A vehicle identification method without vehicle identity

Abstract

The present application relates to a kind of non-vehicle identity vehicle identification method, without needing vehicle identity information, with driver identity information in combination with vehicle identification, the uniqueness of vehicle is determined as itself, and then instead vehicle identity information, i.e. each unique vehicle to be identified can be determined as itself, can be completed at any time when driver identity information, vehicle identification determination, realize non-vehicle identity vehicle identification.The present application identifies vehicle by non-vehicle identity, not only easy to realize, high identification accuracy, but also lower requirement for implementation scene, less affected by environment.The present application is based on driver identity information in combination with vehicle determination, for the identification of vehicle behavior, can be clearly attributed to the driver belonging to driver identity information, i.e. if there is no driver, vehicle behavior certainly will not occur, when vehicle behavior occurs, driver certainly exists, therefore, the legal event involved in the behavior of vehicle can be directly directed to the driver determined by identity.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and more specifically, to a vehicle identification method that does not involve vehicle identification. Background Technology

[0002] Existing vehicle identification technologies first establish that a vehicle possesses identity information, then perform identification to obtain this information, thereby completing vehicle identification. Typically, a unique license plate is used as vehicle identification information, and this is the basis for other methods, such as video tracking, to identify vehicles while they are in motion. While such indirect identification methods in existing technologies may not directly obtain vehicle identification information during implementation, their effectiveness fundamentally depends on the vehicle possessing identity information; that is, the vehicle must first possess vehicle identification information before it can be obtained, regardless of whether this is done directly or indirectly. If this prerequisite of identifying and obtaining vehicle identification information cannot be achieved, any subsequent related identification operations will be ineffective. For example, even if a vehicle is tracked via video while in motion, its identity information cannot be obtained. This is especially true in cases of counterfeit license plates, cloned license plates, or license plates that are unrecognizable due to damage or intentional / unintentional obstruction; all of these situations prevent the vehicle from possessing or correctly identifying and obtaining its identity information.

[0003] Using vehicle identification information, such as license plates, for vehicle identification can be problematic because vehicle identification information and driver identification information cannot be perfectly matched. If driver identification is not completed, actions performed by the vehicle cannot be accurately linked to the associated driver based solely on the vehicle identification information. In other words, if the driver's identity cannot be determined, the actions performed by the vehicle cannot definitively identify the driver to whom they belong. Furthermore, binding vehicle identification information to driver identification information poses a significant risk of privacy breaches.

[0004] Therefore, the existence of vehicle identification information, to some extent, limits the choice of technical approaches for vehicle recognition and hinders the effective implementation of other related recognition methods. Summary of the Invention

[0005] The purpose of this invention is to overcome the shortcomings of the prior art and provide a vehicle identification method that does not rely on vehicle identity. This method cancels the vehicle identity information and uses driver identity information combined with vehicle identification to uniquely identify the vehicle as itself, thereby replacing the vehicle identity information. This method can complete the vehicle identification without vehicle identity at any time when driver identity information and vehicle identification are completed simultaneously, and its effect on identifying the vehicle is equivalent to that of vehicle identity information.

[0006] The technical solution of the present invention is as follows:

[0007] A vehicle identification method without vehicle identity binding driver identity information to the vehicle, wherein the vehicle is not assigned a vehicle identity, and the feature information of the vehicle is extracted based on image recognition as an identification sample; when the following conditions are met simultaneously, the vehicle to be identified is determined to be itself, and the vehicle identification is completed.

[0008] The conditions mentioned include, but are not limited to:

[0009] Condition 1: Identify the driver in the driver's seat of the vehicle and obtain the driver's identity information;

[0010] Condition 2: Collect the feature information of the vehicle to be identified, compare it with the identification sample, and determine that the vehicle to be identified and the vehicle corresponding to the identification sample are the same vehicle.

[0011] Preferably, if the conditions are not met simultaneously, the conditions that are met first are retained until all conditions are met, then vehicle identification is completed.

[0012] Preferably, for any condition, if the condition cannot be maintained at a certain time after it is met, then the conditions described at the previous time are determined to be unmet.

[0013] Preferably, at least one type of perception data of the target to be identified is collected at different collection locations. The target to be identified includes, but is not limited to, a user in the driver's seat of a vehicle and the vehicle to be identified. The perception data is continuously collected. When the collected perception data cannot complete the identification of the target to be identified, it is associated with a unique temporary condition pointing to the current target to be identified until the identification of the target to be identified is completed. The temporary condition is then modified to the condition corresponding to the target to be identified. The completion of the identification of the target to be identified includes, but is not limited to, obtaining the driver's identity information as described in condition one and determining that the vehicle to be identified and the vehicle corresponding to the identification sample are the same vehicle as described in condition two.

[0014] As a preferred option, when there is a blind zone between two adjacent acquisition positions and no sensing data is acquired in the blind zone, it is determined whether the sensing data acquired at the first acquisition position and the sensing data acquired at the second acquisition position have a coherent logic. If so, the target to be identified at the second acquisition position inherits the condition of the target to be identified at the first acquisition position.

[0015] The coherent logic is as follows: based on the type of sensing data and the corresponding preset rules, the estimated sensing data for the subsequent collection time can be inferred from the sensing data collected at the earlier collection time, and the estimated sensing data for the subsequent collection time matches the sensing data collected at the subsequent collection time.

[0016] Preferably, driver identification is performed before and / or after the vehicle has traveled.

[0017] The method for driver identification before the vehicle is driven is as follows: when a user is identified to enter the vehicle from the driver's seat, the user is determined to be the driver and driver identification is performed; or, when two consecutive users are identified to enter the vehicle from the passenger side, the first user to enter is determined to be the driver and driver identification is performed.

[0018] The method for driver identification after the vehicle has finished driving is as follows: when it is detected that a user gets off the vehicle from the driver's seat, the user is identified as the driver and driver identification is performed; or, when it is detected that two consecutive users get off the vehicle from the passenger side, the second user who gets off is identified as the driver and driver identification is performed.

[0019] Preferably, the vehicle's feature information includes, but is not limited to, vehicle model, color, and modifications. The vehicle model is confirmed based on image recognition, and the vehicle model parameters are obtained from a preset vehicle model database.

[0020] As a preferred method, during vehicle operation, blockchain is used to store identification samples, continuously collected feature information of the vehicle to be identified, continuously acquired identification results of the vehicle to be identified, and driver identity information.

[0021] This invention utilizes a peer-to-peer computing system to perform non-specific feature recognition and location recognition of targets, including vehicles and drivers;

[0022] The peer-to-peer computing system includes multiple node devices, and there is no hierarchy among the node devices. Each node device is equipped with a data acquisition device and a computing module. The data acquisition device includes at least one type of sensor, including an image acquisition device, for collecting different types of sensing data. Node devices located at different acquisition positions collect at least one point sample of the target, and the point sample is sensing data of the corresponding sensor type.

[0023] For a given node device, the collected sensing data is processed to obtain result data, which is then propagated to other node devices. Other node devices that receive the result data use it as one of the original data collected, and the result data influences the result data of other node devices. Based on this, without needing to obtain the target's identity information, multiple node devices in the peer-to-peer computing system perform collaborative computing to determine that each unique target is itself, achieving non-specific feature recognition and target location identification.

[0024] As a preferred embodiment, during vehicle operation, a peer-to-peer computing system is used to perform non-media storage of continuously collected raw data based on result data calculations. The raw data includes sensing data and result data received by node devices.

[0025] The non-media storage method is as follows: Each node device processes the collected raw data through its own deployed data processing model, calculates the result data, and propagates the result data to other node devices; other node devices that receive the result data use the result data as one of the collected raw data and use their data processing model to calculate the current node device's result data; specifically, the current node device receives the preceding result data output by the preceding node device, and combines it with the current raw data collected by the current node device to calculate the current result data, and transmits it to the subsequent node device; and so on, the result data is transmitted between node devices. The result data represents the sensing state perceived by the sensors connected to the node device, the node device's own state, and the device state of the execution device connected to the node device. Through the calculation of the result data and the connection relationship between node devices, the non-media storage of the sensing state perceived by the sensors connected to each node device, the own state of each node device, and the device state of the execution device connected to each node device is completed.

[0026] The beneficial effects of this invention are as follows:

[0027] The vehicle identification method without vehicle identity as described in this invention can identify each unique vehicle without requiring vehicle identity information, thus achieving vehicle identification without vehicle identity. This invention eliminates vehicle identity information, using driver identity information combined with vehicle identification to uniquely identify a vehicle, thereby replacing vehicle identity information. This allows for vehicle identification without vehicle identity at any time, with the identification effect equivalent to that of vehicle identity information. This method of vehicle identification without vehicle identity is not only easy to implement and has high accuracy, but also has lower requirements for the implementation scenario and is less affected by the environment.

[0028] This invention is based on driver identity information combined with vehicle identification. The identification of vehicle behavior can be clearly attributed to the driver to whom the driver identity information belongs. That is, if there is no driver, the vehicle will not take any action. When the vehicle takes action, there must be a driver. Therefore, the legal events involved in the vehicle's behavior can be directly attributed to the identified driver.

[0029] This invention, based on coherent logic, enables the inheritance of already satisfied conditions for the target to be identified (including the user in the driver's seat and the vehicle to be identified) in the presence of blind spots. This achieves a certain degree of fault tolerance in driver identification and vehicle identification, avoiding repeated execution of vehicle identification.

[0030] This invention can identify a vehicle (excluding the vehicle's identity) at any time when the driver's identity information and vehicle identification are completed simultaneously. This allows for retrospective evidence collection at earlier moments when the driver's identity information and vehicle identification have not been completed, presuming that the perception results of the driver and vehicle at those earlier moments are valid.

[0031] This invention, based on collaborative computing within a peer-to-peer computing system, performs non-specific feature recognition and location identification for drivers and vehicles. Without requiring specific features or detailed identity information, it can determine that each unique object to be identified is itself, achieving non-specific feature recognition. This invention performs target identification, identity verification, ownership calculation, or event monitoring through non-specific feature recognition, resulting in high accuracy and precise location identification. Furthermore, this invention enables the recognition of identity and ownership relationships without relying on specific features, and also benefits privacy protection.

[0032] This invention employs non-specific feature recognition, effectively preventing risks caused by theft or counterfeiting of specific features, thus significantly enhancing security. It utilizes a non-contact, passive method for seamless identification of the object, greatly improving ease of execution. Based on the aforementioned peer-to-peer computing system, this invention can be easily deployed over coverage areas ranging from hundreds of meters to hundreds of kilometers, making it suitable for various geographical areas.

[0033] This invention identifies targets, confirms their identity and ownership, or monitors events without transmitting or storing raw data, thus eliminating the risk of privacy information leakage; correspondingly, it can resolve the contradiction between public safety and privacy protection.

[0034] This invention also provides tamper-proof and consistent evidence storage for identification samples, continuously collected feature information of the vehicle to be identified, continuously acquired identification results of the vehicle to be identified, and driver identity information. This includes storage using blockchain or non-media storage of perceived data and result data using a peer-to-peer computing system. On the one hand, it can be used to prove the binding relationship between the driver and the vehicle, such as the driver not getting out of the car or the vehicle not being switched. On the other hand, it can be used to ensure the legal validity of evaluating the driver's driving behavior, and further extend to other services, such as calculating car insurance premiums, calculating carbon emissions, promoting safer driving behavior, and rewarding excellent driving behavior.

[0035] The non-media storage method described in this invention, through the transmission and calculation of result data, enables the calculated result data to characterize the sensing states perceived by multiple sensors connected to a node device, the self-state of the node device, and the device states of multiple execution devices connected to the node device. Through the calculation of result data and the connection relationships between node devices, non-media storage of the sensing states perceived by sensors connected to each node device, the self-state of each node device, and the device states of execution devices connected to each node device is achieved. That is, the node device does not store the original data (unless, according to legal or regulatory requirements, sensors may be configured to store data for traditional evidence preservation), nor does it perform point-based identification using the collected original data on a single node device. In this invention, the peer-to-peer computing system does not actually store data with explicit meaning. Instead, when there is a need, the result data (representing the sensing states of each node device) is transmitted and calculated between node devices to obtain a result corresponding to the need, thus achieving a similar "storage" effect. In other words, the existence of the "result" corresponding to the need is not generated and stored through a medium, but rather obtained through calculation based on the need. The computation described in this invention is a causal and continuous computation performed by a massive number of node devices with massive logic (the entire peer-to-peer computing system will process massive demands at the same time). Hackers’ intrusion, hijacking, or tampering with any node device will cause errors in the interaction computation between the intruded, hijacked, or tampered node device and other node devices, thereby discovering the abnormality of the node device, activating corresponding measures for repair, or replacing the faulty node device.

