Access restriction of vehicle-associated information

Abstract

The present invention relates to access restrictions for vehicle-associated information. In some examples, a system receives vehicle-associated information from a data source associated with a vehicle and restricts access to the vehicle-associated information based on at least one privacy criterion selected from a machine learning usage criterion relating to use of the vehicle-associated information by a machine learning model, a vehicle motion criterion relating to a state of motion of the vehicle, or a personnel identity criterion relating to an identity of a person in the vehicle.

Description

Background Technology

[0001] Vehicles may include or receive data from sources that provide vehicle association information. The data source can be inside or outside the vehicle. Entities, whether inside or outside the vehicle, may attempt to access the vehicle association information. Attached Figure Description

[0002] Some implementations of this disclosure are described with reference to the following figures.

[0003] Figure 1 This is a block diagram of a filtering engine based on some examples of vehicles and vehicle association information.

[0004] Figure 2 It is a flowchart based on some examples.

[0005] Figure 3 It is a block diagram based on some example computer systems.

[0006] Throughout the accompanying drawings, the same reference numerals denote similar but not necessarily identical elements. The drawings are not necessarily drawn to scale, and the dimensions of some parts may be enlarged to more clearly illustrate the examples shown. Furthermore, the drawings provide examples and / or implementations consistent with the description; however, the description is not limited to the examples and / or implementations provided in the drawings. Detailed Implementation

[0007] In this disclosure, the terms “a,” “an,” or “the” are also intended to include the plural form unless the context clearly indicates otherwise. Furthermore, the terms “includes,” “including,” “comprises,” “comprising,” “have,” or “having” as used in this disclosure specify the presence of the stated element but do not preclude the presence or addition of other elements.

[0008] It may be desirable to restrict access to vehicle-related information to protect privacy (e.g., the privacy of occupants in the vehicle, the privacy of the vehicle owner, etc.), to provide security, or for other purposes. In some examples, access to vehicle-related information may be based on a license associated with the requesting entity (such as a user, program, or machine). The access license may specify whether the entity is permitted to access the vehicle-related information. The access license may be full or no; that is, the requesting entity may have full access to the vehicle-related information or no access at all. Coarse-grained access control using licenses may result in the exposure of too much vehicle-related information or the provision of too little.

[0009] Examples of vehicle-related information may include any or a combination of the following: data from vehicle sensors (e.g., speedometers that measure vehicle speed, accelerometers that measure vehicle acceleration, vehicle fluid monitoring sensors, tire pressure sensors, temperature sensors, pressure sensors, humidity sensors, engine speed sensors, etc.), vehicle location data, (interior or exterior) static or video images, data stored in storage media within the vehicle, identification information related to vehicle occupants (driver or passenger), data related to the use of vehicle safety equipment (e.g., seat belts, anti-lock brakes, collision avoidance control systems, etc.), data related to facial recognition or other biometric data, data related to vehicle occupants (e.g., weight, seat position, etc.), or any other vehicle-related information.

[0010] Vehicle-related information can be generated from data sources within the vehicle, or alternatively or additionally from data sources outside the vehicle. For example, the data source can be part of a roadside unit (RSU), a remote server, etc. Examples of RSUs can include traffic lights, electronic road signs, or any other electronic equipment located on or near the road where the vehicle is traveling. A remote server can refer to a computer system (including one or more computers), such as a computer system that is part of a cloud, data center, web environment, etc.

[0011] Figure 1 This is a block diagram of an example arrangement of a vehicle 102, which has a vehicle association information filtering engine 104, according to some implementations of this disclosure. The vehicle association information filtering engine 104 is used to control access to vehicle association information from various data sources by entities (users, programs, and / or machines). Entities that may request access to vehicle association information may be entities inside or outside the vehicle 102. Access control for various entities accessing vehicle association information can provide some or all of the following benefits: enhanced data privacy by reducing the sampling rate at which vehicle association information is provided to entities, enhanced data privacy by controlling when vehicle association information is available to entities based on one or more criteria (e.g., the vehicle's motion or ignition state, vehicle speed, time, location, occupant identity, etc.).

[0012] although Figure 1 The vehicle association information filtering engine 104 is shown inside the vehicle 102, but in other examples, the vehicle association information filtering engine 104 may be located outside the vehicle 102. For example, the vehicle association information filtering engine 104 may be part of an RSU 116, a remote server 118, etc.

