Occupant digital profile approach for seamless experience and operable insights

Abstract

An occupant profile method for seamless experience and operable insights. The method includes: receiving occupant data relating to an occupant of the primary traffic vehicle; generating, using the vehicle body, a knowledge graph based on the occupant data; querying the knowledge graph to generate an occupant profile including desires and demands of the occupant of the primary traffic vehicle; transmitting the occupant profile to the target traffic vehicle; comparing the functional component of the target traffic vehicle with the functional component of the main traffic vehicle to determine a compatible functional component; and, based on the occupant profile, configuring at least one compatible functional component of the target traffic vehicle. The compatible functional component is a traffic vehicle functional component existing in both the target traffic vehicle and the main traffic vehicle. The occupant data includes metadata relating to an occupant of the primary traffic vehicle, a customer-oriented functional component of the primary traffic vehicle, and interaction data between the occupant and the customer-oriented functional component of the primary traffic vehicle.

Description

Technical Field

[0001] The present disclosure relates to autonomous transportation vehicles; more particularly, to an occupant digital profiling method for seamless experience and actionable insights. Background Art

[0002] Automated vehicles use driver assistance technologies to reduce or eliminate the need for human operators. Such driver assistance technologies may include automated driving systems (ADS) and advanced driver assistance systems (ADAS). ADS is capable of operating a vehicle from a partially automated driving mode to a fully automated driving mode according to the SAE J3016 driving automation classification from level 0 (no driving automation) to level 5 (full driving automation). ADAS uses sensors in the vehicle (such as radar and cameras) to perceive the world around it and then either provides information to the driver or takes automatic actions based on what it perceives. Some ADAS include a driver information acquisition system (DIAS), which is configured to collect information related to the driver's status to improve vehicle safety. DIAS monitors the driver's alertness using sensors such as optical cameras set in the cabin of the automated vehicle. If DIAS determines that the driver's alertness is below a predetermined alertness threshold, the DIAS can alert the driver and / or instruct the ADS to take over the operation of the vehicle.

[0003] DIAS may collect information related to a specific driver's vehicle component setting preferences to enhance the specific driver's driving experience. Such vehicle component setting preferences include, but are not limited to: infotainment configuration, cabin temperature settings, steering wheel position, seat position, and instrument lighting settings. Each time DIAS recognizes a specific driver, DIAS may configure the vehicle component settings to the specific driver's preferences. This is particularly useful when there are multiple drivers for the same vehicle.

[0004] Therefore, while current DIAS achieve their intended purpose, current DIAS systems are limited to the specific driver and the features of the specific vehicle in which the driver information is collected. A method is needed to generate a portable digital occupant profile to provide a seamless experience when transitioning to other autonomous vehicles with various features. A method is also needed to stitch multiple occupant profiles for generating actionable insights. Summary of the invention

[0005] According to several aspects, an occupant profile method for seamless experience and actionable insights is disclosed. The method includes receiving occupant data related to an occupant of a host vehicle; generating a knowledge graph based on the occupant data using a vehicle ontology; querying the knowledge graph to generate an occupant profile including preferences of the occupant of the host vehicle; transmitting at least a portion of the occupant profile to a target vehicle; comparing the features of the target vehicle with the features of the host vehicle to determine at least one compatible feature, wherein at least one compatible feature is a vehicle feature that exists in both the target vehicle and the host vehicle; and configuring at least one compatible feature of the target vehicle based on the occupant profile.

[0006] In another aspect of the present disclosure, the method also includes comparing functional components of the target transportation vehicle with functional components of the host transportation vehicle to determine at least one incompatible functional component, wherein the at least one incompatible functional component is a transportation vehicle functional component that exists in the target transportation vehicle but is missing in the host transportation vehicle; and recommending coaching related to the transportation vehicle functional component that exists in the target transportation vehicle but is missing in the host transportation vehicle.

[0007] In another aspect of the present disclosure, the method also includes comparing the functional components of the target transportation vehicle with the functional components of the host transportation vehicle to determine at least one incompatible functional component, wherein the at least one incompatible functional component is a transportation vehicle functional component that exists in the host transportation vehicle but is missing in the target transportation vehicle; and recommending an upgrade to the target transportation vehicle to include the transportation vehicle functional component that exists in the host transportation vehicle.

[0008] In another aspect of the present disclosure, the missing vehicle feature in the target vehicle is a subscription to a software-enabled feature in the target vehicle. Recommending an upgrade to the target vehicle includes: based on the occupant's preference, recommending activation of the subscription to the software-enabled feature in the target vehicle.