[0036] When it is necessary to trace the behavior, attributes, state, or event of a target, the current node device takes the need for target tracing as one of the inputs, and adds it to the calculation of the current result data along with other preceding result data and current raw data, and then sends it to other node devices. In this way, based on the preceding result data and current raw data received by the current node device, the relationship between the preceding state value represented by the behavior, attributes, state, or event of the target in the raw data collected by the preceding node devices and the current state value represented by the current result data of the current node device is calculated through the calculation of the current result data. This drives the execution device connected to the current node device to respond, thereby achieving a similar effect of "backward calculation", and thus achieving a similar effect of "reading" from "storage". That is, the "reading" is not a "result" that already exists and is stored using a medium, but is obtained by calculation according to the demand.

[0037] In this invention, state values ​​are used to represent various types of perceived data for a target, simplifying the amount of information corresponding to the data being transmitted. The data is transmitted based on a peer-to-peer computing system. Within the coverage of the peer-to-peer computing system, the execution devices connected to the node devices can respond to the behavior, attributes, states, or events of multiple targets during the transmission of result data.

[0038] This invention also simplifies various types of perception data (i.e., result data) of the target through multi-dimensional matrix transcoding to obtain corresponding state values. Furthermore, as the result data is transmitted, the state is continuously superimposed and updated. Thus, each time the result data output by any node device contains the event sequence and the representation of details in various dimensions over a considerable period of time covered by the peer computing system. When needed, the node device at the receiving end performs calculations and transmissions through a series of node devices. One or more node devices drive their connected execution devices to respond based on the preceding result data received by the current node device. The response actions or results of the execution devices simulate the effect of "backward calculation". Therefore, it is not necessary to store the complete original data of various types of perception data of the corresponding target, and it is no longer necessary to rely on a large and slow database to record massive targets and complex spatiotemporal correlations, which greatly reduces data storage costs, greatly improves operating efficiency, and saves computing power.

[0039] The "storage" and "retrieval" described in this invention actually involve calculating the result data, which contains a massive set of past features, to achieve the effect of "recalling" (i.e., "reverse-engineering") historical states. This differs from single-point calculations that retrieve records from historical moments. Therefore, it does not require, and can exist, a storage medium to "store" data with clear meaning for "retrieval." The "recall" of events or information in this invention is the result of real-time calculation. All "recall" results are obtained from calculations by a massive number of node devices. The calculation results of these massive node devices at each unit of time are a common intermediate calculation, encompassing a high-dimensional, mixed calculation result that describes the historical state of the target.

[0040] When the storage capacity of the medium is limited, the technical solution of connection relationship and result data calculation provided by this invention is used to replace the traditional medium storage. Assuming a directed acyclic transmission mode, if any one of the 100,000 node devices initiates "non-medium storage" or "non-medium reading", it is at the level of 100,000 factorials, and its state value has 456,573 bits (i.e. 10,456,573 state values). When multiple node devices initiate, 10,456,573 changes are formed accordingly. Therefore, the capacity of a city-level infrastructure can be regarded as close to infinite (in fact, the node devices in the path can choose the preceding node device as the subsequent node device, which is a cyclic transmission mode, and a node device can transmit the result data to multiple node devices, so the expression sequence is infinite).

[0041] Because individual node devices do not store data with explicit meaning, and the results of "storage" or "reading" are obtained through real-time calculation, this invention cannot be intruded upon, hijacked, have information stolen, or tampered with unless all node devices are hijacked simultaneously. In city-level deployments, the number of node devices can reach hundreds of thousands or even hundreds of millions. Existing hacking methods cannot simultaneously intrude into all node devices. Therefore, this invention is immune to intrusion, hijacking, information theft, and tampering by existing hacking methods. Detailed Implementation

[0042] The present invention will be further described in detail below with reference to the embodiments.

[0043] To address the shortcomings of existing vehicle identification technologies, such as limited application scenarios, accuracy affected by environmental factors, and inability to prevent identity spoofing, this invention provides a non-vehicle identity-based vehicle identification method. This method can identify each unique vehicle without requiring vehicle identity information, thus achieving non-vehicle identity-based vehicle identification. This invention eliminates vehicle identity information, using driver identity information combined with vehicle identification to uniquely identify a vehicle, thereby replacing vehicle identity information. Non-vehicle identity-based vehicle identification can be completed at any time, with the identification effect equivalent to that of vehicle identity information. This invention, using non-vehicle identity for vehicle identification, is not only easier to implement and has high accuracy, but also has lower requirements for implementation scenarios and is less affected by environmental factors.

[0044] The vehicle identification method without vehicle identity described in this invention binds driver identity information to the vehicle. By identifying and acquiring the driver's identity information, and identifying and determining the vehicle bound to the driver (or vehicle determination, i.e., determining that the vehicle to be identified is the vehicle bound to the driver), it is clear which driver is driving which vehicle, thus determining that the vehicle currently driven by the current driver is unique and specific. When the vehicle is determined to be unique and specific, even if the vehicle does not have vehicle identity information, the driver's identity information combined with vehicle determination can replace the vehicle identity information. Therefore, the vehicle does not have a vehicle identity. Especially when the node devices of the peer-to-peer computing system provided by this invention cover all roads, controlling the car transporter to automatically store and retrieve vehicles can even be done without any vehicle perception, thus avoiding privacy leaks due to license plate recognition.

[0045] This invention identifies vehicles based on image recognition. Specifically, it extracts the feature information of the vehicle based on image recognition and uses it as a recognition sample. When identifying the vehicle to be identified, it can be compared and identified through image recognition.

[0046] In this invention, the vehicle to be identified is determined to be itself when the following conditions are met simultaneously, thus completing vehicle identification;

[0047] The conditions mentioned include, but are not limited to:

[0048] Condition 1: Identify the driver in the driver's seat of the vehicle and obtain the driver's identity information;

[0049] Condition 2: Collect the feature information of the vehicle to be identified, compare it with the identification sample, and determine that the vehicle to be identified and the vehicle corresponding to the identification sample are the same vehicle.

[0050] In this embodiment, when the above two conditions are met, it is possible to clearly identify which driver is driving and which vehicle the driver is driving, which is equivalent to achieving the identification effect of vehicle identity information.

[0051] In practical implementation, when multiple conditions are set, if multiple conditions cannot be met simultaneously, this invention provides a compatible method. That is, when the conditions are not met simultaneously, the conditions that are met first are retained until all conditions are met, thus completing vehicle recognition. Based on this, even if multiple conditions cannot be met simultaneously, it will not affect the accuracy of vehicle recognition, and it is more in line with practical application scenarios, facilitating the implementation of this invention.

[0052] This invention also provides an error correction method to correct deficiencies caused by factors such as misidentification or changes in the driver-vehicle binding (e.g., vehicle-driver swapping, driver-vehicle swapping), which prevent the vehicle identity information from being definitively replaced by the driver's identity information. Specifically, for any condition, if the condition cannot be maintained at a subsequent time after it is met, then the conditions stated in the preceding time steps are determined to be unmet, to avoid incorrect replacement of vehicle identity information due to misidentification. In this case, vehicle identification needs to be performed again, including determining the binding relationship between driver identity information and vehicle, identifying the driver, and determining the vehicle. If related applications were performed after the preceding vehicle identification, it is necessary to reconfirm whether the validity of the related applications is retained; the specific implementation can be selected according to the implementation requirements.

[0053] In this invention, at least one type of sensory data of the target to be identified is collected at different collection locations. The target to be identified includes, but is not limited to, a user in the driver's seat of a vehicle and the vehicle to be identified. The sensory data is not necessarily used to identify the target; it may also include sensory data corresponding to the behavior of the target to be identified, used to determine the state and behavior of the target. When the target to be identified has not completed identification, all sensory data is temporarily stored or temporarily assigned, which can be used to confirm the state and behavior of the target after identification is completed. For example, after the target has completed identification, it can confirm the state or behavior at a previous moment. Based on this, traceability and evidence collection of the target to be identified can be achieved. Specifically, when the continuously collected sensory data cannot complete the identification of the target, it is associated with a unique temporary condition pointing to the current target until the target is identified. The temporary condition is then modified to the condition corresponding to the target. The completion of the target identification includes, but is not limited to, obtaining driver identity information as described in condition one, and determining that the vehicle to be identified and the vehicle corresponding to the identification sample are the same vehicle as described in condition two. When applied to source tracing and evidence collection, this is known as reverse-engineering evidence collection. For example, if you want to collect evidence of a vehicle's violation at 9:00, but vehicle identification was not completed at that time, but the video footage from 9:00 to 9:15 is continuous and constitutes a valid data chain, and vehicle identification was completed at 9:15, then the continuous video footage from 9:00 to 9:15 can be used for source tracing and evidence collection, ensuring the validity of the evidence.

[0054] This invention, based on coherent logic, enables the inheritance of satisfied conditions between consecutive data collection locations when blind spots exist for the target to be identified (including the user in the driver's seat and the vehicle to be identified). This provides a degree of fault tolerance for driver identification and vehicle identification, avoiding repeated execution of vehicle identification. Specifically, when a blind spot exists between two adjacent data collection locations, and no perception data is collected within the blind spot, it is determined whether the perception data collected at the first and second data collection locations has coherent logic. If so, the target to be identified inherits the condition satisfied at the first data collection location in the second data collection location.

[0055] The coherent logic is as follows: based on the type of perceived data and according to the corresponding preset rules, the estimated perceived data for a later acquisition time can be inferred from the perceived data collected at an earlier acquisition time, and the estimated perceived data for a later acquisition time matches the perceived data collected at that later acquisition time. If the perceived data collected at a later acquisition time does not match the estimated perceived data for that later acquisition time, it is determined that the perceived data collected at a later acquisition time does not have a coherent logic with the perceived data collected at an earlier acquisition time. Correspondingly, the condition that the target to be identified is located at the acquisition position corresponding to the later acquisition time is not met.

[0056] Regarding the identification and acquisition of driver identity information, this invention can perform driver identification before and / or after the vehicle has traveled.

[0057] The method for driver identification before vehicle movement is as follows: when a user is detected entering the vehicle from the driver's seat, the user is identified as the driver, and driver identification is performed; or, when two consecutive users are detected entering the vehicle from the passenger side, the first user to enter is identified as the driver, and driver identification is performed. This method of driver identification before vehicle movement is typically implemented in scenarios where identification can be completed within a short timeframe before, during, or after the driver gets into the vehicle.

[0058] For scenarios where driver identification cannot be completed before, during, or shortly after the driver gets in the vehicle, driver identification can be performed after the vehicle has finished driving. Specifically, the method for driver identification after the vehicle has finished driving is as follows: if a user is identified as getting off the vehicle from the driver's seat, that user is determined to be the driver, and driver identification is performed; or, if two consecutive users are identified as getting off the vehicle from the passenger side, the second user is determined to be the driver, and driver identification is performed. Since driver identification is performed after the vehicle has finished driving, the conditions met during the vehicle's journey cannot be temporarily associated with the identified driver. Correspondingly, if driver identification is performed after the vehicle has finished driving, a temporary driver can be set up first, and the temporary driver's identity information can be obtained. After the vehicle starts driving, the met conditions are combined with the temporary driver's identity information to form a temporary vehicle identification; after the vehicle has finished driving and driver identification is performed, the temporary vehicle identification is converted into the final vehicle identification.