[0013] As used herein, "engine" can refer to hardware processing circuitry, which may include any or a combination of the following: a microprocessor, a core of a multi-core microprocessor, a microcontroller, a programmable integrated circuit, a programmable gate array, or other hardware processing circuitry. Alternatively, "engine" can refer to a combination of hardware processing circuitry and machine-readable instructions (software and / or firmware) executable on the hardware processing circuitry.

[0014] Figure 1 The document describes various example data sources for vehicle-related information. These data sources can be located within vehicle 102 and include sensors 106, cameras 108, data loggers 109, a Global Positioning System (GPS) receiver 110 that receives location data from GPS satellites, a storage system 112, or other data sources. External data sources capable of providing vehicle-related information from vehicle 102 can be part of an RSU 116, a remote server 118, etc.

[0015] Sensors 106 in vehicle 102 may be coupled to corresponding vehicle subsystems 107, such as any or some combinations of the following: vehicle engine, vehicle transmission, brakes, tires, audio system, battery, suspension, navigation system, climate control system, seat belts, airbags, collision avoidance system, or any other vehicle subsystem.

[0016] Sensor 106 is used to measure metrics that represent the characteristics of vehicle subsystem 107, including characteristics related to the operation of vehicle subsystem 107, wear of vehicle subsystem 107, errors or malfunctions in vehicle subsystem 107, etc.

[0017] In another example, sensor 106 can be used to perform environmental measurements of the environment inside and / or outside vehicle 102. Examples of environmental metrics that can be measured by the environmental sensor include any or a combination of the following: temperature, pressure, humidity, etc. The sensor can also be used to detect road conditions on which vehicle 102 is traveling, such as the presence of potholes, whether the road is paved or unpaved, etc.

[0018] The camera 108 of vehicle 102 is used to capture images, including still images and / or video images. The camera 108 can be used to capture images of objects inside vehicle 102 (e.g., occupants in vehicle 102, inanimate objects in vehicle 102, etc.) or objects outside vehicle 102 (e.g., the environment around vehicle 102, including any other vehicles or people that may be near vehicle 102).

[0019] The data logger 109 may include hardware or machine-readable instructions to record various data of the vehicle 102, such as data related to the operation of the vehicle. The data logger 109 may store the recorded data in the storage system 112.

[0020] GPS receiver 110 can provide location data associated with vehicle 102. Although depicted as a GPS receiver, in other examples, the location receiver (different from a GPS receiver) can receive location data from other types of satellites or from other positioning systems, such as base stations of cellular networks.

[0021] The storage system 112 of vehicle 102 can be used to store vehicle-related information 114, which can be provided by various data sources, including sensors 106, cameras 108, data loggers 109, and GPS receivers 110. The storage system 112 can be implemented using one or more storage devices, such as disk-based storage devices, solid-state drives, etc.

[0022] The vehicle association information 114 stored in storage system 112 may also include vehicle association information received from data sources outside vehicle 102, including RSU 116 and remote server 118. Examples of vehicle association information that can be received from RSU 116 include any or some combinations of the following: the current state of traffic lights (e.g., whether the lights are red, yellow, or green), images captured by cameras of RSU 116, road-related traffic information, etc. RSU 116 may be located at an intersection of multiple roads, along one side of a road, or at other locations near a road.

[0023] The vehicle-related information that can be provided by the remote server 118 may include any or some combinations of the following: road traffic information, control information for various aspects of controlling vehicle 102, etc.

[0024] RSU 116 and remote server 118 can communicate with vehicle 102 via their respective wireless links 120 and 122 (such as cellular network, wireless local area network (WAN) etc.).

[0025] Vehicle 102 includes a communication interface 124 that allows vehicle 102 to communicate wirelessly with other endpoints, such as RSU 116 and remote server 118. Communication interface 124 may include a transceiver for transmitting and receiving signals, and one or more protocol layers for managing information transmitted over wireless links 120 and 122.

[0026] Vehicle 102 includes vehicle network 125 to allow various components of vehicle 102 to communicate with each other. Vehicle network 125 may include wired network and / or wireless network.

[0027] According to some implementations of this disclosure, the vehicle association information filtering engine 104 controls access to vehicle association information from various data sources (including any of the foregoing).

[0028] The vehicle association information filtering engine 104 can use access control rule information 126 stored in the memory 128 of the vehicle 102. The memory 128 can be implemented using one or more memory devices, such as dynamic random access memory (DRAM) devices, static random access memory (SRAM) devices, flash memory devices, etc.

