Methods for locating a stolen vehicle

Abstract

The present invention relates to a method for locating a stolen vehicle (2). When a vehicle (2) is stolen, a search request (SA) to search for the vehicle (2) is transmitted to a backend server (3), and a collection instruction (A) to collect sensor data (D1~Dn) that includes location-specific information (OI) and is detected using the vehicle's on-board sensors (5.1~5.n) is transmitted from the backend server (3) to the vehicle (2). Based on this, the on-board sensors (5.1~5.n) are used to detect sensor data (D1~Dn) and transmit it to the backend server (3). The backend server (3) extracts location-specific information (OI) from the transmitted sensor data (D1~Dn, D1'~Dn'), and the location of the vehicle (2) is determined by comparing the location-specific information (OI) with location information (KOI) stored in a digital map (7).

Description

Technical Field

[0001] The present invention relates to a method for identifying the position of a stolen vehicle.

Background Art

[0002] From German Patent Application Publication No. 102009028297, a method for identifying the position of a stolen or accident - affected vehicle is known. In this method, the position of the vehicle is identified using position - detecting means arranged in the vehicle, the quality of the position identification is evaluated, and the identified position and the evaluated quality are transmitted to a control center. This transmission is performed using a device arranged in the vehicle for automatically transmitting the position of the vehicle.

[0003] From German Patent Application Publication No. 19649136, a position - identification system for discovering a stolen vehicle is known. This position - identification system includes the following combination. That is, - A satellite station in space equipped with a receiving part and a transmitting part, using a data - transmission connection for transmitting and receiving vehicle - identification data between a computer on the search side and for transmitting and receiving vehicle - identification data between a decodable microchip of a vehicle to be searched, - A transmitting part and a receiving part inside the vehicle, which are installed inside the vehicle as inconspicuous as possible to a third party and are in the form of a microchip for coding vehicle identification, and are operated by a current - supply part inside the vehicle, the transmitting part and the receiving part inside the vehicle, and - A computer for storing and providing data for vehicle identification, equipped with a transmitting part and a receiving part for transmitting data to the satellite station and receiving data from the satellite station.

[0004] Furthermore, German Patent Application Publication No. 102008023970 provides a known method for identifying traffic-related information in a moving vehicle, in which image data from a visual sensor and map data from a navigation system are pre-evaluated for identification, and the results of the pre-evaluations are integrated and interpreted. In the first scene interpretation, the traffic-related information contained therein is retrieved from the image data and compressed into related image data. In the second scene interpretation, the traffic-related information contained therein is retrieved from the map data and compressed into related map data. The related image data and related map data are supplied to a control mechanism and a state machine for interpreting the image data and map data. The control mechanism evaluates the image data and map data and transfers the actions to be performed to the state machine. Based on the actions to be performed, the state machine performs transitions in a state space including predefined states and outputs information associated with those states or transitions. [Overview of the Initiative] [Problems that the invention aims to solve]

[0005] The present invention aims to provide a novel method for locating a stolen vehicle. [Means for solving the problem]

[0006] This objective is achieved, according to the present invention, by a method having the features described in claim 1.

[0007] Advantageous embodiments of the present invention are subject to the dependent claims.

[0008] In a method for locating a stolen vehicle, according to the present invention, when a vehicle is stolen, a search request for the vehicle is transmitted to a backend server, and a collection instruction is transmitted from the backend server to the vehicle for collecting sensor data that includes location-specific information and is detected using the vehicle's onboard sensors. Based on this, sensor data is detected using the onboard sensors and transmitted to the backend server. The backend server extracts location-specific information from the transmitted sensor data, and the vehicle's location is determined by comparing the location-specific information with location information stored in a digital map.

[0009] After a vehicle is stolen, rapid and reliable location tracking is necessary to recover it. However, vehicle thieves often use so-called GPS jammers to prevent GPS data-based location tracking. This method allows for easy and reliable location tracking of a vehicle even when such GPS jammers are in use. A particularly advantageous method for detecting sensor data is the use of sensors already installed on the vehicle. This makes it possible to implement this method with very few additional hardware requirements on the vehicle.

