RFID data acquisition method and system based on artificial intelligence

Abstract

The invention relates to an RFID data acquisition method and system based on artificial intelligence, and the method comprises the steps: calculating a local entropy value relative quantity reflecting the local entropy value change condition at each acquisition position, and calculating a global entropy value relative quantity reflecting the global entropy value change condition between the acquisition positions; and comparing the global entropy change condition with the local entropy change condition to determine whether to update the acquisition path or update the acquisition frequency or maintain the current acquisition strategy. According to the method, rapid dynamic acquisition strategy updating of complex RFID communication types in a large-scale field environment can be supported, so that the acquisition flexibility and the acquisition efficiency are greatly improved on the basis of compatibility with an existing path planning strategy.

Description

Technical Field

[0001] The present invention belongs to the technical field of RFID data collection based on artificial intelligence, and in particular relates to an RFID data collection method and system based on artificial intelligence. Background Art

[0002] The rapid development of the Internet of Things industry has driven the development of the RFID industry. At present, RFID has been applied in finance, mobile payment, urban public utilities, transportation, medical care, food safety and commodity anti-counterfeiting. The further increase in RFID application pilot projects is the main driving force for the development of RFID applications in my country. In particular, the government's support for major application projects in the industry such as financial IC cards, financial social security cards, mobile payments, urban public utility cards, resident health cards, electronic passports, and library applications has gradually increased, which has promoted the opening of large-scale RFID applications. At the same time, due to factors such as increased labor costs, the demand for enterprises to use information technology to improve management levels and efficiency has increased, and the driving force for implementing RFID projects has increased. The number of RFID projects generated by the needs of enterprises in the fields of logistics and warehousing management, asset management, commodity anti-counterfeiting, security and access control has further increased, which has effectively promoted the development of RFID applications.

[0003] RFID technology is used in production lines to track material information, improve production automation and efficiency, and monitor abnormal conditions in the production process; RFID technology is used to monitor the location and quantity of goods in real time, improve logistics efficiency and accuracy, and reduce the risk of goods loss; RFID technology is divided into two types: near field (NearField) and far field (FarField) communication. Whether near field acquisition is required depends on the type of RFID system used and the application scenario. Near field RFID, also known as high frequency (HF) RFID, usually operates at a frequency of 13.56MHz. NFC technology requires communication at a very short distance (usually a few centimeters), so it can be considered as near field acquisition. Far field RFID includes low frequency (LF), high frequency (HF) and ultra high frequency (UHF) RFID, with operating frequencies ranging from 125kHz to 960MHz. Far field RFID systems can read tags at longer distances, ranging from a few meters to tens of meters, without the need for near field acquisition.

[0004] It can be seen that whether it is near-field collection or far-field collection, the collection range is limited. In a large-scale collection environment, RFID tags need to be widely set up, and such tags often do not have their own power supply. They obtain energy from the radio waves emitted by the RFID reader, collect energy through electromagnetic induction, and then use this energy to activate the chip and send signals back to the reader. Although, passive RFID tags do not need to consume their own energy during the collection process, they are energy self-sufficient as long as they are within the signal range of the reader. Especially those scenarios where a large number of tags are required and the tags are not changed frequently. By default, the collection device is often used to collect the default type of RFID tags. Although it can be designed to collect different types of RFID tags at the same time, it needs to be set to multi-band support. The reader must be able to support multiple bands in order to communicate with different types of RFID tags. However, supporting multiple bands brings collection overhead, which reduces the collection speed and increases the collection energy consumption. Therefore, it is necessary to reasonably plan the collection strategy to improve the collection efficiency;

[0005] The term artificial intelligence was first proposed at the Dartmouth Society in 1956. Since then, researchers have developed many theories and principles, and the concept of artificial intelligence has also expanded. It is a new technical science that studies and develops theories, methods, technologies and application systems for simulating, extending and expanding human intelligence. Robotics is an important field that is booming. It integrates optics, electronics, measurement and control technology, automatic control theory, information technology, software technology and computer technology to form a comprehensive new technology. Due to the development and popularization of computer technology, human production has gradually transitioned from mechanization and automation to the "intelligent" era worldwide. Using intelligent acquisition equipment for automatic RFID data acquisition is a commonly used method at present. How to enable artificial intelligence acquisition equipment to quickly adapt to complex acquisition environments and realize lightweight acquisition strategy updates is a technical problem to be solved. Based on the above problems, the present invention can support fast dynamic acquisition strategy updates in large-scale field environments and complex RFID communication types, thereby greatly improving acquisition flexibility and efficiency on the basis of compatibility with existing path planning strategies. Summary of the invention