[0029] Access control rule information 126 includes various access rules that manage access to vehicle-related information provided by any or a combination of internal or external data sources. In response to a request for vehicle-related information for a given item from an entity (internal or external to vehicle 102), or when an entity is notified of a change in the status of vehicle-related information, vehicle-related information filtering engine 104 accesses one or more access rules in access control rule information 126 to determine whether and when to authorize the entity to access the vehicle-related information for the given item. In some cases, the entity may be an application or program internal or external to vehicle 102. Access control rule information 126 may include separate access control rules for different entities.

[0030] Access control rule information 126 can be provided to vehicle 102 from an external source such as remote server 118. In some examples, access control rule information 126 can be dynamically updated over time.

[0031] Examples of various access rules are provided below.

[0032] Access rules for machine learning operations

[0033] Vehicle 102 also includes a computer 130 in which a machine learning (ML) model 132 can be executed. A machine learning model is a model capable of making predictions based on input data, such as input vehicle association information 134 from one or more data sources. Machine learning model 132 can produce outputs containing predictions based on the input vehicle association information 134. For example, machine learning model 132 can classify the input vehicle association information 134 into a selected category among several different categories. As another example, machine learning model 132 can produce output values ​​that make predictions (e.g., vehicle 102 may be experiencing a malfunction, vehicle 102 may be about to collide with another object, whether an object is near vehicle 102, the identity or presence of occupants in vehicle 102, instructions for actions to be taken (e.g., making payment when vehicle 102 is near a parking lot exit door, activating the brakes, etc.), and so on.

[0034] Machine learning model 132 can be trained to perform its predictions. Training can be based on training data, which includes various vehicle association information items and labels assigned to these items. The assigned labels can be provided by humans or other entities and can indicate the category associated with each vehicle association information item, the output value associated with each item, etc. In one example, training data can be provided to computer 130, and machine learning model 132 can be trained based on the training data. Machine learning model 132 can be continuously trained and updated.

[0035] According to some implementations of this disclosure, the vehicle association information filtering engine 104 can control the characteristics of the input vehicle association information 134, which is provided as input to the machine learning model 132, based on access rules related to the machine learning model of the access control rule information 126. For example, the access rules related to the machine learning model may include access rules that specify that only a subset of the specified vehicle association information is provided to the machine learning model 132.

[0036] In some use cases, the machine learning model 132 need not be provided with all the specified vehicle association information to perform the computation of the machine learning model 132. As an example, the specified vehicle association information may include images (static and / or video images) of the environment surrounding the vehicle 102. The machine learning model 132 may be a theft detection machine learning model to predict whether a theft of the vehicle 102 is taking place.

[0037] To predict whether a theft is occurring, a theft detection machine learning model can detect the presence of one or more persons around the vehicle based on images acquired by camera 108, and can receive sensor data indicating intrusion, such as audible data indicating broken glass, opening doors without a key or smart key (FOB), etc. The output provided by the theft detection machine learning model is a theft indicator, which can be set to a "true" value (indicating that a vehicle theft has occurred) or a "false" value (indicating that a vehicle theft has not occurred). For it to make predictions, the theft detection machine learning model does not necessarily need to be fed all images or all sensor data acquired by camera 108. Instead, the theft detection machine learning model may be able to make its predictions effectively based on a sample of image and / or sensor data. "Sample" of data refers to a selection of fewer than all available data samples. Feeding the theft detection machine learning model a sample of data reduces the amount of personal information available to the application running the theft detection machine learning model, while still allowing the theft detection machine learning model to provide its predictions.

[0038] As an example, access rules associated with the machine learning model can specify the sampling rate of the designated vehicle association information (e.g., including the aforementioned image and / or sensor data) to be provided to the theft detection machine learning model. For example, the sampling rate can be defined as a percentage; for instance, the vehicle association information filtering engine 104 can select a specified percentage of all samples of the designated vehicle association information to be provided to the theft detection machine learning model. For example, if the access rules associated with the machine learning model specify a 10% sampling rate, the vehicle association information filtering engine 104 provides one sample out of every ten samples of the designated vehicle association information as input vehicle association information 134 to the theft detection machine learning model. In some cases, the sampling rate can vary based on factors such as the current time, vehicle location, or whether there are moving objects around the vehicle. If the vehicle is in an area with a high crime rate, or if the current time falls within a period when vehicle theft frequently occurs (e.g., 12 am to 5 am), the sampling rate can be increased. Otherwise, a low sampling rate can be used. If the vehicle sensors detect moving objects around a parked, vacant vehicle, the sampling rate can be increased to better capture potential theft events.

