trained using and / or on a large number of computing units 8, which are spatially distributed across a large number of workshops. For the training, data and / or information from all repairs of defective components 4 of different vehicles 2 of the same manufacturer that have already been carried out and / or supported by the repair support system are collected, provided as feedback, and used for the training 14 of the artificial intelligence. In this process, the artificial intelligence is trained and / or taught to recognize a defective component 4 in the vehicle 2 as such.to identify, determine and order the new component 12 for the defective component 4, and furthermore to monitor and / or detect the correct removal of the defective component 4 and the installation of the new component 12. The artificial intelligence continuously learns from each completed repair, and after training 14 (arrow 26) it is updated and / or optimized for future repairs and diagnoses. REFERENCE MARK: 2 vehicles 4 components 6 cameras 8 computing units 10 Source 12 components 14 Training 20, 22, 24, 26 Arrow S1, S2, S3, S4 step QUOTES INCLUDED IN THE DESCRIPTION

[0000] This list of documents cited by the applicant was automatically generated and is included solely for the reader's convenience. The list is not part of the German patent or utility model application. The DPMA accepts no liability for any errors or omissions. Cited patent literature

[0000] DE 10 2023 200 016 A1

[0002] DE 10 2014 226 554 A1

[0003] US 2002 / 044104 A1

[0004]

Claims

Repair support system comprising at least one computing unit (8) which, supported by artificial intelligence, is trained to automatically analyze at least one image depicting a vehicle (2) and to identify at least one defective component (4) of the vehicle (2) depicted in the at least one image, to propose at least one new component (12) for replacing the at least one defective component (4) and to request it from a source (10) for new components (12). Repair support system according to claim 1, comprising at least one camera (6) configured to capture at least one image of the vehicle (2). Repair support system according to claim 1 or 2, wherein the artificial intelligence is trained to automatically analyze and assess, after replacement of the at least one defective component (4) by the at least one new component (12), at least one new image depicting the vehicle (2), whether the at least one new component (4) is correctly installed in the vehicle (2). Repair support system according to claim 3, wherein the artificial intelligence is trained to document a condition of the vehicle (2) by means of at least one image taken before and after replacement of the at least one defective component (4) with the at least one new component (12). Method for supporting the repair of a vehicle (2) in which at least one computing unit (8), supported by artificial intelligence, automatically analyzes at least one image depicting the vehicle (2) and identifies at least one defective component (4) of the vehicle (2) depicted in the at least one image, wherein at least one new component (12) is proposed and requested from a source (10) for new components (12) to replace the at least one defective component (12). Method according to claim 5, wherein, in the event that the at least one new component (12) has been incorrectly installed after replacement, the artificial intelligence makes a suggestion for the correct installation of the new component (12). Method according to claim 5 or 6, wherein the artificial intelligence supports multiple repairs of vehicles (2), wherein in each vehicle (2) at least one defective component (4) of the vehicle (2) is replaced with at least one new component (12) during a repair, wherein the artificial intelligence is trained on the basis of repairs already carried out to support future repairs. Method according to one of claims 5 to 7, wherein the artificial intelligence communicates with at least one device of the vehicle (2). Method according to one of claims 5 to 8, wherein the artificial intelligence proposes maintenance after replacement of the component (4, 12) for the vehicle (2).

Images