[0009] In another aspect of the present disclosure, the occupant data includes metadata related to the occupants of the host transportation vehicle, the customer-facing features of the host transportation vehicle, and interaction data between the occupants and the customer-facing features of the host transportation vehicle. The metadata related to the occupants includes: demographic data, physiological data, biometric data, psychological data, and social data.

[0010] In another aspect of the present disclosure, the method also includes: receiving multiple sets of occupant data related to multiple occupants of multiple host transportation vehicles; generating a separate knowledge graph for each of the multiple sets of occupant data; generating a graph embedding for each separate knowledge graph; generating a graph embedding concatenation by concatenating the graph embeddings; and training a machine learning architecture using the graph embedding concatenation to generate actionable insights.

[0011] In another aspect of the present disclosure, graph embedding is generated for each individual knowledge graph at one of a node level, a link level, and a graph level.

[0012] In another aspect of the present disclosure, the host vehicle is an autonomous vehicle having an occupant profile system configured to perform the method.

[0013] According to several aspects, a tangible, non-transitory machine-readable medium includes machine-readable instructions that, when executed by a processor, cause the processor to: receive occupant data; generate a knowledge graph based on the occupant data using a vehicle ontology; and query the knowledge graph to generate an occupant profile including preferences of the occupant. The occupant data includes metadata related to an occupant of a host transportation vehicle, a customer-facing functional component of the host transportation vehicle, and interaction data between the occupant and the customer-facing functional component of the host transportation vehicle.

[0014] In another aspect of the present disclosure, metadata related to the occupant includes social data of the occupant, the social data including at least one of family relationships, friends, acquaintances, and social media accounts.

[0015] In another aspect of the present disclosure, when the machine-readable instructions are executed by the processor, the processor is further caused to: compare the functional components of the target transportation vehicle with the functional components of the host transportation vehicle to determine at least one compatible functional component, wherein the at least one compatible functional component is a transportation vehicle functional component that exists in both the target transportation vehicle and the host transportation vehicle; and send instructions to the target transportation vehicle to configure at least one compatible functional component of the target transportation vehicle according to the preferences of the occupant.

[0016] In another aspect of the present disclosure, when the machine-readable instructions are executed by the processor, the processor also causes the processor to: compare the functional components of the target transportation vehicle with the functional components of the host transportation vehicle to determine the transportation vehicle functional components that exist in the target transportation vehicle but are missing in the host transportation vehicle; and, in response to determining the transportation vehicle functional components that exist in the target transportation vehicle but are missing in the host transportation vehicle, recommend coaching related to at least one incompatible functional component.

[0017] In another aspect of the present disclosure, when the machine-readable instructions are executed by the processor, the processor is further caused to: compare the functional components of the target transportation vehicle with the functional components of the host transportation vehicle to determine the transportation vehicle functional components that exist in the host transportation vehicle but are missing in the target transportation vehicle; and, in response to determining the transportation vehicle functional components that are missing in the target transportation vehicle but exist in the host transportation vehicle, recommend upgrading the target transportation vehicle to include the transportation vehicle functional components that exist in the host transportation vehicle.

[0018] According to several aspects, a tangible, non-transitory machine-readable medium includes machine-readable instructions that, when executed by a processor, cause the processor to: receive multiple sets of occupant data; generate a separate knowledge graph for each of the multiple sets of occupant data; generate a graph embedding for each separate knowledge graph; generate a graph embedding concatenation by concatenating the graph embeddings; and train a machine learning architecture using the graph embedding concatenation to generate actionable insights. The separate knowledge graphs are based on a vehicle ontology.

[0019] Further areas of applicability will become apparent from the description provided herein.It should be understood that the description and specific examples are intended for purposes of illustration only and are not intended to limit the scope of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] The drawings described herein are for illustration purposes only and are not intended to limit the scope of the present disclosure in any way.

