Intelligent cabin digital twinning recommendation system, method and equipment based on block chain

Abstract

The invention relates to the field of block chains, in particular to an intelligent cabin digital twin recommendation system, method and device based on a block chain, and the method comprises the steps: obtaining data collected by an OBD sensor, a camera, an environment sensor and a biosensor; carrying out fuzzy processing on the collected data through edge calculation; performing feature extraction on the data after fuzzy processing, and generating a six-dimensional feature vector based on a context awareness recommendation algorithm; according to the scene label distribution weight of the six-dimensional feature vector, determining a recommendation priority; obtaining corresponding twinborn decision data according to the priority score; calculating a feature hash value of the twinborn decision data by adopting an SHA-3 algorithm, and generating a data fingerprint; dividing the data fingerprints into vehicle condition data and user data according to data types, and performing cross-chain storage; and performing credibility evaluation and screening on the multi-source data stored in a cross-chain manner, and feeding back the data to a user side. According to the scheme, a data island is broken, the security protection of privacy is enhanced, and the credibility and transparency of information recommendation are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of blockchain, in particular to a smart cockpit digital twin recommendation system based on blockchain, a method and equipment. BACKGROUND

[0002] In the field of smart cockpit technology, with the rapid development of vehicle-mounted sensors, user interaction systems and environmental perception devices, traditional cockpit data processing and service recommendation modes gradually expose many problems.

[0003] First, the data island phenomenon is significant. Vehicle-mounted sensors, user behavior data and environmental information are often stored in different systems, forming fragmented data pools, making it difficult to efficiently integrate multi-source data and unable to provide coherent service support for users.

[0004] Second, centralized data processing has privacy leakage risks. User profiles, driving habits and other sensitive data are stored in a single platform, which is easy to become an attack target, and data control is concentrated in service providers, making it difficult for users to protect their personal information security.

[0005] Third, the transparency of the recommendation algorithm is insufficient. Existing recommendation systems mostly use black box models, and users cannot understand the recommendation logic, which can easily lead to a trust crisis and reduce user acceptance when the recommended results do not match the needs.

[0006] Finally, the dynamic scene adaptability is poor. Traditional systems are difficult to respond in real time to changes in vehicle status, road conditions and environment, for example, when the vehicle's power supply drops sharply or enters an unfamiliar area, the recommendation strategy cannot be adjusted in time, resulting in insufficient service timeliness.

[0007] These problems undoubtedly restrict the intelligent upgrade of the smart cockpit, so there is an urgent need for a solution that can achieve data fusion, privacy protection, transparent recommendation and dynamic adaptation. SUMMARY

[0008] To solve the above problems, the present application provides a smart cockpit digital twin recommendation system based on blockchain, a method and equipment to solve the problem of cockpit data processing and service recommendation in the prior art, which is significant data island phenomenon, privacy leakage risk of centralized data processing, insufficient transparency of recommendation algorithm and poor dynamic scene adaptability.

[0009] To achieve the above purposes, the present application adopts the following technical solutions:

[0010] In a first aspect, the present application provides a smart cockpit digital twin recommendation system based on blockchain, which comprises a three-dimensional data fusion architecture; the three-dimensional data fusion architecture is composed of a physical entity layer, a virtual model layer and a blockchain layer;

[0011] The physical entity layer includes a synchronous acquisition module connected with an OBD sensor, a camera, an environment sensor and a biological sensor; the synchronous acquisition module is configured to acquire data collected by the OBD sensor, the camera, the environment sensor and the biological sensor; and further includes an edge preprocessing module configured to perform fuzzy processing on the collected data through edge computing.

[0012] The virtual model layer includes a digital twin modeling module configured to perform feature extraction on the fuzzy-processed data, and generate a six-dimensional feature vector based on a context-aware recommendation algorithm; assign weights according to scene labels of the six-dimensional feature vector, and determine a recommendation priority; and obtain corresponding twin decision data according to the priority score.

[0013] The blockchain layer includes a blockchain cross-chain routing module configured to calculate a feature hash value of the twin decision data using an SHA-3 algorithm, and generate a data fingerprint; and divide the data fingerprint into vehicle condition data and user data according to data types and store the data across chains.

[0014] An intelligent recommendation service module is configured to perform credibility evaluation and screening on the cross-chain stored multi-source data, and feed back to a user end.

