Method and apparatus for distributed scenario engine for vehicles

Abstract

The present application relates to a kind of method and device for the distributed scene engine of vehicle, the method includes: obtaining the context information of vehicle;Current scene where vehicle is located is judged to be comprehensive scene or pure in-vehicle scene based on context information and predetermined rule set;When current scene is judged to be comprehensive scene, in-vehicle scene signal is generated for in-vehicle scene in comprehensive scene based on context information;And in-vehicle scene signal is sent to cloud server.

Description

Methods and apparatus for a distributed scene engine for vehicles Technical Field

[0001] This invention relates to a method and apparatus for a distributed scene engine for vehicles. Background Technology

[0002] In connected vehicle scenarios, many services need to be presented to users appropriately and at the right time. Therefore, a scenario engine is defined to support the recommendation of appropriate content or services to drivers within specific contexts. These scenarios are typically predefined by operators, product managers, and so on. To support maximum flexibility, rule engines are often used. These rule engines can be used to support business events, service recommendations, and a better user experience. Scenario engines are often used by current connected vehicle systems to provide artificial intelligence (AI) technology. Scenario engines can support intelligent vehicles with internet connectivity, and not only achieve physical connectivity but also provide an intelligent mobility experience by connecting the vehicle to cloud servers (which may also be referred to as the "cloud" in this document).

[0003] However, current scene engines still process human-defined scenarios, which is difficult to implement when supporting intelligence while also considering the vehicle's physical characteristics, thus limiting flexibility. From a performance and efficiency perspective, current scene engines prioritize network information over vehicle-specific information. They are primarily rule-based, meaning scene definitions are often not fixed and frequently change depending on requirements. Vehicle scene engines need to collect a wide range of information across the vehicle and cloud servers. To enable the scene engine to understand the current situation (or context) and trigger relevant actions, the rule engine typically needs to collect diverse information across the vehicle and cloud servers, such as vehicle speed, fuel level, headlight status, and on-chip system information at the vehicle level, and weather, flight information, and order status at the cloud server level.

[0004] As information within vehicles becomes increasingly complex, in-vehicle infotainment systems struggle to handle computationally intensive tasks. However, transmitting large amounts of vehicle-side information to cloud servers for processing would place an excessive burden on cloud servers, leading to high costs and difficulty in supporting real-time tasks, while also increasing the cost of information transmission. Many existing scene engines organize rule parameters obtained from cloud servers or transmit all relevant vehicle parameters (such as sensor data) from the vehicle to the cloud server. However, this incurs significant data transmission costs and puts considerable pressure on the network. Some other scene engines trigger logic from the in-vehicle infotainment system when certain conditions are met (e.g., navigating to a specific point of interest, fuel level below a certain threshold), sending a request for action to the cloud server. However, this results in hard-coded trigger logic that cannot adapt to flexibility requirements.

[0005] Therefore, there is a need and a challenge for a system that is both flexible and efficient. Summary of the Invention

[0006] This invention is provided to introduce a set of concepts, which will be further described in the following detailed description. This invention is not intended to identify key or essential features of the protected subject matter, nor is it intended to limit the scope of the protected subject matter.

[0007] The purpose of this invention is to provide a method and apparatus for a distributed scene engine for vehicles, and in particular a method and apparatus for a two-stage collaborative distributed scene engine for vehicle networking systems, so as to at least partially overcome the defects of the prior art.

[0008] An embodiment of the present invention provides a method for a distributed scene engine for a vehicle, the method being executed at an in-vehicle scene engine, comprising: obtaining context information of the vehicle; determining, based on the context information and a predetermined rule set, whether the current scene in which the vehicle is located is a comprehensive scene or a pure in-vehicle scene; when the current scene is determined to be the comprehensive scene, generating an in-vehicle scene signal for the in-vehicle scene in the comprehensive scene based on the context information; and sending the in-vehicle scene signal to a cloud server.

[0009] Embodiments of the present invention also provide a method for a distributed scene engine for a vehicle, the method being executed at a cloud-based scene engine, comprising: obtaining external information of the vehicle; receiving in-vehicle scene signals from the vehicle; identifying a comprehensive scene based on the external information and the in-vehicle scene signals according to a predetermined cloud rule set; generating a comprehensive scene signal based on the identified comprehensive scene; and sending the comprehensive scene signal to the vehicle.

[0010] Embodiments of the present invention also provide an apparatus for a distributed scene engine for a vehicle. The apparatus, located at an in-vehicle scene engine, includes: an acquisition module for acquiring context information of the vehicle; a judgment module for determining, based on the context information and a predetermined rule set, whether the current scene in which the vehicle is located is a comprehensive scene or a pure in-vehicle scene; a generation module for generating an in-vehicle scene signal based on the context information when the current scene is determined to be the comprehensive scene; and a sending module for sending the in-vehicle scene signal to a cloud server.

[0011] Embodiments of the present invention also provide an apparatus for a distributed scene engine for a vehicle. The apparatus, located at a cloud-based scene engine, includes: an acquisition module for acquiring external information of the vehicle; a receiving module for receiving in-vehicle scene signals from the vehicle; an identification module for identifying a comprehensive scene based on the external information and the in-vehicle scene signals according to a predetermined cloud rule set; a generation module for generating a comprehensive scene signal based on the identified comprehensive scene; and a sending module for sending the comprehensive scene signal to the vehicle.

[0012] An apparatus for a distributed scene engine for a vehicle according to an embodiment of the present invention includes: a processor; and a memory for storing executable instructions, wherein the executable instructions, when executed, cause the processor to perform the aforementioned method.

[0013] A machine-readable medium according to an embodiment of the present invention stores executable instructions thereon, wherein, when executed, the executable instructions cause a machine to perform the aforementioned method.

[0014] As can be seen from the above description, the solution of this embodiment of the invention can enable independent closed-loop processing of scenarios in the vehicle and the cloud server by deploying a two-stage distributed scene engine at the vehicle and the cloud server respectively. This allows some scenarios that can be processed on the vehicle side to be processed by the in-vehicle scene engine, while information on scenarios that cannot or are difficult to complete in the vehicle (such as scenarios that require a lot of processing resources) is transmitted to the cloud server for processing. This can improve real-time response capability and reduce the amount of data transmitted to the cloud server.

[0015] It should be noted that one or more of the above aspects include the detailed description below and the features specifically pointed out in the claims. The following description and drawings set forth certain illustrative features of the one or more aspects in detail. These features merely indicate various ways in which the principles of each aspect can be implemented, and this disclosure is intended to include all such aspects and their equivalents. Attached Figure Description

[0016] The following description will take into account several aspects disclosed, which are provided to illustrate rather than limit the aspects disclosed.

[0017] Figure 1 shows a schematic diagram of the architecture of a distributed scene engine system for a vehicle according to an embodiment of the present invention;

[0018] Figure 2 shows a schematic diagram of a vehicle-side scene processing architecture including an in-vehicle scene engine in a distributed scene engine, according to an embodiment of the present invention.