[0059] In this embodiment, the vehicle's feature information includes, but is not limited to, vehicle model, color, and modifications. The vehicle model is confirmed based on image recognition, and the vehicle model parameters are obtained from a preset vehicle model database. Because vehicle modifications, upgrades, or alterations can change the original vehicle's appearance and model parameters to some extent, or due to uneven lighting in the indoor environment, wall obstructions, etc., it can be difficult to accurately identify the vehicle model from certain angles. Therefore, to ensure the acquisition of vehicle model parameters, this invention uses image recognition of multiple monitoring images to identify and accumulate partial vehicle model parameters from each monitoring image. Multiple monitoring devices simultaneously acquire multiple monitoring images from different angles, gradually accumulating the vehicle's model parameters and feature information, i.e., accumulating and identifying the vehicle's model parameters and feature information.

[0060] In practical implementation, traditional single-point identification methods can be used to identify drivers or vehicles at designated locations to confirm their identities and associate them with location information. Alternatively, the peer-to-peer computing system provided by this invention can be used for non-specific feature-based identity recognition. The peer-to-peer computing system of this invention is based on collaborative computing, does not rely on single-point identification, and distributes computing functions across the entire network, reducing the hardware and software requirements of single-point computing, resulting in high execution efficiency and significantly improved anti-attack capabilities. The system maintains a relatively symmetrical information state among node devices, making it immune to illegal data tampering. Even if a single node device is physically compromised and its transmitted data is altered, the network-wide computation is a highly redundant and complex calculation with numerous multi-dimensional verifications. Therefore, the alteration of data transmitted by a single node device does not affect the overall network computation results. Furthermore, it allows for rapid location of faulty and tampered node devices, ensuring the reliability of the overall network computation results. This, in turn, resolves the conflict between data sharing and information security between departments.

[0061] The result data transmitted between node devices can be the processing result of information rather than the information itself. Therefore, the raw data collected (i.e., perceived data) does not need to be stored. Node devices only receive the calculation results output by other node devices and send out their own calculation results. The amount of information contained in a single calculation result is insufficient to reconstruct any event or target information. A definite result can only be obtained by joint calculation of the calculation results of the entire peer-to-peer computing system, multi-dimensional data matrix elements, and physical space and facility correspondence. The collaborative calculation has less dependence on the information transmitted by a few node devices, thus fundamentally changing the nature of traditional information technology's single-point security sensitivity.

[0062] In this invention, the acquisition of the identity information and location information of the target (i.e., driver, vehicle) can be achieved through collaborative computing using the peer-to-peer computing system provided by this invention. Specifically, the peer-to-peer computing system is used to perform non-specific feature recognition and location recognition on the target. The term "non-specific feature recognition" differs from the common understanding of "recognition" in a strict conceptual definition. Commonly, "recognition" refers to determining the concrete form or specific identity information of the target, such as who it is (including name, specific information indicating the target's identity), or what it is (e.g., car, person). However, the "recognition" in the "non-specific feature recognition" of this invention refers to identifying each unique target as itself; that is, for a given object to be identified, its existence is unique. After implementing "non-specific feature recognition," this invention determines that the object to be identified (i.e., the target that has not been identified or had its identity confirmed) is itself, and not other objects to be identified. The result of "non-specific feature recognition" does not require determining the specific characteristics of the object to be identified, nor does it require determining the identity information or concrete form of the object to be identified. For example, if a person is considered object A to be verified, and an object is considered object B to be verified, then after implementing "non-specific feature recognition," it is not necessary to identify whether object A is a person or what their specific identity is, nor is it necessary to identify whether object B is an object or what kind of object it is; rather, it is necessary to determine that object A is object A itself, and object B is object B itself. Then, corresponding services or controls can be provided for object A or object B.

[0063] The peer-to-peer computing system comprises multiple node devices, all without a hierarchy, forming a decentralized network and computing architecture. Unlike traditional single-point aggregation computing models, the data transmission direction between node devices in this invention does not have a fixed, predetermined path relationship. In the peer-to-peer computing system described in this invention, a particular node device processes the collected raw data to obtain result data, and then propagates the result data to other node devices. Other node devices that receive the result data use it as one of their collected raw data sets, thus influencing the result data of other node devices. For ease of description, the aforementioned "particular node device" is referred to as the "current node device," and the "other node devices" are referred to as "subsequent node devices." One aspect of this influence is that the result data obtained by subsequent node devices is not entirely determined by their own collected raw data, but rather by the result data output by the current node device. Specifically, the result data output by the current node device may alter the data processing model and parameters used by subsequent node devices to calculate the result data, thereby affecting the result data of subsequent node devices. For example, if the output data of the current node device is correlated with the raw data collected by subsequent node devices, it is necessary to consider the impact of the output data of the current node device on the accuracy of the output data of the subsequent node devices. Specifically, for the perception of a specific target, if the result data is calculated based solely on the raw data collected by subsequent node devices, it can only reflect the real-time (including real-time location and time) single-point result judgment of the target within the perception range of the subsequent node devices. However, the output data of the current node device reflects the direct perception data and result judgment of the target at other locations and at other times, or other indirectly related perception data and result judgments, which helps to improve the accuracy and comprehensiveness of the result data of the subsequent node devices, including superimposed calculations of the same dimension and correlation references of different dimensions.

[0064] Because there is no master-slave relationship between nodes in a peer-to-peer computing system, point-to-point transmission is possible. Therefore, for a given calculation result corresponding to a specific perceived data point of a target, as reflected in the output data of one node, the information is relatively symmetrical among other nodes receiving that result data. Other nodes use the received result data as input, combining it with their own sensor data to calculate their own result data. Their own result data naturally encompasses both the received result data and the information reflected by their own sensors, and is transmitted to other nodes in the next layer. Thus, for a specific perceived data point of a target, information is relatively symmetrical across all nodes. This prevents the impact of tampering or falsification of the calculation process and results of a single node on the result data. It also serves as a means to detect faulty, tampered, or non-compliant node devices. This fundamentally solves the inherent hidden dangers of traditional information technology, namely, the false, falsified, and erroneous information caused by information asymmetry, which becomes a point of entry for fraud and cyberattacks. It also addresses the problems of poor accuracy, excessive time consumption, low credibility, and poor responsiveness in complex integrated applications. Therefore, it can truly become the information infrastructure for comprehensive management of large areas and the infrastructure for the digital economy. Unlike blockchain technology, which relies on independent computation by each node to determine the result and emphasizes the preservation of original data, this invention focuses on peer-to-peer collaborative computation among node devices. Through this collaborative computation, each node device can adjust its own data processing model (i.e., the algorithm for calculating the result data) and parameters when processing data. This adjustment is a feedback mechanism from all node devices, transforming the computation of all node devices into a unified whole. Instead of individual nodes performing calculations independently, all node devices collaboratively complete the computation. The adjustments to the node device's data processing model are objectively real and will impact subsequent data processing iterations.

[0065] Node devices are equipped with data acquisition devices (in specific implementations, these may include one or more of the following: image acquisition devices, audio acquisition devices, temperature measurement devices, vibration frequency sensing devices, lidar, chemical sensors, and electromagnetic induction devices) and a computing module. The data acquisition devices include one or more types of sensors or data acquisition devices, including image acquisition devices, used to collect sensing data of different corresponding types. The computing module calculates the resulting data based on a data processing model. Node devices located at different acquisition positions (i.e., at different physical installation locations) collect at least one point sample of the target; the point sample is sensing data of the corresponding sensor type. Based on this, without needing to obtain the target's identity information, multiple node devices in the peer-to-peer computing system perform collaborative computation to determine that each unique target is itself, achieving non-specific feature recognition and target location identification.

[0066] Specifically, taking a given node device as the current node device, and considering the data transmission between its preceding and subsequent node devices (in this invention, "preceding node device" and "subsequent node device" only describe their sequential relationship with the current node device in the current calculation and data transmission process, and do not imply any necessary sequential or priority relationship between them), the current node device receives the result data output by other node devices (including preceding node devices), and subsequent node devices receive the result data output by other node devices (including the current node device). For the current node device, the collected sensing data is combined with the result data from other node devices (including preceding node devices) to calculate the result data of the current node device, and this result data is sent to other node devices (including subsequent node devices). Similarly, the working process of subsequent node devices is the same as that of the current node device, and preceding node devices also receive the result data from the preceding node devices of their predecessors and perform the same working process as the current node device; that is, the node devices in the peer-to-peer computing system perform the same working process. Furthermore, the node devices in the peer-to-peer computing system perform collaborative calculations as sensing data is collected and result data is calculated. In this process, the output data of a certain node device is only received and used as input by the subsequent layer of node devices, and the output data of the subsequent layer of node devices will cover the output data of the preceding layer of node devices (including the aforementioned node device).

[0067] In a peer-to-peer computing system, all events are processed synchronously, and it is not necessarily necessary to explicitly produce staged results such as what event was discovered or what the specific content of the event is. In a peer-to-peer computing system, only the sensor's perception and the corresponding execution device (in this invention, the guiding device) respond are explicit. All other intermediate processes are processed simultaneously through collaborative computing. That is, during the operation of this invention, the intermediate process of event discovery is imperceptible. As collaborative computing proceeds and the node device obtains the result data, the corresponding execution device automatically responds and executes.

[0068] To further ensure the trustworthiness of the data source and computation process, in this invention, all node devices encrypt their computational results based on an encrypted consensus mechanism, obtaining encrypted results, which are then sent to other node devices. The encrypted consensus mechanism includes one or more consensus mechanisms, with different mechanisms corresponding to changes in the encryption algorithm structure and parameters of the node devices.

[0069] Node devices communicate using standard-sized data packets (i.e., result data or calculation results). In this invention, the node devices in the peer-to-peer computing system are similar to human neurons. Just as each neuron does not transmit specific data directly describing external events, the node devices do not output raw data. Instead, they process the raw data acquired by connected sensors and data acquisition devices into standard-sized data packets (i.e., result data or calculation results, similar to nerve impulses in neurons) based on their own data processing models (similar to the biological characteristics of nerve cells). The information contained in a single data packet is insufficient to reconstruct any event or target information. A definite result can only be obtained through collaborative computation involving the calculation results across the entire peer-to-peer computing system, multi-dimensional data matrix elements, and the correspondence between physical space and facilities. Collaborative computation has little dependence on the data output by a few node devices, and it simultaneously processes all requests received or initiated by all node devices. It is a collaborative verification computation of highly multi-dimensional related information, thus fundamentally changing the traditional single-point security sensitivity of information systems.

[0070] To ensure data integrity and the effective execution of collaborative computing, this invention deploys a QoS mechanism in the peer-to-peer computing system. The QoS mechanism prioritizes ensuring the transmission quality of result data between node devices.

[0071] In practical implementation, the peer-to-peer computing system can be networked using one or more combinations of 4G, 5G, or MESH modes to suit different application scenarios. The optimal solution is achieved by considering factors such as feasibility and cost. The MESH mode is based on the LTE standard, communicating at the LTE physical layer. Data is carried by a customized frame structure, and interaction is performed using a dedicated wireless communication protocol. Customizing the frame structure to suit peer-to-peer computing and employing a proprietary wireless communication protocol developed for urban cluster peer-to-peer computing further enhances its security and reliability. Furthermore, the wireless algorithm is fully adapted to the multipath channel environment controlled by a consensus mechanism required for peer-to-peer computing, achieving communication distances of 100 meters to 10 kilometers within cities and 120 kilometers in the field using omnidirectional antennas. In this embodiment, the Mesh network communication distance is 50-150 meters between indoor nodes and 50 meters to 120 kilometers between outdoor nodes, with each node capable of connecting to 65,535 nodes. In addition, when networking in 4G and 5G modes, there is no limit to the communication distance, and the number of node devices that can be connected depends on the computing power of the computing chip and the communication latency.

[0072] In a peer-to-peer computing system, for a specific point sample of an object to be identified, the result data transmitted from the node device that collected the point sample to other node devices allows subsequent node devices to adjust their perceptual attention based on the features of that point sample (it is not necessary for the result data to contain the features of that point sample, but rather that the features of that point sample participate in the computation of the preceding node device, so that the result data of the preceding node device can be used as input to the data processing model of the subsequent node device, allowing the subsequent node device's data processing model to achieve the effect of adjusting perceptual attention during computation); or, the features of that point sample can be reported for subsequent node devices to adjust their perceptual attention (the features of that point sample are directly described in the result data). If other subsequent node devices do not detect the features of that point sample, but can determine from the features of other point samples that the undetected features of that point sample still belong to the object to be identified, then the features of that undetected point sample are continued to be described in the result data of the current node device and transmitted to other node devices. For example, if a preceding node device senses the color of an object A to be identified, but the current node device does not sense the color of the object A to be identified, but it can be determined from the sensing data of other node devices that there is another object A to be identified besides other objects to be identified, then the color of the object A to be identified that has not been sensed will still be represented in the result data of the current node device.