[0006] In order to solve the above problems in the prior art, the present invention proposes an RFID data collection method and system based on artificial intelligence, the method comprising:

[0007] Step S1: Obtain the RFID data collected within the most recent T consecutive unit time intervals at each collection location and represent it as a time-based current collection data sequence Where: t = 1 ~ T, p = 1 ~ P, k = 1 ~ K; t is the time number, p is the acquisition location number, k is the communication type number;

[0008] Step S2: respectively calculating the local entropy value relative amount reflecting the change of the local entropy value at each collection position, and calculating the global entropy value relative amount and the global entropy value absolute amount reflecting the change of the global entropy value between the collection positions;

[0009] Step S3: Compare the current acquisition data sequence and the previous data collection sequence The global entropy value change and the local entropy value change are used to determine whether to update the acquisition strategy. If so, directly update the acquisition strategy or enter step S4; otherwise, maintain the current acquisition strategy and return to step S1; specifically: if for all acquisition positions, the current acquisition data sequence and the previous data collection sequence The relative value of the global entropy value SH p If the changes between the two are less than the change threshold, the current collection strategy is maintained and the process returns to step S1; otherwise, according to the relative relationship or absolute relationship between the changes, the collection path in the collection strategy is selected to be maintained, and the collection frequency, the residence time of the collection device at the collection location and / or the proportion of the retention time of the collection device in each communication type in the collection strategy are directly updated, or the process goes to step S4 to adjust the collection path in the collection strategy;

[0010] When the relationship between the relative amounts of the global entropy values ​​is consistent, the collection frequency is updated; when the relationship between the relative amounts of the local entropy values ​​is consistent, the residence time of the collection device at the collection location and / or the ratio of the length of time the collection device is kept in each type of communication are maintained; when the relationship between the absolute amounts of the global entropy values ​​is consistent and the relationship between the relative amounts of the local entropy values ​​is inconsistent, the residence time of the collection device at the collection location and / or the ratio of the length of time the collection device is kept in each type of communication are updated; when the relationship between the absolute amounts of the global entropy values ​​and the relative amounts of the global entropy values ​​are inconsistent, step S4 is entered; wherein: the consistency is that the proportional relationship is consistent or the difference in the numerical relationship is within a preset range;

[0011] Step S4: Determine or redetermine the collection path or the collection strategy including the collection path; and send the collection path / collection strategy to the collection device; the collection strategy includes the collection path, the collection frequency, the residence time of the collection device at the collection location and / or the proportion of the retention time of the collection device in each communication type.

[0012] Furthermore, the server obtains the real-time collection data sent by the collection device, updates the collection strategy by analyzing the collection data, and sends the updated collection strategy to the collection device.

[0013] Furthermore, when some parameters in the acquisition strategy change, only the changed parameters are updated.

[0014] Further, the partial parameters are the collection frequency, the residence time of the collection device at the collection location and / or the ratio of the retention time length of the collection device in each communication type.

[0015] Furthermore, the artificial intelligence server determines the collection strategy based on the distribution of collection locations in the current field environment, the amount of RFID data that needs to be collected in each collection location and the communication type used.

[0016] An artificial intelligence-based RFID data collection system is used to implement the artificial intelligence-based RFID data collection method.

[0017] Furthermore, a neural network model is set in the artificial intelligence server.

[0018] An artificial intelligence-based RFID data collection device, wherein the artificial intelligence-based RFID device is used to implement the artificial intelligence-based RFID data collection method.

[0019] An artificial intelligence-based RFID data collection platform, the artificial intelligence-based RFID platform comprising an artificial intelligence-based RFID data collection module; the module is used to implement the artificial intelligence-based RFID data collection method.

[0020] A computer-readable storage medium, characterized in that it includes a program, which, when running on a computer, enables the computer to execute the artificial intelligence-based RFID data collection method.

[0021] The beneficial effects of the present invention include:

[0022] (1) Realize lightweight collection strategy updates. According to the global and local relative and local changes of entropy information, the collection strategy can be adjusted in a lightweight manner by updating some parameters of the collection strategy. This can support fast dynamic collection strategy updates in large-scale field environments and complex RFID communication types, thereby greatly improving collection flexibility and efficiency while being compatible with existing path planning strategies.