[0019] Figure 3 shows a schematic diagram of a cloud-side scene processing architecture including a cloud-based scene engine in a distributed scene engine, according to an embodiment of the present invention.

[0020] Figure 4 shows a flowchart of a method for a distributed scene engine for a vehicle according to an embodiment of the present invention;

[0021] Figure 5 shows a flowchart of an example method for classifying factors of vehicle context information according to an embodiment of the present invention.

[0022] Figure 6 shows an example diagram of a method for a distributed scene engine for a vehicle executed at an in-vehicle scene engine according to an embodiment of the present invention.

[0023] Figure 7 shows an example diagram of a method for a distributed scene engine for a vehicle executed at a cloud scene engine according to another embodiment of the present invention;

[0024] Figure 8 shows a schematic diagram of a distributed scene engine for a vehicle at an in-vehicle scene engine according to an embodiment of the present invention.

[0025] Figure 9 shows a schematic diagram of a distributed scene engine for a vehicle at a cloud scene engine according to an embodiment of the present invention;

[0026] Figure 10 shows a schematic diagram of a device for a distributed scene engine for a vehicle according to an embodiment of the present invention. Detailed Implementation

[0027] The present disclosure will now be discussed with reference to various exemplary embodiments. It should be understood that this discussion of embodiments is merely intended to enable those skilled in the art to better understand and implement the embodiments of the present disclosure, and is not intended to teach any limitation on the scope of the present disclosure.

[0028] The various embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0029] Figure 1 illustrates a schematic architecture of a system 100 for a distributed scene engine for a vehicle according to an embodiment of the present invention. As shown in Figure 1, the exemplary system 100 for a distributed scene engine for a vehicle may include a network 110, a vehicle 120, and a cloud server 130.

[0030] In Figure 1, network 110 is used to interconnect vehicle 120 and cloud server 130, enabling vehicle 120 and cloud server 130 to exchange data or information via network 110. Network 110 can be any type of network capable of interconnecting network entities, and can be a single network or a combination of various networks.

[0031] Vehicle 120 may be an intelligent vehicle. Vehicle 120 may include an in-vehicle scene engine 122, which is used to determine whether the current scene of the vehicle is a comprehensive scene, and to process the in-vehicle scene in a closed loop and / or process the comprehensive scene in collaboration with a cloud scene engine. In this document, a comprehensive scene may include a combination of in-vehicle scenes and cloud scenes. In the case of a comprehensive scene, the in-vehicle scene engine 122 can process in-vehicle information or data related to the in-vehicle scene portion of the comprehensive scene in a closed loop, generate in-vehicle scene signals for the in-vehicle scene, and transmit the in-vehicle scene signals to the cloud server 130 via network 110. Furthermore, the in-vehicle scene engine 122 can also perform operations based on the identified in-vehicle scene, such as issuing warnings or prompts to the user or driver and / or controlling vehicle components to automatically perform actions (such as automatic deceleration or braking). In some examples, the in-vehicle scene engine 122 can respond to the vehicle's current state information, identify the in-vehicle scene based on predetermined rules, and perform related operations. In some examples, the vehicle's current state information may include, but is not limited to, various information related to the vehicle itself, such as vehicle status, remaining fuel level, speed, battery level, and vehicle malfunctions. In some cases, the in-vehicle scene engine 122 can identify in-vehicle scenes based on a predefined set of in-vehicle rules or algorithms.

[0032] The cloud server 130 can receive signals from the vehicle 120 or optionally obtain vehicle-related information. It can also obtain external information from the network or other entities, such as weather, user flight, order or express package information, vehicle trip or navigation information (e.g., destination, distance from the destination, arrival time, points of interest along the way, etc.), city traffic restriction information, user profile (e.g., driving habits, preferred gas station brands), user traffic violation information, gas station or charging pile information, etc.

[0033] As shown in Figure 1, a cloud scene engine 132 is configured in the cloud server 130. In some examples, the cloud scene engine 132 receives in-vehicle scene signals (also known as scene codes) from the vehicle, obtains corresponding context information (e.g., external information), and, based at least on the in-vehicle scene signals and the corresponding context information, identifies a comprehensive scene and generates a comprehensive scene signal according to predetermined rules. In some examples, the cloud scene engine 132 can identify cloud scenes and generate cloud scene definitions based on predetermined cloud rule sets or algorithms and external information. In some examples, the cloud scene engine 132 can also identify comprehensive scenes and generate comprehensive scene signals based on cloud scene definitions and in-vehicle scene signals. The comprehensive scene signal can be fed back or sent to the vehicle 120 via the network 110, enabling the vehicle 120 to perform operations based on the comprehensive scene signal, including but not limited to issuing prompts to the vehicle's user and / or controlling vehicle components to automatically perform actions.

[0034] It should be understood that all network entities shown in Figure 1 are exemplary, and system 100 may include any other network entities depending on specific application requirements. Furthermore, although only one vehicle 120 and one cloud server 130 are shown in Figure 1, any number of vehicles and / or any number of cloud servers may exist depending on specific application requirements, and the present invention does not impose any limitations in this regard.

[0035] Figure 2 shows a schematic diagram of a vehicle-side scene processing architecture 200 including an in-vehicle scene engine in a distributed scene engine, according to an embodiment of the present invention.

[0036] The scene processing architecture 200 on the vehicle side includes at least a rule set definer 202, a data synthesizer 204, a receiver 206, a sensor 208, an in-vehicle scene engine 210, and a scene executor 216. In some examples, the in-vehicle scene engine 210 includes, but is not limited to, a scene interpreter 212 and a scene recognizer 214.

[0037] In some examples, the rule set definer 202 is used to pre-set or define several rule sets (or algorithms), such as a scene rule set for determining whether the current scene of the vehicle is a comprehensive scene or a purely in-vehicle scene, an in-vehicle rule set for identifying in-vehicle scenes, and a cloud rule set for identifying cloud scenes or comprehensive scenes. In some examples, the scene rule set can be in the form of <condition or information, scene category>, such as <pure in-vehicle information, pure in-vehicle scene>, <contextual information or in-vehicle information + external information, comprehensive scene>. For example, a scene rule can be <current vehicle state, in-vehicle scene>, <current vehicle state + gas station promotional information, comprehensive scene>, etc. It should be understood that some of the scene rules listed here are only exemplary; in practical applications, scene rules of any form or content can be used.