[0073] In this embodiment, the method for reporting the features of the point sample for subsequent node devices to adjust the perceptual attention is as follows: adjusting the parameters of the data processing model of the subsequent node device based on the features of the point sample provided by the preceding node device, so that the subsequent node device can improve the computing power of the point sample to identify its features; or, the subsequent node device uses the perceptual attention model to match the features of the received point sample to adjust the computing power.

[0074] The “feature” mentioned above has a different meaning from the “feature recognition” in the prior art. The “feature recognition” in the prior art usually refers to information that can determine the identity of a target, while the “feature” in this invention represents a kind of perceived data belonging to the object to be identified, such as coordinates, colors belonging to the object to be identified, etc. The “non-specific feature recognition” of the object to be identified cannot be directly completed by the “feature” perceived by a single point.

[0075] In this embodiment, the method for reporting the features of the point sample for subsequent node devices to adjust the perceptual attention is as follows: based on the result data expressing the features of the point sample provided by the preceding node device (in this invention, the features of the point sample are usually not provided themselves, but expressed in the result data), or the features of the point sample (i.e. the features of the point sample itself), the parameters of the data processing model of the subsequent node device are adjusted so that the subsequent node device can improve the computing power of the point sample to identify its features; or, the subsequent node device uses the perceptual attention model to match the features of the received point sample or the result data expressing the features of the point sample to adjust the computing power.

[0076] When a node device processes the output data from several preceding node devices, based on the data processing model, if the objects to be identified described by several preceding node devices can be determined to be the same target through certain common point sample features, the point sample features and other information described by each node device are merged into the same target. For example, point sample features in physical space that almost completely overlap at the same time can be determined to be the same target.

[0077] When the result data received by a node device indicates that the flag used by the current node device to identify the object to be identified before the current reception of result data is different from the flags used by other node devices to identify the object to be identified, and the flags assigned to the object by other node devices have been updated, then the flag used by the current node device to identify the object to be identified before the current reception of result data is converted. Specifically, the method for converting the flag used by the current node device to identify the object to be identified before the current reception of result data is as follows:

[0078] The flag used by the current node device to identify the object to be identified before the current reception of result data is replaced with the latest flag assigned to the object by other node devices; this is a simpler implementation of the present invention.

[0079] Alternatively, the conversion relationship between the flag used by the current node device to identify the object to be identified before the current receiving result data and the updated flag assigned to the object by other node devices can be recorded, and the conversion can be performed when the current node device's current receiving result data needs to be referenced; this is a relatively complex implementation method provided by the present invention.

[0080] Alternatively, the node device can deploy a conversion model to perform corresponding conversions on the labels of multiple objects to be identified based on the input raw data or result data; this is a more complex implementation provided by the present invention.

[0081] In this invention, in order to improve the effectiveness of "non-specific feature recognition", for one or more point samples collected successively by node devices at different collection locations, if the feature values ​​of one or more point samples at different collection locations meet the preset similarity conditions or are determined by a specific model to have a correlation threshold, and are unique at each collection location, then it is determined that the point samples at different collection locations are correlated.

[0082] On the other hand, for one or more point samples collected simultaneously by node devices at different collection locations, if the node devices at different collection locations collect data on the same spatial field, and there is only one object to be identified in the spatial field, or the collected point sample can correctly point to one of the multiple objects to be identified, then for a certain object to be identified, one or more point samples collected by node devices at different collection locations are correlated.

[0083] In this invention, the data acquisition device of the node device includes one or more combinations of an image acquisition device, an electromagnetic induction device, a temperature measurement device, and a vibration frequency sensing device, and a lidar. The data acquired by the aforementioned devices (i.e., one or more combinations of the image acquisition device, electromagnetic induction device, temperature measurement device, and vibration frequency sensing device) and the three-dimensional point cloud acquired by the lidar, or the point cloud generated from images acquired by multiple image acquisition devices, are jointly calculated to obtain three-dimensional points with data. The image color, contour, lines, reflectivity, motion trend, electromagnetic characteristics, temperature, temperature change trend, vibration frequency, and vibration frequency change trend based on two-dimensional perception are used as additional attributes of the corresponding three-dimensional points to constitute an attributed three-dimensional point cloud. Combining electromagnetic induction, temperature patterns, vibration frequency change characteristics, motion correlation (different motion correlations exhibited by different materials such as ropes and fabrics), and reflectivity, the correspondence between each region of the attributed three-dimensional point cloud and each part or related part of the 3D appearance of the object to be identified is determined. This embodiment utilizes the attributes and correlations of attributed 3D point clouds to determine the relationships between points, the correspondence between the regions to which each related point belongs and each part or related part of the 3D appearance of the object to be identified, and can more accurately determine the point sample features belonging to the object to be identified, thereby improving the efficiency and accuracy of "non-specific feature recognition".

[0084] In the process of "non-specific feature recognition," this invention can also acquire the identity information of the object to be identified when necessary. Specifically, when it is determined that the identity information of the object to be identified needs to be acquired, an identity information acquisition command is triggered. This command is used as one of the inputs in the calculation of the result data of the node device. By driving the node device in the peer-to-peer computing system, which is connected to the barrier-free data collection conditions capable of obtaining the identity information of the object to be identified, to respond with the corresponding result data, the identity information of the object to be identified is acquired. The acquisition of identity information is also the result of collaborative computing; that is, the acquisition of identity information is triggered by the determination that it needs to be acquired, rather than by an additional triggering through a specific request command. Based on this invention, if permission calculation is triggered by a request command, in most cases it can be completed without acquiring identity information. Only in a few cases, when it is found that permission calculation cannot be completed without acquiring identity information, will the determination that it is necessary to acquire identity information be generated according to implementation requirements. For example, if collaborative computing reveals that a person's identity information exists in several location-based QR code registration systems, package pickup registration systems, or consumer registration systems, and prior authorization from the person or legal access to these systems is obtained, then the peer-to-peer computing system can drive node devices connected to these systems via barrier-free data collection. The obtained information is then sent to the peer-to-peer computing system through each node device for information comparison and to provide accurate identity information. Based on this, the present invention can also minimize the possibility of identity tampering with a system.

[0085] Specifically, the peer-to-peer computing system determines the permissions of an object by verifying the authenticity of its identity information. In this system, node devices capable of acquiring identity information may not provide the identity information (or may provide it depending on implementation requirements), but instead express the verification result in their own result data based solely on the verification requirements for the authenticity of the identity information within the received result data. That is, in this invention, even when a node device capable of acquiring identity information does not provide it, the verification result is expressed in its own result data based solely on the verification requirements for the authenticity of the identity information within the received result data.

[0086] When a node device in a peer-to-peer computing system that can obtain identity information does not provide identity information, the information source device that drives the provision of identity information establishes an encrypted file transmission channel with the input terminal of the node device that needs to obtain identity information, or establishes an encrypted information transmission channel using other network communication modes; and uses the identity information as one of the inputs of the node device.

[0087] When necessary, in order to meet the needs of other traditional computing modes for raw data, such as the need for evidence preservation in traditional evidence presentation, in this embodiment, the node settings can be equipped with a data storage device for storing the raw data sensed by the sensor.

[0088] In practical implementation, the node device can also be equipped with leakage protection and other functions in its power supply. The node device can also provide various communication interfaces, including fiber optic interfaces and wireless communication interfaces; it can also provide a data interface for connecting external storage devices. The node device can be powered by solar energy or mains power. When implemented outdoors, the node device can be installed on poles such as streetlights (without crossarms, mounted on the main pole, or integrated into the lampshade); in pole-less areas, if implemented indoors, it can be wall-mounted or integrated into the ceiling.

[0089] When this invention is implemented indoors and outdoors, the node devices, as artificial intelligence facilities installed in public spaces, can serve as digital economy infrastructure for urban clusters, providing 24 / 7 seamless coverage. Through collaborative computing across node devices, vehicle identification at any location within the coverage area can achieve near 100% accuracy, with location identification accuracy related to sensor accuracy.

[0090] In this invention, the architecture of a peer-to-peer computing network consists of nodes of the same type and function. Each node dynamically adjusts its data processing model in real time according to the network's consensus mechanism. The raw data collected by the data acquisition devices (including sensors, cameras, etc.) connected to each node is processed and encrypted by the node according to its own data processing model, generating byte-level processing and encryption results (i.e., result data). This result data is then sent to other node devices (the computational and encryption results received by the current node from other node devices are also considered part of the raw data collected by the current node). Therefore, the effect of the raw data sensed by each sensor will propagate exponentially among a massive number of peer-to-peer node devices. If each node sends its result data to 100 surrounding node devices, after four units of time, hundreds of millions of node devices will be affected by the event sensed by that sensor. In this computing model, information is relatively symmetrical and immune to tampering and forgery. It fundamentally solves the inherent hidden dangers of traditional information technology, namely, the false, forged, and erroneous information caused by information asymmetry, which in turn become entry points for fraud and cyberattacks, as well as the problems of long cycles, poor accuracy, and poor adaptability in complex and integrated applications. In turn, it truly becomes an information infrastructure for comprehensive management of large areas and a digital economy infrastructure.

[0091] This invention utilizes the collaborative computing of a peer-to-peer computing system. When the results of this collaborative computing can identify an event, the event is detected. In this embodiment, the event detection by the peer-to-peer computing system includes the event's content, its location, and the corresponding response. In a peer-to-peer computing system, all events are processed synchronously; it is not necessary to explicitly produce staged outputs such as what event was detected or its specific content. In a peer-to-peer computing system, only the sensor's perception and the corresponding execution device's response are explicitly defined. All other intermediate processes are handled simultaneously by collaborative computing. That is, during the operation of this invention, the intermediate process of event detection is imperceptible; it is achieved as collaborative computing progresses, the node devices acquire their result data, and the corresponding execution devices automatically respond and execute. Furthermore, unlike the single-objective calculation of existing technologies, this invention calculates multiple objectives simultaneously. For example, in toll calculation, if the single-objective calculation of existing technologies is used, the specific cost is calculated for a particular vehicle's route. This invention, however, can use 5,000 node devices to calculate the cost for all 35,000 vehicles that need to pay, which is equivalent to simultaneously calculating the routes, matching toll conditions, and calculating costs for all 50,000 vehicles. This is reflected in directly driving the corresponding terminal (i.e., the execution device is the payment terminal) to implement contactless payment deduction or displaying a toll QR code based on the corresponding vehicle's location (i.e., the execution device is the display terminal). During this process, all the above objectives and logic are calculated simultaneously, and there is no calculation mode where a node calculates the route information, toll standard, location judgment, and terminal interactive control for a particular vehicle.

[0092] In a peer-to-peer computing system, the result data calculated and output by node devices can be implemented as a state corresponding to the perceived data (i.e., the raw data), which can be represented using state values. Therefore, node devices do not need to store and transmit the raw data. In this embodiment, the data or elements in the multidimensional matrix are related to the installation location, attributes, etc., of each node device. Therefore, when transmitting the result data, what is actually transmitted is the transcoded result after transcoding multiple sets of parameters. A multidimensional matrix is ​​actually a combination of multiple sets of parameters. For example, if the path to a target is from abcd, and the physical locations of the abcd node devices are fixed, then the sequence abcd can be expressed using a single character or a similar concept during multi-parameter transcoding and transmission.

[0093] Based on the technical characteristics of peer-to-peer computing, it can be applied to various application scenarios that provide targeted services or control for a specific target or event. Since the data transmitted between node devices is the result of information processing, rather than the information itself, the raw data collected (i.e., perceived data) does not need to be stored. Node devices only receive the computation results output by other node devices and send out their own computation results. The information contained in a single computation result is insufficient to reconstruct any event or target information; a definite result can only be obtained through collaborative computation involving the computation results across the entire peer-to-peer computing system, multi-dimensional data matrix elements, and the correspondence between physical space and facilities. Collaborative computation has less dependence on the information transmitted by a few node devices, thus fundamentally changing the traditional single-point security sensitivity of information systems.