[0021] Figure 1 is a functional diagram of an autonomous vehicle according to an exemplary embodiment;

[0022] Figure 2 According to an exemplary embodiment Figure 1 A schematic diagram of a passenger compartment of an autonomous transportation vehicle;

[0023] Figure 3 is a functional diagram of a system for digital occupant profiling according to an exemplary embodiment;

[0024] Figure 4 is a block diagram of a method for generating a digital occupant profile using a knowledge graph to enable seamless transition and actionable insights according to an exemplary embodiment;

[0025] Figure 5 is a knowledge graph based on automobile ontology according to an exemplary embodiment;

[0026] Figure 6 is a functional diagram of a method for seamlessly transferring an occupant from a host transportation vehicle to a target transportation vehicle using a digital occupant profile based on a knowledge graph according to an exemplary embodiment;

[0027] 7A to 7C are examples of knowledge graph embedding at the node level, link level, and graph level, respectively, according to an exemplary embodiment; and

[0028] Figure 8 is a block diagram of a method of stitching multiple graph embeddings of a knowledge graph of occupant data to predict actionable insights according to an exemplary embodiment. DETAILED DESCRIPTION

[0029] The following description is merely exemplary in nature and is not intended to limit the present disclosure, application, or use. The illustrated embodiments are disclosed with reference to the accompanying drawings, wherein like reference numerals represent corresponding parts in the several drawings. The drawings are not necessarily drawn to scale, and some features may be exaggerated or minimized to show details of particular features. The specific structural and functional details disclosed are not intended to be interpreted as limiting, but rather as a representative basis for teaching those skilled in the art how to practice the disclosed concepts.

[0030] As used herein, the terms "module", "control module" or "controller" refer to any hardware, software, firmware, electronic control component, processing logic and / or processor device, alone or in any combination, including but not limited to: application-specific integrated circuits (ASICs), electronic circuits, processors (shared, dedicated or grouped) and memories that execute one or more software or firmware programs, combinational logic circuits and / or other suitable components that provide the described functionality.

[0031] Embodiments of the present disclosure may be described herein in terms of functional and / or logical block components and various processing steps. It should be understood that such block components may be implemented by any number of hardware, software, and / or firmware components configured to perform specified functions. For example, embodiments of the present disclosure may employ various integrated circuit components (e.g., memory elements, digital signal processing elements, logic elements, lookup tables, etc.); various integrated circuit components may perform various functions under the control of one or more microprocessors or other control devices. In addition, those skilled in the art will appreciate that embodiments of the present disclosure may be practiced in conjunction with any number of systems, and the systems described herein are merely exemplary embodiments of the present disclosure.

[0032] The connecting lines shown in the various figures included herein are intended to represent example functional relationships and / or physical couplings between the various elements. Conventional techniques may be used for signal processing, data transmission, signaling, control, and other functional aspects of the system (and the individual operating components of the system) may not be described in detail herein. It should be noted that many alternative or additional functional relationships or physical connections may exist in the embodiments of the present disclosure.

[0033] Figure 1A functional block diagram schematic diagram of an autonomous vehicle 100 capable of operating according to the SAE J3016 driving automation classification from level 0 (no driving automation) to level 5 (full driving automation) is shown. The vehicle 100 generally includes a body 106, front wheels 108, and rear wheels 110. The body 106 substantially surrounds the components of the vehicle 100. The front wheels 108 and the rear wheels 110 are each rotatably coupled to the body 106 near a respective corner of the body 106. As shown, the vehicle 100 generally includes a vehicle sensor system 116, an object detection system 118, a propulsion system 120, a transmission system 122, a steering system 124, a braking system 126, an actuator system 128, a vehicle communication system 130, an occupant profile system 132, and at least one vehicle controller 136. Although the vehicle 100 is depicted as a sedan in the illustrated embodiment, other examples of autonomous vehicles include, but are not limited to, motorcycles, trucks, sport utility vehicles (SUVs), recreational vehicles (RVs), boats, and airplanes.

[0034] The vehicle sensor system 116 includes: one or more vehicle external sensing devices 140A-140N, also referred to as external sensors 140A-140N, which are configured to sense observable conditions of the external environment of the vehicle 100; and internal sensing devices 152A-152N (in Figure 2 ), also referred to as interior sensors, are configured to sense observable conditions within the passenger compartment of the transportation vehicle 100. Examples of transportation vehicle external sensing devices 140A-140N include, but are not limited to, radar 140A, lidar 140B, camera 140C, ultrasonic sensor 140D, and other external sensors 140N capable of collecting and / or receiving information related to the external environment of the transportation vehicle. Examples of transportation vehicle internal sensing devices 152A-152N include, but are not limited to, optical and thermal cameras 152A, low frequency radar 152B, and other sensors 140N capable of collecting and / or receiving information related to occupants and transportation vehicle component settings within the transportation vehicle cabin. Occupants include the driver or operator of the transportation vehicle, as well as any passengers within the transportation vehicle cabin.