[0039] Providing a theft detection machine learning model with all available samples of a given vehicle's association information can raise privacy concerns, as it may expose the model to too much information. Furthermore, theft detection machine learning models can operate within applications that secretly collect information for other purposes (e.g., integrated sensors). By limiting the data available to that application, better privacy is achieved. A theft detection machine learning model can accurately predict the presence of a theft based solely on a subset of all available samples of a given vehicle's association information.

[0040] In other examples, computer 130 may execute other types of machine learning models 132, for which corresponding access rules are defined in access control rule information 126. Examples of other types of machine learning models 132 may include machine learning models for detecting the presence and / or identity of occupants in vehicle 102, or machine learning models for making decisions about making payments, etc. Access control rule information 126 may include access rules specifying different sampling rates for different machine learning models 132 executed by computer 130.

[0041] In other use cases, such as if the machine learning model 132 is used to detect whether a collision with another vehicle or human is imminent, the machine learning model 132 may be provided with a continuous stream of vehicle association information (i.e., all available samples of vehicle association information, such as images of objects in front of the vehicle 102 captured by the camera 108).

[0042] In another example, alternative or additional access rules may be included in the access control rule information 126 for the machine learning model. As an example, the access rules for the machine learning model may include any or a combination of the following: the type of vehicle association information to be provided to the machine learning model (e.g., providing a first type of vehicle association information but not a second type), when to provide the vehicle association information to the machine learning model, start and stop criteria for providing the vehicle association information to the machine learning model (e.g., providing the vehicle association information to the machine learning model when the start criterion is met, and not providing the vehicle association information when the stop criterion is met), and so on. For example, the access rules for the machine learning model may specify that specified vehicle association information is not provided during daytime hours but will be provided during nighttime hours to protect the privacy of the driver or other users of vehicle 102 when using vehicle 102 during daytime hours.

[0043] Motion-based access rules

[0044] In a further example, access control rule information 126 may include access rules based on the motion state of vehicle 102 (“motion-based access rules”). For example, an application (which may run inside or outside vehicle 102) may operate based on detecting objects inside or outside the vehicle when it is stopped. For example, the application may record images collected by camera 108 when vehicle 102 is stopped (e.g., parked at a parking spot, etc.). As another example, an application (such as a machine learning model) may implement theft detection of vehicle 102 by observing the behavior of people moving around vehicle 102 when vehicle 102 is stopped.

[0045] In such an example, motion-based access rules could stipulate that when vehicle 102 is not moving, an application (e.g., a theft detection application) will be provided with images collected by camera 108, but when vehicle 102 is moving or when ignition is on (but the vehicle is not moving), the application will not be provided with images collected by camera 108.

[0046] The vehicle association information filtering engine 104 can monitor the speed of vehicle 102 based on sensor data (e.g. from a speedometer), and can only provide images (or other vehicle association information) to the application when vehicle 102 is not moving (i.e., when the speed of vehicle 102 is zero).

[0047] In other examples, if vehicle 102 is moving (e.g., at a speed greater than a predetermined threshold), another application (which may operate inside or outside vehicle 102) can perform calculations. For example, the other application could calculate the average speed of vehicle 102 while it is moving (but would not include the zero speed of vehicle 102 when it is stationary in the calculated average speed). In such examples, motion-based access rules could specify the speed data to be provided to the other application when the speed of vehicle 102 exceeds a predetermined threshold.

[0048] Access rules based on driver identity

[0049] In a further example, a driver behavior program (executed inside or outside vehicle 102) can track the driving behavior of the driver of vehicle 102. The driver's driving behavior can be based on measurement data from sensors 106 of vehicle 102 (e.g., sensors 106 can indicate the speed and acceleration of vehicle 102, which can indicate driver aggression). The driver behavior program can also track the position of vehicle 102 to determine where the driver has driven vehicle 102.

[0050] In some examples, the ability to track a driver's driving behavior can be based on the driver's identity. Access control rule information 126 may include driver identity-based access rules that prevent the driver behavior program from accessing specified vehicle-related information (such as speed, acceleration, and location) if the driver has a primary identity (e.g., an adult in the family), but allow access to the specified vehicle-related information for any other driver with a different identity (e.g., a minor in the family). In this scenario, parents could use the driver behavior program to monitor whether their minor is exhibiting good driving behavior.

[0051] Therefore, in response to detecting that the driver has a primary identity (e.g., based on driver facial recognition, biometric data such as driver fingerprints, etc.), the vehicle association information filtering engine 104 can use access rules based on driver identity to prevent specified vehicle association information from being provided to the driver behavior program. In response to detecting that the driver has an identity different from the primary identity, the vehicle association information filtering engine 104 can use access rules based on driver identity to transmit specified vehicle association information to the driver behavior program.