[0094] In this invention, since the output data of each node device reflects the state evolution of the output data of the preceding node devices, the behavior, attributes, state, or events of the target when it was perceived by the preceding node devices can be inferred based on the output data received by the current node device. For example, when it is necessary to find the location of target 'a' 15 minutes ago, the location of the node device that perceived target 'a' can be obtained at the current moment, thus inferring the location of target 'a'. Then, based on the transmission path of the output data, it can be inferred back to 15 minutes ago to estimate the location of target 'a' 15 minutes ago (determined by the node device that perceived target 'a'). Furthermore, the node device does not need to store the original data about target 'a'. That is, based on this invention, it is not necessary to identify the original data to find target 'a', but rather to first infer the node device that perceived target 'a', and if necessary, obtain the original data about target 'a' at the time when it needs to be found from the storage device connected to the node device.

[0095] This invention also provides tamper-proof and continuous evidence storage for identification samples, continuously collected feature information of the vehicle to be identified, continuously acquired identification results of the vehicle to be identified, and driver identity information. This includes using blockchain to store identification samples, continuously collected feature information of the vehicle to be identified, continuously acquired identification results of the vehicle to be identified, and driver identity information during vehicle operation; or using a peer-to-peer computing system to perform non-media storage of continuously collected raw data based on result data calculations during vehicle operation. The raw data includes sensor data and result data received by node devices. On one hand, it can be used to prove the binding relationship between the driver and the vehicle, such as the driver not getting out of the vehicle or the vehicle not being switched. On the other hand, by identifying the vehicle's position during operation through a peer-to-peer computing system, the vehicle's driving behavior can be identified and converted into the driver's driving behavior. After evaluating the driver's driving behavior, the evaluation results are stored non-medially to ensure the legal validity of the evaluation results, and can be extended to other services, such as calculating car insurance premiums, calculating carbon emissions, promoting safer driving behavior, and rewarding excellent driving behavior.

[0096] Due to the shortcomings of existing technologies, such as single-point identification and storage of data with clear meaning through storage media, which suffer from high storage overhead and easy data leakage, this invention provides a non-media storage method for result data computation based on a peer-to-peer computing system. Instead of simply storing data with clear meaning through storage media, it adopts a non-media storage real-time result data computation technical solution. In addition to achieving a similar "storage" effect, it also fundamentally changes the traditional single-point security sensitive nature of information technology based on a peer-to-peer computing system. The data source and computation process are reliable, the execution efficiency is high, the hardware requirements are low, and the identification accuracy is high. It can achieve non-specific feature identification of visitors without the need for specific feature identification or obtaining the visitor's identity information.

[0097] Regarding the definitions of "media storage" and "non-media storage" in this invention, specifically, "media storage" refers to storing data (or data content) with a definite meaning through a storage medium, while "non-media storage" refers to storing data (or data content) with a definite meaning without a storage medium. "Having a definite meaning" means that a definite meaning can be determined from the data, such as user identity information, product type, product quantity, sensor-collected perception data, judgment results or recognition results obtained based on perception data, etc.; that is, the meaning corresponding to the data itself can be determined solely by the data itself, and the content that the data itself expresses can be known. Furthermore, "media storage" and "non-media storage" are not actually custom antonyms, but rather indicate a fundamental difference in technical solutions, representing completely different technical means based on entirely different technical principles. The "non-media storage" described in this invention does not mean that no storage medium is used at all during implementation, nor does it mean that no data is stored using any storage medium. Rather, in terms of technical features that achieve a similar effect to "storage" and "reading," it means that data content with explicit meaning is no longer stored on a storage medium. The original data is not stored, and simple single-point calculations are not performed on the original data beforehand. For example, for the number "1," the data corresponding to the number "1" is no longer recorded on the storage medium, either directly or after encryption. When "reading" is needed, the action to be performed after "reading" the number "1" is directly responded to through real-time calculation, thereby achieving a similar effect to "reading."

[0098] The non-media storage described in this invention is based on a peer-to-peer computing system. Each node device processes the collected raw data using its own deployed data processing model, calculates the result data, and propagates the result data to other node devices. Other node devices receiving the result data use it as one of the collected raw data sets and utilize their data processing models to calculate the current node device's result data. The result data itself does not directly show, represent, or express any explicit meaning; that is, the result data itself does not require other processing to reproduce or interpret a definite meaning. In practice, the result data in this invention represents intermediate, high-dimensional, mixed computational results encompassing the historical state description of the target. To obtain a result with a definite meaning, such as confirmation of an event or information, this invention obtains it during the transmission and calculation of the result data, without needing to form any data content with a definite meaning, let alone store it. This is the definition of "non-media storage" in this invention.

[0099] Specifically, the current node device receives the preceding result data output by the preceding node device, and calculates the current result data by combining it with the current raw data collected by the current node device, and transmits it to the subsequent node device. In this way, the result data is transmitted between the node devices. The result data represents the sensing state perceived by the sensors connected to the node device, the state of the node device itself, and the state of the execution device connected to the node device. Through the calculation of the result data and the connection relationship between the node devices, the non-media storage of the sensing state perceived by the sensors connected to each node device, the state of each node device itself, and the state of the execution device connected to each node device is completed.

[0100] The "storage" and "retrieval" described in this invention actually involve calculating the result data, which contains a massive set of past features, to achieve an effect similar to "recalling" (i.e., "reverse deduction") historical states. This differs from single-point calculations that retrieve records from historical moments. Therefore, it does not require, and can exist, a storage medium to "store" data with clear meaning for "retrieval." The "recall" of events or information in this invention is the result of real-time calculation. All "recall" results are obtained from calculations by a massive number of node devices. The calculation results of these massive node devices at each unit of time are a common intermediate calculation, encompassing a high-dimensional, mixed calculation result that describes the historical state of the target.

[0101] In this invention, the result data transmitted between node devices can be result data for multiple targets (i.e., result data), specifically a comprehensive state of one or more combinations of the behaviors, attributes, states, or events of multiple targets. In this invention, sensors connected to the node devices are used to sense their coverage area to obtain full-volume sensing data corresponding to the current time and current location within their coverage area (which necessarily includes the behaviors, attributes, states, or events of multiple targets). Specifically, some or all node devices are equipped with at least one type of sensor to collect corresponding different types of sensing data and calculate the current result data. For any node device where a target is sensed, the current node device receives the preceding result data output by the preceding node device, combines it with the current raw data collected by the current node device, calculates the current result data, and transmits it to the subsequent node device; this process continues to complete the transmission of result data between node devices. In other words, in each transmission of result data by a node device, the received data is combined with the current raw data it has collected, and calculations are performed based on the data processing model to obtain result data. Then, the result data is sent to other node devices in the next layer (in most cases, this includes multiple subsequent node devices, each of which corresponds to multiple current node devices, and each node device corresponds to multiple preceding node devices; the logical principle is the same throughout the network). That is, a node device completes one transmission of result data.

[0102] In this invention, the current raw data represents all the data output by the sensors connected to the current node device, including the current state of the behavior, attributes, status, or events of multiple targets within its coverage area (i.e., the perception state sensed by the sensors connected to the node device, the node device's own state, and the device state of the execution device connected to the node device can correspond to the state of the target's behavior, attributes, status, or events; or, when the target is an object sensed by the sensors, or a node device or an execution device connected to the node device, then the perception state sensed by the sensors connected to the node device, the node device's own state, and the device state of the execution device connected to the node device include the state of the target's behavior, attributes, status, or events). The preceding result data represents all the data output by the sensors connected to the preceding node devices, including the preceding comprehensive state of the target's preceding behavior, attributes, status, or events. The current result data represents all the data output by the sensors connected to the current node device and the result data received from other node devices, including the current comprehensive state of the target's current behavior, attributes, status, or events. The current comprehensive state is a combination of the current state and the preceding comprehensive state. For example, in a simple single-dimensional, single-target scenario, the preceding node device outputs the preceding result data as ab (i.e., the result data output by the previous-level node device is a, the perceived data of the preceding node device is b, and the combined current result data is ab), and the current node device's current raw data is c, then the current result data is abc. The same logic applies to other complex multi-dimensional, multi-target scenarios.

[0103] In this invention, the preceding node device, current node device, and subsequent node device are not necessarily in a fixed order. For example, in a peer-to-peer computing system, three node devices x, y, and z may transmit result data in a specific instance: node device x transmits result data x to node device y, and node device y transmits result data y to node device z. However, in another instance, these three node devices may not have a data transmission relationship. The same applies to all node devices in the network. Correspondingly, the transmission path in a single instance of result data transmission does not mean that the node devices constituting the transmission path are on the same transmission path in every instance of result data transmission. The transmission path is obtained by summarizing the result data transmissions that have occurred, rather than a pre-planned path (including fixed and dynamic paths) or a calculated path. The transmission path in this invention only represents a summarizing result and does not imply its existence. In this embodiment, the data processing model of the current node device determines the subsequent node device that receives the current result data.

[0104] Different node devices collect raw data at their respective times and / or in their respective spaces. If it is necessary to trace the behavior, attributes, state, or event of a target, the current node device takes the need for target tracing as one of the inputs, and adds it to the calculation of the current result data along with other preceding result data and the current raw data, and then sends it to other node devices. Similarly, based on the preceding result data and the current raw data received by the current node device, the relationship between the preceding state values ​​represented by the target's behavior, attributes, state, or event in the raw data collected by the preceding node devices and the current state values ​​represented by the current result data of the current node device is calculated, thus completing "non-media reading" through the calculation of the result data. In practical implementation, preceding node devices may not be able to discover or determine the target's behavior, attributes, state, or events. However, the raw data collected by preceding node devices that cover the target inevitably includes sensor perception data of the target. After obtaining the relationship between the preceding state values ​​representing the target's behavior, attributes, state, or events in the raw data collected by each preceding node device and the current state values ​​representing the current result data in the current node device, a "backward deduction" effect can be achieved based on the relationship between the result data that can discover or determine the target's behavior, attributes, state, or events, through the calculation of subsequent result data and the response of the execution device. In fact, the so-called "backward deduction," "recall," and "tracing" in this invention are all similar effects, which are actually obtaining new calculation results, i.e., new result data. Based on the technical solution of this invention, it is known that the required result is obtained through the calculation of result data, and the final effect is exactly similar to the traditional understanding of "backward deduction," "recall," and "tracing." Traditional understanding of "reverse engineering," "recalling," and "tracing" historical data recorded through storage media involves actually performing a data search operation. The "reverse engineering," "recalling," and "tracing" achieved by this invention involves the current node device calculating the result data to generate current result data corresponding to the "original data collected by previous node devices regarding the target's behavior, attributes, state, or events" that have existed or occurred in the past. This drives the execution device connected to the current node device to respond, and the execution device performs the corresponding action.Specifically, for example, node a (the current node device) determines the perception state of the target by preceding node m (i.e., the previous node device) at past moments by adding a specific description to the result data calculated by node a. This description is converted into a state value in the output matrix (i.e., the expression form of the result data). Then, the result data calculated by node a is transmitted to node m, or sent to other node devices, participating in the calculations of other node devices one after another, and finally reaching node m. Node m, using the same principle, adds a corresponding description to the result data calculated by node a. This description is converted into a state value in the output matrix. Similarly, the execution device of the node device corresponding to the "traceability" requirement, such as node z1, z2, etc., outputs a response, which can be a response action or a response result. This response action or response result is the result corresponding to the "traceability" requirement. When the result is obtained, a similar "reverse reasoning" effect is achieved. Therefore, the "non-media reading" described in this invention is actually based on "non-media storage," acquiring the result of "non-media storage."