[0010] In one possible embodiment of this method, filtering of sensor data occurs before transmission from the vehicle to the backend server. - Personal data of other passengers and / or - Vehicle-related data of other traffic participants This is removed. This allows for the protection of personal data and / or vehicle-related data, and ensures that vehicle location can be reliably achieved without violating data protection requirements.

[0011] In one possible embodiment of this method, image data detected using at least one camera is used as sensor data. Image data of the vehicle's surrounding environment detected using a camera is particularly suitable for reliably identifying location-specific information. Cameras are robust, inexpensive, and often already equipped on vehicles. Furthermore, in image data detected using a camera, personal data and / or vehicle-related data can be detected and filtered particularly easily and reliably through image recognition.

[0012] In one possible embodiment of this method, location-specific information is extracted from traffic signs represented in the image data. This can be done with particular reliability using a so-called traffic sign recognition algorithm.

[0013] In one possible embodiment of this method, as a traffic sign, at least - Signposts and / or - Regional signs and / or - Place name signs and / or - Road name signs and / or - Tourist information signs This is taken into consideration. This type of traffic sign enables easy and reliable vehicle location identification by providing information about geographical location and / or distance to a specific geographical location.

[0014] In particular, it becomes possible to pinpoint the region where a stolen vehicle is located, and if the vehicle is outside the country, it becomes possible to initiate a search for the vehicle across borders.

[0015] In one other embodiment of this method, the location-specific information is validated and / or its accuracy is improved in the backend server using other sensor data detected by the in-vehicle sensor. -As separate sensor data, other image data detected using at least one camera is used, in this case, - The content of traffic signs that regulate the maximum speed limit, and / or - Regulations for traffic signs to identify the start and / or intermediate points and / or end of road construction sites, and / or - Information regarding lane markings, and / or - Information regarding traffic signals, and / or - Information about structures This is extracted from the image data and used as validation information, and / or -As additional sensor data, radar data detected using at least one radar is used, in which case information regarding the start and / or end of the tunnel extracted from the radar data is used as validation information, and / or -As additional sensor data, chassis data detected using at least one chassis sensor and / or at least one accelerometer is used, in which case information regarding the elevation of the road surface and / or information regarding depressions on the road surface extracted from the chassis data is used as validation information and / or -As additional sensor data, data detected using at least one inertial sensor system and / or wheel rotation speed sensors of the vehicle is used, in which case the vehicle trajectory calculated from the data is used as validation information.

[0016] Such validation and / or accuracy improvements enhance the reliability of identifying the vehicle's geographical location and improve vehicle positioning. In particular, when location-specific information identified from sensor data is ambiguous, the actual geographical location of the vehicle can be determined with greater reliability.

[0017] In one possible embodiment of this method, validation information is compared with information stored on a backend server, and location-specific information is identified based on this comparison, thereby simplifying and / or improving the accuracy of validating location-specific information identified from sensor data.

[0018] In one possible embodiment of this method, the sensor data is processed before transmission from the vehicle to the backend server. - The calendar date of the detection of the sensor data and / or - The time of the detection of the sensor data and / or - The section where the vehicle has traveled after being last activated after theft is associated and transmitted to the backend server together with the sensor data. Thereby, it is possible to achieve verification of the validity and / or improvement of the accuracy of the location-specific information identified from the sensor data. Furthermore, it is possible to obtain the time passage of the movement of the vehicle after the theft of the vehicle.

[0019] In another possible embodiment of the method, based on the geographical location of the vehicle identified in the vehicle location determination, the driving route of the vehicle is obtained, and based on the driving route, at least one possible destination of the vehicle is predicted. By this prediction of the destination, measures at the destination can be taken early to retrieve the vehicle again.

[0020] In another possible embodiment of the method, the identification of the location-specific information is automatically performed by a machine-learned algorithm. Such identification of the location-specific information can be performed particularly advantageously very fast, and thus at least approximately in real time.

[0021] Hereinafter, embodiments of the present invention will be described in detail based on the drawings.