[0052] Speed-based access rules

[0053] As a further example, access control rule information 126 may include speed-based access rules that disable access to vehicle speed information (from the speedometer) if the speed of vehicle 102 exceeds a first threshold. In other words, according to the speed-based access rules, if the speed of vehicle 102 does not exceed the first threshold, the vehicle association information filtering engine 104 may allow the speed monitoring program (executing in or outside vehicle 102) to access the vehicle speed information. If the speed of vehicle 102 exceeds the first threshold, the vehicle association information filtering engine 104 may disable access to the speed information by the speed monitoring program.

[0054] In other examples, access control rule information 126 may include speed-based access rules that disable access to vehicle speed information (from the speedometer) if the speed of vehicle 102 is less than a second threshold. In some cases, different applications / programs may have different access control rules (e.g., different filtering requirements). Each application / program can send its filtering requirements (e.g., filter start / stop criteria, sampling rate) to access control rule information 126, allowing vehicle association information filtering engine 104 to provide corresponding filtered data. Some applications / programs may not be associated with access control rules, causing such applications / programs to receive unfiltered data (i.e., data not filtered by vehicle association information filtering engine 104).

[0055] Access rules based on passenger identity

[0056] In a further example, an occupant tracking procedure (executed inside or outside vehicle 102) can track the identity of occupants within vehicle 102, such as for identification purposes. In some cases, it may not be desirable to track the identity of certain occupants of vehicle 102, such as for privacy reasons.

[0057] Access control rule information 126 may include access rules based on occupant identity, which stipulate that if an occupant in vehicle 102 has a first identity, then user identity data (e.g., images of the interior of vehicle 102, user identification information derived from biometric or facial recognition processes, etc.) shall not be provided to occupant tracking (program).

[0058] Therefore, in response to detecting that an occupant in vehicle 102 has a first identity, vehicle association information filtering engine 104 can use access rules based on occupant identity to prevent specified vehicle association information from being provided to the occupant tracking program. In response to detecting that all occupants in vehicle 102 have identities different from the first identity, vehicle association information filtering engine 104 can use access rules based on occupant identity to transmit specified vehicle association information to the occupant tracking program.

[0059] Seatbelt Information Access Rules

[0060] In a further example, a seatbelt tracking program (performed inside or outside vehicle 102) may want to verify whether occupants in vehicle 102 are wearing their seatbelts. In some examples, certain categories of drivers or passengers may be exempt from having to wear seatbelts.

[0061] In such an example, access control rule information 126 may include a seatbelt information access rule that specifies that seatbelt information should not be provided to drivers or passengers of a specified category.

[0062] The vehicle association information filtering engine 104 can identify any occupant of vehicle 102 that falls into a specified category, and can prevent seat belt information (e.g., information from seat belt sensors indicating whether the seat belt is fastened) from being transmitted to the seat belt tracking program for any occupant in the specified category by using seat belt information access rules. The vehicle association information filtering engine 104 can implement occupant identification and, based on the identified identity, determine (e.g., using correlation information) whether an occupant with that identity falls into a specified category.

[0063] The vehicle association information filtering engine 104 can allow seat belt information of any occupants who do not fall into a specified category to be transmitted to the seat belt tracking program.

[0064] Geofencing-based access rules

[0065] In some examples, a location determination program (executing inside or outside vehicle 102) may query location information (e.g., GPS information) to determine the location of vehicle 102. For privacy reasons, it may not be desirable to allow the location of vehicle 102 to be transmitted to this location determination program unless vehicle 102 is inside (or outside) a geofence. A "geofence" can refer to information that defines a geographic region (which may consist of a single geographic area or multiple geographic areas).

[0066] Access control rule information 126 may include geofence-based access rules, which control the transmission of vehicle location data to the location determination program based on the relationship between the current location of vehicle 102 and the geofence. For example, a geofence-based access rule may stipulate that if vehicle 102 is inside or outside the geofence, vehicle location data will not be transmitted to the location determination program.

[0067] If vehicle 102 is currently inside or outside a geofence, vehicle association information filtering engine 104 can use geofence-based access rules to prevent vehicle location data from being transmitted to the location determination program. However, if vehicle 102 is currently outside or inside a geofence, vehicle association information filtering engine 104 can use geofence-based access rules to allow vehicle location data to be transmitted to the location determination program. Restricting the provision of vehicle location data to the location determination program based on geofence-based access rules can achieve the goal of preventing the location determination program from receiving data to determine certain locations, such as home, office, etc.