[0015] Optionally, the synchronous acquisition module includes:

[0016] A first data reading unit is configured to read vehicle parameters collected by an on-board OBD through a CAN bus; wherein the vehicle parameters include vehicle speed, tire pressure and battery temperature; and a second data reading unit is configured to read temperature and humidity and PM2.5 data recorded by an environment sensor through an I 2 C bus.

[0017] A first data receiving unit is configured to receive a video stream acquired by a camera; the video stream includes a driver's face image.

[0018] A second data receiving unit is configured to receive physiological signals collected by a biological sensor; wherein the physiological signals include heart rate / respiratory rate data.

[0019] Optionally, the edge preprocessing module includes:

[0020] A first processing unit is configured to smooth sudden jumps and jitter of the vehicle parameters based on Kalman filtering; a second processing unit is configured to perform Gaussian blur processing on the collected driver's face image through a MobileNetV3 model deployed by an edge node

[0021] to output an emotional label in real time; and a third processing unit is configured to filter abnormal values in the temperature and humidity and PM2.5 data and the physiological signals.

[0022] Optionally, the digital twin modeling module includes:

[0023] a feature space construction unit configured to determine user preferences based on historical database statistics of user stop selection ratios, and to call a high

[0024] a precise map API configured to obtain lane-level coordinates and a real-time vehicle position, to analyze a vehicle state by calculating a vehicle health index, and to construct a six-dimensional feature vector based on user preferences, the real-time vehicle position, and the vehicle state;

[0025] The vehicle health index is calculated by the following formula:

[0026] The vehicle health index = (current tire pressure / standard tire pressure) x 80 + (battery health / 100) x 20. Optionally, the digital twin modeling module further comprises a context classification unit and a weight allocation unit.

[0027] The context classification unit is configured to input the six-dimensional feature vector into an SVM classifier and output a scenario label.

[0028] The weight allocation unit is configured to call a dynamic weight function according to the scenario label, to determine a recommended priority based on a weight allocation strategy of the scenario label dynamic weight function.

[0029] The weight allocation strategy defines a weight vector and determines a recommended priority.

[0030] The weight allocation strategy based on the scenario label dynamic weight function is as follows:

[0031] In a first scenario, the vehicle state weight is greater than or equal to a first percentage.

[0032] In a second scenario, the real-time position weight is greater than or equal to a second percentage. In a third scenario, the user preference weight is greater than or equal to a third percentage.

[0033] Optionally, the blockchain cross-link routing module comprises:

[0034] A data fingerprint generation unit is configured to calculate a feature hash of twin decision data using an SHA-3 algorithm: hash = SHA3_256(VIN + timestamp + score). In the formula, Vin is the input of the hash function SHA-256, score represents the score of each data node, and timestamp is the time stamp.

[0035] A cross-chain storage decision unit is configured to identify vehicle condition data and user data by judging whether the vehicle safety parameters are included in the feature hash.

[0036] The vehicle condition data is stored in the VeChain chain, written into the Fabric channel, and the smart contract is triggered.

[0037] The smart contract includes: automatically pushing a charging station reservation service when the vehicle enters a preset range and the tire pressure is less than a preset threshold; storing user data to the IOTA node, generating an IOTA transaction and settling it.

[0038] Optionally, the intelligent recommendation service module includes:

[0039] The data credibility assessment unit is used to assess the credibility of multi-source data using an improved PageRank mechanism: CredScore = α + β + γ*;

[0040] The filtering unit is used to filter recommended data based on predefined elimination rules and credibility assessment results.

[0041] Where CredScore represents the node credit score of multi-source data; α, β, and γ are reliability parameters, representing data freshness, source reliability, and cross-validation matching degree, respectively; reliability parameter = (number of historical recommendations adopted / total number of recommendations) × 100%;

[0042] The predefined elimination rule is as follows: when the credibility is lower than the second preset threshold, the corresponding data will not be included in the recommendation.

[0043] Optionally, the intelligent recommendation service module further includes: a content recommendation unit, used to match users' historical preference tags to filter content and obtain recommended content; a collaborative filtering unit, used to cluster and select car owners with similar characteristics from the same car model / region; a safety rule interception unit, used to filter recommended content based on predefined traffic rules; and a service execution unit, used to call third-party services through API and output the filtering results to the cockpit screen.

[0044] Secondly, the present invention also provides a blockchain-based smart cockpit digital twin recommendation method, the method comprising:

[0045] Acquire data from OBD sensors, cameras, environmental sensors, and biosensors;

[0046] The collected data is blurred using edge computing;

[0047] Feature extraction is performed on the fuzzy data, and a six-dimensional feature vector is generated based on a context-aware recommendation algorithm; weights are assigned to the scene labels of the six-dimensional feature vector to determine the recommendation priority; and corresponding twin decision data is obtained based on the priority scores.