Brief Description of the Drawings

[0022] [Figure 1] It is a schematic block diagram of an apparatus for identifying the location of a stolen vehicle. [Figure 2] It is a schematic diagram of a plurality of traffic signs.

Modes for Carrying Out the Invention

[0023] In any of the figures, the same reference numerals are given to the corresponding parts.

[0024] Figure 1 shows a block diagram of a possible embodiment of the device 1 for locating the stolen vehicle 2.

[0025] If a report of theft of vehicle 2 is received, that is, if a report is received that vehicle 2 has been stolen, this information is transmitted to backend server 3 via search request SA. Subsequently, backend server 3 sends a corresponding notification to vehicle 2. This notification is transmitted via information I over the mobile radio network and received by vehicle 2's mobile radio module 4.

[0026] Information I includes location-specific information OI and collection instruction A for collecting sensor data D1 to Dn detected using on-board sensors 5.1 to 5.n of vehicle 2.

[0027] The mobile radio module 4 of vehicle 2 transmits information I, including collection instruction A, to the control device 6 of vehicle 2. This control device 6 is, for example, a central driver assistance system control device, also known as an ADAS control device (ADAS = Advanced Driver Assistance System).

[0028] Then, the surrounding environment of the vehicle 2 is detected using the on-board sensors 5.1 to 5.n connected to the control device 6, and the detected sensor data D1 to Dn is transmitted to the control device 6. The on-board sensors 5.1 to 5.n include, for example, at least one camera, radar, at least one chassis sensor, at least one acceleration sensor, wheel rotation speed sensor and / or inertial sensor system. The sensor data D1 to Dn detected using the on-board sensors 5.1 to 5.n are subsequently supplied to the control device 6.

[0029] Since the sensor data D1 to Dn may contain personal data and / or vehicle-related data of other traffic participants, in one possible embodiment, such personal data and / or vehicle-related data contained in the sensor data D1 to Dn are removed from the sensor data D1 to Dn in filtering performed using the control device 6.

[0030] For example, image data detected using at least one camera shows traffic signs V1 to V7, as detailed in Figure 2. Here, the image data may contain unrelated individuals and / or other vehicles, such as their public identification information, which represents so-called personal data and / or vehicle-related data. Filtering is performed, for example, using traffic sign recognition, which selects the relevant image portions in the image data where the corresponding traffic signs V1 to V7 exist, and removes the unnecessary image portions from the sensor data D1 to Dn.

[0031] The filtered sensor data D1'~Dn' is then transmitted to the backend server 3 using the mobile wireless module 4, and the backend server 3 extracts location-specific information OI from the transmitted sensor data D1'~Dn'.

[0032] In the data evaluation DA performed using backend server 3, a comparison is made between location-specific information (OI) and location information (KOI) stored in the digital map 7 of the road network. Based on the identification results of this comparison, the location of vehicle 2 is determined. In other words, the location of vehicle 2 is determined based on the results of the comparison.

[0033] For example, location-specific information OI is extracted from traffic signs V1 to V7 represented in the transmitted and filtered sensor data D1' to Dn', and this location-specific information OI is identified, for example, from road signs and / or local signs and / or place name signs and / or road name signs and / or tourist information signs.

[0034] Furthermore, the location-specific information OI is compared with other sensor data D1-Dn detected using the in-vehicle sensors 5.1-5.n in the backend server 3, particularly to confirm its validity and / or improve its accuracy.

[0035] For example, other sensor data D1-Dn may be other image data detected using at least one camera, in which case the validation information may be the content of traffic signs not shown in detail for regulating the maximum speed limit, and / or the content of traffic signs not shown in detail for identifying the start and / or intermediate points and / or end of a road construction site, extracted from the image data. Furthermore, the validation information may include information about lane markings, particularly the number of lane markings, the type of lines, and / or the relative position of the vehicle 2 to those lane markings. Additionally, the validation information may include information about traffic signals and / or structures. In this case as well, in one possible embodiment, the other sensor data D1-Dn are filtered using the control device 6 so that personal data and / or vehicle-related data contained in the other sensor data D1-Dn are removed before transmission to the backend server 3.

