A body-aware agent location perception method and system based on wireless radio frequency fingerprints

Abstract

This invention discloses an embodied intelligent agent position perception method and system based on wireless radio frequency fingerprinting, belonging to the field of embodied intelligent agent wireless position perception technology. The method includes: using a full-band multi-directional wireless signal acquisition device to collect radio frequency signals in a time-division manner and obtain digital signals containing radio frequency fingerprints; clustering and grouping the digital signals, assigning identity authentication IDs, and removing invalid data; training a classification model using the grouped data and realizing ID recognition; determining relative position by combining angle arrival algorithm and signal power strength; dynamically sensing relative position time changes based on its own position changes; and performing the clustering to dynamic sensing steps in a long-cycle loop to realize the dynamic discovery and deletion of intelligent agents. This invention is based on physical layer autonomous perception, requires no central node, and has the characteristics of low latency, low power consumption, and high accuracy. It is suitable for the dynamic collaborative needs of multiple intelligent agents in power scenarios and can be widely used in embodied intelligent agent collaborative operations in power scenarios such as substations and converter stations.

Description

Technical Field

[0001] This invention belongs to the field of integrated wireless communication sensing and embodied intelligence technology, specifically relating to an embodied intelligent agent location sensing method and system based on wireless radio frequency fingerprinting. Background Technology

[0002] As the power industry transforms and upgrades from traditional to new power systems, a large number of data acquisition, monitoring, and transmission devices are needed in power scenarios such as substations, converter stations, and transmission lines to improve management efficiency. Using traditional equipment would inevitably increase the workload of frontline workers. To alleviate the workload of frontline personnel and improve the efficiency of substation inspections, a large number of drones, quadruped robots, and wheeled robots with embodied intelligence are currently deployed in power scenarios for scene inspections. However, the currently deployed embodied intelligent devices mainly rely on pre-set locations and logic to operate, resulting in limited flexibility and practicality. They cannot effectively achieve rapid collaboration between multiple embodied intelligent agents, nor can they efficiently utilize their respective advantages to form a regional intelligent management system. This limits the working capacity and efficiency of these embodied intelligent devices, leading to resource waste and failing to effectively replace manual labor. Furthermore, embodied intelligent devices have blind spots in their inspection field of view, and their inspection granularity is limited. Therefore, improving the collaboration of embodied intelligent devices is both urgent and imperative.

[0003] The prerequisite for multi-agent collaboration is that agents can perceive each other's presence and know each other's location. Interaction, communication, negotiation, and data sharing only occur after perceiving each other's presence. Therefore, location awareness technology for multi-agents is a prerequisite for transforming multi-agent intelligence from individual intelligence to collaborative intelligence. Currently, communication between agents is diverse, differing at the physical layer, data link layer, network layer, and protocol layer. For example, physical layer wireless carrier frequencies include 2.4GHz and 5.8GHz for local area networks, and 700MHz, 900MHz, 1800MHz, and 2100MHz for wide area networks, as well as other frequencies. Carrier bandwidths also vary, such as 25kHz, 200kHz, 1.4MHz, 5MHz, 10MHz, and 20MHz. Coding and modulation methods are also diverse, including BPSK, QPSK, QAM, OFDM, and DSSS. Communication protocols are also diverse. These characteristics dictate that using traditional communication protocol conversion methods to achieve location awareness for multiple embodied agents leads to various problems due to excessively long awareness times. Furthermore, various business processes consume significant hardware resources, and increased power consumption causes embodied agents to deplete their energy rapidly, ultimately affecting their long-term reliable operation. The mobile nature of embodied agents means that their relative positions change in real time, necessitating rapid location awareness among them to prevent accidental collisions due to delayed awareness and ensure smooth subsequent business collaboration. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to provide a method and system for embodied intelligent agents to perceive their location based on wireless radio frequency fingerprinting, which addresses the shortcomings of the prior art. By processing the physical layer data of embodied intelligent agents, the invention enables rapid discovery and location perception among multiple embodied intelligent agents, and reduces the resource consumption and energy consumption within the embodied intelligent agents. This invention solves the technical problems of excessively long dynamic perception time, high processing resource consumption, and excessive energy consumption of embodied intelligent agents using existing location perception technologies in power scenarios.

[0005] The present invention adopts the following technical solution: A method for location perception of an embodied intelligent agent based on radio frequency fingerprinting includes the following steps: Using a full-band multi-directional wireless signal acquisition device deployed on an embodied intelligent agent, full-band wireless radio frequency signals are acquired in a time-division manner to obtain digital signals containing the wireless hardware radio frequency fingerprint characteristics of the embodied intelligent agent. The digital signals are the data of the full-band wireless radio frequency signals after analog-to-digital conversion. The digital signals are clustered and grouped, and an identity authentication ID is assigned to each group of data; A classification model is trained using the grouped data and their corresponding identity authentication IDs, and the trained classification model is used to identify the identity authentication IDs of newly collected digital signals. The system selects the identified identity authentication ID data, uses the angle arrival algorithm combined with the wireless signal power strength to determine the direction and relative distance of the surrounding embodied intelligent agents, and calculates the relative position coordinates of the surrounding embodied intelligent agents relative to the machine. Based on the identified identity authentication ID corresponding to the embodied intelligent agent, wireless signal characteristics are collected in real time. Combined with the changes in the position of the embodied intelligent agent itself, the relative position coordinates change over time are used to dynamically perceive the changes in the relative position with the surrounding embodied intelligent agents over time, and generate the motion trajectory and velocity vector of the surrounding embodied intelligent agents.

[0006] Preferably, the full-band multi-directional wireless signal acquisition device includes N groups of radio frequency link devices, where N is an integer greater than or equal to 1; each group of radio frequency link devices is independently configured to process signals from different directions or different frequency bands simultaneously or in a time-division manner.

[0007] Preferably, each group of the radio frequency links includes: The antenna is a full-band high-gain directional antenna used to collect radio frequency signals in a specific direction; A power detector, connected to the output of the antenna, is used to detect the power value of a preset frequency band; The local oscillator can be controlled to generate a local oscillator signal; A mixer, whose input terminals are respectively connected to the output terminal of the antenna and the output terminal of the controllable local oscillator, is used to mix the radio frequency signal with the local oscillator signal to obtain a low-frequency carrier signal; An analog-to-digital converter, connected to the output of the mixer, is used to convert the low-frequency carrier signal into a digital signal by analog-to-digital conversion. The local oscillator controller is communicatively connected to both the power detector and the controllable local oscillator.

[0008] Preferably, the local oscillator controller executes the following control logic: Receive power value results from the power detector; set the local amplitude value according to the received power value; send control information to the controllable local oscillator to set the local oscillator frequency and amplitude in real time during data acquisition.

[0009] Preferably, the time-division acquisition of full-band wireless radio frequency signals includes a spatial dimension acquisition strategy: Within the same time period, control multiple groups of radio frequency link devices to simultaneously acquire signals in the same frequency band but in different spatial directions; Alternatively, multiple sets of RF link devices can be controlled to simultaneously acquire signals from different directions and different frequency bands; The above strategy is used to obtain raw sampling data containing spatial orientation information and frequency band information.

[0010] Preferably, clustering and grouping the digital signals specifically includes: The radio frequency fingerprint feature vectors are extracted from the digital signal, and the feature vectors are clustered using an unsupervised clustering algorithm; Feature vectors belonging to the same cluster are grouped together, and a unique identity authentication ID is generated for this group of data. Digital signals with excessive noise or whose feature vectors are more than a preset threshold from any cluster center are identified as unclassifiable data and removed to eliminate interference from non-smart signals.

[0011] Preferably, the number of sampling points for clustering is dynamically adjusted based on the number of known embodied agents in the current environment; When the number of embodied agents in the environment increases, the number of sampling points is automatically increased to improve clustering accuracy; when the number of embodied agents in the environment decreases, the number of sampling points is automatically reduced to reduce processing power consumption.

[0012] Preferably, the training classification model specifically includes: The grouped data is used as input samples, and the assigned identity authentication ID is used as output label to construct a training dataset. The classifier is trained using the training dataset. The classifier supports incremental learning and can incorporate newly collected digital signal features into the existing model parameters to update the classification model.