[0048] The SHA-3 algorithm is used to calculate the feature hash value of twin decision data to generate data fingerprints; the data fingerprints are then divided into vehicle condition data and user data according to data type and stored across chains.

[0049] After the credibility of the multi-source data stored across chains is assessed and filtered, the results are fed back to the user.

[0050] Thirdly, the present invention provides an electronic device, the electronic device comprising:

[0051] At least one processor; and,

[0052] A memory communicatively connected to the at least one processor; wherein the memory stores a computer program executable by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the steps of the method described in the second aspect.

[0053] The beneficial effects of this invention are reflected in:

[0054] The aforementioned blockchain-based intelligent cockpit digital twin recommendation system, method, and device include a three-dimensional data fusion architecture. This architecture comprises a physical entity layer, a virtual model layer, and a blockchain layer. The physical entity layer includes a synchronous acquisition module connected to OBD sensors, cameras, environmental sensors, and biosensors. This synchronous acquisition module is used to acquire data collected by the OBD sensors, cameras, environmental sensors, and biosensors. The system also includes an edge preprocessing module for performing fuzzy processing on the acquired data using edge computing.

[0055] The virtual model layer includes: a digital twin modeling module, which extracts features from the fuzzy data and generates a six-dimensional feature vector based on a context-aware recommendation algorithm; assigns weights to the scene labels of the six-dimensional feature vector to determine the recommendation priority; and obtains the corresponding twin decision data based on the priority score.

[0056] The blockchain layer includes: a blockchain cross-chain routing module, used to calculate the feature hash value of twin decision data using the SHA-3 algorithm to generate data fingerprints; and to divide the data fingerprints into vehicle condition data and user data according to data type and store them across chains;

[0057] The intelligent recommendation service module is used to evaluate and filter the credibility of multi-source data stored across chains and then feed it back to the user.

[0058] The above-mentioned invention provides multi-dimensional optimizations to address the pain points of existing intelligent cockpit systems, resulting in significant beneficial effects:

[0059] First, it solves the problem of data silos. The three-dimensional data fusion architecture (physical entity layer, virtual model layer, and blockchain layer) breaks down the fragmented barriers between vehicle sensors, user behavior, and environmental data. It synchronously collects vehicle parameters, environmental data, and user characteristics through multiple sources such as CAN bus and cameras. After constructing a six-dimensional feature space (PLVEST) through digital twin modeling, it achieves efficient cross-system data collaboration and lays the data foundation for accurate service recommendations.

[0060] Secondly, it strengthens privacy and security protection. Based on the privacy computing mechanism of zero-knowledge proof, combined with edge node preprocessing, it ensures the privacy of sensitive information; by utilizing the distributed storage of blockchain technology, it avoids the risks of centralized data storage; and the data fingerprint generated by SHA-3 hash and asymmetric encryption technology further ensure data integrity and confidentiality, significantly improving users' sense of privacy and security.

[0061] Third, it enhances the credibility and transparency of recommendations. Dynamically weighted recommendation data is adaptively adjusted through initial weights and scenarios (e.g., increased weight for vehicle status in emergency scenarios), making the recommendation logic traceable; the improved PageRank credibility assessment algorithm eliminates low-quality data, and combined with multimodal interactive feedback, allows users to clearly perceive the basis for recommendations, significantly enhancing trust. Attached Figure Description

[0062] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the accompanying drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. In all the drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, the elements or parts are not necessarily drawn to scale.

[0063] Figure 1 This is a schematic diagram of the structure of the blockchain-based smart cockpit digital twin recommendation system provided by the present invention;

[0064] Figure 2 This is a flowchart of the blockchain-based smart cockpit digital twin recommendation method provided by the present invention;

[0065] Figure 3 This is a schematic diagram of the data flow process of the synchronous acquisition module provided by the present invention;

[0066] Figure 4 This is an internal structural diagram of the electronic device provided by the present invention. Detailed Implementation

[0067] The embodiments of the technical solution of the present invention will now be described in detail with reference to the accompanying drawings. These embodiments are only used to more clearly illustrate the technical solution of the present invention and are therefore merely examples, and should not be construed as limiting the scope of protection of the present invention.

[0068] It should be noted that, unless otherwise stated, the technical or scientific terms used in this application should have the ordinary meaning as understood by one of ordinary skill in the art to which this invention pertains.