[0036] Alternatively or additionally, radar data detected using at least one radar, for example, a wide-area radar, may be used as other sensor data D1 to Dn, in which case information about the start and / or end of the tunnel extracted from the radar data is used as validation information. In this case as well, in one possible embodiment, the other sensor data D1 to Dn are filtered using the control device 6 so that any personal data and / or vehicle-related data contained in the other sensor data D1 to Dn is removed before transmission to the backend server 3.

[0037] Similarly, alternatively or additionally, chassis data detected using at least one chassis sensor and / or at least one acceleration sensor of the vehicle 2 may be used as other sensor data D1-Dn, in which case information on road elevation and / or depressions on the road surface extracted from the chassis data may be used as validation information. In this case as well, in one possible embodiment, the other sensor data D1-Dn may be filtered using the control device 6 so that any personal data and / or vehicle-related data contained in the other sensor data D1-Dn may be removed before transmission to the backend server 3.

[0038] Alternatively or additionally, data detected using at least one inertial sensor system and / or wheel rotation speed sensors of vehicle 2 may be used as other sensor data D1-Dn, in which case the vehicle trajectory calculated from the data may be used as validation information. The calculation of the vehicle trajectory is performed in particular after the theft of vehicle 2 is recognized. In this case as well, in one possible embodiment, the other sensor data D1-Dn may be filtered using the control device 6 so that any personal data and / or vehicle-related data contained in the other sensor data D1-Dn may be removed before transmission to the backend server 3.

[0039] The validation information is also compared with the information stored in the digital map 7 and / or the information stored in the backend server 3. In this case, location-specific information (OI) is identified based on the comparison, and validation is performed based on this location-specific information (OI).

[0040] Additionally, the calendar date of detection of sensor data D1~Dn, and / or the time of detection of sensor data D1~Dn, and / or the section traveled by vehicle 2 since it was last activated after the theft, are associated with all detected sensor data D1~Dn, or all filtered D1'~Dn', before transmission from vehicle 2 to backend server 3, and are transmitted to backend server 3 together with sensor data D1~Dn or filtered sensor data D1'~Dn'.

[0041] Based on the sensor data D1~Dn and D1'~Dn' transmitted to the backend server 3, and based on the aforementioned information, the estimated location of vehicle 2 can be determined by comparing it with a digital map 7 of the road network and with other information from the backend 3, such as known information about road construction sites derived from construction site identification by highly automated driving vehicles, as well as information about depressions or uplifts.

[0042] This comparison can be performed by an authorized individual or automatically in real time by an appropriate algorithm, such as a machine learning algorithm. The algorithm is configured to identify place names, distances to geographical locations, road names, etc., from sensor data D1-Dn, D1'-Dn' and additional information using so-called OCR recognition. Furthermore, the algorithm is configured to determine the possible vehicle position at each point in time by comparing a series of observations contained in the sensor data D1-Dn, D1'-Dn' with a digital map 7, such as one used in a navigation system. In addition to the sensor data D1-Dn, D1'-Dn', the aforementioned additional sensor data D1-Dn, such as lane markings, traffic signs, construction sites, and tunnel identification, is used to validate the location-specific information OI identified from the sensor data D1-Dn, D1'-Dn', and therefore to further refine the results and eliminate possible ambiguities.

[0043] From the estimated location of the vehicle identified as described above, clues can be obtained about a reasonable route taken by vehicle 2 and its destination.

[0044] Figure 2 illustrates several traffic signs V1 to V7 that can be extracted from sensor data D1 to Dn to identify location-specific information OI.

[0045] Traffic sign V1 includes, for example, multiple road markers. Traffic sign V2 is also a road marker. Traffic signs V3 and V5 are tourist information signs. Traffic sign V4 is a place name sign, and traffic signs V6 and V7 are road name signs.