[0068] Fuzzy Personal Identification Information (PII)

[0069] In a further example, the vehicle association information filtering engine 104 may blur a portion of personally identifiable information (PII) when certain criteria are met. Examples of PII may include any or some combinations of the following: the user's identity, an image of the user's face, the user's biometric data, etc. Blurring a portion of the PII may include deleting a portion of the PII, replacing a portion of the PII with different data, hiding a portion of the PII, and so on. For example, if the PII falls into a specified category, a portion of the PII is blurred.

[0070] Other examples

[0071] Note that the vehicle association information filtering engine 104 can apply multiple different access rules (e.g., any of the aforementioned access rules) in the access control rule information 126 to restrict the transmission of vehicle association information to entities, whether inside or outside the vehicle 102.

[0072] Figure 2 This is a flowchart of process 200 based on some examples, which can be implemented by a computer system. The computer system may include a single computer or multiple computers, and may be inside or outside a vehicle (e.g., vehicle 102).

[0073] Process 200 includes (at 202) receiving vehicle association information from a data source associated with the vehicle. The data source associated with the vehicle can be internal or external to the vehicle. It should also be noted that the vehicle association information from the data source may be stored in a storage system before being used by entities internal or external to the vehicle.

[0074] Process 200 includes (at 204) receiving a request from an entity for accessing vehicle association information. In some examples, the entity may include... Figure 1 The machine learning model 132, or any of the programs discussed above.

[0075] In response to the request, process 200 includes (at 206) retrieving access control rule information (e.g., Figure 1 In step 126) and (in step 208) determine whether the entity is authorized to access the vehicle association information based on access control rule information. For example, tasks 206 and 208 can be implemented by the vehicle association information filtering engine 104.

[0076] Based on this determination, process 200 can (at 210) restrict access to vehicle-related information according to access control rule information. In some examples, the access control rule information includes at least one privacy criterion selected from: machine learning use criteria related to the use of vehicle-related information by a machine learning model, vehicle motion criteria related to the vehicle's movement state, or person identity criteria related to the identity of persons in the vehicle. In other examples, the access control rule information may include additional or alternative criteria related to access to vehicle-related information.

[0077] In some examples, machine learning usage criteria may include access rules for machine learning operations, such as any of the access rules discussed above. Vehicle motion criteria may include motion-based access rules, such as any of the access rules discussed above. Personnel identity criteria may include driver-based access rules and / or occupant-based access rules, such as any of the access rules discussed above.

[0078] In a further example, access control rule information includes location-based criteria, such as geofence-based access rules. Process 200 can restrict access to vehicle association information based on location-based criteria, which is done by blocking access to vehicle association information if the vehicle has a defined relationship relative to a geofence (inside or outside).

[0079] In such a further example, process 200 may receive a request from an entity to access vehicle association information, determine whether the entity has permission to access the vehicle association information, and in response to determining that the entity has permission, determine whether the vehicle has a prescribed relationship relative to a geofence, and if the vehicle has a prescribed relationship relative to a geofence, prevent the entity from accessing the vehicle association information.

[0080] If a vehicle does not have a defined relationship with respect to a geofence, process 200 allows entities to access vehicle association information.

[0081] In some examples, process 200 restricts access to vehicle-associated information based on vehicle motion criteria by blocking access to vehicle-associated information while the vehicle is moving. In further examples, process 200 restricts access to vehicle-associated information based on vehicle motion criteria by blocking access to vehicle-associated information while the vehicle is not moving.

[0082] In some examples, process 200 restricts access to vehicle-related information based on vehicle motion criteria by blocking access to vehicle-related information when the vehicle's speed has a prescribed relationship with a speed threshold (exceeding or falling below the speed threshold).

[0083] In some examples, process 200 restricts access to vehicle-related information based on a person identity criterion by blocking access to vehicle-related information if the identity of a person in the vehicle matches a prescribed identity. If the identity of a person in the vehicle differs from the prescribed identity, process 200 allows access to the vehicle-related information.

[0084] In some examples, process 200 restricts access to seat belt information based on seat belt information access guidelines that specify that seat belt information is not provided to a particular category of driver or passenger.

[0085] Figure 3 This is a block diagram of a computer system 300, which may be a vehicle (such as...). Figure 1 The computer system 300 is part of or outside the vehicle 102. The computer system 300 includes one or more hardware processors 302. The hardware processor may include a microprocessor, the core of a multi-core microprocessor, a microcontroller, a programmable integrated circuit, a programmable gate array, or other hardware processing circuitry.