[0069] In existing intelligent automotive cockpit scenarios, in-vehicle terminals need to collect terabytes of heterogeneous data daily from hundreds of sensors, cameras, and cloud services for personalized recommendations such as navigation, entertainment, air conditioning, and safety reminders. However, current mainstream solutions generally adopt a centralized architecture of "vehicle-side data collection - cloud-based centralized processing," which brings irreconcilable pain points: First, data silos are severe, with OBD, cameras, in-vehicle apps, and third-party services operating independently, lacking a unified semantic model, resulting in fragmented information that cannot be integrated and biased recommendation results; Second, all user profiles and original trajectories are centrally stored on the servers of car manufacturers or internet platforms, posing a single point of leakage risk, and several recent cases of car owner privacy leaks have raised concerns from both regulators and users; Third, recommendation algorithms operate completely in a black box, lacking auditable mechanisms, making it impossible for car owners to verify the basis and source of information pushes, leading to a continuous decline in trust; Finally, existing model weights are fixed and cannot be dynamically adjusted according to real-time vehicle conditions. When a vehicle experiences a sudden drop in tire pressure or a critical low battery, the system still prioritizes coffee shops based on historical preferences, missing crucial roadside assistance services.

[0070] To address these shortcomings, this invention provides a blockchain-based smart cockpit digital twin recommendation system, method, and device.

[0071] The embodiments of the present invention will now be described with reference to the accompanying drawings:

[0072] Please refer to Figure 1 ,like Figure 1 It is a blockchain-based smart cockpit digital twin recommendation system, which includes a three-dimensional data fusion architecture; the three-dimensional data fusion architecture consists of a physical entity layer 110, a virtual model layer 120 and a blockchain layer 130;

[0073] The physical entity layer includes: a synchronous acquisition module 15 connected to the OBD sensor 11, camera 12, environmental sensor 13, and biosensor 14; the synchronous acquisition module 15 is used to acquire data collected by the OBD sensor 11, camera 12, environmental sensor 13, and biosensor 14; it also includes: an edge preprocessing module 16, used to perform fuzzing processing on the acquired data through edge computing;

[0074] The virtual model layer 120 includes: a digital twin modeling module 21, used to extract features from the fuzzy data and generate a six-dimensional feature vector based on a context-aware recommendation algorithm; assign weights to the scene labels of the six-dimensional feature vector to determine the recommendation priority; and obtain corresponding twin decision data based on the priority score.

[0075] The blockchain layer 130 includes: a blockchain cross-chain module 31, which is used to calculate the feature hash value of twin decision data using the SHA-3 algorithm to generate data fingerprints; and to divide the data fingerprints into vehicle condition data and user data according to data type and store them across chains.

[0076] The intelligent recommendation service module 32 is used to evaluate and filter the credibility of multi-source data stored across chains and then feed it back to the user.

[0077] In the above embodiments, such as Figure 2 As shown, the synchronous acquisition module includes:

[0078] The first data reading unit is used to read vehicle parameters collected by the on-board OBD via the CAN bus; wherein, the vehicle parameters include vehicle speed, tire pressure, and battery temperature; the second data reading unit is used to read vehicle parameters collected by the on-board OBD via I 2 The C-bus reads temperature, humidity, and PM2.5 data recorded by environmental sensors;

[0079] The first data receiving unit is used to receive a video stream acquired by the camera; the video stream includes a driver's facial image.

[0080] The second data receiving unit is used to receive physiological signals collected by the biosensor; wherein the physiological signals include heart rate / respiratory rate data.

[0081] In one embodiment, the synchronous acquisition module acquires data synchronously (100ms / time):

[0082] 1. Read 20+ vehicle parameters (vehicle speed, tire pressure, battery temperature, etc.) via CAN bus.

[0083] 2. The camera captures the driver's facial image → the edge nodes use the MobileNetV3 model to output emotion labels (fatigue / distraction) in real time;

[0084] 3. Biosensor collects heart rate / respiratory rate → filters out abnormal values ​​(±3σ principle);

[0085] In the above embodiments, the edge preprocessing module includes:

[0086] The first processing unit is used to smooth sudden jumps in vehicle parameters (such as sudden speed jumps) based on Kalman filtering; the second processing unit is used to process the acquired driver facial image through a MobileNetV3 model deployed at edge nodes.