[0038] In this paper, the rule set definer 202 transmits a predefined rule set to the vehicle or the cloud for use by the in-vehicle scene engine and the cloud scene engine to identify in-vehicle scenes, cloud scenes, or combined scenes, respectively. In some examples, depending on the requirements, the distinction between in-vehicle closed-loop scenes and cloud closed-loop scenes in a combined scene can be determined through logical definitions in the rule platform or tool. The rules in the rule set are predefined, compiled, and dynamically distributed to the in-vehicle infotainment system to instruct the system on how to perform operations. For example, the rule set can be defined by specifying what actions the system should take under what conditions. In this paper, "action" includes actions in both in-vehicle and cloud scenes. In other examples, the in-vehicle rule set can be in the form of <condition or information, specific scene or action>. For example, a rule in the in-vehicle rule set could be <low fuel level, refueling scene or refueling>, etc. In still other examples, the cloud rule set can be in the form of <in-vehicle scene signal + one or more cloud conditions or information, specific scene or action>. For example, a rule in the cloud rule set could be <a signal indicating "needs refueling" + gas station promotional information, refueling scene or refueling>, etc. It should be understood that the in-vehicle rules and cloud rules listed here are merely examples. In actual applications, in-vehicle rules and cloud rules of any form or content can be used.

[0039] The vehicle-side data synthesizer 204 can obtain a predetermined rule set from the rule set definer 202. In some examples, the rule set can be obtained in advance from the rule set definer 202 and stored in the data synthesizer 204. Furthermore, the data synthesizer 204 can obtain external vehicle information or corresponding context information from the cloud from the receiver 206, and internal vehicle information or corresponding context information from the sensor 208 or the vehicle's memory (not shown). In this document, internal and external vehicle information can be combined to form the vehicle's context information. In some examples, internal vehicle information may include, but is not limited to, vehicle location information, vehicle speed information, vehicle driving time, vehicle geofence, remaining fuel or battery level, etc.; external vehicle information may include, but is not limited to, the type of gas station near the vehicle's current location (whether there are promotions, whether it is a user-preferred brand, etc.), whether there are delivery services at the vehicle's location, etc.

[0040] The data synthesizer 204 transmits the acquired rule set and context information to the in-vehicle scene engine 210 for closed-loop scene processing, thereby generating in-vehicle scene signals to be sent to the cloud server, or to take corresponding actions directly within the vehicle based on the in-vehicle scene. It should be understood that although the data synthesizer 204 is shown outside the in-vehicle scene engine 210 in Figure 2, it can also be located inside the in-vehicle scene engine 210. In some examples, the rule set can be transmitted from the rule set definer 202 to the in-vehicle scene engine 210 in real time, or pre-transmitted to and stored in the in-vehicle scene engine 210.

[0041] The in-vehicle scene engine 210 is configured to process in-vehicle scenes in a closed loop and / or process comprehensive scenes in collaboration with a cloud-based scene engine, and it can be an example of the in-vehicle scene engine 122 of Figure 1.

[0042] In one embodiment, the scene interpreter 212 in the in-vehicle scene engine 210 is used to determine whether the current scene of the vehicle is a pure in-vehicle scene or a comprehensive scene based on the scene rule set and the currently received data or information. Specifically, it can determine whether the current scene of the vehicle is a comprehensive scene based on whether the received information includes cloud factors.

[0043] For example, if the context information of the current scene only includes factors from the vehicle's infotainment system, the current scene can be determined as a purely in-vehicle scene. The information can then be processed only within the vehicle to identify the current scene and perform in-vehicle operations based on this identified scenario, without needing to provide the information or signals generated from it to the cloud. For instance, when the information only includes factors such as the vehicle's remaining fuel level being below a threshold, the information can be processed to identify the purely in-vehicle scene as a "refueling scenario" and prompt the user that the vehicle needs refueling. In these exemplary purely in-vehicle scenes, the system does not need to send in-vehicle information or generated signals to the cloud.

[0044] In other examples, if the contextual information of the current scene includes a combination of in-vehicle and cloud factors, the current scene can be classified as a composite scene. In this case, the in-vehicle scene is processed at the vehicle, and only the in-vehicle scene signal generated from this processing is transmitted to the cloud, without needing to transmit a large amount of information related to the in-vehicle scene. For example, if the information in the current scene includes in-vehicle factors indicating that the remaining fuel level is below a threshold and cloud factors indicating information about various gas stations (such as promotional information), then the current scene is classified as a composite scene.

[0045] In one embodiment, the scene recognizer 214 is used to identify the current scene based on received context information and a predetermined in-vehicle rule set, such as a vehicle start-up scene, a refueling / charging scene, a deceleration scene, a parking scene, an acceleration scene, an air conditioning start-up / stop scene, etc. For example, when "low fuel" information is received from sensor 208, based on the rule <low fuel, refueling scene or refueling> in the predetermined in-vehicle rule set, the current scene is identified as a "refueling scene" or the action to be performed is "refueling".

[0046] In some examples, based on the category and identified current scene, the scene actuator 216 can generate an in-vehicle scene signal under a comprehensive scene category or perform an operation for the current scene under a purely in-vehicle scene category. In some examples, the performed operations include, but are not limited to, issuing prompts to the vehicle user and / or controlling vehicle components to automatically perform actions. For example, if the current scene is a purely in-vehicle scene and the identified purely in-vehicle scene is a refueling scene, the scene actuator 216 can issue a prompt signal to the user: "The vehicle needs refueling and can travel X kilometers." As another example, if the current scene is a purely in-vehicle scene and the identified scene is a deceleration scene, the scene actuator 216 can issue a prompt to the driver: "Please slow down" or directly control the brakes to slow down in autonomous driving mode. In another example, if the current scenario is a composite scenario and the identified composite scenario is a refueling scenario, the scenario actuator 216 can generate an in-vehicle scenario signal (e.g., a signal or scenario code indicating "the vehicle needs refueling and can travel X kilometers") and send this signal to the cloud server, without having to transmit in-vehicle information related to the refueling scenario, such as remaining fuel level, vehicle speed, and fuel consumption per kilometer, to the cloud server. This can speed up the vehicle's response or reduce the amount of information transmitted under different scenario categories. Although the above description uses a vehicle refueling scenario as an example, in practical applications, any other scenario can be processed or any other appropriate information or prompts can be used.

[0047] It is important to understand that although the scene executor 216 is shown outside the in-vehicle scene engine 210 in Figure 2, it can also be set inside the in-vehicle scene engine 210. Furthermore, although the rule set definer 202 is shown as being set within the scene processing architecture on the vehicle side in Figure 2, it can also be set in a cloud server, within another entity, or stored as a module in a separate device, depending on the actual application requirements.

[0048] Figure 3 shows a schematic diagram of a cloud-side scene processing architecture 300 including a cloud-based scene engine in a distributed scene engine, according to an embodiment of the present invention.

[0049] The cloud-side scene processing architecture 300 includes at least a rule set definer 302, a cloud scene engine 304, a receiver 306, and a scene executor 316. In some examples, the cloud scene engine 304 may include, but is not limited to, a signal recognizer 308, (optional) a scene supplement unit 310, a scene selection unit 312, and a scene element extraction unit 314.