[0013] Preferably, the method of using the angle arrival algorithm in conjunction with the wireless signal power strength to determine the direction and relative distance of surrounding embodied intelligent agents specifically includes: Based on the phase difference of the same identity authentication ID signal received by antennas in different directions in the full-band multi-directional wireless signal acquisition device, the azimuth angle of the surrounding embodied intelligent agents relative to the local machine is calculated using the angle arrival algorithm; based on the received wireless signal power intensity attenuation model, the relative distance between the surrounding embodied intelligent agents and the local machine is calculated; combining the azimuth angle and the relative distance, the two-dimensional or three-dimensional relative position coordinates of the surrounding embodied intelligent agents in the local coordinate system are calculated.

[0014] Preferably, the dynamic perception and the change in the relative position with the surrounding embodied intelligent agents over time specifically include: Within consecutive time slices, continuously track the relative position coordinate sequence of embodied intelligent agents with the same identity authentication ID; By combining the machine's own motion state data, the relative position coordinate sequence is differentially processed or trajectory fitted to generate the motion trajectory and velocity vector of the surrounding embodied intelligent agents, thereby realizing real-time dynamic position perception.

[0015] Preferably, by utilizing the sequence of changes in the relative position coordinates over time, the relative position with the surrounding embodied intelligent agents is dynamically perceived to change over time. According to a preset long period T, at the beginning of each long period T, the clustering process in step S2 is re-executed to identify whether a new embodied intelligent agent has joined and formed a new cluster. If no data corresponding to a certain identity authentication ID is detected within multiple consecutive long periods T, it is determined that the embodied intelligent agent has left the environment, and the identity authentication ID and its associated historical data are deleted from the system's identity authentication ID list.

[0016] Secondly, embodiments of the present invention provide an embodied intelligent agent location sensing system based on wireless radio frequency fingerprinting, comprising: The signal acquisition module is used to acquire full-band wireless radio frequency signals in a time-division manner using a full-band multi-directional wireless signal acquisition device deployed on the embodied intelligent agent, and to obtain a digital signal containing the wireless hardware radio frequency fingerprint characteristics of the embodied intelligent agent. The digital signal is the data after analog-to-digital conversion of the full-band wireless radio frequency signal. The clustering and grouping module is used to cluster and group the digital signals, and assign an identity authentication ID to each group of data. The model training and recognition module is used to train a classification model using the grouped data and its corresponding identity authentication ID, and to recognize the identity authentication ID of newly collected digital signals using the trained classification model. The location determination module is used to select the data of the identified identity authentication ID, use the angle arrival algorithm combined with the wireless signal power strength to determine the direction and relative distance of the surrounding embodied intelligent agents, and calculate the relative position coordinates of the surrounding embodied intelligent agents relative to the local machine. The dynamic perception module is used to collect wireless signal features in real time based on the identified embodied intelligent agent corresponding to the identity authentication ID, and combine the changes in the position of the embodied intelligent agent itself with the changes in the relative position coordinates over time to dynamically perceive the changes in the relative position with the surrounding embodied intelligent agents over time, and generate the motion trajectory and velocity vector of the surrounding embodied intelligent agents.

[0017] Preferably, in the signal acquisition module, the full-band multi-directional wireless signal acquisition device includes N groups of radio frequency link devices, where N is an integer greater than or equal to 1; each group of radio frequency link devices is independently configured to process signals from different directions or different frequency bands simultaneously or in a time-division manner. Each group of radio frequency links includes: The antenna is a full-band high-gain directional antenna used to collect radio frequency signals in a specific direction; A power detector, connected to the output of the antenna, is used to detect the power value of a preset frequency band; The local oscillator is controllable and used to generate a local oscillator signal. The local oscillator controller executes the following control logic: Receive power value results from the power detector; set the local amplitude value according to the received power value; send control information to the controllable local oscillator to set the local oscillator frequency and amplitude in real time during data acquisition; A mixer, whose input terminals are respectively connected to the output terminal of the antenna and the output terminal of the controllable local oscillator, is used to mix the radio frequency signal with the local oscillator signal to obtain a low-frequency carrier signal; An analog-to-digital converter, connected to the output of the mixer, is used to convert the low-frequency carrier signal into a digital signal by analog-to-digital conversion. The local oscillator controller is communicatively connected to both the power detector and the controllable local oscillator.

[0018] Preferably, in the signal acquisition module, the time-division acquisition of full-band wireless radio frequency signals includes a spatial dimension acquisition strategy: Within the same time period, control multiple groups of radio frequency link devices to simultaneously acquire signals in the same frequency band but in different spatial directions; Alternatively, multiple sets of RF link devices can be controlled to simultaneously acquire signals from different directions and different frequency bands; The above strategy is used to obtain raw sampling data containing spatial orientation information and frequency band information.

[0019] Preferably, in the clustering and grouping module, the clustering and grouping of the digital signals specifically includes: The radio frequency fingerprint feature vectors are extracted from the digital signal, and the feature vectors are clustered using an unsupervised clustering algorithm; Feature vectors belonging to the same cluster are grouped together, and a unique identity authentication ID is generated for this group of data. Digital signals with excessive noise or whose feature vectors are more than a preset threshold away from any cluster center are identified as unclassifiable data and removed to eliminate interference from non-intelligent signals. The number of sampling points for clustering is dynamically adjusted based on the number of known embodied agents in the current environment; When the number of embodied agents in the environment increases, the number of sampling points is automatically increased to improve clustering accuracy; when the number of embodied agents in the environment decreases, the number of sampling points is automatically reduced to reduce processing power consumption.

[0020] Preferably, in the model training and recognition module, the training classification model specifically includes: The grouped data in the clustering module is used as input samples, and the assigned identity authentication ID is used as output label to construct a training dataset. The classifier is trained using the training dataset. The classifier supports incremental learning and can incorporate newly collected digital signal features into the existing model parameters to update the classification model.

[0021] Preferably, in the location determination module, the determination of the direction and relative distance of surrounding embodied intelligent agents using an angle arrival algorithm combined with wireless signal power strength specifically includes: Based on the phase difference of the same identity authentication ID signal received by antennas in different directions in the full-band multi-directional wireless signal acquisition device, the azimuth angle of the surrounding embodied intelligent agents relative to the local machine is calculated using the angle arrival algorithm; based on the received wireless signal power intensity attenuation model, the relative distance between the surrounding embodied intelligent agents and the local machine is calculated; combining the azimuth angle and the relative distance, the two-dimensional or three-dimensional relative position coordinates of the surrounding embodied intelligent agents in the local coordinate system are calculated.

[0022] Preferably, in the dynamic perception module, the change in the relative position between the dynamic perception and the surrounding embodied intelligent agents over time specifically includes: Within consecutive time slices, continuously track the relative position coordinate sequence of embodied intelligent agents with the same identity authentication ID; By combining the machine's own motion state data, the relative position coordinate sequence is differentially processed or trajectory fitted to generate the motion trajectory and velocity vector of the surrounding embodied intelligent agents, thereby realizing real-time dynamic position perception.

[0023] Preferably, the system further includes a cyclic update module, wherein the step of triggering the clustering grouping module, model training and recognition module, location decision module, and dynamic perception module to repeat their work according to a preset long period T specifically includes: At the beginning of each long period T, clustering is re-executed to identify whether new embodied agents have joined and formed new clusters. If no data corresponding to a certain identity authentication ID is detected within several consecutive long periods T, it is determined that the embodied agent has left the environment, and the identity authentication ID and its associated historical data are deleted from the system's identity authentication ID list.

[0024] Thirdly, a computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the computer program, implements the steps of the above-described embodied intelligent agent location perception method based on radio frequency fingerprinting.

[0025] Fourthly, embodiments of the present invention provide a computer-readable storage medium including a computer program, which, when executed by a processor, implements the steps of the above-described method for location perception of an embodied intelligent agent based on radio frequency fingerprinting.

[0026] Fifthly, a chip includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the computer program, implements the steps of the above-described embodied intelligent agent location perception method based on radio frequency fingerprinting.

[0027] In a sixth aspect, embodiments of the present invention provide an electronic device including a computer program, which, when executed by the electronic device, implements the steps of the above-described method for location perception of an embodied intelligent agent based on radio frequency fingerprinting.