[0087] Gaussian blurring erases privacy and outputs emotion tags in real time; that is, after Gaussian blurring, facial images retain only emotion tags. The third processing unit is used to filter out abnormal values ​​in temperature, humidity, PM2.5 data, and physiological signals.

[0088] In the above embodiments, the context-aware recommendation algorithm in the digital twin modeling module constructs a six-dimensional feature space including: user preferences (P), real-time location (L), vehicle status (V), environmental factors (E), service inventory (S), and traffic rules (T), resulting in a recommendation weight matrix W = [0.3P, 0.2L, 0.15V, 0.15E, 0.1S, 0.1T];

[0089] Specifically, the six-dimensional feature space is constructed as follows:

[0090] User preferences (P): Historical behavior clustering (collaborative filtering + time decay factor);

[0091] Real-time location (L): High-precision map matching (lane-level positioning, error <0.5m0);

[0092] Vehicle Status (V): OBD data is mapped to a health index (0-100 points);

[0093] Environmental factors (E): PM2.5 classification triggers air purification, etc.;

[0094] Service Inventory (S): Real-time query of charging station / parking space availability via cloud service API;

[0095] Traffic Rules (T): Traffic Regulations Knowledge Graph (Restricted / No-Entry Zone Rule Engine);

[0096] The weight matrix is ​​dynamically adjusted: the initial weights W=[0.3P,0.2L,0.15V,0.15E,0.1S,0.1T] are adaptively adjusted according to changes in the situation.

[0097] In the above embodiments, the digital twin modeling module includes:

[0098] Feature space construction unit, used to determine user preferences by statistically analyzing the proportion of user docking choices based on historical databases; calls high...

[0099] The fine map API obtains lane-level coordinates to get the real-time location of vehicles; it analyzes vehicle status by calculating vehicle health index; and it constructs a six-dimensional feature vector based on user preferences, real-time vehicle location, and vehicle status.

[0100] The vehicle health index is calculated using the following formula:

[0101] Vehicle health index = (current tire pressure / standard tire pressure) × 80 + (battery health / 100) × 20. Furthermore, the digital twin modeling module also includes a context classification unit and a weight allocation unit;

[0102] The context classification unit is used to input the six-dimensional feature vector into the SVM classifier and output the scene label;

[0103] The weight allocation unit is used to call a dynamic weight function based on the scene label, and the dynamic weight function based on the scene label...

[0104] The weight allocation strategy defines a weight vector and calls the dynamic weight function: weight_adapt("EMERGENCY") → [0.1P, 0.3L, 0.4V, ...] to determine the recommendation priority;

[0105] The weight allocation strategy based on the scene label dynamic weight function is as follows:

[0106] In the first scenario, the vehicle state weight is greater than or equal to the first percentage.

[0107] In the second scenario, the weight of real-time location is greater than or equal to the second percentage; in the third scenario, the weight of user preference is greater than or equal to the third percentage.

[0108] like Figure 2 As shown in the above embodiment, the blockchain cross-link module combines the distributed storage function of blockchain technology; specifically, the blockchain is a Hyperledger Fabric consortium blockchain structure; including:

[0109] The data fingerprint generation unit is used to calculate the feature hash of the twin decision data using the SHA-3 algorithm.

[0110] hash = SHA3_256(VIN + timestamp + score); where Vin is the input to the hash function SHA-256, and score is the input to the hash function SHA-256.

[0111] This represents the score for each data node, with timestamp being the timestamp.

[0112] The cross-chain storage decision unit is used to identify vehicle condition data by determining whether the feature hash contains vehicle safety parameters.

[0113] User data and vehicle condition data are stored on the VeChain chain and written into the Fabric channel, triggering a smart contract.

[0114] The smart contract triggering mechanism enhances the ability to adapt to dynamic scenarios; it can respond to changes in vehicle condition, location, and environment in milliseconds, and quickly adjust recommendation strategies in scenarios such as congested navigation and emergency vehicle situations, with service timeliness and accuracy significantly superior to traditional systems.

[0115] The smart contract triggering mechanism includes: automatically pushing when the vehicle enters a preset range and the tire pressure is less than a preset threshold.

[0116] Offers charging station reservation service; stores user data in IOTA nodes, generates IOTA transactions, and settles them.

[0117] Specifically, when generating fingerprint data, the SHA-3 algorithm is used to generate a feature hash, which is then stored across chains: vehicle condition data is stored to...

[0118] VeChain chain stores entertainment preference data on the IOTA chain. Example of smart contract triggering conditions: When a vehicle enters the scenic area and the remaining battery power is >50%, a charging station reservation service is automatically pushed.