[0050] In some examples, rule set definer 302 may be the same as or similar to rule set definer 202 in Figure 2, used to pre-set or define several rule sets, such as a scene rule set for determining whether the current scene of the vehicle is a comprehensive scene or a purely in-vehicle scene, an in-vehicle rule set for identifying in-vehicle scenes, and a cloud rule set for identifying cloud scenes or comprehensive scenes. In other examples, rule set definer 302 may differ from rule set definer 202 in Figure 2, for example, defining only the cloud rule set for identifying cloud scenes or comprehensive scenes. For ease of description, only the functions or parts of the rule set definer 302 involving cloud rule sets are described in Figure 3. Rule set definer 302 transmits the pre-set cloud rule sets to cloud scene engine 304, enabling cloud scene engine 304 to identify cloud scenes or comprehensive scenes based on the cloud rule sets.

[0051] In the example shown in Figure 3, the cloud scene engine 304 receives in-vehicle scene signals from a vehicle (e.g., an in-vehicle scene engine) and external information or corresponding context information from the network or other entities via a network through a receiver 306. In this example, the signal recognizer 308 analyzes the received in-vehicle scene signals to identify them. The cloud scene engine 304 can identify cloud scenes based on external information and a predetermined set of cloud rules. The scene supplementation unit 310 supplements the identified cloud scenes based on a scene forest. In the embodiments described herein, the scene forest can be in the form of a scene tree, where each scene corresponds to several sub-scenes, and each sub-scene corresponds to several conditions or rules.

[0052] Based on the identified in-vehicle scene signals and / or supplementary cloud-based scenes, the scene selection unit 312 can identify or select matching scenes from the scene forest as a comprehensive scene. Optionally, the element extraction unit 314 extracts scene elements from the selected scenes to generate the comprehensive scene signal. In some examples, scene elements include various elements involving contextual information, including but not limited to vehicle speed, fuel / battery level, vehicle malfunction, weather, user preferences, user profiles, information for order information, user traffic violation information, city traffic restriction information, charging station availability, navigation destination, etc.

[0053] In some examples, scene executor 316 can generate a composite scene signal based on the identified composite scene. In further examples, scene executor 316 can process scene elements extracted from the scene to generate a composite scene signal. It should be understood that although scene executor 316 is shown outside the cloud scene engine 304 in Figure 3, it can also be located inside the cloud scene engine 304.

[0054] Although the rule set definer 302 is shown in Figure 3 as being set within the scene processing architecture on the cloud side, it can also be set within another entity or stored as a module in a separate device, depending on the actual application requirements.

[0055] Figure 4 shows a flowchart illustrating a method 400 for a distributed scene engine for a vehicle according to an embodiment of the present invention. The method 400 shown in Figure 4 can be implemented by, for example, system 100.

[0056] As shown in Figure 4, in box 402, rule sets are predefined by the rule set definer, such as scene rule sets, in-vehicle rule sets, cloud rule sets, etc.

[0057] In box 404, a rule set can be obtained from the rule set definer. In this document, the obtained rule set may include one or more of the scenario rule set, the in-vehicle rule set, and the cloud rule set. In some examples, these rule sets may be provided in advance to the vehicle or the cloud by the rule set definer, or they may be requested from the rule set definer in real time by the vehicle or the cloud.

[0058] In box 406, the vehicle's contextual information can be obtained, where the contextual information can be in the form of factors and each factor can be identified or classified as a vehicle-mounted factor or a cloud-based factor.

[0059] In box 408, the current scenario is determined to be either a comprehensive scenario or a purely in-vehicle scenario based on a predetermined set of scenario rules. In this paper, if all factors in the obtained context information are determined to be in-vehicle factors, the current scenario can be classified as a purely in-vehicle scenario; if the factors in the obtained context information include a combination of in-vehicle and cloud factors, the current scenario can be classified as a comprehensive scenario. In some examples, a comprehensive scenario refers to a scenario that includes both in-vehicle and cloud scenarios, or involves both in-vehicle processing and cloud server processing. If it is determined not to be a comprehensive scenario (i.e., a purely in-vehicle scenario), the process proceeds to box 410.

[0060] In box 410, based on the in-vehicle rule set obtained from the rule set definer in box 404 and the context information obtained in box 406, a specific scenario within a purely in-vehicle setting can be identified. In some examples, identifying a specific scenario may further include the following operations: determining a specific rule in the in-vehicle rule set that matches the context information (e.g., vehicle-mounted factors) based on the obtained context information and the in-vehicle rule set; and identifying the specific purely in-vehicle scenario based on that specific rule.

[0061] In box 412, in-vehicle operations are performed based on the purely in-vehicle scene identified in box 410. In some examples, in-vehicle operations may include prompting the user or instructing in-vehicle components to automatically perform actions, such as accelerating, decelerating, parking, turning the air conditioning on / off, etc.

[0062] In one embodiment, if it is determined in box 408 that the current scene is a composite scene, the process proceeds to boxes 414 and 422.

[0063] In box 414, based on the in-vehicle rule set obtained from the rule set definer in box 404 and the vehicle-mounted factors in the context information obtained in box 406, the specific in-vehicle scene under the comprehensive scenario can be identified.

[0064] In box 416, based on the identified in-vehicle scene, an in-vehicle scene signal is generated according to factors from the vehicle's infotainment system.

[0065] In box 418, the generated in-vehicle scene signal is sent to the cloud server.

[0066] In box 420, the signal is analyzed on the cloud server to identify in-vehicle scene signals, for example, by a signal recognizer in the cloud server, and the identified in-vehicle scene signals are provided to the cloud scene engine.

[0067] In box 422, cloud scenarios can be identified based on the cloud rule set obtained from the rule set definer in box 404 and cloud factors in the context information obtained in box 406.

[0068] In box 424, a definition of the cloud scene can be generated based on the identified cloud scene and provided to the cloud scene engine.

[0069] In box 426, the cloud scene engine receives the definition of the cloud scene, the in-vehicle scene signal, and optional cloud scene context information to identify the specific comprehensive scene.

[0070] In box 428, the cloud-based scene engine or cloud server generates a comprehensive scene signal based on the identified comprehensive scene and sends the comprehensive scene signal to the vehicle or the vehicle's user. For example, it may send a prompt signal to the user such as "The vehicle needs to refuel, and the gas station 2.5 kilometers away is having a promotion" or "The vehicle needs to refuel, and the gas station 3 kilometers away is your favorite gas station brand", etc.

[0071] Based on the flowchart in Figure 4, the specific process of the distributed scene engine method for vehicles will be described here using "pure in-vehicle refueling scenario", "comprehensive refueling scenario" and "comprehensive scenario of receiving express parcels" as examples, as shown below.

[0072] For any of the scenarios described above, the operations in boxes 402 and 404 can be the same. For example, in box 402, the scenario rules predefined in the rule set definer can include, but are not limited to, <remaining fuel level below threshold, pure in-vehicle scenario>, <remaining fuel level below threshold + gas station promotional information, combined scenario>, and <entering home via geofence + receiving package information, combined scenario>. In-vehicle rules can include, but are not limited to, <remaining fuel level below threshold, refueling scenario> and <geofence information, location scenario>; while cloud-based rules can include, but are not limited to, <gas station promotional information, refueling scenario> and <package information, location scenario>.