[0105] In this invention, the "reverse reasoning" effect is not necessarily the work of a single node device. In most cases, it's the overall result achieved by multiple node devices through connections and the transmission of result data. When a result not stored in media is needed, an extraction command (including a need for target tracing) is issued to the node devices. The node device receiving the extraction command processes the data (i.e., a type of raw data) related to the person, device, and scenario that issued the command, similar to raw data collected by sensors. The calculated result data is then sent to other node devices. Through collaborative computation among the node devices, one or more appropriate node devices are driven to respond. The data processing models of each node device can, in terms of structure and training, encompass permission judgments based on real identity, scenario, and complex rules.

[0106] As the result data of this invention is transmitted, its state is continuously superimposed and updated. Thus, each output result data from any node device contains an event sequence and details of various dimensions over a considerable period covered by the peer-to-peer computing system. When needed, from the node device at the receiving end, through a series of node devices, calculations and transmissions are performed. One or more node devices, based on the preceding result data received by the current node device, drive their connected execution devices to respond through collaborative computation, achieving an effect similar to "backward calculation" (in most cases, exhaustive enumeration is not required; instead, result data that meets the conditions is output by the corresponding node device. The effect of collaborative computation by multiple node devices is to filter out suitable result data and output it on the node devices that meet the conditions. This filtering is not from all explicit data, but rather an output during the process of multiple node devices calculating and transmitting comprehensive calculation result data covering massive events). Furthermore, the result data is a simplified state value. After receiving the result data in state value form, the node device can directly perform combined calculations without first achieving a "backward calculation" to obtain the target's past historical state. In practical implementation, a "backward calculation" approach is only used when the target's behavior, attributes, state, or event needs to be traced and output, and the target cannot be reached based on simplified state values. (In this invention, "non-media storage" actually achieves a similar "backward calculation" effect through computation, using the multi-dimensional matrix provided by this invention to refine the calculation of the result data, which is one of the effects of massive node devices performing result data calculations.) Especially in peer-to-peer computing systems, result data often does not return to the initiating point. The purpose of initiating tracing is to determine what purpose needs to be achieved based on the tracing result. For example, it may be necessary to determine to whom the tracing result that meets the conditions should be given, i.e., to "backward calculate" the node device that meets the tracing result. This requirement to provide the tracing result will be added to its own result data, which will then drive the terminal devices connected to it or the terminal devices of several other node devices to output.

[0107] This invention, by being able to operate independently of a specific transmission path, distinguishes itself from existing technologies that rely on a specific path to determine events (a specific path is typically a description of a single target, usually required for single-point identification in event determination). It achieves event determination or response from a different technical perspective. The response includes the response from the execution device connected to the node device. In this embodiment, the node device is equipped with one or more output modes, including hardware / software interface output, screen output, indicator light signals, and electromagnetic wave signals. Each output module is implemented as a corresponding execution device.

[0108] In this implementation, a given node device can selectively send its result data to specific other node devices, depending on the implementation requirements. In practice, the result data of a particular node device does not necessarily need to be received by all other node devices; some node devices can be selected as the next layer of subsequent node devices to receive it.

[0109] In this invention, node devices in a peer-to-peer computing system collect any perceptible data (i.e., raw data) within their sensing range using all the sensors they are equipped with. Simultaneously, each node device also receives result data output from other node devices, calculates its own result data, and outputs it. Thus, the entire peer-to-peer computing system continuously senses, collects, inputs, outputs, and calculates data. Based on this, result data propagates among the node devices in the peer-to-peer computing system. Based on this result data, multiple node devices perform collaborative computation, enabling the execution devices connected to the node devices to respond to one or more combinations of behaviors, attributes, states, or events of one or more targets.

[0110] This invention differs from traditional methods in that what is actually transmitted between node devices (including between node devices in a peer-to-peer computing system) is a state representation (such as a representation through state values), not the state data itself. In this invention, the state representation can be a simplified transcoding, which can discard and compress some original information while achieving the accuracy of the event target, or it can be simplified transcoding to maintain the original accuracy, thereby enabling communication between node devices with very small bytes, while ensuring that a massive number of node devices can accurately calculate and obtain the behavior, attributes, states, or events of massive targets when needed (wherein, the acquisition is often unnecessary).

[0111] For example, in a simplified embodiment, target a is sensed sequentially by node devices 1, 2, and 3, all the way to node device 9. The target sensing data output by node device 2 to node device 3 is target a from node device 1 to node device 2, i.e., 1-2, which can be described by a single state value in a matrix. Subsequently, node device 3 also senses target a, and its sensing result is added in a matrix, i.e., 1-2-3, and sent to the target node device. 1-2-3 is still a state value in the matrix. Similarly, the target sensing data transmitted by node device 9 to the next group of node devices is target a from node device 1 to node device 9, i.e., node device 1-2-3-4-5-6-7-8-9, which is also still just a single state value in the matrix. In practice, an intermediate judgment result for target 'a' is not typically generated. Instead, node device 1 processes its sensor perception along with the result data received from preceding node devices to obtain a matrix-form expression value (containing state values ​​representing the category perception data of target 'a'), and then passes it to node device 2. Similarly, node device 2 calculates the result data, processes it to obtain a matrix-form expression value (containing state values ​​representing the category perception data of target 'a'), and then passes it to node device 3. This process encompasses the transmission path from node devices 1-2-3-4-5-6-7-8-9 (perceiving target 'a'), and also covers various other perceptible details, as well as multiple states and characteristics of other targets.

[0112] In a full coverage scenario, the path information involved in the node devices is often multi-dimensional. Correspondingly, the node devices will receive multiple result data. For example, node device x may receive tracking sequences of target a from node devices corresponding to paths 123, 825, abc, dbm, lmn, etc. Node device x will aggregate these result data into a matrix for expression. Therefore, although these node devices cover target a from their own different paths, at the convergence point of all paths, node device x, the result data of all node devices for target a will be expressed in matrix form at the position of node device x. The expression result is still a state value.

[0113] Although this invention does not require determining the target's behavior, attributes, state, or events by generating a path, the result data output by each node device reflects the preceding result data, that is, the preceding behavior, attributes, state, or events of the target are integrated into the preceding state. Therefore, based on the preceding result data received by the current node device and the current raw data perceived by itself, calculations are performed to drive the execution device connected to the current node device to respond, thus achieving an effect similar to that obtained by "reverse calculation". For example, when it is necessary to find the location of target 'a' 15 minutes ago, the node device receiving the search request adds the search request to its current result data calculation and transmission. Then, the result data calculated by the node device will likely include node devices that previously sensed target 'a', until all node devices that sensed target 'a' 15 minutes ago have corresponding representations in their result data. After transmission by several node devices, the node device suitable for outputting to the requester outputs the result data that meets the requirements. Furthermore, the node device does not need to store the original data about target 'a'. That is, based on this invention, it is not necessary to identify the original data to find target 'a', but rather to first obtain the location of target 'a' 15 minutes ago through a process similar to "backtracking", and then feed it back to the requester through the execution device on some node devices.

[0114] In practice, different node devices may collect different types of sensor data due to the different sensor types they are configured with. Consequently, the types of sensing data collected by different node devices may also be different. Therefore, if the current node device does not have one or more types of sensors configured by the preceding node device, the current node device will still use the preceding result data containing the corresponding type of sensing data sent by the preceding node device as input and process it synchronously with the sensing data collected by the current node device's sensors to avoid the loss of some types of sensing data during the transmission of result data, which would affect the integrity of the result data.

[0115] In this invention, the resulting data is a simplified transcoding format of a multidimensional matrix. Specifically, it involves: simplifying and transcoding different types of perception data belonging to different targets collected by different sensors on the node device according to transcoding rules to obtain matching perception codes; constructing a perception multidimensional matrix using the perception codes corresponding to all current raw data collected by the node device; and combining the perception multidimensional matrix with the preceding result data in multidimensional matrix form to calculate the current result data in multidimensional matrix form, representing the current comprehensive state of the target's current behavior, attributes, and status. The preceding result data in multidimensional matrix form is also implemented using the method of "simplifying and transcoding different types of perception data belonging to different targets collected by different sensors on the node device according to transcoding rules to obtain matching perception codes; and constructing a perception multidimensional matrix using the perception codes corresponding to all current raw data collected by the node device."

[0116] In specific implementation, the multidimensional matrix obtained by the node device, which contains the target's behavior, attributes, state, and events, is simplified and transcoded. This simplification is then repeated. Furthermore, to further simplify the complex original data and facilitate the transcoding of the multidimensional matrix, this embodiment exhaustively enumerates all perceptual abbreviations and the perceptual multidimensional matrices they form, assigning a specific abbreviation value to each perceptual multidimensional matrix. An M-bit N-ary sequence is constructed, with each bit of the N-ary sequence assigned one of K different symbols. Each specific value of the M-bit N-ary sequence is matched one-to-one with each abbreviation value; where the value of M×N is greater than the number of abbreviation values. In practical implementation, once the peer-to-peer computing system is built, parameters such as the location, time, sensor type, and quantity of node devices can be clearly defined, both at the time of construction and during subsequent maintenance and updates. Therefore, enumerating all sensing codes and their corresponding multidimensional sensing matrices is quite simple, and the data volume is not large. For example, if there are 100 million possibilities, 10,000 different symbols can be used to construct a base-10,000 sequence, i.e., a 2-digit base-10,000 sequence, such as X1X0. X1 and X0 can each be assigned one of the 10,000 different symbols, resulting in a total of 100 million values ​​for X1X0, each corresponding to one of the 100 million possibilities. If a larger amount of information is required during implementation, the number of digits in the base-10,000 sequence can be increased according to implementation needs, such as X3X2X1X0, which has a total of 10... 16 Each value corresponds to one out of a quadrillion possibilities. The matrix identifier sequence can be distributed and encrypted using non-volatile storage on various node devices, expressed through machine learning models, or represented through chip circuit structures.

[0117] In this embodiment, one implementation of the transcoding rule is as follows: different types of sensing data belonging to different targets collected by different sensors set on the node device are predefined into several levels, and all levels are divided into several level segments; a corresponding level segment identifier is assigned to each level segment, and a predefined type identifier is assigned to each type of sensing data. The type identifier + level segment identifier constitutes the original code parameter; the difference between the level of the sent type of sensing data and the first value of its corresponding level segment is set as a correction value.

[0118] Multiple sets of original code parameters that need to be sent simultaneously are arranged into an original code parameter string, and the correction values ​​are arranged into a correction value string. The original code parameter string is converted into a matching abbreviated code value, and the correction value string is converted into a matching correction code. The abbreviated code value + correction code is sent to the subsequent node device.

[0119] After receiving the abbreviated code value and correction code, subsequent node devices can translate the abbreviated code value back into the original code parameter string, and then decompose the original code parameter string into several original code parameters corresponding to the current node device; they can also translate the correction code back into a correction value string, and then decompose the correction value string into several correction values ​​corresponding to the current node device; finally, they add the correction values ​​to the first value of the level segment corresponding to the original code parameter to obtain the level of the perception data of the type sent by the current node device. (In most cases, subsequent node devices will directly process the abbreviated code value and correction code as one of their inputs, without the need for translation and decomposition.)

[0120] The simplified transcoding approach described in this invention first simplifies the complex data before restoring it. In this invention, all original code parameter strings of the current node device are enumerated first, and a specific simplified code value is assigned to each original code parameter string. In subsequent node devices, the correspondence between the original code parameter strings and simplified code values ​​is the same as that of the current node device. Similarly, all correction value strings of the current node device are enumerated first, and a specific correction code is assigned to each correction value string. In subsequent node devices, the correspondence between the correction value strings and correction codes is the same as that of the current node device. Compared with existing technologies, only the most concise simplified code values ​​are transmitted during transmission, greatly saving bandwidth utilization. Without expansion, it can be used to transmit more complex and larger amounts of data. Especially for complex data with a large number of enumerated parameters, the method of segmenting and then correcting, compared to defining a parameter level for each level of parameter and then enumerating the original code parameters, avoids the deficiency of insufficient simplification of simplified code values ​​when the latter is applied to complex data with a large number of enumerated parameters.

[0121] To achieve the goal of simplifying the transmitted content, the word length of the abbreviated code value is less than the word length of the original code parameter string, and the word length of the correction code is less than the word length of the correction value string. In this invention, the number of bits in the abbreviated code value and the correction code is determined based on the number of exhaustively enumerated original code parameter strings and correction value strings, so that the number of original code parameter strings and correction value strings that match the abbreviated code value and the correction code is not less than the number of exhaustively enumerated original code parameter strings and correction value strings.