[0086] Computer system 300 includes a non-transitory machine-readable or computer-readable storage medium 304 storing machine-readable instructions executable on one or more hardware processors 302 to perform various tasks. The machine-readable instructions include vehicle association information filtering instructions 306 to implement access control for vehicle association information. For example, vehicle association information filtering instructions 306 could be... Figure 1 The instructions of the vehicle association information filtering engine 104.

[0087] The vehicle association information filtering instruction 306 can implement access control of vehicle association information based on the access control rule information 308 stored in the memory 310 of the computer system 300.

[0088] Storage media (e.g., 304) may include any or a combination of the following: semiconductor storage devices such as dynamic or static random access memory (DRAM or SRAM), erasable and programmable read-only memory (EPROM), electrically erasable and programmable read-only memory (EEPROM), and flash memory or other types of non-volatile storage devices; magnetic disks such as fixed disks, floppy disks, and removable disks; additional magnetic media including magnetic tape; optical media such as compact discs (CDs) or digital video discs (DVDs); or other types of storage devices. Note that the instructions discussed above may be provided on a computer-readable or machine-readable storage medium, or alternatively, may be provided on multiple computer-readable or machine-readable storage media distributed across a large system that may have multiple nodes. Such one or more computer-readable or machine-readable storage media are considered part of an article of manufacture (or article of manufacture). An article of manufacture or article of manufacture may refer to any single or multiple manufactured components. One or more storage media may be located in a machine that executes the machine-readable instructions, or at a remote site from which the machine-readable instructions can be downloaded via a network for execution.

[0089] In the foregoing description, numerous details have been set forth to provide an understanding of the subject matter disclosed herein. However, implementations can be practiced without some of these details. Other implementations may include modifications and variations to the foregoing details. The appended claims are intended to cover such modifications and variations.

Claims

1. A non-transitory machine-readable storage medium, comprising instructions that, when executed, cause the system to: Receive vehicle-related information from data sources associated with the vehicle; and Access to the vehicle-related information is restricted based on at least one privacy criterion, the at least one privacy criterion including a machine learning usage criterion related to the use of the vehicle-related information by a machine learning model, wherein the machine learning usage criterion includes controlling the sampling rate of the vehicle-related information input to the machine learning model, and wherein restricting the machine learning model's access to the vehicle-related information includes: Based on the aforementioned machine learning usage guidelines, it is permissible to provide information of the first type as input to the machine learning model at the aforementioned sampling rate, and Based on the aforementioned machine learning usage guidelines, providing the machine learning model with second-type information, which differs from the first-type information, is prohibited. The sampling rate is controlled based on the vehicle's location using the following: Based on the vehicle's first location, a first sampling rate is set for the vehicle association information input into the machine learning model, and Based on the second location of the vehicle, a second sampling rate is set for the vehicle association information input into the machine learning model, wherein the second location is different from the first location, and the second sampling rate is different from the first sampling rate.

2. The non-transitory machine-readable storage medium according to claim 1, wherein, When executed, the instruction causes the system to further restrict access to vehicle association information based on location-based criteria by blocking access to the vehicle association information if the vehicle has a defined relationship with respect to a geofence.

3. The non-transitory machine-readable storage medium according to claim 2, wherein, When the instruction is executed, it causes the system to: Receive a request from the entity to access the vehicle association information; Determine that the entity has permission to access the vehicle-related information; In response to determining that the entity has the permission, determine whether the vehicle has a prescribed relationship with respect to the geofence; as well as If the vehicle has the specified relationship relative to the geofence, then access to the vehicle association information by the entity is prevented.

4. The non-transitory machine-readable storage medium according to claim 3, wherein, When the instruction is executed, it causes the system to: If the vehicle does not have the specified relationship relative to the geofence, then the entity is allowed to access the vehicle association information.

5. The non-transitory machine-readable storage medium according to claim 1, wherein, When executed, the instruction causes the system to further restrict access to the vehicle-associated information based on vehicle motion criteria by blocking access to the vehicle-associated information while the vehicle is moving, wherein the vehicle motion criteria are included in the at least one privacy criterion.

6. The non-transitory machine-readable storage medium according to claim 1, wherein, When executed, the instruction causes the system to further restrict access to the vehicle-associated information based on vehicle motion criteria by preventing access to the vehicle-associated information when the vehicle is not moving, wherein the vehicle motion criteria are included in the at least one privacy criterion.

7. The non-transitory machine-readable storage medium according to claim 1, wherein, When executed, the instruction causes the system to further restrict access to the vehicle-associated information based on vehicle motion criteria by blocking access to the vehicle-associated information when the vehicle's speed has a predetermined relationship with respect to a speed threshold, wherein the vehicle motion criteria are included in the at least one privacy criterion.