[0023] (2) The collected data sequence supports multiple communication types and the collection strategy based on the collected data sequence is simple to update. It can run independently in a mobile collection device without the support of big data. On the basis of being compatible with existing path planning strategies, it greatly improves the collection flexibility and compatibility. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of the present application, but do not constitute an improper limitation of the present invention. In the drawings:

[0025] Figure 1 This is a schematic diagram of the artificial intelligence-based RFID data collection method provided by the present invention. DETAILED DESCRIPTION

[0026] The present invention will be described in detail below in conjunction with the accompanying drawings and specific embodiments, wherein the illustrative embodiments and descriptions are only used to explain the present invention but are not intended to limit the present invention.

[0027] The present invention proposes an artificial intelligence-based RFID data collection system, comprising a collection device and an RFID tag; the RFID tag is set in a collection environment or on an object in the collection environment as required; the collection device uses communication methods of different communication types to collect RFID data based on a collection strategy; wherein: the collection strategy includes a collection path, a collection frequency, a residence time of the collection device at a collection location, and / or a ratio of the length of time the collection device stays in each communication type;

[0028] Preferably: the acquisition device is an artificial intelligence acquisition device;

[0029] That is to say, various types of RFID tags are arranged in the collection site, and each type of RFID tag communicates with the collection device using the communication type corresponding to the type; the collection device moves between the collection positions along the collection path to collect RFID data in the collection environment; PM RFID data from different RFID tags are collected at each collection position p; wherein: p = 1 to P; P is the total number of collection positions; of course, the arrangement of the collection positions is closely related to the specific collection environment;

[0030] Preferably, the PM is a dynamic value, and as the deployment of the on-site RFID tags changes continuously, the PM value also changes continuously;

[0031] Preferably: the RFID tag includes low frequency (LF), high frequency (HF), ultra high frequency (UHF) and microwave (MW) frequency bands; therefore, the corresponding communication type is to use low frequency (LF), high frequency (HF), ultra high frequency (UHF) and microwave (MW) frequency bands for communication; the artificial intelligence acquisition device needs to switch the communication type to complete the acquisition of different types of RFID tag data;

[0032] Based on the same inventive concept, the present invention proposes an artificial intelligence-based RFID data collection method and system, as shown in the attached Figure 1As shown, the method comprises the following steps:

[0033] Step S1: Obtain the RFID data collected within the most recent T consecutive unit time intervals at each collection location and represent it as a time-based current collection data sequence Where: t = 1 ~ T, p = 1 ~ P, k = 1 ~ K; t is the time number, p is the acquisition location number, k is the communication type number;

[0034] Preferably: the server obtains the real-time collection data sent by the collection device, updates the collection strategy by analyzing the collection data, and sends the updated collection strategy to the collection device to collect the collection strategy adopted by the real-time artificial intelligence collection device;

[0035] Preferably: when some parameters in the collection strategy change, only the changed parameters are updated; the partial parameters are the collection frequency, the residence time of the collection device at the collection location and / or the proportion of the retention time length of the collection device in each communication type;

[0036] Preferably: before executing step S1, the collected RFID data is preprocessed such as normalization;

[0037] Step S2: Calculate the local entropy value relative amount SH reflecting the change of the local entropy value at each sampling position p，k , and the relative amount of global entropy SH that calculates the change in global entropy between the collection locations p ; The entropy value is used to indicate the change of RFID data at the collection location, which contains the quantitative value of the frequency and degree of change;

[0038] The change of the local entropy value at each collection position is calculated respectively as follows: the relative amount SH of the local entropy value of the kth communication type at the collection position p is calculated using the following formulas (1) to (3): p，k ;

[0039]

[0040] The calculation of the change of the global entropy value between the acquisition positions is specifically as follows: the relative amount SH of the global entropy value at the acquisition position p is calculated using the following formulas (4) to (6): p ;

[0041]

[0042] Alternatively, the following formulas (6) to (8) are used to calculate the relative amount of global entropy value SH at the acquisition position p: p ; When the communication type switching overhead of the acquisition device is large, this method can increase the relative amount of communication type information;

[0043]