[0119] Optionally, if the data contains vehicle safety parameters (such as abnormal tire pressure), it is routed to the VeChain chain.

[0120] Write to Fabric channel (Organization: Automakers + Insurance Companies)

[0121] Trigger the smart contract: Notify the 4S store when the tire pressure is <1.8 bar;

[0122] ELSE IF Entertainment data (such as recommended restaurants) → Route to IOTA Tangle

[0123] Generate IOTA transactions (zero transaction fees) and settle coupon points with the merchant system.

[0124] In the above embodiments, the intelligent recommendation service module includes:

[0125] The data credibility assessment unit is used to evaluate the credibility of multi-source data using an improved PageRank mechanism: CredScore = α (data freshness) + β (source reliability) + γ* (cross-validation matching degree).

[0126] The filtering unit is used to filter recommended data based on predefined elimination rules and credibility assessment results.

[0127] Where CredScore represents the node credit score of multi-source data; α, β, and γ are all reliability parameters; reliability parameter = (number of historical recommendations adopted / total number of recommendations) × 100%;

[0128] The predefined elimination rule is as follows: when the credibility is lower than the second preset threshold, the corresponding data will not be included in the recommendation.

[0129] In addition, the intelligent recommendation service module also includes: a content recommendation unit, used to match users' historical preference tags to filter content and obtain recommended content; a collaborative filtering unit, used to cluster and select car owners with similar characteristics from the same car model / region; a safety rule interception unit, used to filter recommended content based on predefined traffic rules; and a service execution unit, used to call third-party services through API and output the filtering results to the cockpit screen.

[0130] In one embodiment, 1. Credibility assessment (removing low-credit data): Calculate CredScore = 0.4*e^(-0.1·Δt) + 0.4*Reliability + 0.2*Consistency

[0131] Elimination rule: Data with a CredScore < 0.8 will not be included in the recommendations.

[0132] 2. Hybrid recommendation strategy:

[0133] Content recommendation (80% of scenarios): Matching users' historical preference tags (such as "coffee shop")

[0134] Collaborative filtering (20% of scenarios): Clustering similar car owner selections (same car model / same region)

[0135] 3. Security and compliance interception:

[0136] Traffic rule filter: IF Recommended content is video AND vehicle speed > 30km / h THEN Block

[0137] 4. Service Execution:

[0138] Calling third-party services (such as charging station reservations) via RESTful API

[0139] The results are output to the cockpit screen (voice + graphical interface).

[0140] Based on the same inventive concept, this application also provides a blockchain-based smart cockpit digital twin recommendation system and a blockchain-based smart cockpit digital twin recommendation method. The solution provided by this method is similar to the implementation scheme described in the above-described embodiments. Therefore, the specific limitations of one or more blockchain-based smart cockpit digital twin recommendation method embodiments provided below can be found in the limitations of the blockchain-based smart cockpit digital twin recommendation system described above, and will not be repeated here.

[0141] Please refer to Figure 4, Figure 4 This invention provides a blockchain-based digital twin recommendation method for smart cockpits, the method specifically including the following steps:

[0142] S101 acquires data collected by OBD sensors, cameras, environmental sensors, and biosensors;

[0143] S102 performs fuzzy processing on the collected data through edge computing;

[0144] S103 extracts features from the fuzzy data and generates a six-dimensional feature vector based on a context-aware recommendation algorithm; it assigns weights to the scene labels of the six-dimensional feature vector to determine the recommendation priority; and it obtains the corresponding twin decision data based on the priority score.

[0145] S104 uses the SHA-3 algorithm to calculate the feature hash value of twin decision data and generate data fingerprints; the data fingerprints are divided into vehicle condition data and user data according to data type and stored across chains;

[0146] S105 performs credibility assessment and screening on the multi-source data stored across chains, and then feeds the results back to the user.

[0147] Example 1: This example further combines real-world scenarios with a blockchain-based smart cockpit digital twin recommendation method to optimize cross-domain navigation during peak hours:

[0148] 1. Vehicle sensors detect congestion and encrypt the data before uploading it to the blockchain;

[0149] (1) Real-time data collection by vehicle-mounted sensors (GPS, radar, camera):

[0150] ① GPS: Vehicle speed < 20km / h for 5 minutes

[0151] ②Radar: Vehicle spacing < 5 meters

[0152] ③ Camera: AI identifies lane occupancy rate > 80%

[0153] ④ Determining congestion based on a sliding window

[0154] (2) Data encryption and on-chain:

[0155] Use asymmetric encryption (RSA-2048):

[0156] 1) Vehicle data is signed with a private key (tamper-proof)

[0157] 2) Encrypt data using blockchain public keys (to prevent theft)

[0158] 3) Upload to the blockchain (e.g., Hyperledger Fabric)

[0159] 2. The digital twin model uses data from roadside cameras to verify the authenticity of traffic congestion.