[0035] The object detection system 118 is configured to receive object data from the traffic vehicle sensor system 116, detect one or more candidate objects in the object data, and predict an object type for each detected object. The object detection system 118 is pre-trained to predict predefined object types by using pre-defined concepts associated with each pre-defined object type. The object detection system 118 is also pre-trained to predict pre-defined behaviors and preferences of the occupant by using pre-defined concepts associated with each pre-defined behavior and preference of the occupant.

[0036] In various embodiments, the propulsion system 120 may include an internal combustion engine, an electric machine (such as a traction motor), and / or a fuel cell propulsion system. The transmission system 122 is configured to transmit power from the propulsion system 120 to the front wheels 108 and the rear wheels 110 of the vehicle according to a selectable speed ratio. According to various embodiments, the transmission system 122 may include a stepped automatic transmission, a continuously variable transmission, or other suitable transmission.

[0037] The braking system 126 is configured to provide braking torque to the front wheels 108 and the rear wheels 110. In various embodiments, the braking system 126 may include: friction brakes, brake-by-wire brakes, a regenerative braking system (such as an electric motor), and / or other suitable braking systems. The steering system 124 affects the position of the front wheels 108 and the rear wheels 110. Although depicted as including a steering wheel for illustrative purposes, in some embodiments contemplated within the scope of the present disclosure, the steering system 124 may not include a steering wheel.

[0038] The actuator system 128 includes one or more actuator devices 142 configured to control one or more vehicle features, such as, but not limited to, the propulsion system 120, the transmission system 122, the steering system 124, and the braking system 126. The one or more actuator devices 142 may be configured to control one or more vehicle features located within the passenger cabin, such as steering wheel position, seat position, infotainment settings, and cabin environment setting sensors.

[0039] The vehicle communication system 130 is configured to wirelessly transmit information to and from other entities (vehicle-to-everything ("V2X") communication). For example, the vehicle communication system 130 is configured to: wirelessly transmit information to and from other vehicles (vehicle-to-vehicle ("V2V") communication); wirelessly transmit communication information to and from driving system infrastructure (vehicle-to-infrastructure ("V2I") communication); and wirelessly transmit communication information to and from remote systems and / or personal devices. In an embodiment, the vehicle communication system 130 is a wireless communication system configured to communicate via a wireless local area network (WLAN) using the IEEE 802.11 standard or by using cellular data communications. However, additional or alternative communication methods such as dedicated short-range communication (DSRC) channels are also considered to be within the scope of the present disclosure. A DSRC channel refers to a unidirectional or bidirectional short- to medium-range wireless communication channel designed specifically for automotive use, and a set of corresponding protocols and standards.

[0040] The occupant profile system 132 is configured to receive information about the occupants of the transportation vehicle from internal sensors, the transportation vehicle communication system 130, and other information sources about the occupants. The occupants may include drivers, operators, or passengers of the transportation vehicle and are users of the occupant profile system 132. Occupants are also referred to as users or customers. Information about occupants is also referred to as occupant information.

[0041] Occupant information includes, but is not limited to, metadata about occupants and features, and interaction data between occupants and customer-facing features within a transportation vehicle. Occupant information is used to generate an occupant knowledge graph, which is used to generate occupant profiles. Occupant profiles are used to enhance the occupant experience by enabling a seamless transition from a primary transportation vehicle to a target transportation vehicle. A primary transportation vehicle is a transportation vehicle where information about an occupant is collected. A target transportation vehicle is a transportation vehicle to which an occupant transitions or migrates from a primary transportation vehicle. Occupant profiles are extensible, meaning that occupant profiles can be combined with other occupant profiles from various transportation vehicles for machine learning to generate actionable insights. Actionable insights include, but are not limited to: quantifiable Net Promoter Score (NPS) at feature level, vehicle level, and fleet level; most commonly used features; preferences for features and services, and matching customers to the features they are likely to want most; customer segmentation (segmentation) for targeted marketing, retention, and variant reduction; features used together, services purchased together, common claims, etc.; and anomaly detection.

[0042] The transportation vehicle controller 136 includes at least one processor 144 and a computer readable storage device 146. The computer readable storage device 146 may also be referred to as a computer readable medium 146. The computer readable storage device 146 is capable of storing data, some of which represents instructions executable by the processor 144, including algorithms for operating various transportation vehicle systems. The processor 144 and the storage device 146 may be part of the transportation vehicle controller 136, separate from the transportation vehicle controller 136, or part of a separate transportation vehicle system. Although Figure 1 Only one vehicle controller 136 is shown, but alternative embodiments of the autonomous vehicle 10 may include any number of controllers that communicate via any suitable communication medium or combination of communication media and collaborate to process sensor signals, execute logic, calculations, methods and / or algorithms, and generate control signals to automatically control functional components of the autonomous vehicle 100.