[0073] For the example of a "pure in-vehicle refueling scenario," when the vehicle context information obtained at box 406 only includes in-vehicle factors, such as low fuel level information obtained from the fuel tank sensor (e.g., remaining fuel level is below a threshold, belonging to in-vehicle factors), then based on the scenario rule <in-vehicle factors, pure in-vehicle scenario> obtained from the rule set definer (as shown in box 404), it can be determined at box 408 that the current scenario is a pure in-vehicle scenario rather than a comprehensive scenario (i.e., the judgment result is "no"), and the process can proceed to box 410. Based on the in-vehicle rule <remaining fuel level below threshold, refueling scenario> obtained at box 404, and the low fuel level information obtained at box 406, the current specific scenario can be identified (as shown in box 410) as a "refueling scenario." Based on the identified "refueling scenario," operations can be performed in the vehicle, such as issuing a prompt to the user such as "refueling is needed."

[0074] For the example of a "comprehensive refueling scenario," when the vehicle context information obtained at box 406 includes both vehicle-side factors and cloud-side factors—for example, low fuel level information obtained from the fuel tank sensor (a vehicle-side factor) and promotional information from one or more nearby gas stations obtained from the cloud server (a cloud-side factor)—then based on the scenario rule <vehicle-side factors + cloud-side factors, comprehensive scenario> obtained from the rule set definer (as shown in box 404), it can be determined at box 408 that the current scenario is a comprehensive scenario (i.e., the judgment result is "yes"), and the process can proceed to boxes 414 and 422. Based on the in-vehicle rules obtained at box 404, such as, but not limited to: <remaining fuel level below threshold, refueling scenario>, and the vehicle-side factors (low fuel level information) in the context information, the current specific in-vehicle scenario can be identified (as shown in box 414) as a "refueling scenario." Based on the identified "refueling scenario," an in-vehicle scenario signal can be generated (as shown in box 416). This signal can indicate that the vehicle needs refueling, and optionally, it can also indicate vehicle location information, vehicle ID, user ID, etc. This signal (e.g., in the form of a scenario code) is sent to the cloud (as shown in box 418). The cloud server receives and analyzes the signal to identify that the information indicated by the signal is "the vehicle needs refueling" (as shown in box 420). Furthermore, based on the cloud factor "gas station promotional message" in the context information and the cloud rule <gas station promotional information, refueling scenario> obtained at box 404, the specific cloud scenario can be identified as "refueling scenario" (as shown in box 422). The identified cloud scene definition, such as "refueling scene", and the corresponding context information (e.g., cloud factors "gas station promotion message" and the vehicle's current distance from the gas station) are provided to the cloud scene engine (as shown in box 424) to identify the specific comprehensive scene based on the in-vehicle scene signal and the cloud scene definition (as shown in box 426), thereby generating a comprehensive scene signal (e.g., the signal indicating "the vehicle needs refueling, and a gas station 2.5 kilometers away is having a promotion") and sending the comprehensive scene signal to the vehicle or the vehicle's user (as shown in box 428).

[0075] For the example of a "comprehensive scenario of receiving a package," when the vehicle context information obtained at box 406 includes both vehicle-mounted factors and cloud-based factors—for example, information about the vehicle entering a home's electronic fence obtained from a location sensor (belonging to vehicle-mounted factors) and package information related to the user obtained from a cloud server (belonging to cloud-based factors)—then based on the scenario rule <vehicle-mounted factors + cloud-based factors, comprehensive scenario> obtained from the rule set definer (as shown in box 404), it can be determined at box 408 that the current scenario is a comprehensive scenario (i.e., the judgment result is "yes"), and the process can proceed to box 414. Based on the in-vehicle rule <electronic fence information, location scenario> obtained at box 404, and the vehicle-mounted factor "entering the home's electronic fence" in the context information, the current specific in-vehicle scenario can be identified (as shown in box 414) as a "location scenario." Based on the identified "location scenario," an in-vehicle scenario signal can be generated (as shown in box 416). This signal can indicate that the vehicle has entered the home's electronic fence, and optionally, it can also indicate the vehicle's detailed location information, vehicle ID, user ID, etc. The signal (e.g., in the form of a scene code) is sent to the cloud (as shown in box 418). The cloud server receives and analyzes the signal (as shown in box 420) to identify the information indicated by the signal as "vehicle entering the home's electronic fence". Furthermore, based on the cloud factor "user-related parcel information" in the context information and the cloud rule <parcel information, location scene> obtained at box 404, the cloud scene can be identified as a "location scene" (as shown in box 422). The identified cloud scene definition, such as "location scene", and the corresponding context information (e.g., the cloud factor "user-related parcel information" and / or information about whether there is a parcel to be picked up at the current location) are provided to the cloud scene engine (as shown in box 424) to identify the specific comprehensive scene (as shown in box 426), thereby generating a comprehensive scene signal (e.g., a signal indicating "a parcel to be picked up at a parcel locker near home") and sending this comprehensive scene signal to the vehicle or the vehicle's user (as shown in box 428).

[0076] It should be understood that the above scenario examples are merely illustrative descriptions for the purpose of understanding the process described in Figure 4, and are not intended to limit the pure in-vehicle scenario or the comprehensive scenario.

[0077] Figure 5 shows a flowchart of an example method 500 for classifying factors of vehicle context information according to an embodiment of the present invention. This method 500 may be a process performed before executing the comprehensive scene determination operation at box 408 in Figure 4.

[0078] At box 502, the source of the obtained context information is determined based on predetermined rules to determine whether the obtained context information originates from inside the vehicle. In this document, the obtained context information can be in the form of factors and may include in-vehicle information and / or external information. For example, the source of information obtained through the vehicle's sensors can be determined as an in-vehicle source and such information can be identified as in-vehicle information; the source of information queried or obtained from the cloud can be determined as a cloud source and such information can be identified as external information.

[0079] If the result of the judgment at box 502 is yes, the process can proceed to box 504, where the in-vehicle rule engine (also known as the vehicle-mounted rule engine) further determines whether the factors in the context information are vehicle-mounted factors based on the vehicle context information.

[0080] If the result of the judgment at box 504 is yes, the process can proceed to box 508, where the obtained factors can be identified as vehicle-mounted factors.

[0081] If the result of the judgment at box 504 is negative, the process can proceed to box 506, where it can be determined whether the obtained factors are vehicle-side factors based on static data. In this article, static data can refer to fundamentally unchanging factors related to the vehicle, such as, but not limited to, the vehicle model or ID, vehicle component information, user ID, geofence data, and holiday times.