[0122] As an easy-to-implement implementation method, in this invention, if the number of exhaustively enumerated original code parameter strings and correction value strings is less than 256, then both the simplified code value and the correction code are represented using ASCII code, with each original code parameter string and correction value string matched with one ASCII code. One ASCII code consists of 8 binary number combinations, representing 256 possible characters. Each character can be used to represent one original code parameter string or one correction value string. One original code parameter string plus one correction value code represents a combination of multiple sets of parameters transmitted in parallel each time, based on a predetermined number of parameters.

[0123] Furthermore, if the number of exhaustively enumerated original code parameter strings and correction value strings is greater than the number of ASCII codes, then each original code parameter string and correction value string matches at least two ASCII code bits, and these multiple ASCII code bits are arranged to form an ASCII code string. Specifically, the ASCII code string includes a certain number of ASCII code bits, and the number of original code parameter strings and correction value strings matched by the ASCII code string is P = 256. n Where n = 1, 2, 3, ... That is, in order to ensure that the number of possible characters included in the ASCII code string is greater than the number of exhaustive original code parameter strings and correction value strings, in practice, the number of bits in the ASCII code string is determined based on the number of exhaustive original code parameter strings and correction value strings, so that the number of original code parameter strings and correction value strings matched by the ASCII code string is not less than the number of exhaustive original code parameter strings and correction value strings.

[0124] The above embodiments using ASCII code are merely a convenient example for illustrating the present invention. The simplified code value and correction code described in this invention can also be other binary codes with a preset number of bits (such as 8 bits or 1 byte in ASCII code), and the number of bits in the binary code forms a 2... n Count of characters.

[0125] For example, consider sending performance status parameters for three devices a, b, and c. The performance status of devices a, b, and c is collected. Each device has 30 levels of parameters, from 0 to 29. Exhaustively listing the original parameter strings yields 27,000 combinations. The acquired performance status data for the three devices is represented by 9 characters, such as a15b27c18. In computer storage and transmission, these 9 characters require 9 eight-bit binary codes, meaning a total of 9 bytes (72 bits) of information need to be transmitted.

[0126] In this embodiment, the parameters of each device are classified into six segments, such as a0-4, a5-9, a10-14, a15-19, a20-24, and a24-29. The same applies to devices b and c. A parameter segment identifier is generated based on which segment the parameter level to be sent falls into. This results in 216 possible combinations of parameter segments for the three devices, which can be represented using only one ASCII character. A parameter correction table is predefined in the current node device. The correction value represents the numerical increment of the parameter level within its corresponding segment, i.e., the difference between the parameter level and the first value of the segment. For example, the correction values ​​for a15b27c18 are 0, 3, and 3 respectively, so the correction value string is 033. The parameter correction table is an exhaustive combination of correction value strings. In this embodiment, there are 125 possible correction value strings, which can also be represented using one ASCII character. It can be seen that this invention can transmit 9 bytes of information using only two bytes. For example, the combination a15b27c18 corresponds to the character "p" in the transcoding mapping table and the character "k" in the parameter correction table. "pk" is transmitted to subsequent node devices. These subsequent node devices pre-set a transcoding mapping table and a parameter correction table with content completely identical to the current node device. The transcoding mapping table shows that "p" corresponds to the level segments a15-19, b24-29, and c15-19, while the parameter correction table shows that "k" corresponds to 0, 3, and 3. The correction value is added to the first value of the level segment to reconstruct a15b27c18.

[0127] Through the above process, 9 bytes of information can be transmitted using only 2 bytes, thus achieving the purpose of this invention.

[0128] Combining multidimensional matrix encoding, the above example can be evolved as follows: the xyz coordinates of the abnormal temperature point A that protrudes from the background observed by thermal camera 1 can be represented by abc (equivalent to recording data for the target in only three dimensions, and so on for multidimensional ones). After simplified transcoding, the data of A in the multidimensional matrix is ​​pk. In the multidimensional matrix output by the node device, the observation result of thermal camera 1 is Apkmw, the observation result of thermal camera 2 is Cabdef, and the data of other sensors are similar, ultimately forming a multidimensional matrix.

[0129] Based on the simplified transcoding described above, in specific implementation, according to implementation requirements, the result data obtained by the node device in the multidimensional matrix can be represented as the data of the multidimensional matrix, the state value of the multidimensional matrix data, or the result of simplified transcoding of the multidimensional nested matrix data.

[0130] Node devices perceive targets and obtain result data covering the target's behavior, attributes, state, or events. The result data obtained by each node device includes multiple target behaviors, attributes, states, or events with corresponding time and / or spatial parameters; time and / or space are used as dimensions in a multidimensional matrix. In this embodiment, relatively stable basic data such as the installation location, sensor type, surrounding facilities and environment, terrain, purpose, and attributes of each node device can actually serve as the basis for multidimensional matrix calculation. The multidimensional matrix outputs a label of these basic data, without needing to put these basic data into the multidimensional matrix; furthermore, the combinations of label values ​​are finite, and a transformation value can be output by combining them to participate in the calculation of result data by other node devices.

[0131] In this invention, the sensors of the node devices perceive the target, and the perception time is used as one dimension of a multidimensional matrix. All node devices are synchronized in time to ensure consistency in perception time and minimize the impact of time errors on the accuracy of the resulting data. Alternatively, each type of data perceived by each type of sensor can be used as one dimension of the multidimensional matrix (depending on the sensor, some sensors can perceive multiple types of data), or the types of sensors that meet the correlation criteria can be combined as one dimension of the multidimensional matrix.

[0132] In this invention, the current node device selectively sends its current result data to specific other node devices. The principle of selective sending is based on preset fixed rules, dynamic rules, or a model trained through machine learning to determine the specific other node devices. The model trained through machine learning is a reinforcement training model based on the node itself, using the previous result data of the node device as annotations for reinforcement training.

[0133] Based on the principle of selectivity, the node devices that need to transmit result data are added to the node list. Then, the result data calculated by the node device (i.e., after simplified transcoding) is sent to the node device selected from the list of communicably reachable nodes according to the principle of selective transmission.

[0134] If the execution device connected to the node device needs to obtain basic data outside the peer-to-peer computing system to determine the response to one or more combinations of behaviors, attributes, states, or events of one or more targets, the node device connected to the peer-to-peer computing system with accessibility data acquisition conditions (including one or more combinations of system departments, operation locations, authorized personnel identities, triggering conditions, time, and collected values ​​corresponding to the response, all serving as the basis of a multidimensional matrix, which is a measure of these basic elements) performs basic data acquisition. The data acquired through accessibility data acquisition is then used as input to the node device for calculation to obtain the current result data. Based on this, the present invention can solve effects that existing technologies cannot achieve. For example, existing methods stipulate that if a hospital receives more than 10 cases of food poisoning, it is considered an emergency event and needs to be reported. Without specific aggregation, if 5 hospitals receive a total of 10 food poisoning cases from the same school, an emergency response will not be triggered. However, the present invention, based on the peer-to-peer computing system, can detect if 5 hospitals have received a total of 10 food poisoning cases from the same school, thereby triggering an emergency response.

[0135] This invention utilizes deep learning technology to train data processing models for node devices via neural networks. Based on task performance requirements, these models receive result data from other node devices and real-time dynamic environmental changes captured by connected sensors. They then collaborate with other node devices to perform cognition and response. During this process, each node device automatically optimizes the computation, structure, and parameters of its own deep learning neural network (i.e., the corresponding data processing model), thereby achieving environmental adaptation and continuous evolution of the data processing model. In other words, in this invention, each node device automatically optimizes the computation, structure, and parameters of its data processing model each time it processes data and calculates results. That is, the data processing model changes its computation, structure, or parameters to achieve environmental adaptation and continuous evolution. Here, a change in the data processing model (i.e., no change) is one effect of optimization; even a change in the data processing model (i.e., no change) can be considered a form of change.

[0136] In this invention, to accurately reflect the mutual influence between node devices and determine whether the data processing model of the current node device is suitable for its connected sensors and actuators, the data processing model loaded on the node device performs adaptive perceptual calculations on the data processing models of preceding node devices. Specifically, when a node device processes result data from other node devices, the adaptive perceptual calculations include verifying the result data of the current node device with that of other relevant node devices and / or verifying the historical result data of the current node device to obtain an adaptability value or as input to the adaptability judgment model. In this invention, based on collaborative computing, the results of the verification calculations are reflected in the result data of the node devices, enabling the result data of a series of nodes to drive the model training facility to make greater adjustments to the data processing model of the problematic node device. The implementation of the adaptability value can serve as an alternative to the aforementioned adjustment method based on collaborative computing.

[0137] When the data processing model of a certain node device cannot adapt to the sensors and actuators it is connected to, the adaptive value of its adaptive perception calculation has the following impact on the current result data of the current node device: a preset flag bit is assigned a value in the current result data of the current node device; or, other related node devices perform comprehensive calculations on the current result data of the current node device based on their own data processing models, and when the resulting data can be used as input by subsequent node devices, the resulting data is used by other node devices for calculation, driving the adjustment of the data processing model of the current node device.

[0138] For data processing models, adaptive perceptual computing is encompassed within other computing (including the calculation of result data, the adjustment of data processing models, etc.). In other words, a single round of computing includes both other types of computing and adaptive perceptual computing, meaning that adaptive perceptual computing and the calculation of result data are performed simultaneously.

[0139] In this invention, the adjustment of the data processing model occurs during the current calculation of the result data or during several data transfers and calculations of the result data. Specifically, the data processing model of the node device should cover the following modes:

[0140] Mode 1: When the adaptability value of a node device's data processing model exceeds a threshold, the node device's output data is added to the list of receiving nodes for the connected model training facility. Upon receiving this data, the connected node device will include a driving command for the model training facility in its current calculation result data. The threshold value can be a preset value or a dynamic value, depending on the implementation requirements. If the threshold value is dynamic, it is determined by a pre-set program or by a dynamically updated data processing model. In this invention, based on collaborative computing, the model training facility can be driven to adjust the data processing model of the problematic node device during the node device's calculation of the result data. The implementation of the driving command can serve as an alternative to the aforementioned collaborative computing-based adjustment method.

[0141] Mode 2: The model training facility is regarded as a type of sensor. The signals emitted by it are added to the connected node devices. After the node devices calculate and send out the result data, the corresponding node devices in the peer computing system will drive the data storage module of the connected sensor after calculating the result data. By establishing a file transfer channel between the node devices in the peer computing system or by establishing a file transfer channel through other network communication modes, the sensing data specified by the model training facility is sent to a specific location to participate in model training and improvement. The improved data processing model will be deployed in the same way to node devices that the original data processing model cannot adapt to.

[0142] In another implementation, the model training facility can also drive multiple node devices to generate new data processing models and update the data processing models by using their own perception data in a federated computing manner.

[0143] In this invention, each node device automatically optimizes its own data processing model's operations, structure, and parameters each time it processes data and calculates the result data. That is, the data processing model changes its own operations, structure, or parameters. These changes originate from the entire peer-to-peer computing system's comprehensive perception of various targets, events, and their combinations. Therefore, this change is a kind of "memory" of the entire peer-to-peer computing system for various targets, events, and their combinations, which can improve the computational effect of "recalling" and "tracing".

[0144] The above embodiments are merely illustrative of the present invention and are not intended to limit the invention. Any changes or modifications to the above embodiments based on the technical essence of the present invention will fall within the scope of the claims of the present invention.