[0043] Figure 2A schematic diagram of a passenger compartment 150 of a transportation vehicle 100 is shown. The passenger compartment 150 is also referred to as a transportation vehicle cabin 150. The transportation vehicle cabin 150 includes transportation vehicle features that can be electronically adjusted or reconfigured by an occupant of the transportation vehicle 100. Such transportation vehicle features include, but are not limited to: infotainment systems, transportation vehicle software applications, environmental systems (including heating and cooling), electronic human machine interfaces (such as backup camera displays and instrument clusters), seats, mirrors, sunroofs, and other electronically adjustable components. The transportation vehicle cabin 150 includes one or more actuator devices 142 that are configured to receive instructions from the profile system 132 and / or the transportation vehicle controller 136 to reconfigure these transportation vehicle features based on the occupant profile.

[0044] The transportation vehicle cabin 150 is equipped with a plurality of transportation vehicle component status and occupant sensors 152A-152N, collectively referred to as interior sensors 152A-152N, and configured to collect data related to at least one occupant 153 in the transportation vehicle cabin 150. The occupant 153 may be a driver, operator, or passenger of the transportation vehicle 100. The interior sensors include, but are not limited to: optical and thermal cameras 152A, low frequency radars 152B, cabin environment sensors (such as temperature and humidity sensors) 152C, infotainment settings sensors 152D, seat position sensors 152E, steering wheel adjustment sensors 152F, instrument display adjustment sensors 152G, pedal position sensors 152H, and other sensors 152N that may collect data about the occupant. The collected data about the occupant 153 (also referred to as occupant data), as well as any other information about the occupant received by the transportation vehicle communication system 130, is processed by the profiling system 132 to generate an occupant profile.

[0045] Figure 3is a functional diagram of the occupant profile system 132. The occupant profile system 132 includes a system controller 202 that communicates with at least one or more of the traffic vehicle sensor system 116, the object detection system 118, the actuator system 128, the traffic vehicle communication system 130, and the traffic vehicle controller 136. The system controller 202 includes at least one system processor 204 and a system non-transitory computer-readable storage device or medium 206. The system processor 204 can be any custom or commercially available processor, a central processing unit (CPU), a graphics processing unit (GPU), an auxiliary processor in a number of processors associated with the system controller 202, a semiconductor-based microprocessor (in the form of a microchip or chipset), a macroprocessor, a combination thereof, or generally a device for executing instructions. The system computer-readable storage device or medium 206 may include volatile and non-volatile storage devices such as read-only memory (ROM), random access memory (RAM), and keep-alive memory (KAM). KAM is a persistent or non-volatile memory that can be used to store various operating variables when the system processor 204 is powered off. The system computer readable storage device or medium 206 of the system controller 202 may be implemented using a plurality of storage devices, such as a programmable read only memory (PROM), an electrically programmable read only memory (EPROM), an electrically erasable programmable read only memory (EEPROM), flash memory, or other electrical, magnetic, optical, or combination storage devices capable of storing data, some of which represents executable instructions. The system non-transitory computer readable storage device or medium 206 may store occupant data received from one of the internal sensors 152A-152N and from the transportation vehicle communication system 130.

[0046] The instructions may include one or more separate programs, each of which includes an ordered list of executable instructions for implementing logical functions. When the instructions are executed by the system processor 204, the processor receives and processes signals from the internal sensors 152A-152N and the vehicle communication system 130, and executes logic, calculations, methods and / or algorithms to generate an occupant knowledge graph, an occupant profile and / or control signals to actuators to automatically control the vehicle functional components of the host vehicle 100 or transmit at least a portion of the occupant profile to the target vehicle.

[0047] Although a single system controller 202 is shown, the profile system 132 may include one or more system controllers 202 located on or off the transportation vehicle 100 in a cloud-based server, configured to communicate via a suitable communication medium or combination of communication media and to cooperate to process sensor signals, execute logic, calculations, methods, and / or algorithms, and generate control signals to automatically control functional components of the profile system 132. In various embodiments, one or more instructions of the system controller 202 are embedded in the profile system 132. The system non-transitory computer-readable storage device or medium 206 and / or the transportation vehicle non-transitory computer-readable storage device or medium 206 include machine-readable instructions that, when executed by one or more system processors 204, cause the system processors 204 to perform the methods and processes described below.