[0044] Step S3: Compare the current acquisition data sequence and the previous data collection sequence (or default data collection sequence Where: bs is the global entropy value change and local entropy value change of the acquisition data sequence number under the default condition) to determine whether to update the acquisition strategy. If so, directly update the acquisition strategy or enter step S4; otherwise, keep the current acquisition strategy and return to step S1; specifically: if for all acquisition positions, the current acquisition data sequence and the previous data collection sequence The relative value of the global entropy value SH p If the changes between the two are less than the change threshold, the current collection strategy is maintained and the process returns to step S1; otherwise, according to the relative relationship or absolute relationship between the changes, the collection path in the collection strategy is selected to be maintained, and the collection frequency, the residence time of the collection device at the collection location and / or the proportion of the retention time of the collection device in each communication type in the collection strategy are directly updated, or the process goes to step S4 to adjust the collection path in the collection strategy;

[0045] The step S3 specifically includes the following steps:

[0046] Step S31: Calculate the current acquisition data sequence and the previous data collection sequence (or default data collection sequence The absolute value of global entropy SHA p and (or ); Specifically: use the following formula (9) or (10) to calculate the absolute value of the global entropy value; where: SH p , (or ) are the relative values ​​of their global entropy values;

[0047]

[0048] Step S32: Compare the current acquisition data sequence and the previous data collection sequence (or default data collection sequence The relationship between the absolute amount of global entropy values, the relative amount of global entropy values ​​and / or the relative amount of local entropy values ​​between them is used to determine the way to update the collection strategy; when the relationship between the relative amounts of global entropy values ​​is consistent, the collection frequency is updated; when the relationship between the relative amounts of local entropy values ​​is consistent, the residence time of the collection device at the collection location and / or the proportion of the length of the collection device's retention time in each type of communication is maintained; when the relationship between the absolute amounts of global entropy values ​​is consistent and the relationship between the relative amounts of local entropy values ​​is inconsistent, the residence time of the collection device at the collection location and / or the proportion of the length of the collection device's retention time in each type of communication is updated; when the relationship between the absolute amounts of global entropy values ​​and the relative amounts of global entropy values ​​are inconsistent, step S4 is entered; wherein: the consistency is that the proportional relationship is consistent or the difference in the numerical relationship is within a preset range;

[0049] The step S32 is specifically as follows: based on the satisfaction of conditions (R1) to (R8), make the following judgments 1 to 4; if none of judgments 1 to 4 is true, proceed to step S4; wherein: SH p and are the relative amounts of the global entropy values ​​of the current data collection sequence and the previous data collection sequence, respectively; QTr is the numerical threshold of the relative amount of the global entropy value; PTr is the proportional threshold of the relative amount of the global entropy value;

[0050] Decision 1: If the condition (R1) (or (R1-)) (R2) (R3) is satisfied, the current acquisition strategy is maintained; and the process returns to step S1;

[0051] Decision 2: If condition (R1) (or (R1-)) (R2) (R4) is met, then record the acquisition position that meets condition (R4) and execute step S33;

[0052] Decision 3: If conditions ((R5) or (R6) or (R7)) and (R8) are met, proceed to step S34;

[0053] Decision 4: If conditions (R1) and ((R5) or (R6) or (R7)) are satisfied, proceed to step S35;

[0054]

[0055] Preferably: wherein judgment 1 to judgment 4 are all judgments or selected judgments; in the case of all judgments, judgment 1 to judgment 4 are executed;

[0056] Interchangeable: Judgment 1 to Judgment 4 are in a selection relationship;

[0057] Preferably: QTr and PTr are preset values;

[0058] Preferably: the above ≈ is close in value and can be limited by a threshold value;

[0059] Step S33: The acquisition position that meets the condition (R4) is called sp, and the acquisition strategy at the acquisition position sp is adjusted; specifically, the relative amount of local entropy value SH is calculated. p，k Communication type ratio Pat k,sp , based on the communication type ratio, set or adjust the proportion of the length of time Pat_tim of the acquisition device's holding time in each communication type when the acquisition device is located at the acquisition position sp k,sp ;

[0060] Preferably: the communication type ratio Pat is calculated using the following formula (11): k,sp ;

[0061]

[0062] The ratio of the holding time length of the acquisition device in each communication type is set or adjusted based on the ratio when the acquisition device is located at the acquisition position sp; specifically: setting Pat_tim k，sp =Pat k,sp ; Among them: Pat_tim k,sp is the ratio of the holding time length of each communication type at the collection location sp of the collection device;

[0063] Alternatively, set Pat_tim k,sp =Pat_tim k，sp ×(1+Pat k，sp );