[0160] (1) Digital twin model verifies the authenticity of congestion

[0161] Blockchain-triggered verification: The smart contract listens for new blocks and automatically calls the digital twin platform API after detecting a congestion event;

[0162] (2) Verification of multi-source data fusion

[0163] Digital twin model:

[0164] 1) Access roadside cameras (5G real-time video transmission)

[0165] 2) Access the meteorological bureau's API (excluding weather influences)

[0166] 3) Obtain data from the traffic control center's loop detectors.

[0167] 3. The smart contract triggers the navigation system to replan the route and synchronize it with surrounding vehicles:

[0168] Triggered when contract conditions are met (when the digital twin returns "confirmed").

[0169] Dynamic route planning algorithm: Graph Neural Network (GNN) optimization algorithm, input: real-time traffic map, output: K optimal routes.

[0170] 4. The user's cabin automatically adjusts to "Efficient Driving Mode"; entertainment functions are reduced, and road condition prompts are enhanced.

[0171] Mode switching trigger condition: When the vehicle receives a new route, the CAN bus sends a command to trigger mode switching;

[0172] Adaptive cockpit control (reduce music, enhance traffic alerts, activate seat support, and prioritize voice commands for traffic > entertainment > notifications).

[0173] In one embodiment, the present invention also provides an electronic device, which may be a terminal, and its internal structure diagram may be as follows. Figure 4As shown. The electronic device includes a processor, memory, communication interface, display screen, and input device connected via a system bus. The processor provides computing and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, NFC (Near Field Communication), or other technologies. When the computer program is executed by the processor, it implements the blockchain-based smart cockpit digital twin recommendation method described in any of steps S101 to S105. The display screen can be a liquid crystal display (LCD) or an e-ink display. The input device can be a touch layer covering the display screen, buttons, a trackball, or a touchpad mounted on the device's casing, or an external keyboard, touchpad, or mouse.

[0174] Those skilled in the art will understand that Figure 4 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0175] In one embodiment, the present invention also provides a computer-readable storage medium storing a computer program thereon, characterized in that the computer program, when executed by a processor, implements the steps of any one of steps S101 to S105 described above.

[0176] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0177] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0178] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0179] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0180] The above are merely embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention are included within the scope of the claims of the present invention pending approval.

Claims

1. A blockchain-based intelligent cockpit digital twin recommendation system, characterized in that, The system includes a three-dimensional data fusion architecture; The three-dimensional data fusion architecture consists of a physical entity layer, a virtual model layer, and a blockchain layer. The physical entity layer includes: a synchronous acquisition module connected to the OBD sensor, camera, environmental sensor, and biosensor; the synchronous acquisition module is used to acquire data acquired by the OBD sensor, camera, environmental sensor, and biosensor; it also includes: an edge preprocessing module, used to perform fuzzing processing on the acquired data through edge computing; The virtual model layer includes: a digital twin modeling module, used to extract features from the fuzzy data and generate a six-dimensional feature vector based on a context-aware recommendation algorithm; assign weights to the scene labels of the six-dimensional feature vector to determine the recommendation priority; and obtain corresponding twin decision data based on the priority score. The blockchain layer includes: a blockchain cross-chain routing module, used to calculate the feature hash value of twin decision data using the SHA-3 algorithm to generate data fingerprints; and to divide the data fingerprints into vehicle condition data and user data according to data type and store them across chains. The intelligent recommendation service module is used to evaluate and filter the credibility of multi-source data stored across chains and then feed it back to the user.

2. The system according to claim 1, characterized in that, The synchronous acquisition module includes: The first data reading unit is used to read vehicle parameters collected by the on-board OBD via the CAN bus; wherein, the vehicle parameters include vehicle speed, tire pressure, and battery temperature; The second data reading unit is used to read data via I. 2 The C-bus reads temperature, humidity, and PM2.5 data recorded by environmental sensors; The first data receiving unit is used to receive a video stream acquired by the camera; the video stream includes a driver's facial image. The second data receiving unit is used to receive physiological signals collected by the biosensor; wherein the physiological signals include heart rate / respiratory rate data.