8. The non-transitory machine-readable storage medium according to claim 1, wherein, When executed, the instructions cause the system to: restrict access to the vehicle association information based on the machine learning usage criteria by controlling the sampling rate of the vehicle association information input to the machine learning model, and wherein controlling the sampling rate includes adjusting the sampling rate of the vehicle association information input to the machine learning model according to the current time.

9. The non-transitory machine-readable storage medium according to claim 1, wherein, When executed, the instructions cause the system to: restrict access to the vehicle association information based on the machine learning usage criteria by controlling the sampling rate of the vehicle association information input to the machine learning model, wherein controlling the sampling rate includes: Based on the vehicle sensors detecting moving objects around the vehicle, a first sampling rate is set for the vehicle association information input into the machine learning model; Based on the fact that the vehicle sensors did not detect the moving object around the vehicle, a second sampling rate is set for the vehicle association information input to the machine learning model, the second sampling rate being different from the first sampling rate.

10. The non-transitory machine-readable storage medium according to claim 1, wherein, When the instruction is executed, the system further restricts access to vehicle-related information based on personnel identity criteria by controlling access to vehicle-related information based on a comparison of the identity of the person in the vehicle with a prescribed identity, wherein the personnel identity criteria are included within the at least one privacy criterion.

11. The non-transitory machine-readable storage medium according to claim 10, wherein, The specified identity in the personnel identity criteria is the driver or passenger of the vehicle.

12. The non-transitory machine-readable storage medium according to claim 1, wherein, When executed, the instruction causes the system to restrict access to seatbelt information based on seatbelt information access criteria, which stipulate that the seatbelt information is not provided to specific categories of drivers or passengers.

13. The non-transitory machine-readable storage medium according to claim 1, wherein, When executed, the instruction causes the system to restrict access to the vehicle association information by obscuring the personally identifiable information within the vehicle association information.

14. The non-transitory machine-readable storage medium according to claim 1, wherein, The system is either part of the vehicle or located remotely from the vehicle.

15. The non-transitory machine-readable storage medium according to claim 1, wherein, When executed, the instructions cause the system to: restrict access to the vehicle association information based on the machine learning usage criteria by controlling the sampling rate of the vehicle association information input to the machine learning model, wherein controlling the sampling rate includes: The detection showed that the vehicle was parked and vacant, and Based on the detection that the vehicle is parked and vacant, the sampling rate of the vehicle association information input to the machine learning model is increased.

16. The non-transitory machine-readable storage medium according to claim 1, wherein, The data source is located inside or outside the vehicle.

17. A computer system comprising: One or more hardware processors; as well as A non-transitory storage medium, wherein the non-transitory storage medium stores instructions executable on one or more hardware processors to: Receive vehicle-related information from data sources associated with the vehicle; as well as Access to the vehicle-related information is restricted based on at least one privacy criterion, the at least one privacy criterion including a machine learning usage criterion related to the use of the vehicle-related information by a machine learning model, wherein restricting the machine learning model's access to the vehicle-related information includes: obfuscating personally identifiable information in the vehicle-related information; and controlling the sampling rate of the vehicle-related information input to the machine learning model according to the current time, wherein controlling the sampling rate includes: Based on the current time, within a first time period, a first sampling rate is set for the vehicle association information input to the machine learning model; and Based on the current time, a second sampling rate is set for the vehicle association information input to the machine learning model within a second time period, wherein the second time period is different from the first time period, and the second sampling rate is different from the first sampling rate.

18. The computer system of claim 17, wherein the instructions are executable on one or more hardware processors to adjust the sampling rate based on vehicle sensors detecting moving objects around the vehicle.

19. A method for a computer system, comprising: The computer system receives vehicle association information from a data source associated with the vehicle. as well as The computer system restricts access to the vehicle-related information based on at least one privacy criterion, the at least one privacy criterion including a machine learning usage criterion related to the use of the vehicle-related information by a machine learning model, wherein restricting the machine learning model's access to the vehicle-related information includes: obfuscating personally identifiable information in the vehicle-related information; and controlling the sampling rate of the vehicle-related information input to the machine learning model, wherein controlling the sampling rate includes: Based on the vehicle sensors detecting moving objects around the vehicle, a first sampling rate is set for the vehicle association information input into the machine learning model; Based on the fact that the vehicle sensors did not detect the moving object around the vehicle, a second sampling rate is set for the vehicle association information input to the machine learning model, the second sampling rate being different from the first sampling rate.

Images