[0028] Compared with the prior art, the present invention has at least the following beneficial effects: A method for embodied agent location awareness based on radio frequency fingerprinting is proposed. First, it acquires rich radio frequency fingerprint features through multi-directional time-division acquisition across the entire frequency band, eliminating the need for additional hardware and solving the problem of single signal sources. Second, a unique clustering-then-classification strategy automatically eliminates environmental noise and non-agent interference when the device identity is unknown, significantly improving the system's robustness. Third, it utilizes physical layer signals for direct angle arrival and power intensity fusion positioning, avoiding complex application layer protocol interactions and significantly reducing perception latency. Finally, a long-period dynamic update mechanism T is introduced, enabling the system to adapt to the addition and departure of devices in the environment, achieving true real-time dynamic perception. This method does not rely on communication handshakes between agents or coordination by a central node, significantly reducing the time overhead and hardware resource consumption caused by protocol conversion and data parsing, laying a technical foundation for subsequent rapid collaborative operations, and fundamentally solving the problem of difficult agent discovery in heterogeneous network environments.

[0029] Furthermore, the design of N independently configured groups enables the acquisition device to possess multi-dimensional signal acquisition capabilities. The acquisition strategy can be flexibly adjusted according to scenario requirements, avoiding blind spots in unidirectional acquisition and adapting to different frequency bands of wireless signals. This effectively addresses the heterogeneity of wireless signals from embodied intelligent agents in power scenarios. Simultaneously, the independent link design reduces interference between signals, improves the purity of the original acquired signals, and provides a high-quality data foundation for subsequent RF fingerprint feature extraction and location determination, significantly enhancing the robustness of the entire sensing system.

[0030] Furthermore, the full-band high-gain directional antenna ensures efficient acquisition of radio frequency signals from different directions. The coordinated operation of the power detector, local oscillator controller, and controllable local oscillator enables dynamic control of the acquisition process. The mixer and analog-to-digital converter complete the frequency reduction and digital conversion of the signal, providing a standard digital signal for subsequent algorithm processing.

[0031] Furthermore, the local oscillator controller dynamically adjusts the amplitude and frequency of the local oscillator by receiving the power value results from the power detector, allowing the acquisition device to optimize the acquisition parameters in real time according to the strength of the actual radio frequency signal. This ensures the integrity of the signal acquisition while avoiding distortion caused by excessively strong signals and acquisition failure caused by excessively weak signals.

[0032] Furthermore, by employing a multi-directional acquisition strategy within the same frequency band or different frequency bands in different directions, the acquisition device can acquire multi-dimensional, multi-frequency signal data within the same time period, significantly improving signal acquisition efficiency and shortening the duration of a single acquisition. Simultaneously, the acquired raw data contains both spatial direction and frequency band information, providing crucial spatial dimension data for subsequent angle-based arrival algorithms to determine the agent's orientation, making the location determination more comprehensive and the results more accurate.

[0033] Furthermore, the extraction of radio frequency fingerprint feature vectors provides accurate feature basis for clustering, and unsupervised clustering algorithms can quickly group signals of similar agents together, achieving initial differentiation of agents. Simultaneously, by setting thresholds to remove noise and invalid data, interference from non-agent signals is effectively eliminated, improving the quality of the dataset for subsequent model training.

[0034] Furthermore, as the number of agents increases, the number of sampling points is automatically increased to improve the accuracy of feature extraction and ensure the accuracy of clustering results; when the number of agents decreases, the number of sampling points is automatically reduced to decrease data processing volume and computational energy consumption. This dynamic adjustment strategy allows the system to optimize the allocation of computing resources in real time according to environmental changes, avoiding feature loss due to too few sampling points and preventing resource waste due to too many sampling points. It perfectly adapts to the dynamic changes in the number of embodied agents in power scenarios, further improving the system's energy efficiency and adaptability, and extending the working time of embodied agents.

[0035] Furthermore, by using grouped data as input samples and identity authentication IDs as output labels, a standardized training dataset was constructed, ensuring the effectiveness of model training. The incremental learning mechanism allows the model to incorporate newly acquired signal features into existing parameters, achieving dynamic model updates without retraining the entire model, significantly shortening model update time and reducing computational resource consumption.

[0036] Furthermore, based on the phase difference of antennas in different directions, combined with the angle arrival algorithm, the azimuth angle determination has a precise physical basis. The power intensity attenuation model realizes the quantitative calculation of relative distance. The combination of the two can solve for the precise relative position coordinates, realizing the accurate calculation of the two-dimensional / three-dimensional relative position of the intelligent agent, and solving the problems of low accuracy and single dimension of traditional position determination methods.

[0037] Furthermore, by using the position coordinate sequence of continuous time slices, the position change trend of the agent can be clearly reflected. Combined with the differential processing and trajectory fitting of the local motion state data, the quantitative analysis of motion trajectory and velocity vector is realized, enabling the system to predict the direction of motion of the agent and achieve accurate capture of the agent's motion trajectory and velocity vector. This solves the problem that traditional methods can only achieve static position perception and cannot track dynamic trajectories.

[0038] Furthermore, re-performing clustering in each long cycle can quickly identify newly added agents and form new clusters, ensuring the timely discovery of new agents; if there is no data for several consecutive cycles, it is determined that the agent has left, and the relevant ID and data are automatically deleted, avoiding the accumulation of invalid data. This achieves standardization and automation of dynamic agent discovery and deletion, solving the problem that traditional methods cannot effectively adapt to the dynamic changes in the number of agents.

[0039] It is understood that the beneficial effects of the second to sixth aspects mentioned above can be found in the relevant descriptions in the first aspect mentioned above, and will not be repeated here.

[0040] In summary, this invention achieves high-precision dynamic position perception with no central node and low energy consumption through full-band multi-directional acquisition and a clustering-then-classification strategy; it has strong anti-interference, adaptive environmental change and real-time motion compensation capabilities, and can independently complete multi-agent collaborative positioning without additional infrastructure, significantly improving the autonomy and security of embodied intelligent agents.

[0041] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0042] Figure 1 This is a schematic diagram of the radio frequency fingerprint location sensing framework of the present invention; Figure 2 This is a schematic diagram of the multi-directional full-band radio frequency acquisition device for the embodied intelligent agent of the present invention; Figure 3 This is a schematic diagram illustrating the intelligent agent label allocation for data collection using a clustering algorithm, as described in this invention. Figure 4 This is a schematic diagram illustrating device identification via radio frequency fingerprinting according to the present invention; Figure 5This is a schematic diagram of the relative position perception process of the present invention, wherein (a) is the multi-directional signal acquisition process of the main embodied intelligent agent, and (b) is the direction distance determination and relative position calculation process; Figure 6 This is a schematic diagram illustrating the dynamic position perception of the embodied intelligent agent according to the present invention; Figure 7 This is a flowchart of the method of the present invention; Figure 8 A schematic diagram of a computer device provided in an embodiment of the present invention; Figure 9 This is a block diagram of a chip provided according to an embodiment of the present invention.

[0043] Among them, 60. Computer equipment; 61. Processor; 62. Memory; 63. Computer program; 600. Electronic device; 610. Processing unit; 620. Storage unit; 6201. Random access memory unit; 6202. Cache memory unit; 6203. Read-only memory unit; 6204. Program / utility; 6205. Program module; 630. Bus; 640. Display unit; 650. Input / output interface; 660. Network adapter; 700. External device. Detailed Implementation

[0044] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0045] In the description of this invention, it should be understood that the terms "comprising" and "including" indicate the presence of the described features, integrals, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or collections thereof.

[0046] It should also be understood that the terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the invention. As used in this specification and the appended claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise.

[0047] It should also be further understood that the term "and / or" as used in this specification and the appended claims refers to any combination and all possible combinations of one or more of the associated listed items, and includes such combinations. For example, A and / or B can represent three cases: A alone, A and B simultaneously, and B alone. Additionally, the character " / " in this invention generally indicates that the preceding and following objects have an "or" relationship.

[0048] It should be understood that although terms such as first, second, third, etc., may be used in the embodiments of the present invention to describe the preset range, these preset ranges should not be limited to these terms. These terms are only used to distinguish the preset ranges from one another. For example, without departing from the scope of the embodiments of the present invention, the first preset range may also be referred to as the second preset range, and similarly, the second preset range may also be referred to as the first preset range.

[0049] Depending on the context, the word "if" as used here can be interpreted as "when," "when," "in response to determination," or "in response to detection." Similarly, depending on the context, the phrase "if determination" or "if detection (of the stated condition or event)" can be interpreted as "when determination," "in response to determination," "when detection (of the stated condition or event)," or "in response to detection (of the stated condition or event)."