[0048] Figure 4 The present invention is a block diagram of a method 400 for generating an occupant profile using an occupant knowledge graph for seamless transition of an occupant from a host transportation vehicle to a target transportation vehicle and / or for machine learning to generate actionable insights. The host transportation vehicle refers to a transportation vehicle in which information about the occupant is collected to develop an occupant knowledge graph and from which the occupant profile is generated. The target transportation vehicle is a transportation vehicle to which the occupant transitions from the host transportation vehicle.

[0049] Method 400 includes: box 402, collecting and receiving occupant data; box 404, generating an occupant knowledge graph based on the occupant data; box 405, generating an occupant profile based on the knowledge graph; box 406, generating a graph embedding based on the knowledge graph; box 408, training a machine learning (ML) architecture related to the graph embedding or the concatenation of multiple graph embeddings to generate actionable insights.

[0050] In box 402, occupant data related to the behavior and preferences of the occupant (also called the user) is collected by internal sensors. Additional occupant data can be received from external sources (such as social media accounts, subscription services, dealer information, etc.). The occupant data to be included in the knowledge graph includes: metadata related to the occupant and the functional components, and interaction data between the occupant and the customer-facing functional components within the transportation vehicle. Examples of occupant data include, but are not limited to, occupant metadata (such as demographic data, physiological data, biometric data, psychological data, and social data), as shown in Table 1; occupant-specific service and customer-facing functional component metadata, as shown in Table 2; and occupant interaction data, as shown in Table 3. Occupant data can be further classified into: constant or slowly changing variables, and rapidly changing variables.

[0051]

[0052] Table 1

[0053]

[0054] Table 2

[0055]

[0056] Table 3

[0057] In block 404, a knowledge graph is generated based on the ontology of the collected occupant data. Figure 5 , based on the car ontology, the collected occupant data is contextualized into a knowledge graph 500 of meaningful information. The knowledge graph 500 is an evolving graph-structured data type that encodes both metadata (related to occupants / users and customer-facing features within the vehicle) and interaction data between occupants / users, customer-facing features within the vehicle, and other interesting extra-vehicle entities represented as nodes (such as dealers, car insurance, homes, social media, etc.).

[0058] The occupant knowledge graph includes three main parts: nodes 502, edges 504, and attributes 506. Nodes 502 represent real-world entities, such as users and customer-facing functional components. Edges 504 or relationships define the connections between these nodes. Attributes 506 can be associated with these nodes 502 and edges 504 to represent defined metadata and interaction data about entities of interest. The expected benefits of using a knowledge graph include: capturing dynamic interactions between different entities of interest and supporting semantic search through contextual background.

[0059] Knowledge graph technologies include Resource Description Framework (RDF) and Labeled Property Graph (LPG) methods. RDF and SPARQL query languages ​​are defined by the World Wide Web Consortium (W3C). The LPG method and Cypher query language are adopted by graph technology vendors. There are also some formal ontologies that can be used to define a set of concepts and types to be included in the occupant knowledge graph, such as the automotive ontology taught by the automotive ontology working group (TheAutomotive Ontology working Group: "Automotive Data Interoperability at WebScate". Website: www.automotive-ontology.org. Access date: August 14, 2023).

[0060] In block 405, an occupant profile may be generated based on querying the knowledge graph (using, for example, the SPARQL query language or the Cypher query language). The RDF and SPARQL query languages ​​are defined by the W3C. Graph technology vendors employ the LPG approach and the Cypher query language. The occupant profile may be used in method 600 of seamlessly transferring an occupant from a host vehicle to a target vehicle.

[0061] In block 406, a graph embedding is generated based on the knowledge graph from block 404 and prepared for machine learning. This technique is described in Dai, Yuanfei; Wang, Shiping; Xiong, Neal N.; Guo, Wenzhong, et al., May 2020: "A Survey on Knowledge Graph Embedding: Approaches, Applications and Benchmarks", Electronics, Vol. 9, No. 5, p. 750, doi: 10.3390 / electronics9050750, the entire contents of which are incorporated herein by reference.

[0062] Referring to FIG. 7 , graph embedding can be performed at different progressive levels, for example, Fig. 7A The node level 706A shown is as follows Figure 7B Link level 706B as shown, or as Figure 7C The segmented graph level 706C is shown. Node level embedding Is categorized (e.g., feature-level utilization, NPS for specific features, feature adoption). Link-level embedding Is link prediction (e.g. two features are used together or eliminate unsubscribe links / services). Graph-level embedding are stratification (e.g. customer stratification) and anomaly detection.