[0046] Furthermore, although not illustrated in detail, traffic signs V1 to V7 may also be signs indicating other geographical locations, such as local signs or other types of signs. [Prior art documents] [Patent Documents]

[0047] [Patent Document 1] German Patent Application Publication No. 102009028297 [Patent Document 2] German Patent Application Publication No. 19649136 [Patent Document 3] German Patent Application Publication No. 102008023970

Claims

1. In a method for locating the stolen vehicle (2), - When the vehicle (2) is stolen, a search request (SA) to search for the vehicle (2) is transmitted to the backend server (3). - A collection instruction (A) for collecting sensor data (D1 to Dn) that includes location-specific information (OI) and is detected using on-board sensors (5.1 to 5.n) of the vehicle (2) is transmitted from the backend server (3) to the vehicle (2). - Based on the collection instruction (A), sensor data (D1 to Dn) is detected using the on-board sensors (5.1 to 5.n). - The sensor data (D1 to Dn, D1' to Dn') is transmitted to the backend server (3), - Using the backend server (3), location-specific information (OI) is extracted from the transmitted sensor data (D1 to Dn, D1' to Dn'), - A method characterized in that the location of the vehicle (2) is determined by comparing the location-specific information (OI) with location information (KOI) stored in a digital map (7).

2. Before the transmission from the vehicle (2) to the backend server (3), filtering of the sensor data (D1 to Dn) is performed. - Personal data of other transport participants and / or - Vehicle-related data of other traffic participants The method according to claim 1, characterized in that the substance is removed.

3. The method according to claim 1 or 2, characterized in that image data detected using at least one camera is used as sensor data (D1 to Dn).

4. The method according to claim 3, characterized in that the location-specific information (OI) is extracted from traffic signs (V1 to V7) represented in the image data.

5. As traffic signs (V1 to V7), at least - Signposts and / or - Local signs and / or - Place name signs and / or - Road name signs and / or - Tourist information sign The method according to claim 4, characterized in that the following is taken into consideration.

6. In the backend server (3), the location-specific information (OI) is validated and / or the accuracy of the location-specific information (OI) is improved using other sensor data (D1 to Dn, D1' to Dn') detected using the on-board sensors (5.1 to 5.n). - As additional sensor data (D1 to Dn, D1' to Dn'), other image data detected using at least one camera is used. - The content of traffic signs that regulate the maximum speed limit, and / or - Regulations for traffic signs used to identify the start and / or intermediate points and / or end of road construction sites, and / or - Information regarding lane markings, and / or - Information regarding traffic signals, and / or - Information regarding the structure is extracted from the image data and used as validation information, and / or - Radar data detected using at least one radar is used as additional sensor data (D1 to Dn, D1' to Dn'), and information regarding the start and / or end of the tunnel extracted from the radar data is used as validation information, and / or - Chassis data detected using at least one chassis sensor and / or at least one acceleration sensor is used as additional sensor data (D1 to Dn, D1' to Dn'), and information regarding the elevation of the road and / or depressions on the road surface extracted from the chassis data is used as validation information, and / or The method according to any one of claims 1 to 5, characterized in that data detected using at least one inertial sensor system of the vehicle (2) and / or a wheel rotation speed sensor of the vehicle (2) is used as other sensor data (D1 to Dn, D1' to Dn'), and the vehicle trajectory calculated from the data is used as validation information.

7. The method according to claim 6, characterized in that the validation information is compared with information stored in the backend server (3), and location-specific information (OI) is identified based on the comparison.

8. Before the transmission from the vehicle (2) to the backend server (3), the sensor data (D1 to Dn, D1' to Dn') is processed as follows: - The calendar day and / or of the detection of the sensor data (D1 to Dn, D1' to Dn') - The detection time and / or of the sensor data (D1 to Dn, D1' to Dn') - The section that the vehicle (2) traveled since it was last activated after the theft. The method according to any one of claims 1 to 7, characterized in that the data is associated with the sensor data (D1 to Dn, D1' to Dn') and transmitted to the backend server (3) together with the sensor data.

9. The method according to any one of claims 1 to 8, characterized in that a driving route for the vehicle (2) is determined based on the geographical location of the vehicle (2) identified in the location identification of the vehicle (2), and at least one possible destination for the vehicle (2) is predicted based on the driving route.

10. The method according to any one of claims 1 to 9, characterized in that the identification of the location-specific information (OI) is performed automatically by a machine learning algorithm.

Images