[0050] The accompanying drawings illustrate various structural schematic diagrams according to embodiments disclosed in this invention. These drawings are not to scale, and some details have been enlarged for clarity, and some details may have been omitted. The shapes of the various regions and layers shown in the drawings, as well as their relative sizes and positional relationships, are merely exemplary and may deviate from reality due to manufacturing tolerances or technical limitations. Furthermore, those skilled in the art can design regions / layers with different shapes, sizes, and relative positions as needed.

[0051] This invention provides a method for embodied intelligent agent location awareness based on wireless radio frequency fingerprinting. Unlike traditional solutions that rely on base stations (such as UWB and Bluetooth beacons) or active handshakes between devices (such as Wi-Fi RTT), this invention relies solely on the embodied intelligent agent's own acquisition device. It achieves awareness by passively listening to the radio frequency fingerprints of surrounding devices, eliminating the need for a central node or mutual information exchange, thus completely solving the infrastructure dependency problem. Unlike single-frequency or omnidirectional acquisition, this invention employs a time-division acquisition method of full-band multi-directional wireless signals, combined with adaptive local oscillator control, effectively solving the problem of missed detections caused by the diversity of frequency bands and strong directionality of embodied intelligent agent wireless signals, achieving broad-spectrum coverage. Unlike direct feature matching, this invention innovatively utilizes a clustering-then-classification approach to automatically eliminate non-embodied intelligent agent signal interference (such as background WiFi and noise) during the full-band acquisition process without prior knowledge, significantly improving the accuracy of identity recognition. Unlike complex protocol parsing at the application layer, this invention utilizes only radio frequency signals carrying the fingerprint characteristics of embodied intelligent agents for data analysis at the physical layer. This significantly reduces computational resource consumption and processing time, facilitating rapid agent discovery and millisecond-level positional awareness. Unlike static point-based positioning, this invention achieves relative position and trajectory awareness between embodied intelligent agents over time through real-time data acquisition combined with motion compensation. This provides dynamic data assurance for collision avoidance and collaborative operations during high-speed movement of embodied intelligent agents.

[0052] Please see Figure 7 The present invention discloses a method for location perception of an embodied intelligent agent based on radio frequency fingerprinting, comprising the following steps: S1. Deploy a full-band multi-directional wireless signal acquisition device on the embodied intelligent agent and use the acquisition device to collect full-band wireless radio frequency signals in a time-division manner. A single embodied intelligent agent dynamically collects wireless signals across the entire frequency band in a time-division manner, performs analog-to-digital conversion on the collected signals, and uses the collected time-domain signal fingerprints for feature classification. The agent then dynamically and in real-time determines the relative position information of the classified data, confirms the embodied intelligent agent based on the characteristics of the position information, and achieves dynamic and real-time position perception for the confirmed embodied intelligent agent.

[0053] The structural shape of the full-band multi-directional wireless signal acquisition device varies depending on the direction of the wireless signals acquired. To enable the embodied agent to simultaneously acquire more signals, multiple independent wireless radio frequency acquisition devices are deployed on the embodied agent to collect data during the acquisition of surrounding wireless signals. Each acquisition device uses a directional antenna to acquire data from different directions, reducing the impact of differences in acquisition time on the final decision. Specific acquisition devices include... Figure 1 and Figure 2 As shown. The embodied intelligent agent wireless signal acquisition device includes NThe system consists of RF link components such as an antenna, power detector, controllable local oscillator, mixer, analog-to-digital converter, and local oscillator controller. The antenna is a full-band high-gain directional antenna used to acquire RF signals from different directions. The power detector detects the power value of a preset frequency band, communicates with the local oscillator controller, receives the acquisition frequency band information from the local oscillator controller to set the detection frequency band, and transmits the detected power value result to the local oscillator controller for setting the local oscillator amplitude. The controllable local oscillator receives control information from the local oscillator controller to set the local oscillator frequency and amplitude during real-time acquisition. The mixer mixes the local oscillator frequency with the transmitted RF signal to obtain a low-frequency carrier signal. The analog-to-digital converter performs analog-to-digital conversion on the acquired low-frequency carrier signal for further processing.

[0054] When in use, the power of different frequency bands is collected in a time-division manner according to preset rules and formed into digital signals. It can collect power from multiple directions in the same frequency band at the same time, or collect power from different frequency bands in different directions at the same time.

[0055] S2. Utilize the characteristic of the collected wireless radio frequency signals carrying the wireless hardware radio frequency fingerprint of the embodied intelligent agent to cluster and group the wireless radio frequency signals and assign device identity tags to each group, and remove wireless signal data that cannot be classified. Please see Figure 3 Radio frequency fingerprints are the inherent characteristics of various radio frequency hardware modules in a wireless communication system, which are reflected in the wireless transmission signal due to differences in component parameter tolerances, circuit deviations, and welding processes. Different hardware generates different radio frequency fingerprints, which have spatial uniqueness and temporal stability. Therefore, radio frequency fingerprints can be used as device identification IDs for mutual identification between devices.

[0056] When embodied intelligent agents interact via wireless signals, the wireless signals they emit can serve as their specific identity features for rapid identification between agents. The digital signal obtained after analog-to-digital conversion contains the hardware characteristics of the embodied intelligent agent, and its uniqueness can represent the embodied intelligent agent itself. Therefore, using the collected wireless signal as a radio frequency fingerprint can be used for identity authentication of the embodied intelligent agent.

[0057] The acquisition time of wireless signals is highly correlated with the recognition accuracy of RFID fingerprints. If the acquisition time is too short, it will be impossible to effectively distinguish the characteristics of each embodied intelligent agent; if the acquisition time is too long, it will cause excessive computational burden. The acquisition time is equal to the sampling period of the digital-to-analog converter multiplied by the number of sampling points. Since the sampling period of the digital-to-analog converter is known in the system, the sampling time is set by setting the number of sampling points.

[0058] In this invention, the number of sampling points is dynamically adjusted according to the number of embodied agents within the region, and also takes into account the different wireless modulation methods on the embodied agents. Set to: , The number of known embodied intelligent agents within the region.

[0059] S3. Use grouped data and corresponding labels to train the classification model, and use the trained classification model to classify and identify the identity labels of newly collected data. Please see Figure 4 The system clusters the data combinations of sampling points collected at different time periods and their corresponding local oscillator frequency information. The clustering results are used to assign device identification IDs to the data combinations formed by the sampling point data and the local oscillator information. Unclassifiable sampling point signals are discarded. The input signal is formed by combining the sampling points and the set local oscillator frequency information, with the device identification ID as the output. Multiple sets of data are collected from the same agent to train a classifier model. Common classifier types such as Support Vector Machines, Random Forests, Decision Trees, Naive Bayes Networks, and Neural Networks can be used. The trained classifier model is then used to classify newly collected wireless signal data and determine the device identification ID. The model parameters during the parameter training phase can be dynamically updated to the classifier model parameters during the device classification phase.

[0060] S4. The embodied intelligent agent takes the data and identity tags that have been classified and identified, and uses the angle arrival algorithm combined with the wireless signal power strength to determine the direction and relative distance of the surrounding intelligent agents, and then determines the relative position of the surrounding embodied intelligent agents. Please see Figure 5 The process of relative position perception between two embodied intelligent agents is described for ease of explanation. For example, the embodied intelligent agent performing the active discovery process is called the "master embodied intelligent agent," and the other is called the "slave embodied intelligent agent." After the master embodied intelligent agent identifies the ID of the slave embodied intelligent agent, it uses the known characteristics of the slave's wireless carrier signal to set the parameters of its signal acquisition device, ensuring consistency in the local oscillator signals from different directions. This allows it to acquire short-term wireless radio frequency signal characteristics of the slave embodied intelligent agent from multiple directions. The direction and relative position of the slave embodied intelligent agent are then determined based on the intensity of the signals from different directions. The direction and distance determination algorithm can be implemented using a combination of arrival time and wireless signal power strength algorithms, neural network classification algorithms, etc.

[0061] The specific directions of the embodied intelligent agents surrounding the judgment include: Acquire short-time radio frequency signal characteristics of the slave embodied intelligent agent from multiple directions; calculate the phase difference or time difference between signals received by antennas in different directions; and calculate the azimuth angle of the slave embodied intelligent agent relative to the master embodied intelligent agent based on the phase difference or time difference using an angle arrival algorithm. The relative distances of the embodied intelligent agents surrounding the judgment specifically include: Extract the signal strength values ​​from different directions; based on the wireless signal propagation loss model, estimate the distance between the embodied agent and the main embodied agent using the received signal strength indication; or, use a neural network classification algorithm to directly calculate the relative distance based on the signal strength features regression.