[0063] Return to reference Figure 4 In block 408, the graph embedding is provided to a machine learning framework to generate actionable insights. The graph embedding may be concatenated with other graph embeddings from the same vehicle type or different vehicle types for the ML framework to predict actionable insights. The ML framework may invoke one or more deep learning architectures (e.g., deep neural networks, convolutional deep neural networks, deep belief networks, and / or recurrent neural networks) to learn and predict actionable insights based on the graph embedding or concatenation of graph embeddings.

[0064] Figure 6is a functional diagram illustrating a method 600 for seamlessly transferring an occupant from a host transportation vehicle 602 to a target transportation vehicle 626. The occupant profiling system 132 of the host transportation vehicle 602 collects occupant data to generate an occupant knowledge graph. The occupant knowledge graph implements an automotive ontology-driven, transferable, self-describing, intrinsically interpretable, contextually relevant occupant digital profile that encodes multimodal metadata related to the occupant and the transportation vehicle functional components and the interactions therebetween. Based on the occupant's preferences (including their wishes and needs), the occupant knowledge graph is queried with a predetermined query to generate an occupant profile.

[0065] Moving to block 604, the occupant profile is uploaded to the cloud server using the transport vehicle communication system 130. Blocks 608 to 618 may be performed by the system controller 202 on the host transport vehicle 602 or in the cloud server 604. Moving to block 608, the system controller 202 or the cloud server 604 validates the request initiated by the occupant or a third party service (on behalf of the occupant to transfer the occupant profile to the target transport vehicle 626). Moving to block 610, if more than one profile exists, the requester selects the occupant profile, or transfers a portion of the occupant profile to the target transport vehicle 626.

[0066] Moving to box 612, the system controller 202 or cloud server 604 confirms whether the target transportation vehicle 626 includes transportation vehicle features that are compatible with the transportation vehicle features of the host transportation vehicle 602. Compatible means that the target transportation vehicle includes transportation vehicle components, features, and / or software applications that are substantially the same and / or functionally similar (with respect to the occupant profile) as the host transportation vehicle.

[0067] Moving to box 614, if compatible, the method 600 moves to box 616 and transfers the occupant profile to the target transportation vehicle 626. The profile system of the target transportation vehicle 626 receives the occupant profile and instructs the actuators of the target transportation vehicle 626 to configure compatible transportation vehicle features according to the occupant preferences (e.g., personalization) identified in the occupant profile. Non-limiting examples of occupant preferences include personalized ergonomic settings (e.g., adjusting seat and steering wheel height) and visibility settings (e.g., adjusting mirrors, rearview camera zoom).

[0068] Referring back to block 614, if incompatible, meaning that the target transportation vehicle 626 does not include certain transportation vehicle components, functional components, and / or software applications that are substantially the same and / or functionally similar to the host transportation vehicle 602, the method 600 moves to block 618. In block 618, the system controller 202 determines whether the target transportation vehicle 626 is missing transportation vehicle components, functional components, and / or software applications of the host transportation vehicle 602, or whether the target transportation vehicle 626 includes new or updated components, functional components, and / or software applications.

[0069] At block 618, if the system controller determines that the target transportation vehicle 626 lacks transportation vehicle components, features, and / or software applications that are substantially the same and / or functionally similar to the host transportation vehicle 602, the method 600 moves to block 622. At block 622, the system controller 132 of the host transportation vehicle 602 or the target transportation vehicle 626 recommends upgrading the target transportation vehicle 626 that lacks transportation vehicle components, features, and / or software applications that are substantially the same and / or functionally similar to the host transportation vehicle 602.

[0070] Referring back to block 618, if the system controller 132 determines that the target transportation vehicle 626 includes transportation vehicle components, features, and / or software applications that are not present in the host transportation vehicle 602 (i.e., new features are present in the target transportation vehicle), the method 600 moves to block 624. In block 624, the system controller 132 of the host transportation vehicle 602 or the target transportation vehicle 626 recommends to the occupant that in-vehicle coaching be provided regarding components, features, and / or software applications of the target transportation vehicle 626 that are not present in the host transportation vehicle 602. The coaching may be positive or proactive.