[0082] If the judgment result at box 506 is yes, the process can proceed to box 508, where the obtained information can be identified as a vehicle-mounted factor. If the judgment result at box 506 is no, the process can proceed to box 510. At box 510, the method can further determine whether the obtained factor is a vehicle-mounted factor based on the dynamic changes in the factor value. If the judgment result at box 510 is no, that is, the dynamic change in the factor value is below a threshold, for example, the dynamic change in the factor value is small and easily perceptible, the process can proceed to box 508, where the obtained factor can be identified as a vehicle-mounted factor.

[0083] If the judgment result at box 510 is yes, that is, the dynamic change of the factor value is greater than or equal to the threshold, for example, the dynamic change of the factor value is large and the change is difficult to perceive, then the process can go to box 512, where the obtained factor can be identified as a cloud factor.

[0084] Furthermore, if the judgment result at box 502 is negative, meaning the obtained information or factor originates from outside the vehicle, the process can proceed to box 510 to determine whether the dynamic change in the value of the obtained factor is greater than or equal to a threshold. If the judgment result is positive, meaning the dynamic change in the value of the factor is greater than or equal to the threshold (e.g., the dynamic change in the value of the factor is large and difficult to perceive), the process can proceed to box 512, where the factor originating from outside the vehicle is identified as a cloud-based factor. If the judgment result is negative, meaning the dynamic change in the value of the factor is less than the threshold (e.g., the dynamic change in the value of the factor is small and easily perceived), the process can proceed to box 508, where the factor originating from outside the vehicle can be identified as a vehicle-mounted factor.

[0085] In a further embodiment, if all the factors in the obtained context information are vehicle-mounted factors, the current scenario can be judged as a pure in-vehicle scenario; if the obtained factors are a combination of vehicle-mounted factors and cloud factors, the current scenario can be judged as a comprehensive scenario.

[0086] Based on the factor classification process in Figure 5, different contextual information or factors can be classified. For example, when the obtained contextual information is that the remaining fuel level is below a threshold, it can be determined at box 502 that the factor originates from inside the vehicle and the process proceeds to box 504; in box 504, based on the context, it can be determined that the factor is a vehicle-to-everything (V2X) factor, and the process can proceed to box 508. In another example, when the obtained contextual information is a geofence factor, it can be determined at box 502 that the factor originates from inside the vehicle and the process proceeds to box 504; in box 504, based on the context, it can be determined that the factor is not a V2X factor, and the process can proceed to box 506; in box 506, based on pre-deployed static data (e.g., geofence data) to the vehicle, it can be determined that the factor is a V2X factor, and the process then proceeds to box 508.

[0087] In another example, when the obtained context information is the current city (e.g., Beijing), such as through in-vehicle GPS, the judgment result at box 502 is yes, and the process proceeds to box 504. At box 504, based on the context, it can be determined that the factor "current city (Beijing)" is not a vehicle-mounted factor, so the process continues to box 506. At box 506, based on the distributed static data, it can be determined that this factor is not a vehicle-mounted factor, and the process proceeds to box 510 to determine whether the dynamic change of the factor's value is greater than or equal to a threshold. Since the value of the factor "current city" remains basically unchanged or changes little, the judgment result at box 510 is no, and the process proceeds to box 508 to classify this factor as a vehicle-mounted factor.

[0088] In some cases, if the obtained contextual information or factor is user preference or payment service (such as activating password-free service), it can be determined at box 502 that the factor does not originate from inside the vehicle, i.e., the judgment result is negative, and the process proceeds to box 510. Since the values ​​of user preference or activation of password-free service remain basically unchanged or change little, the judgment result at box 510 is negative (i.e., the dynamic change of the factor value is less than the threshold), and the process then proceeds to box 508 to classify the factor as a vehicle-mounted factor.

[0089] In some other examples, if the obtained contextual information or factor is user violation information, it can be determined at box 502 that the factor does not originate from inside the vehicle, i.e., the judgment result is negative, and the process proceeds to box 510. Since the value of user violation information is variable (difficult to perceive) or varies greatly, the judgment result at box 510 is positive (i.e., the dynamic change of the factor value is greater than or equal to the threshold), and the process then proceeds to box 512 to classify the factor as a cloud factor.

[0090] It should be understood that the above examples are merely illustrative descriptions for the purpose of understanding the process described in Figure 5, and are not intended to limit the obtained contextual information or factors or the corresponding classification process.

[0091] Figure 6 shows an example diagram of a method 600 for a distributed scene engine for a vehicle, executed at an in-vehicle scene engine according to an embodiment of the present invention.

[0092] In box 602, vehicle context information can be obtained. In some examples, the context information includes in-vehicle information and out-of-vehicle information, and obtaining the vehicle context information further includes: obtaining in-vehicle information from the vehicle's sensors and receiving out-of-vehicle information from the cloud. In some examples, each factor in the obtained context information can be categorized into vehicle-mounted factors and cloud factors.

[0093] At block 604, the current scenario of the vehicle can be determined as either a comprehensive scenario or a purely in-vehicle scenario based on context information and a predetermined set of rules. In one embodiment, a comprehensive scenario includes a combination of in-vehicle and cloud scenarios. In one example, when the current scenario is determined to be a purely in-vehicle scenario, in-vehicle operations are performed at least based on context information and a predetermined set of in-vehicle rules. In a further example, the method further includes: determining specific rules in the in-vehicle rule set that match the context information; identifying a purely in-vehicle scenario based on the specific rules; and performing in-vehicle operations based on the purely in-vehicle scenario, wherein the in-vehicle operations include issuing prompts to the user of the vehicle or controlling vehicle components to automatically perform actions. In some embodiments, determining whether the current scenario is a comprehensive scenario or a purely in-vehicle scenario further includes: determining the current scenario as a purely in-vehicle scenario when all factors in the context information are vehicle-mounted factors, and determining the current scenario as a comprehensive scenario when the context information includes a combination of vehicle-mounted factors and cloud factors.

[0094] In box 606, when the current scene is determined to be a comprehensive scene, an in-vehicle scene signal can be generated for the in-vehicle scene within the comprehensive scene based on context information. In some examples, generating the in-vehicle scene signal further includes: identifying the in-vehicle scene based on vehicle-mounted factors and a predetermined in-vehicle rule set; and generating the in-vehicle scene signal based on the definition of the identified in-vehicle scene.

[0095] In box 608, the in-vehicle scene signal can be sent to the cloud server. In some embodiments, the in-vehicle scene signal can be used to enable the cloud server to identify a comprehensive scene based at least on the in-vehicle scene signal and the corresponding context information, and to generate a comprehensive scene signal based on the identified comprehensive scene.

[0096] Optionally, the method 600 may further include: receiving a generated integrated scene signal from a cloud server; and performing an operation based on the integrated scene signal, wherein the operation includes prompting a user of the vehicle and / or controlling components of the vehicle to automatically perform actions. Although not shown in the figures, it should be understood that any other operations described herein may be used to add, remove, or modify steps in the method 600.