3. The system according to claim 1, characterized in that, The edge preprocessing module includes: The first processing unit is used to smooth sudden jitters in vehicle parameters based on Kalman filtering; The second processing unit is used to perform Gaussian blur processing on the collected driver facial images through the MobileNetV3 model deployed at the edge nodes, and output emotion labels in real time. The third processing unit is used to filter out abnormal values ​​in temperature, humidity, PM2.5 data, and physiological signals.

4. The system according to claim 1, characterized in that, The digital twin modeling module includes: The feature space construction unit is used to determine user preferences by statistically analyzing the proportion of user parking choices based on historical databases; to obtain lane-level coordinates by calling the high-precision map API to obtain the real-time vehicle location; to analyze vehicle status by calculating the vehicle health index; and to construct a six-dimensional feature vector based on user preferences, real-time vehicle location, and vehicle status. The vehicle health index is calculated using the following formula: Vehicle health index = (current tire pressure / standard tire pressure) × 80 + (battery health / 100) × 20.

5. The system according to claim 4, characterized in that, The digital twin modeling module also includes a context classification unit and a weight allocation unit; The context classification unit is used to input the six-dimensional feature vector into the SVM classifier and output the scene label; The weight allocation unit is used to call a dynamic weight function based on scene tags, define a weight vector based on the weight allocation strategy of the scene tag dynamic weight function, and determine the recommendation priority. The weight allocation strategy based on the scene label dynamic weight function is as follows: In the first scenario, the vehicle state weight is greater than or equal to the first percentage. In the second scenario, the weight of real-time location is greater than or equal to the second percentage; in the third scenario, the weight of user preference is greater than or equal to the third percentage.

6. The system according to claim 1, characterized in that, The blockchain cross-link module includes: The data fingerprint generation unit is used to calculate the feature hash of the twin decision data using the SHA-3 algorithm: hash = SHA3_256(VIN + timestamp + score); where Vin is the input of the hash function SHA-256, score represents the score of each data node, and timestamp is the timestamp; The cross-chain storage decision unit is used to identify vehicle condition data and user data by determining whether the feature hash contains vehicle safety parameters; it stores the vehicle condition data in the VeChain chain, writes it into the Fabric channel, and triggers the smart contract. The smart contract includes: automatically pushing a charging station reservation service when the vehicle enters a preset range and the tire pressure is less than a preset threshold; storing user data to the IOTA node, generating an IOTA transaction and settling it.

7. The system according to claim 1, characterized in that, The intelligent recommendation service module includes: The data credibility assessment unit is used to assess the credibility of multi-source data using an improved PageRank mechanism: CredScore = α + β + γ*; The filtering unit is used to filter recommended data based on predefined elimination rules and credibility assessment results. Where CredScore represents the node credit score of multi-source data; α, β, and γ are reliability parameters, representing data freshness, source reliability, and cross-validation matching degree, respectively; reliability parameter = (number of historical recommendations adopted / total number of recommendations) × 100%; The predefined elimination rule is as follows: when the credibility is lower than the second preset threshold, the corresponding data will not be included in the recommendation.

8. The system according to claim 7, characterized in that, The intelligent recommendation service module also includes: The content recommendation unit is used to match users' historical preference tags to filter content and obtain recommended content. The collaborative filtering unit is used to cluster and select car owners with similar characteristics from the same car model or region; The safety rule interception unit is used to filter recommended content based on predefined traffic rules; The service execution unit is used to call third-party services via API and output the filtering results to the cockpit screen.

9. A blockchain-based digital twin recommendation method for smart cockpits, characterized in that, The method includes: Acquire data from OBD sensors, cameras, environmental sensors, and biosensors; The collected data is blurred using edge computing; Feature extraction is performed on the fuzzy data, and a six-dimensional feature vector is generated based on a context-aware recommendation algorithm; weights are assigned to the scene labels of the six-dimensional feature vector to determine the recommendation priority; and corresponding twin decision data is obtained based on the priority scores. The SHA-3 algorithm is used to calculate the feature hash value of twin decision data to generate data fingerprints; the data fingerprints are then divided into vehicle condition data and user data according to data type and stored across chains. After the credibility of the multi-source data stored across chains is assessed and filtered, the results are fed back to the user.

10. An electronic device, characterized in that, The electronic device includes: At least one processor; and, A memory communicatively connected to the at least one processor; wherein the memory stores a computer program executable by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the steps of the method of claim 9.

Images