[0062] S5. After setting the parameters of the known surrounding embodied intelligent agents, collect wireless signal characteristics in real time, determine the position information through the classification and decision algorithm, and combine the changes in its own position to decide the changes in its relative position with the surrounding embodied intelligent agents over time. Please see Figure 6 After the embodied intelligent agent realizes the perception of other embodied intelligent agents in the surrounding area during the movement, it forms a relative position perception map and performs dynamic position perception based on the change of its own position over time and the change of the position of other embodied intelligent agents in the surrounding area over time.

[0063] S6. Repeat steps S2-S5 at long-period intervals of T to achieve dynamic discovery of newly added embodied agents in the environment and deletion of embodied agents leaving the environment.

[0064] The system executes the entire process of signal clustering and grouping, model training and identity recognition, relative position determination, and dynamic position perception in a pre-defined long-cycle T-loop, continuously detecting and analyzing wireless radio frequency signals within the environment. For newly added embodied intelligent agents, a new device identity ID is assigned to them through clustering and grouping, and subsequent position perception is completed. For embodied intelligent agents leaving the environment, their corresponding signal data and identity tags are removed, realizing the dynamic discovery and deregistration of embodied intelligent agents and adapting to the dynamic changes in the number and location of intelligent agents within the environment.

[0065] In another embodiment of the present invention, an embodied intelligent agent position perception system based on radio frequency fingerprinting is provided. This system can be used to implement the above-mentioned embodied intelligent agent position perception method based on radio frequency fingerprinting. Specifically, the embodied intelligent agent position perception system based on radio frequency fingerprinting includes a signal acquisition module, a clustering and grouping module, a model training and recognition module, a position decision module, a dynamic perception module, and a cyclic update module.

[0066] Among them, the signal acquisition module is used to acquire full-band wireless radio frequency signals in a time-division manner using a full-band multi-directional wireless signal acquisition device deployed on the embodied intelligent agent, and obtain a digital signal containing the wireless hardware radio frequency fingerprint characteristics of the embodied intelligent agent. The digital signal is the data after analog-to-digital conversion of the full-band wireless radio frequency signal. In the signal acquisition module, the full-band multi-directional wireless signal acquisition device includes N groups of radio frequency link devices, where N is an integer greater than or equal to 1; each group of radio frequency link devices is independently configured to process signals from different directions or different frequency bands simultaneously or in a time-division manner. Each group of radio frequency links includes: The antenna is a full-band high-gain directional antenna used to collect radio frequency signals in a specific direction; A power detector, connected to the output of the antenna, is used to detect the power value of a preset frequency band; The local oscillator is controllable and used to generate a local oscillator signal. The local oscillator controller executes the following control logic: Receive power value results from the power detector; set the local amplitude value according to the received power value; send control information to the controllable local oscillator to set the local oscillator frequency and amplitude in real time during data acquisition; A mixer, whose input terminals are respectively connected to the output terminal of the antenna and the output terminal of the controllable local oscillator, is used to mix the radio frequency signal with the local oscillator signal to obtain a low-frequency carrier signal; An analog-to-digital converter, connected to the output of the mixer, is used to convert the low-frequency carrier signal into a digital signal by analog-to-digital conversion. The local oscillator controller is communicatively connected to both the power detector and the controllable local oscillator.

[0067] The time-division acquisition of full-band wireless radio frequency signals includes a spatial acquisition strategy: Within the same time period, control multiple groups of radio frequency link devices to simultaneously acquire signals in the same frequency band but in different spatial directions; Alternatively, multiple sets of RF link devices can be controlled to simultaneously acquire signals from different directions and different frequency bands; The above strategy is used to obtain raw sampling data containing spatial orientation information and frequency band information.

[0068] The clustering and grouping module is used to cluster and group the digital signals, assign an identity authentication ID to each group of data, and remove data that cannot be classified. In the clustering and grouping module, the clustering and grouping of the digital signals specifically includes: The radio frequency fingerprint feature vectors are extracted from the digital signal, and the feature vectors are clustered using an unsupervised clustering algorithm; Feature vectors belonging to the same cluster are grouped together, and a unique identity authentication ID is generated for this group of data. Digital signals with excessive noise or whose feature vectors are more than a preset threshold away from any cluster center are identified as unclassifiable data and removed to eliminate interference from non-intelligent signals. The number of sampling points for clustering is dynamically adjusted based on the number of known embodied agents in the current environment; When the number of embodied agents in the environment increases, the number of sampling points is automatically increased to improve clustering accuracy; when the number of embodied agents in the environment decreases, the number of sampling points is automatically reduced to reduce processing power consumption.

[0069] The model training and recognition module is used to train a classification model using the grouped data and its corresponding identity authentication ID, and to recognize the identity authentication ID of newly collected digital signals using the trained classification model. In the model training and recognition module, the training classification model specifically includes: The grouped data in the clustering module is used as input samples, and the assigned identity authentication ID is used as output label to construct a training dataset. The classifier is trained using the training dataset. The classifier supports incremental learning and can incorporate newly collected digital signal features into the existing model parameters to update the classification model.

[0070] The location determination module is used to select the data of the identified identity authentication ID, use the angle arrival algorithm combined with the wireless signal power strength to determine the direction and relative distance of the surrounding embodied intelligent agents, and calculate the relative position coordinates of the surrounding embodied intelligent agents relative to the local machine. In the location determination module, the direction and relative distance of surrounding embodied intelligent agents are determined by combining the angle arrival algorithm with the wireless signal power strength. Specifically, this includes: Based on the phase difference of the same identity authentication ID signal received by antennas in different directions in the full-band multi-directional wireless signal acquisition device, the azimuth angle of the surrounding embodied intelligent agents relative to the local machine is calculated using the angle arrival algorithm; based on the received wireless signal power intensity attenuation model, the relative distance between the surrounding embodied intelligent agents and the local machine is calculated; combining the azimuth angle and the relative distance, the two-dimensional or three-dimensional relative position coordinates of the surrounding embodied intelligent agents in the local coordinate system are calculated.

[0071] The dynamic perception module is used to collect wireless signal features in real time based on the embodied intelligent agent corresponding to the identified identity authentication ID, and combine the changes in the position of the embodied intelligent agent itself with the changes in the relative position coordinates over time to dynamically perceive the changes in the relative position with the surrounding embodied intelligent agents over time, and generate the motion trajectory and velocity vector of the surrounding embodied intelligent agents. In the dynamic perception module, the change of the relative position between the dynamic perception and the surrounding embodied intelligent agents over time specifically includes: Within consecutive time slices, the relative position coordinate sequence of embodied intelligent agents with the same identity authentication ID is continuously tracked; combined with the machine's own motion state data, the relative position coordinate sequence is differentially processed or trajectory fitted to generate the motion trajectory and velocity vector of the surrounding embodied intelligent agents, thereby realizing real-time dynamic position perception.

[0072] The cyclic update module is used to trigger the clustering and grouping module, model training and recognition module, position decision module and dynamic perception module to work repeatedly according to a preset long period T, so as to realize the dynamic discovery of embodied intelligent agents newly added to the environment and the deletion of embodied intelligent agents leaving the environment.

[0073] In the cyclic update module, the step of triggering the clustering grouping module, model training and recognition module, location decision module, and dynamic perception module to repeat their work according to a preset long period T specifically includes: At the beginning of each long period T, clustering is re-executed to identify whether new embodied agents have joined and formed new clusters. If no data corresponding to a certain identity authentication ID is detected within several consecutive long periods T, it is determined that the embodied agent has left the environment, and the identity authentication ID and its associated historical data are deleted from the system's identity authentication ID list.