[0097] Figure 7 shows an example diagram of a method 700 for a distributed scene engine for a vehicle executed at a cloud scene engine according to another embodiment of the present invention.

[0098] In box 702, obtain the vehicle's exterior information.

[0099] In box 704, receive in-vehicle scene signals from the vehicle.

[0100] In block 706, a comprehensive scene is identified based on external vehicle information and in-vehicle scene signals according to a predetermined cloud rule set. In one embodiment, identifying the comprehensive scene further includes: classifying each factor in the external vehicle information as a vehicle-mounted factor or a cloud factor; identifying the cloud scene based on the cloud rule set and cloud factors; generating a definition of the identified cloud scene; analyzing the in-vehicle scene signals; and identifying the comprehensive scene based on the analyzed in-vehicle scene signals and the definition of the cloud scene.

[0101] In box 708, a comprehensive scene signal is generated based on the identified comprehensive scene.

[0102] In block 710, a comprehensive scene signal is sent to the vehicle or the user of the vehicle so that an operation is performed in the vehicle based on the comprehensive scene signal, wherein the operation includes prompting the user.

[0103] Although not shown in the figure, it should be understood that any other operations described herein can be used to add, remove or modify steps in method 700.

[0104] Figure 8 shows a schematic diagram of a distributed scene engine for a vehicle at an in-vehicle scene engine according to an embodiment of the present invention. The device 800 shown in Figure 8 can be implemented using software, hardware, or a combination of both.

[0105] As shown in Figure 8, the device 800 may include an acquisition module 802, a judgment module 804, a generation module 806, and a sending module 808.

[0106] The acquisition module 802 can be used to obtain the context information of the vehicle.

[0107] The judgment module 804 can be used to determine whether the current scene of the vehicle is a comprehensive scene or a pure in-vehicle scene based on context information and a predefined rule set. The comprehensive scene includes a combination of in-vehicle scene and cloud scene.

[0108] The generation module 806 can be used to generate in-vehicle scene signals based on context information for in-vehicle scenes in a comprehensive scene when the current scene is determined to be a comprehensive scene.

[0109] The sending module 808 can be used to send the in-vehicle scene signal to the cloud so that the cloud can identify the comprehensive scene based at least on the in-vehicle scene signal and the corresponding context information, and generate a comprehensive scene signal based on the identified comprehensive scene.

[0110] Optionally, device 800 may further include a classification module for classifying each factor in the context information as either a vehicle-mounted factor or a cloud-based factor. In one example, the judgment module 804 is further configured to: determine the current scene as a purely in-vehicle scene when all factors in the context information are vehicle-mounted factors, based on a predetermined set of rules; and determine the current scene as a combined scene when the context information includes a combination of vehicle-mounted and cloud-based factors. Although not shown in the figures, it should be understood that modules in device 800 can also be added, removed, or modified using modules for performing any other operations described herein.

[0111] Figure 9 shows a schematic diagram of a distributed scene engine for vehicles at a cloud-based scene engine according to an embodiment of the present invention. The device 900 shown in Figure 9 can be implemented using software, hardware, or a combination of both.

[0112] As shown in Figure 9, the device 900 may include an acquisition module 902, a receiving module 904, an identification module 906, a generation module 908, and a sending module 910.

[0113] In one embodiment, the acquisition module 902 can be used to acquire external information of the vehicle. The receiving module 904 can be used to receive in-vehicle scene signals from the vehicle. The identification module 906 can be used to identify a comprehensive scene based on the external information and the in-vehicle scene signals according to a predetermined cloud rule set. The generation module 908 can be used to generate a comprehensive scene signal based on the identified comprehensive scene. The sending module 910 can be used to send the comprehensive scene signal to the vehicle. In one example, the identification module 906 is also used to identify a cloud scene based on the cloud rule set and the external information.

[0114] Optionally, the device 900 may further include: a classification module for classifying each factor in the external information as a vehicle-mounted factor or a cloud factor; another generation module for generating a definition of the identified cloud scene; and an analysis module for analyzing the in-vehicle scene signals. In one example, the identification module 906 further identifies the cloud scene based on a cloud rule set and cloud factors, and identifies a combined scene based on the identified in-vehicle scene signals and the definition of the cloud scene. Although not shown in the figures, it should be understood that modules in the device 900 can be added, removed, or modified using modules for performing any other operations described herein.

[0115] Figure 10 shows a schematic diagram of a device 1000 for a distributed scene engine for a vehicle according to an embodiment of the present invention.

[0116] As shown in FIG10, the device 1000 may include a processor 1002 and a memory 1004, wherein the memory 1004 is used to store executable instructions, which, when executed, cause the processor 1002 to perform method 400 shown in FIG4 and / or method 500 shown in FIG5 and / or method 600 shown in FIG6 and / or method 700 shown in FIG7.

[0117] This invention also provides a machine-readable medium storing executable instructions that, when executed, cause a machine to perform method 400 shown in FIG4 and / or method 500 shown in FIG5 and / or method 600 shown in FIG6 and / or method 700 shown in FIG7.

[0118] It should be understood that all operations in the methods described above are merely exemplary, and this disclosure is not limited to any operation in the methods or the order of such operations, but should cover all other equivalent variations under the same or similar concept. It should also be understood that all modules in the apparatus described above can be implemented in various ways. These modules can be implemented as hardware, software, or a combination thereof. Furthermore, any of these modules can be further functionally divided into submodules or combined together.

[0119] Processors have been described in conjunction with various devices and methods. These processors can be implemented using electronic hardware, computer software, or any combination thereof. Whether these processors are implemented as hardware or software will depend on the specific application and the overall design constraints imposed on the system. As an example, the processors, any portions of processors, or any combinations of processors given in this disclosure can be implemented as microprocessors, microcontrollers, digital signal processors (DSPs), field-programmable gate arrays (FPGAs), programmable logic devices (PLDs), state machines, gate logic, discrete hardware circuits, and other suitable processing units configured to perform the various functions described in this disclosure. The functionality of the processors, any portions of processors, or any combinations of processors given in this disclosure can be implemented as software executed by a microprocessor, microcontroller, DSP, or other suitable platform.

[0120] Those skilled in the art should understand that the various embodiments disclosed above can be modified and varied in various ways without departing from the spirit of the invention. All such modifications and variations should fall within the protection scope of the present invention, and the protection scope of the present invention should be defined by the claims.

Claims

1. A method for a distributed scene engine for a vehicle, the method being executed at an in-vehicle scene engine, comprising: Obtain the vehicle's context information; Based on the context information and the predetermined rule set, it is determined whether the current scene of the vehicle is a comprehensive scene or a pure in-vehicle scene. The comprehensive scene includes a combination of in-vehicle scenes and cloud scenes. When the current scene is determined to be the comprehensive scene, the following operations are performed for the current scene: generating an in-vehicle scene signal for the in-vehicle scene in the comprehensive scene based on the context information; and sending the in-vehicle scene signal to the cloud server. The system receives a comprehensive scene signal from the cloud server. The comprehensive scene signal is generated at the cloud server based on the identified comprehensive scene. The identified comprehensive scene is identified at least based on the in-vehicle scene signal and the corresponding context information. And to perform operations based on the comprehensive scene signals, when the current scene is determined to be a pure in-vehicle scene, to perform in-vehicle operations based at least on the context information and the predetermined in-vehicle rule set.