[0064] Alternatively, set Pat_tim k,sp =Pat k，sp ×∑ k Pat_tim k,sp ;

[0065] Step S34: for each acquisition position, calculate the first adjustment coefficient β1 of the acquisition frequency, and adjust the acquisition frequency in the current acquisition strategy based on the adjustment coefficient; specifically: calculate the first adjustment coefficient β1 of the acquisition frequency based on the following formula (12);

[0066]

[0067] The adjustment of the acquisition frequency of the calculated acquisition position is specifically: setting f=f×β1; wherein: f is the acquisition frequency in the current acquisition strategy;

[0068] The adjustment of the acquisition frequency of the calculated acquisition position is specifically: when β1>1, f=f×β1; when β1≤1, f=f / (1+β1); by frequency doubling, under normal circumstances, the acquisition strategy can be quickly returned to the default setting; when only the acquisition frequency is adjusted, the acquisition path in the acquisition strategy does not change, only a small update amount is required, and the adjustment is relatively flexible;

[0069] Preferably: f is the acquisition frequency reference value applicable to all acquisition locations under the current acquisition strategy;

[0070] Step S35: for each collection location, calculate the second adjustment coefficient β2 of the residence time of the collection device at the collection location, Based on the second adjustment coefficient β2, the residence time sdt of the acquisition device at the acquisition position in the current acquisition strategy is adjusted; sdt=sdt×β2; wherein: sdt is the residence time sdt of the acquisition device at the acquisition position; the adjustment here is a separate adjustment for each independent acquisition position, which is different from the overall frequency adjustment in step S34;

[0071] When updating the collection strategy, the collection path has the highest computational complexity and the most complex iteration. In fact, the better collection path needs to be closely related to the collection environment. Therefore, the collection path may be obtained after manual feedback, and it may be unreasonable to adjust it. In fact, when collecting RFID data, there are many important parameters in the collection strategy. Only fine-tuning the specific parameters in the collection strategy can solve some local or global data problems.

[0072] Step S4: Determine or redetermine a collection path or a collection strategy including the collection path; and send the collection path / collection strategy to the collection device; the collection strategy includes the collection path, the collection frequency, the residence time of the collection device at the collection location, and / or the proportion of the retention time of the collection device in each communication type;

[0073] Preferably: the artificial intelligence server performs path planning based on the path planning algorithm, considering the distribution of collection locations in the current field environment, the amount of RFID data to be collected in each collection location and the communication type adopted, and outputs the collection path, or a collection strategy including the collection path;

[0074] Replaceable: The AI ​​server takes the distribution of collection locations in the current field environment, the amount of RFID data to be collected in each collection location and the communication type adopted as input information, and obtains a collection strategy including a collection path;

[0075] Preferably: the artificial intelligence server is an artificial intelligence server based on a neural network;

[0076] Alternatively: the path planning server performs path planning based on the distribution of collection locations in the current field environment and obtains candidate collection paths. Based on the amount of RFID data to be collected in each collection location and the communication type used, the time limit information of the collection is used to select, constrain and optimize the candidate collection paths to obtain a collection strategy containing the collection paths; specifically: the path planning server determines the starting point and the end point, and uses the shortest path algorithm to search for a path from the starting point to the end point, and obtains multiple candidate collection paths; for each candidate collection path, the path length, the time required to complete the collection of all RFID data at each collection location, and the communication type switching overhead are used as optimization target weights, the collection time limit and the shortest collection time interval are used as constraints, and the shortest collection time is used as the optimization target, and the collection strategy containing the collection path is obtained through selection, constraint and optimization;

[0077] Preferably: according to the communication types involved in the two collection positions passed successively, the order of maintaining each communication type of the collection device is adjusted so that the communication type maintained at the last of the collection position passed earlier and the communication type maintained at the first of the collection position passed later are the same;

[0078] Based on the same inventive concept, the present invention also provides an artificial intelligence-based RFID data collection device, the system is used to implement the above-mentioned artificial intelligence-based RFID data collection method; the collection device is a mobile device, suitable for performing RFID data collection;

[0079] Based on the same inventive concept, the present invention also provides an artificial intelligence-based RFID data collection system, which is used to implement the above-mentioned artificial intelligence-based RFID data collection method;

[0080] Based on the same inventive concept, the present invention also provides an artificial intelligence-based RFID data acquisition server, which is used to implement the above-mentioned artificial intelligence-based RFID data acquisition method;