[0074] This invention provides a terminal device comprising a processor and a memory. The memory stores a computer program, which includes program instructions. The processor executes the program instructions stored in the computer storage medium. The processor may be a Central Processing Unit (CPU), or other general-purpose processors, graphics processing units (GPUs), tensor processing units (TPUs), digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. It is the computing and control core of the terminal, suitable for implementing one or more instructions, specifically suitable for loading and executing one or more instructions to achieve a corresponding method flow or corresponding function. The processor described in this embodiment can be used for the operation of an embodied intelligent agent location perception method based on wireless radio frequency fingerprinting, including: A multi-directional, full-band wireless signal acquisition device deployed on an embodied intelligent agent is used to collect full-band wireless radio frequency signals in a time-division manner, obtaining digital signals containing the wireless hardware radio frequency fingerprint characteristics of the embodied intelligent agent. These digital signals are data obtained after analog-to-digital conversion of the full-band wireless radio frequency signals. The digital signals are then clustered and grouped, with each group assigned an authentication ID. A classification model is trained using the grouped data and their corresponding authentication IDs, and the trained model is used to identify the authentication IDs of newly acquired digital signals. Data with identified authentication IDs is selected, and an angle-of-arrival algorithm combined with wireless signal power intensity is used to determine the direction and relative distance of surrounding embodied intelligent agents, calculating the relative position coordinates of these agents relative to the agent. Based on the identified authentication IDs and the corresponding embodied intelligent agents, wireless signal features are collected in real time. Combined with changes in the embodied intelligent agent's own position, the relative position coordinates over time are dynamically sensed to change with surrounding embodied intelligent agents, generating the motion trajectories and velocity vectors of these agents.

[0075] Please see Figure 8 The terminal device is a computer device. In this embodiment, the computer device 60 includes a processor 61, a memory 62, and a computer program 63 stored in the memory 62 and executable on the processor 61. When executed by the processor 61, the computer program 63 implements the embodied intelligent agent location awareness method based on radio frequency fingerprinting in this embodiment. To avoid repetition, these details are not elaborated here. Alternatively, when executed by the processor 61, the computer program 63 implements the functions of each model / unit in the embodied intelligent agent location awareness system based on radio frequency fingerprinting in this embodiment. To avoid repetition, these details are not elaborated here.

[0076] Computer device 60 can be a desktop computer, laptop, handheld computer, cloud server, or other computing device. Computer device 60 may include, but is not limited to, a processor 61 and a memory 62. Those skilled in the art will understand that... Figure 8 This is merely an example of computer device 60 and does not constitute a limitation on computer device 60. It may include more or fewer components than shown, or combine certain components, or different components. For example, computer device may also include input / output devices, network access devices, buses, etc.

[0077] The processor 61 may be a Central Processing Unit (CPU), or other general-purpose processors, graphics processing units (GPUs), tensor processing units (TPUs), digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor.

[0078] The memory 62 can be an internal storage unit of the computer device 60, such as a hard disk or memory of the computer device 60. The memory 62 can also be an external storage device of the computer device 60, such as a plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, etc. equipped on the computer device 60.

[0079] Furthermore, the memory 62 may include both internal storage units of the computer device 60 and external storage devices. The memory 62 is used to store computer programs and other programs and data required by the computer device. The memory 62 can also be used to temporarily store data that has been output or will be output.

[0080] Please see Figure 9 The terminal device is an electronic device 600, which is manifested in the form of a general-purpose computing device. The components of the electronic device may include, but are not limited to: at least one processing unit 610, at least one storage unit 620, a bus 630 connecting different platform components (including storage unit 620 and processing unit 610), a display unit 640, etc.

[0081] The storage unit stores program code, which can be executed by the processing unit 610 to perform the steps described in the method section of this specification according to various exemplary embodiments of the present invention. For example, the processing unit 610 can perform actions such as... Figure 7 The steps are shown in the figure.

[0082] Storage unit 620 may include a readable medium in the form of a volatile storage unit, such as random access memory (RAM) 6201 and / or cache memory 6202, and may further include a read-only memory (ROM) 6203.

[0083] Storage unit 620 may also include a program / utility 6204 having a set (at least one) program module 6205, such program module 6205 including but not limited to: operating system, one or more application programs, other program modules and program data, each or some combination of these examples may include an implementation of a network environment.

[0084] Bus 630 can represent one or more of several types of bus structures, including a memory cell bus or memory cell controller, a peripheral bus, a graphics acceleration port, a processing unit, or a local bus using any of the multiple bus structures.

[0085] Electronic device 600 can also communicate with one or more external devices 700 (e.g., keyboard, pointing device, Bluetooth device, etc.), and with one or more devices that enable a user to interact with electronic device 600, and / or with any device that enables electronic device 600 to communicate with one or more other computing devices (e.g., router, modem). This communication can be performed via input / output interface 650. Furthermore, electronic device 600 can also communicate with one or more networks (e.g., local area network, wide area network, and / or public network, such as the Internet) via network adapter 660. Network adapter 660 can communicate with other modules of electronic device 600 via bus 630. It should be understood that, although not shown in the figures, other hardware and / or software modules can be used in conjunction with electronic device 600, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage platforms.

[0086] Example 3 This invention also provides a storage medium, specifically a computer-readable storage medium, which is a memory device in a terminal device for storing programs and data. It is understood that the computer-readable storage medium here can include both built-in storage media in the terminal device and extended storage media supported by the terminal device; it can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. The computer-readable storage medium provides storage space that stores the terminal's operating system. Furthermore, the storage space also stores one or more instructions suitable for loading and execution by a processor, which can be one or more computer programs (including program code). More specific examples of the computer-readable storage medium include: an electrical connection with one or more wires, a portable disk, a hard disk, random access memory, read-only memory, erasable programmable read-only memory, optical fiber, portable compact disk read-only memory, optical storage device, magnetic storage device, or any suitable combination thereof.

[0087] Computer-readable storage media also include data signals propagated in baseband or as part of a carrier wave, carrying readable program code. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A readable storage medium can also be any readable medium other than a readable storage medium that can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the readable storage medium can be transmitted using any suitable medium, including but not limited to wireless, wired, optical fiber, radio frequency, etc., or any suitable combination thereof.

[0088] Program code for performing the operations of this invention can be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java and C++, and conventional procedural programming languages ​​such as C or similar languages. The program code can execute entirely on the user's computing device, partially on the user's computing device, as a standalone software package, partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server. In cases involving remote computing devices, the remote computing device can be connected to the user's computing device via any type of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computing device (e.g., via the Internet using an Internet service provider).

[0089] One or more instructions stored in a computer-readable storage medium can be loaded and executed by a processor to implement the corresponding steps of the embodied intelligent agent location awareness method based on radio frequency fingerprinting in the above embodiments; one or more instructions in the computer-readable storage medium are loaded and executed by the processor to perform the following steps: A multi-directional, full-band wireless signal acquisition device deployed on an embodied intelligent agent is used to collect full-band wireless radio frequency signals in a time-division manner, obtaining digital signals containing the wireless hardware radio frequency fingerprint characteristics of the embodied intelligent agent. These digital signals are data obtained after analog-to-digital conversion of the full-band wireless radio frequency signals. The digital signals are then clustered and grouped, with each group assigned an authentication ID. A classification model is trained using the grouped data and their corresponding authentication IDs, and the trained model is used to identify the authentication IDs of newly acquired digital signals. Data with identified authentication IDs is selected, and an angle-of-arrival algorithm combined with wireless signal power intensity is used to determine the direction and relative distance of surrounding embodied intelligent agents, calculating the relative position coordinates of these agents relative to the agent. Based on the identified authentication IDs and the corresponding embodied intelligent agents, wireless signal features are collected in real time. Combined with changes in the embodied intelligent agent's own position, the relative position coordinates over time are dynamically sensed to change with surrounding embodied intelligent agents, generating the motion trajectories and velocity vectors of these agents.

[0090] The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0091] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0092] Example 4 In critical application scenarios such as substations, converter stations, and transmission lines in power systems, mobile embodied intelligent agents are being deployed on a large scale to meet business needs such as equipment inspection, fault diagnosis, and emergency response. These agents generally rely on wireless networks to complete data interaction and command transmission. Due to the differences in hardware architecture among different types of embodied intelligent agents, their wireless communication links exhibit significant heterogeneity in physical layer coding schemes, modulation and demodulation methods, and communication protocols. In traditional technical solutions, location awareness and collaborative localization between agents mainly rely on the information interaction mechanism of upper-layer protocols. This process requires multiple rounds of data transmission, reception, parsing, and verification, which not only generates a large amount of communication overhead but also significantly prolongs the location determination delay between agents.