2. The method of claim 1, further comprising: Each factor in the context information is classified as a vehicle-mounted factor or a cloud-based factor; wherein, determining whether the current scenario is a comprehensive scenario or a purely in-vehicle scenario further includes: according to the predetermined rule set, when all factors in the context information are vehicle-mounted factors, the current scenario is determined to be a purely in-vehicle scenario; and when the context information includes a combination of vehicle-mounted factors and cloud-based factors, the current scenario is determined to be a comprehensive scenario.

3. The method as described in claim 2, wherein, Generating the in-vehicle scene signal further includes: identifying the in-vehicle scene based on the vehicle-mounted factors and a predetermined in-vehicle rule set; and generating the in-vehicle scene signal based on the definition of the identified in-vehicle scene.

4. The method of claim 1, wherein, The operation includes: issuing prompts to the user of the vehicle and / or controlling the components of the vehicle to automatically perform actions.

5. The method of claim 1, wherein, The context information includes in-vehicle information and out-of-vehicle information, and obtaining the vehicle's context information further includes: obtaining the in-vehicle information from the vehicle's sensors and receiving the out-of-vehicle information from the cloud server.

6. The method of claim 1, further comprising: Determine specific rules in the in-vehicle rule set that match the context information; identify the pure in-vehicle scene based on the specific rules; And to perform in-vehicle operations based on the pure in-vehicle scenario, wherein the in-vehicle operations include issuing prompts to the user of the vehicle and / or controlling the automatic execution of actions by the components of the vehicle.

7. A method for a distributed scene engine for vehicles, comprising: At the in-vehicle scene engine, obtain the vehicle's context information; Based on the context information and the predetermined rule set, it is determined whether the current scene of the vehicle is a comprehensive scene or a pure in-vehicle scene. The comprehensive scene includes a combination of in-vehicle scenes and cloud scenes. When the current scene is determined to be the comprehensive scene, the following operations are performed for the current scene: generating an in-vehicle scene signal for the in-vehicle scene in the comprehensive scene based on the context information; and sending the in-vehicle scene signal to the cloud server. The system receives a comprehensive scene signal from the cloud server. The comprehensive scene signal is generated at the cloud server based on the identified comprehensive scene. The identified comprehensive scene is identified at least based on the in-vehicle scene signal and the corresponding context information. And to perform operations based on the comprehensive scene signals, when the current scene is determined to be a pure in-vehicle scene, to perform in-vehicle operations at least based on the context information and the predetermined in-vehicle rule set at the cloud scene engine of the cloud server, and to obtain the vehicle's external information. Based on a predetermined cloud rule set, a comprehensive scene is identified using the external vehicle information and the in-vehicle scene signals received from the vehicle. Generate a comprehensive scene signal based on the identified comprehensive scene; And send the integrated scene signal to the vehicle.

8. The method of claim 7, wherein, The step of identifying the comprehensive scene based on the external vehicle information and the in-vehicle scene signals received from the vehicle further includes: classifying each factor in the external vehicle information as a vehicle-mounted factor or a cloud-based factor; identifying a cloud-based scene based on the cloud-based rule set and the cloud-based factors; generating a definition of the identified cloud-based scene; analyzing the in-vehicle scene signals; and identifying the comprehensive scene based on the analyzed in-vehicle scene signals and the definition of the cloud-based scene.

9. An apparatus for a distributed scene engine for a vehicle, the apparatus being located at an in-vehicle scene engine, comprising: The acquisition module is used to obtain the vehicle's context information; The judgment module is used to determine whether the current scene of the vehicle is a comprehensive scene or a pure in-vehicle scene based on the context information and a predetermined rule set. The comprehensive scene includes a combination of in-vehicle scene and cloud scene. The generation module is used to generate an in-vehicle scene signal based on the context information for the in-vehicle scene in the comprehensive scene when the current scene is determined to be the comprehensive scene. The device also includes a sending module for sending the in-vehicle scene signal to a cloud server. The cloud server identifies a comprehensive scene based at least on the in-vehicle scene signal and corresponding context information, and generates a comprehensive scene signal based on the identified comprehensive scene. The device is further configured to receive the comprehensive scene signal from the cloud server and perform operations based on the comprehensive scene signal, and when the current scene is determined to be a pure in-vehicle scene, perform in-vehicle operations based at least on the context information and a predetermined in-vehicle rule set.

10. The apparatus of claim 9, further comprising: The classification module is used to classify each factor in the context information as either a vehicle-mounted factor or a cloud-based factor; wherein, the judgment module is further used to: determine the current scene as a pure in-vehicle scene when all factors in the context information are vehicle-mounted factors, according to the predetermined rule set; and determine the current scene as a combined scene when the context information includes a combination of vehicle-mounted factors and cloud-based factors.

11. A system for a distributed scene engine for vehicles, comprising: The device at the engine location in an in-vehicle scene as described in claim 9 or 10; And an apparatus for a distributed scene engine for a vehicle at a cloud engine, the apparatus at the cloud engine comprising: an acquisition module for acquiring external information of the vehicle; a receiving module for receiving an in-vehicle scene signal from the vehicle, wherein the in-vehicle scene signal is generated based on the vehicle's context information for an in-vehicle scene within a comprehensive scene when the current scene category of the vehicle is determined to be a comprehensive scene, the comprehensive scene including a combination of an in-vehicle scene and a cloud scene; an identification module for identifying a comprehensive scene based on the external information and the in-vehicle scene signal according to a predetermined cloud rule set; a generation module for generating a comprehensive scene signal based on the identified comprehensive scene; and a sending module for sending the comprehensive scene signal to the vehicle.

12. The system of claim 11, wherein, The identification module is also used to identify cloud scenes based on the cloud rule set and the external vehicle information. The device at the cloud engine further includes: a classification module for classifying each factor in the external vehicle information into vehicle-mounted factors or cloud factors; another generation module for generating a definition of the identified cloud scene; and an analysis module for analyzing the in-vehicle scene signal. The identification module further identifies cloud scenes based on the cloud rule set and the cloud factors, and identifies the comprehensive scene based on the identified in-vehicle scene signal and the definition of the cloud scene.

13. A device for a distributed scene engine for vehicles, comprising: processor; And a memory for storing executable instructions, wherein, when executed, the executable instructions cause the processor to perform the method of any one of claims 1-6 and / or the method of any one of claims 7-8.

14. A machine-readable medium having executable instructions stored thereon, wherein, When executed, the executable instructions cause the machine to perform the method of any one of claims 1-6 and / or the method of any one of claims 7-8.

Images