[0071] Figure 8 A functional block diagram of using a concatenated graph embedding of multiple knowledge graphs based on the same vehicle type or multiple vehicle types to generate actionable analysis or insights. The graph embedding is an occupant knowledge graph generated based on occupant data collected from multiple host vehicles 100A, 100B, 100N and from external sources. In boxes 802A to 802N, occupant data related to occupant behavior and preferences are collected by internal sensors and / or vehicle communication systems. Move to boxes 804A to 804N, and generate a separate occupant knowledge graph based on the vehicle body of the collected occupant data. Move to boxes 806A to 806N, generate a knowledge graph embedding based on the occupant knowledge graph, and prepare the knowledge graph embedding for machine learning. Move to box 908, concatenate the knowledge graph embedding. Move to box 910, use the concatenated knowledge graph embedding in the ML architecture to predict trends based on a set of predetermined training algorithms.

[0072] Such algorithms are designed to generate actionable analytics or insights. Examples of such actionable analytics or insights include, but are not limited to: determining quantifiable Net Promoter Scores (NPS) at the feature level, vehicle level, and fleet level; identifying the most frequently used features; predicting preferences for features and services and matching customers with the features that customers are most likely to want; identifying customer tiers for targeted marketing, retention, and variation reduction; predicting features used together, services purchased together, common claims, etc.; and detecting anomalies.

[0073] The description of the present disclosure is merely exemplary in nature, and variations that do not depart from the general meaning of the present disclosure are intended to fall within the scope of the present disclosure. These variations should not be regarded as departing from the spirit and scope of the present disclosure.

Claims

1. An occupant profiling approach for seamless experiences and actionable insights, including: receiving occupant data associated with an occupant of a host vehicle; Using the vehicle ontology, generating a knowledge graph based on the occupant data; querying the knowledge graph to generate an occupant profile including preferences of an occupant of the host transportation vehicle; transmitting at least a portion of the occupant profile to a target vehicle; comparing the target transportation vehicle's features to the host transportation vehicle's features to determine at least one compatible feature, wherein the at least one compatible feature is a transportation vehicle feature present in both the target transportation vehicle and the host transportation vehicle; as well as At least one compatible feature of the target transportation vehicle is configured based on the occupant profile.

2. The method according to claim 1, further comprising: comparing the target vehicle's features to the host vehicle's features to determine at least one incompatible feature, wherein the at least one incompatible feature is a vehicle feature present in the target vehicle but absent in the host vehicle; and Based on the preferences of the occupant of the host vehicle, tutorials are recommended related to the vehicle features present in the target vehicle but missing from the host vehicle.

3. The method according to claim 1, further comprising: comparing the functional components of the target transportation vehicle with the functional components of the host transportation vehicle to determine at least one incompatible functional component, wherein the at least one incompatible functional component is a transportation vehicle functional component that is present in the host transportation vehicle but is absent in the target transportation vehicle; as well as Based on preferences of an occupant of the host vehicle, an upgrade to the target vehicle is recommended to include vehicle features present in the host vehicle.

4. The method according to claim 2, wherein: The missing vehicle feature in the target vehicle is a subscription to a software-enabled feature in the target vehicle; as well as The recommending to upgrade the target vehicle includes: based on the preference of the occupant of the host vehicle, recommending to activate the subscription of the software-enabled functional component in the target vehicle.

5. The method according to claim 1, wherein: The occupant data includes: occupants of the host transportation vehicle, metadata of customer-facing functional components of the host transportation vehicle, and interaction data between the occupants and the customer-facing functional components of the host transportation vehicle.

6. The method according to claim 5, wherein: The occupant's metadata includes demographic data, physiological data, biometric data, psychological data, and social data.

7. The method according to claim 1, further comprising: receiving a plurality of sets of occupant data associated with a plurality of occupants of a plurality of host vehicles; generating a separate knowledge graph for each of the plurality of sets of occupant data; Generate graph embeddings for each individual knowledge graph; Generate a graph embedding concatenation by concatenating the graph embeddings; as well as Train a machine learning architecture using the graph embedding concatenation to generate actionable insights.

8. The method according to claim 7, wherein: Graph embedding is generated for each individual knowledge graph at one of the following levels: node level, link level, and graph level.

9. The method according to claim 7, wherein: The actionable insights include at least one of: quantifiable Net Promoter Score (NPS) at feature level, vehicle level, and fleet level; commonly used features; customer preferences for features and services; customer segmentation for targeted marketing, vehicle retention, and reduction of vehicle variants; and, vehicle features used together, services purchased together.

10. The method according to claim 1, wherein: The host vehicle is an autonomous vehicle with an occupant profile system.

Images