[0093] Experimental environment and parameter settings Experimental scenario: Simulates a 220kV outdoor substation with a scenario area of ​​200m×200m, including transformer equipment, walls, structures and other obstructions, to simulate a complex electromagnetic environment and physical obstructions; Experimental subjects: Deployed 6 embodied intelligent agents (2 wheeled inspection robots, 2 quadruped inspection robots, and 2 operational drones), supporting 2.4GHz / 5.8GHz dual-band communication, with a wireless radio frequency signal transmission power of 10dBm; System parameters of this invention: The full-band multi-directional wireless signal acquisition device is equipped with 8 sets of radio frequency link devices, sampling frequency of 100MHz, digital-to-analog converter sampling period of 10μs, the number of cluster sampling points is dynamically adjusted according to the number of intelligent agents, the long period T is set to 2s, the classification model adopts support vector machine and incremental learning mechanism, and the position determination adopts a fusion algorithm of angle arrival combined with wireless signal strength. Comparison method: Comparison Method 1: Traditional location-aware methods based on TCP / IP upper-layer protocol interaction; Comparison Method 2: Location method based solely on RSSI wireless signal strength.

[0094] Taking a substation scenario as an example, core tasks such as path collaborative planning of multiple inspection robots, air-ground linkage operations between inspection robots and drones, and precise docking operations between drones and intelligent robotic arms all place stringent requirements on the position synchronization between intelligent agents. Once the position determination delay exceeds the threshold, it can easily lead to problems such as path conflicts, loss of synchronization in operation, and target positioning deviations. In severe cases, it can even cause safety hazards such as equipment collisions and task interruptions.

[0095] The invention proposes a physical layer wireless fingerprint-based technology for acquisition, classification, identification, and dynamic sensing. This technology bypasses the redundant processes of traditional upper-layer protocol interactions, directly extracting the unique fingerprint features of different intelligent agents' wireless signals. A lightweight classification model enables rapid matching of agent identity and location. This significantly reduces the communication energy consumption and location awareness time of intelligent agents, and allows for real-time dynamic tracking of agent location information, accurately capturing their spatial location changes over time. It effectively ensures the efficient collaborative operation of embodied intelligent agents in complex scenarios such as substations and converter stations, providing reliable technical support for the construction of an intelligent operation and maintenance system for power systems.

[0096] In summary, this invention provides an embodied intelligent agent location perception method and system based on wireless radio frequency fingerprinting. It achieves location perception solely through the intelligent agent's own acquisition device, eliminating the need for information interaction between a central node and the intelligent agent, effectively reducing communication overhead and energy consumption. Full-band, multi-directional, time-division acquisition solves the signal heterogeneity problem, and clustering followed by classification eliminates interference from non-intelligent agents. Physical layer signal processing significantly shortens perception latency, and dynamic perception and cyclic updates enable real-time tracking and dynamic management of the intelligent agent's location. This invention significantly improves the efficiency and accuracy of embodied intelligent agent location perception, effectively avoids the risk of intelligent agent collisions, provides reliable technical support for multi-agent collaborative inspection and fault handling in power scenarios such as substations, and promotes the construction of an intelligent operation and maintenance system for power systems.

[0097] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is merely an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. Furthermore, the specific names of the functional units and modules are only for easy differentiation and are not intended to limit the scope of protection of this application. The specific working process of the units and modules in the above system can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0098] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0099] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed in this invention can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0100] In the embodiments provided by this invention, it should be understood that the disclosed devices / terminals and methods can be implemented in other ways. For example, the device / terminal embodiments described above are merely illustrative. For instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0101] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0102] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0103] If the integrated module / unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a portable hard drive, a magnetic disk, an optical disk, a computer memory, a read-only memory (ROM), a random-access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium, etc.

[0104] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus, and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0105] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0106] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0107] The above content is only for illustrating the technical concept of the present invention and should not be construed as limiting the scope of protection of the present invention. Any modifications made to the technical solution based on the technical concept proposed in this invention shall fall within the scope of protection of the claims of this invention.

Claims

1. A method for location perception of an embodied intelligent agent based on radio frequency fingerprinting, characterized in that, Includes the following steps: Using a full-band multi-directional wireless signal acquisition device deployed on an embodied intelligent agent, full-band wireless radio frequency signals are acquired in a time-division manner to obtain digital signals containing the wireless hardware radio frequency fingerprint characteristics of the embodied intelligent agent. The digital signals are the data of the full-band wireless radio frequency signals after analog-to-digital conversion. The digital signals are clustered and grouped, and an identity authentication ID is assigned to each group of data; A classification model is trained using the grouped data and their corresponding identity authentication IDs, and the trained classification model is used to identify the identity authentication IDs of newly collected digital signals. The system selects the identified identity authentication ID data, uses the angle arrival algorithm combined with the wireless signal power strength to determine the direction and relative distance of the surrounding embodied intelligent agents, and calculates the relative position coordinates of the surrounding embodied intelligent agents relative to the machine. Based on the identified identity authentication ID corresponding to the embodied intelligent agent, wireless signal characteristics are collected in real time. Combined with the changes in the position of the embodied intelligent agent itself, the relative position coordinates change over time are used to dynamically perceive the changes in the relative position with the surrounding embodied intelligent agents over time, and generate the motion trajectory and velocity vector of the surrounding embodied intelligent agents. 2.The wireless radio frequency fingerprint-based embodied agent location-awareness method of claim 1, wherein, The full-band multi-directional wireless signal acquisition device includes N groups of radio frequency link devices, where N is an integer greater than or equal to 1; each group of radio frequency link devices is independently configured to process signals from different directions or different frequency bands simultaneously or in a time-division manner. 3.The wireless radio frequency fingerprint-based embodied agent location-awareness method of claim 2, wherein, Each group of radio frequency links includes: The antenna is a full-band high-gain directional antenna used to collect radio frequency signals in a specific direction; A power detector, connected to the output of the antenna, is used to detect the power value of a preset frequency band; The local oscillator can be controlled to generate a local oscillator signal; A mixer, whose input terminals are respectively connected to the output terminal of the antenna and the output terminal of the controllable local oscillator, is used to mix the radio frequency signal with the local oscillator signal to obtain a low-frequency carrier signal; An analog-to-digital converter, connected to the output of the mixer, is used to convert the low-frequency carrier signal into a digital signal by analog-to-digital conversion. The local oscillator controller is communicatively connected to both the power detector and the controllable local oscillator. 4.The wireless radio frequency fingerprint-based embodied agent location-awareness method of claim 3, wherein, The local oscillator controller executes the following control logic: Receive power value results from the power detector; set the local amplitude value according to the received power value; send control information to the controllable local oscillator to set the local oscillator frequency and amplitude in real time during data acquisition.

5. The method for location perception of an embodied intelligent agent based on radio frequency fingerprinting according to claim 1, characterized in that, The time-division acquisition of full-band wireless radio frequency signals includes a spatial acquisition strategy: Within the same time period, control multiple groups of radio frequency link devices to simultaneously acquire signals in the same frequency band but in different spatial directions; Alternatively, multiple sets of RF link devices can be controlled to simultaneously acquire signals from different directions and different frequency bands; The above strategy is used to obtain raw sampling data containing spatial orientation information and frequency band information.

6. The method for location perception of an embodied intelligent agent based on radio frequency fingerprinting according to claim 1, characterized in that, Clustering and grouping the digital signals specifically includes: The radio frequency fingerprint feature vectors are extracted from the digital signal, and the feature vectors are clustered using an unsupervised clustering algorithm; Feature vectors belonging to the same cluster are grouped together, and a unique identity authentication ID is generated for this group of data. Digital signals with excessive noise or whose feature vectors are more than a preset threshold from any cluster center are identified as unclassifiable data and removed to eliminate interference from non-smart signals.

7. The method for location perception of an embodied intelligent agent based on radio frequency fingerprinting according to claim 6, characterized in that, The number of sampling points for clustering is dynamically adjusted based on the number of known embodied agents in the current environment; When the number of embodied agents in the environment increases, the number of sampling points is automatically increased to improve clustering accuracy; when the number of embodied agents in the environment decreases, the number of sampling points is automatically reduced to reduce processing power consumption.

8. The method for location perception of an embodied intelligent agent based on radio frequency fingerprinting according to claim 1, characterized in that, The training classification model specifically includes: The grouped data is used as input samples, and the assigned identity authentication ID is used as output label to construct a training dataset. The classifier is trained using the training dataset. The classifier supports incremental learning and can incorporate newly collected digital signal features into the existing model parameters to update the classification model.

9. The method for location perception of an embodied intelligent agent based on radio frequency fingerprinting according to claim 1, characterized in that, The method of using the angle arrival algorithm combined with the wireless signal power strength to determine the direction and relative distance of surrounding embodied intelligent agents specifically includes: Based on the phase difference of the same identity authentication ID signal received by antennas in different directions in the full-band multi-directional wireless signal acquisition device, the azimuth angle of the surrounding embodied intelligent agents relative to the local machine is calculated using the angle arrival algorithm; based on the received wireless signal power intensity attenuation model, the relative distance between the surrounding embodied intelligent agents and the local machine is calculated; combining the azimuth angle and the relative distance, the two-dimensional or three-dimensional relative position coordinates of the surrounding embodied intelligent agents in the local coordinate system are calculated.