Claims

1. A vehicle identification method of non-vehicle identity, characterized by, Driver identity information is bound to a vehicle, the vehicle is not provided with vehicle identity, feature information of the vehicle is extracted based on image recognition as an identification sample; when the following conditions are met at the same time, the vehicle to be identified is determined to be itself, and vehicle identification is completed; The conditions include: Condition one: driver identity identification is performed on a user located at a vehicle driving position to obtain driver identity information; Condition two: feature information of the vehicle to be identified is collected and compared with the identification sample to determine that the vehicle to be identified and the vehicle corresponding to the identification sample are the same vehicle; When the conditions are not met at the same time, the first satisfied condition is retained until all conditions are met, and then vehicle identification is completed; Non-specific feature identification and location identification of a target are performed by using a peer-to-peer computing system, the target including a vehicle and a driver; The peer-to-peer computing system includes a plurality of node devices, and there is no primary and secondary relationship between all node devices; the node devices are provided with data acquisition devices and operation modules, the data acquisition devices including at least one type of sensor including an image acquisition device, for collecting different corresponding types of sensing data; at least one point sample of the target is collected by node devices arranged at different collection positions, and the point sample is sensing data of a corresponding sensor type; For a certain node device, the collected sensing data is processed to obtain result data, and the result data is propagated to other node devices; the other node devices receiving the result data take the result data as one of the collected original data, and the result data influences the result data of other node devices; based on this, without obtaining the identity information of the target, the plurality of node devices in the peer-to-peer computing system perform cooperative calculation to determine that each unique target is itself, realize non-specific feature identification, and perform location identification on the target; The result data is a format result of simplified transcoding of a multi-dimensional matrix, specifically: different types of sensing data collected by different sensors of the node device belonging to different targets are simplified and transcoded based on a transcoding rule to obtain matched sensing simple codes; a sensing multi-dimensional matrix is constructed based on the sensing simple codes corresponding to all current original data collected by the node device; the sensing multi-dimensional matrix is combined with previous result data in the form of a multi-dimensional matrix to calculate current result data in the form of a multi-dimensional matrix, representing the current comprehensive state of the behavior, property and state of the target.

2. The non-vehicle identity vehicle identification method of claim 1, wherein, For any condition, if the condition is met, and the condition cannot be maintained at a subsequent time, it is determined that the condition at the previous time is not met.

3. The non-vehicle identity vehicle identification method of claim 1, wherein, At least one sensing data of the to-be-identified target is collected at different collection positions, the to-be-identified target including a user at a driving position of the vehicle and the to-be-identified vehicle; the continuously collected sensing data is associated with a unique temporary condition pointing to the current to-be-identified target until the identification of the to-be-identified target is completed, and the temporary condition is modified to the condition corresponding to the to-be-identified target; the identification of the to-be-identified target includes the obtaining of the driver identity information according to condition one and the determination of the to-be-identified vehicle and the vehicle corresponding to the identification sample as the same vehicle according to condition two.

4. The non-vehicle identity vehicle identification method of claim 3, wherein, When there is a blind area between two adjacent collection positions, and no sensing data is collected in the blind area, it is determined whether the sensing data collected at the first collection position and the sensing data collected at the second collection position have a coherent logic, and if so, the to-be-identified target at the second collection position inherits the condition satisfaction of the to-be-identified target at the first collection position. The coherent logic is that according to the type of the sensing data, the sensing data collected at the previous collection time can be inferred to obtain the inferred sensing data at the subsequent collection time, and the inferred sensing data at the subsequent collection time matches the sensing data collected at the subsequent collection time.

5. The non-vehicle identity vehicle identification method of claim 1, wherein, Driver identity identification is performed before and / or after the vehicle travels. The method for performing driver identity identification before the vehicle travels is that when it is identified that a user enters the vehicle from the driving position of the vehicle, it is determined that the user is the driver, and driver identity identification is performed; or when it is identified that two users enter the vehicle from the co-driver side of the vehicle, it is determined that the first entering user is the driver, and driver identity identification is performed. The method for performing driver identity identification after the vehicle travels is that when it is identified that a user gets off from the driving position of the vehicle, it is determined that the user is the driver, and driver identity identification is performed; or when it is identified that two users get off from the co-driver side of the vehicle, it is determined that the second getting-off user is the driver, and driver identity identification is performed.

6. The non-vehicle identity vehicle identification method of claim 1, wherein, The characteristic information of the vehicle includes the vehicle model, color, and modification, the vehicle model is confirmed based on image recognition, and the vehicle model parameters of the vehicle model are obtained from a preset vehicle model database.

7. The vehicle identification method of non-vehicle identity according to any one of claims 1 to 6, characterized in that, During the travel of the vehicle, the identification sample, the characteristic information of the to-be-identified vehicle continuously collected, the identification result of the to-be-identified vehicle continuously obtained, and the driver identity information are stored by using a block chain.

8. The non-vehicle identity vehicle identification method of claim 1, wherein, The current node device receives the result data output by other node devices; for the current node device, the collected sensing data is combined with the result data from other node devices to calculate the result data of the current node device, and the result data is sent to other node devices. The node devices in the peer-to-peer computing system perform collaborative calculation with the collection of sensing data and the calculation of result data.

9. The non-vehicle identity vehicle identification method of claim 8, wherein, In the peer-to-peer computing system, for a certain point sample of a certain target, the subsequent node device adjusts the perception attention according to the characteristics of the point sample from the result data transmitted by the node device collecting the point sample to other node devices, or reports the characteristics of the point sample for the subsequent node device to adjust the perception attention; if the subsequent other node device does not detect the characteristics of the point sample, but can determine from the characteristics of other point samples that the characteristics of the point sample that has failed to be detected still belong to the target, then the characteristics of the point sample that has failed to be detected are still expressed in the result data of the current node device and transmitted to other node devices.

10. The non-vehicle identity vehicle identification method of claim 9, wherein, The method for reporting the characteristics of the point sample for the subsequent node device to adjust the perception attention is: adjusting the parameters of the data processing model of the subsequent node device according to the result data provided by the previous node device expressing the characteristics of the point sample, or the characteristics of the point sample, so that the subsequent node device improves the computing power for identifying the characteristics of the point sample; or the subsequent node device uses the perception attention model to match the received characteristics of the point sample or the result data expressing the characteristics of the point sample for computing power adjustment.

11. The non-vehicle identity vehicle identification method of claim 10, wherein, When the node device processes the result data output by several previous node devices, based on the data processing model, when the targets described by the several previous node devices can be determined as the same target through certain common point sample characteristics, the point sample characteristics and other information described by each node device are combined into the same target.

12. The non-vehicle identity vehicle identification method of claim 11, wherein, When the node device receives the result data indicating that the mark used by the current node device to identify the target before the current time receiving the result data is different from the mark used by other node devices to identify the target, and the mark allocated by other node devices for the target is updated, the mark used by the current node device to identify the target before the current time receiving the result data is converted.

13. The vehicle identification method of non-vehicle identity according to claim 12, characterized in that, The method for converting the mark used by the current node device to identify the target before the current time receiving the result data is: Replacing the mark used by the current node device to identify the target before the current time receiving the result data with the latest mark allocated by other node devices for the target; Or, recording the conversion relationship between the mark used by the current node device to identify the target before the current time receiving the result data and the mark allocated by other node devices for the target after the update, and converting when referring to the result data currently received by the current node device; Or, the node device deploys a conversion model to convert the marks of multiple targets according to the input original data or result data.

14. The non-vehicle identity vehicle identification method of claim 9, wherein, For one or more point samples collected by node devices at different collection positions in sequence, if the characteristic values of a certain point sample or more point samples at different collection positions respectively meet the preset proximity condition or are determined by a certain model to have relevance reaching a threshold, and are unique at each collection position, it is determined that the point sample of the certain type at different collection positions has relevance.

15. The non-vehicle identity vehicle identification method of claim 9, wherein, If one or more point samples collected by node devices at different collection positions are associated, the node devices at different collection positions collect the same space field, and there is only one target in the space field, or the point samples can correctly point to one of the multiple targets to which they belong.

16. The vehicle identification method of non-vehicle identity according to claim 15, wherein, The data collection device of the node device includes one or a combination of several of image acquisition devices, electromagnetic induction devices, temperature measurement devices, vibration frequency sensing devices, and laser radars. The data collected by the above devices and the three-dimensional point cloud collected by the laser radar or generated from the images collected by multiple image acquisition devices are jointly calculated to obtain three-dimensional points with data. The image color, contour, line, reflectivity, motion trend, electromagnetic characteristics, temperature, temperature change trend, vibration frequency, and vibration frequency change trend based on two-dimensional perception are used as additional attributes of the corresponding three-dimensional points to form a three-dimensional point cloud with attributes. The correspondence between each region of the three-dimensional point cloud with attributes and each part or associated part of the 3D appearance of the target is determined based on the electromagnetic induction, temperature law, vibration frequency change characteristics, motion correlation, and reflectivity.

17. The non-vehicle identity vehicle identification method of claim 1, wherein, When the identity information of the target is needed, an identity information acquisition command is triggered, and the identity information acquisition command is used as one of the inputs for the calculation of the result data of the node device. The corresponding result data is obtained by driving the node device connected to the barrier-free data collection condition that can obtain the identity information of the target in the peer-to-peer computing system, thereby achieving the acquisition of the identity information of the target.

18. The non-vehicle identity vehicle identification method of claim 17, wherein, The peer-to-peer computing system verifies the identity information of the target and determines its authority. The node device that can obtain the identity information in the peer-to-peer computing system does not provide the identity information, but only expresses the verification result of the identity information in the result data of the node device based on the verification requirement of the identity information in the received result data.

19. The non-vehicle identity vehicle identification method of claim 18, wherein, The node device that can obtain the identity information in the peer-to-peer computing system does not provide the identity information, and the information source device that provides the identity information is driven to establish an encrypted file transmission channel or an encrypted information transmission channel in other network communication modes between the input terminal of the node device that needs to obtain the identity information. The identity information is used as one of the inputs of the node device.

20. The non-vehicle identity vehicle identification method of claim 1, wherein, The data collection device includes one or more of image acquisition devices, audio acquisition devices, temperature measurement devices, vibration frequency sensing devices, laser radars, chemical sensors, and electromagnetic induction devices.

21. The non-vehicle identity vehicle identification method of claim 1, wherein, During the driving of the vehicle, the peer-to-peer computing system performs non-medium storage of the continuously collected raw data based on the result data calculation. The raw data includes perception data and result data received by the node device. The non-medium storage method is that each node device processes the collected original data by using a data processing model deployed by the node device, calculates result data, and transmits the result data to other node devices; other node devices receiving the result data take the result data as one of the collected original data, and calculate the result data of the current node device by using the data processing model; specifically, the current node device receives the previous result data output by the previous node device, combines the current original data collected by the current node device, calculates the current result data, and transmits the current result data to the subsequent node device; in this way, the result data is transmitted between node devices, and the result data represents the sensing state sensed by the sensor connected to the node device, the state of the node device, and the state of the execution device connected to the node device. Through the calculation of the result data and the connection relationship between the node devices, the non-medium storage of the sensing state sensed by the sensor connected to each node device, the state of each node device, and the state of the execution device connected to each node device is completed.

22. The vehicle identification method of non-vehicle identity according to claim 21, wherein, Different node devices collect original data at different times and / or in different spaces; if the behavior, attribute, state or event of a target needs to be traced back, the current node device takes the requirement of the target tracing back as one of the inputs, and combines the other previous result data and the current original data to calculate the current result data, and then sends the current result data to other node devices; in this way, based on the previous result data and the current original data received by the current node device, the relationship between the previous state value representing the behavior, attribute, state or event of the target in the original data collected by each node device in the previous sequence and the current state value representing the behavior, attribute, state or event of the target in the current result data of the current node device is calculated, and the non-medium reading is completed through the calculation of the result data. The current original data represents the current state of the behavior, attribute, state or event of the target, the previous result data represents the previous comprehensive state of the behavior, attribute, state or event of the target, the current result data represents the current comprehensive state of the behavior, attribute, state or event of the target, and the current comprehensive state is the combination of the current state and the previous comprehensive state.

23. The vehicle identification method of non-vehicle identity according to claim 21 or 22, characterized in that, When the node device collects the original data, the original data collected by each node device has a corresponding time parameter and / or space parameter; the time and / or space are taken as dimensions in a multi-dimensional matrix; the sensor of the node device senses the target, and the sensing time is taken as one dimension in the multi-dimensional matrix; each type of data sensed by each type of sensor is taken as one dimension in the multi-dimensional matrix, or the types of type sensors meeting the correlation condition are combined as one dimension in the multi-dimensional matrix.