10. The method for location perception of an embodied intelligent agent based on radio frequency fingerprinting according to claim 1, characterized in that, The dynamic perception and the change of the relative position with the surrounding embodied intelligent agents over time specifically include: Within consecutive time slices, continuously track the relative position coordinate sequence of embodied intelligent agents with the same identity authentication ID; By combining the machine's own motion state data, the relative position coordinate sequence is differentially processed or trajectory fitted to generate the motion trajectory and velocity vector of the surrounding embodied intelligent agents, thereby realizing real-time dynamic position perception.

11. The method for location perception of an embodied intelligent agent based on radio frequency fingerprinting according to claim 1, characterized in that, By utilizing the sequence of changes in relative position coordinates over time, the relative position with surrounding embodied intelligent agents is dynamically perceived to change over time. At the beginning of each long period T, according to a preset long period T, the clustering process is re-executed to identify whether any new embodied intelligent agents have joined and formed new clusters. If no data corresponding to a certain identity authentication ID is detected within several consecutive long periods T, it is determined that the embodied intelligent agent has left the environment, and the identity authentication ID and its associated historical data are deleted from the system's identity authentication ID list.

12. A location sensing system for an embodied intelligent agent based on radio frequency fingerprinting, characterized in that, include: The signal acquisition module is used to acquire full-band wireless radio frequency signals in a time-division manner using a full-band multi-directional wireless signal acquisition device deployed on the embodied intelligent agent, and to obtain a digital signal containing the wireless hardware radio frequency fingerprint characteristics of the embodied intelligent agent. The digital signal is the data after analog-to-digital conversion of the full-band wireless radio frequency signal. The clustering and grouping module is used to cluster and group the digital signals, and assign an identity authentication ID to each group of data. The model training and recognition module is used to train a classification model using the grouped data and its corresponding identity authentication ID, and to recognize the identity authentication ID of newly collected digital signals using the trained classification model. The location determination module is used to select the data of the identified identity authentication ID, use the angle arrival algorithm combined with the wireless signal power strength to determine the direction and relative distance of the surrounding embodied intelligent agents, and calculate the relative position coordinates of the surrounding embodied intelligent agents relative to the local machine. The dynamic perception module is used to collect wireless signal features in real time based on the identified embodied intelligent agent corresponding to the identity authentication ID, and combine the changes in the position of the embodied intelligent agent itself with the changes in the relative position coordinates over time to dynamically perceive the changes in the relative position with the surrounding embodied intelligent agents over time, and generate the motion trajectory and velocity vector of the surrounding embodied intelligent agents.

13. The embodied intelligent agent location sensing system based on radio frequency fingerprinting according to claim 12, characterized in that, In the signal acquisition module, the full-band multi-directional wireless signal acquisition device includes N groups of radio frequency link devices, where N is an integer greater than or equal to 1; each group of radio frequency link devices is independently configured to process signals from different directions or different frequency bands simultaneously or in a time-division manner. Each group of radio frequency links includes: The antenna is a full-band high-gain directional antenna used to collect radio frequency signals in a specific direction; A power detector, connected to the output of the antenna, is used to detect the power value of a preset frequency band; The local oscillator controller is controllable and used to generate a local oscillator signal. The local oscillator controller executes the following control logic: Receive power value results from the power detector; set the local amplitude value according to the received power value; send control information to the controllable local oscillator to set the local oscillator frequency and amplitude in real time during data acquisition; A mixer, whose input terminals are respectively connected to the output terminal of the antenna and the output terminal of the controllable local oscillator, is used to mix the radio frequency signal with the local oscillator signal to obtain a low-frequency carrier signal; An analog-to-digital converter, connected to the output of the mixer, is used to convert the low-frequency carrier signal into a digital signal by analog-to-digital conversion. The local oscillator controller is communicatively connected to both the power detector and the controllable local oscillator.

14. The embodied intelligent agent location sensing system based on radio frequency fingerprinting according to claim 12, characterized in that, In the signal acquisition module, the time-division acquisition of full-band wireless radio frequency signals includes a spatial dimension acquisition strategy: Within the same time period, control multiple groups of radio frequency link devices to simultaneously acquire signals in the same frequency band but in different spatial directions; Alternatively, multiple sets of RF link devices can be controlled to simultaneously acquire signals from different directions and different frequency bands; The above strategy is used to obtain raw sampling data containing spatial orientation information and frequency band information.

15. The embodied intelligent agent location sensing system based on radio frequency fingerprinting according to claim 12, characterized in that, In the clustering and grouping module, the clustering and grouping of the digital signals specifically includes: The radio frequency fingerprint feature vectors are extracted from the digital signal, and the feature vectors are clustered using an unsupervised clustering algorithm; Feature vectors belonging to the same cluster are grouped together, and a unique identity authentication ID is generated for this group of data. Digital signals with excessive noise or whose feature vectors are more than a preset threshold away from any cluster center are identified as unclassifiable data and removed to eliminate interference from non-intelligent signals. The number of sampling points for clustering is dynamically adjusted based on the number of known embodied agents in the current environment; When the number of embodied agents in the environment increases, the number of sampling points is automatically increased to improve clustering accuracy; when the number of embodied agents in the environment decreases, the number of sampling points is automatically reduced to reduce processing power consumption.

16. The embodied intelligent agent location sensing system based on radio frequency fingerprinting according to claim 12, characterized in that, In the model training and recognition module, the training classification model specifically includes: The grouped data in the clustering module is used as input samples, and the assigned identity authentication ID is used as output label to construct a training dataset. The classifier is trained using the training dataset. The classifier supports incremental learning and can incorporate newly collected digital signal features into the existing model parameters to update the classification model.

17. The embodied intelligent agent location sensing system based on radio frequency fingerprinting according to claim 12, characterized in that, In the location determination module, the direction and relative distance of surrounding embodied intelligent agents are determined by combining the angle arrival algorithm with the wireless signal power strength. Specifically, this includes: Based on the phase difference of the same identity authentication ID signal received by antennas in different directions in the full-band multi-directional wireless signal acquisition device, the azimuth angle of the surrounding embodied intelligent agents relative to the local machine is calculated using the angle arrival algorithm; based on the received wireless signal power intensity attenuation model, the relative distance between the surrounding embodied intelligent agents and the local machine is calculated; combining the azimuth angle and the relative distance, the two-dimensional or three-dimensional relative position coordinates of the surrounding embodied intelligent agents in the local coordinate system are calculated.

18. The embodied intelligent agent location sensing system based on radio frequency fingerprinting according to claim 12, characterized in that, In the dynamic perception module, the change of the relative position between the dynamic perception and the surrounding embodied intelligent agents over time specifically includes: Within consecutive time slices, continuously track the relative position coordinate sequence of embodied intelligent agents with the same identity authentication ID; By combining the machine's own motion state data, the relative position coordinate sequence is differentially processed or trajectory fitted to generate the motion trajectory and velocity vector of the surrounding embodied intelligent agents, thereby realizing real-time dynamic position perception.

19. The embodied intelligent agent location sensing system based on radio frequency fingerprinting according to claim 12, characterized in that, The system also includes a cyclic update module. Within this module, the clustering and grouping module, model training and recognition module, location decision module, and dynamic perception module repeatedly perform tasks according to a preset long period T. Specifically, this includes: At the beginning of each long period T, clustering is re-executed to identify whether new embodied agents have joined and formed new clusters. If no data corresponding to a certain identity authentication ID is detected within several consecutive long periods T, it is determined that the embodied agent has left the environment, and the identity authentication ID and its associated historical data are deleted from the system's identity authentication ID list.

20. A computer-readable storage medium for storing one or more programs, characterized in that, The one or more programs include instructions that, when executed by a computing device, cause the computing device to perform the method of any one of claims 1 to 11.

21. A computing device, characterized in that, include: One or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, the one or more programs including steps for performing the method of any one of claims 1 to 11.

22. A chip, characterized in that, A memory on which computer programs are stored; A processor for executing the computer program in the memory to implement the steps of the method according to any one of claims 1 to 11.

23. An electronic device, characterized in that, Including the chip as described in claim 22.

Images