[0081] Based on the same inventive concept, the present invention also provides an artificial intelligence-based RFID data collection device, which is used to implement the above-mentioned artificial intelligence-based RFID data collection method;

[0082] Based on the same inventive concept, the present invention also provides an artificial intelligence-based RFID data collection platform, which is used to implement the above-mentioned artificial intelligence-based RFID data collection method;

[0083] A computer program (also referred to as a program, software, software application, script, or code) can be written in any form of a programming language, including an assembled or interpreted language, a declarative or procedural language, and it can be deployed in any form, including as a stand-alone program or as a module, component, subroutine, object, or other unit suitable for use in a computing environment. A computer program may, but need not, correspond to a file in a file system. A program can be stored in a portion of a file that holds other programs or data (e.g., one or more scripts stored in a markup language document), in a single file dedicated to the program, or in multiple collaborative files (e.g., files storing one or more modules, subroutines, or code portions). A computer program can be deployed to execute on one computer or on multiple computers located at one site or distributed across multiple sites and interconnected by a communication network.

[0084] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0085] The present invention is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of the processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0086] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0087] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0088] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. An artificial intelligence-based RFID data collection method, characterized in that: The method comprises: Step S1: Obtain the RFID data collected within the most recent T consecutive unit time intervals at each collection location and represent it as a time-based current collection data sequence Where: t = 1 ~ T, p = 1 ~ P, k = 1 ~ K; t is the time number, p is the acquisition location number, k is the communication type number; Step S2: respectively calculating the local entropy value relative amount reflecting the change of the local entropy value at each collection position, and calculating the global entropy value relative amount and the global entropy value absolute amount reflecting the change of the global entropy value between the collection positions; Step S3: Compare the current acquisition data sequence and the previous data collection sequence The global entropy value change and the local entropy value change are used to determine whether to update the acquisition strategy. If so, directly update the acquisition strategy or enter step S4; otherwise, maintain the current acquisition strategy and return to step S1; specifically: if for all acquisition positions, the current acquisition data sequence and the previous data collection sequence The relative value of the global entropy value SH p If the changes between the two are less than the change threshold, the current collection strategy is maintained and the process returns to step S1; otherwise, according to the relative relationship or absolute relationship between the changes, the collection path in the collection strategy is selected to be maintained, and the collection frequency, the residence time of the collection device at the collection location and / or the proportion of the retention time of the collection device in each communication type in the collection strategy are directly updated, or the process goes to step S4 to adjust the collection path in the collection strategy; When the relationship between the relative amounts of the global entropy values ​​is consistent, the collection frequency is updated; when the relationship between the relative amounts of the local entropy values ​​is consistent, the residence time of the collection device at the collection location and / or the ratio of the length of time the collection device is kept in each type of communication are maintained; when the relationship between the absolute amounts of the global entropy values ​​is consistent and the relationship between the relative amounts of the local entropy values ​​is inconsistent, the residence time of the collection device at the collection location and / or the ratio of the length of time the collection device is kept in each type of communication are updated; when the relationship between the absolute amounts of the global entropy values ​​and the relative amounts of the global entropy values ​​are inconsistent, step S4 is entered; wherein: the consistency is that the proportional relationship is consistent or the difference in the numerical relationship is within a preset range; Step S4: Determine or redetermine the collection path or the collection strategy including the collection path; and send the collection path / collection strategy to the collection device; the collection strategy includes the collection path, the collection frequency, the residence time of the collection device at the collection location and / or the proportion of the retention time of the collection device in each communication type.

2. The RFID data collection method based on artificial intelligence according to claim 1 is characterized in that: The server obtains the real-time collection data sent by the collection device, updates the collection strategy by analyzing the collection data, and sends the updated collection strategy to the collection device.

3. The RFID data collection method based on artificial intelligence according to claim 2 is characterized in that: When some parameters in the acquisition strategy change, only the changed parameters are updated.

4. The RFID data collection method based on artificial intelligence according to claim 3 is characterized in that: The partial parameters are the collection frequency, the residence time of the collection device at the collection location and / or the proportion of the retention time length of the collection device in each communication type.

5. The artificial intelligence-based RFID data collection method according to claim 4, characterized in that: The artificial intelligence server determines the collection strategy based on the distribution of collection locations in the current field environment, the amount of RFID data that needs to be collected in each collection location and the communication type used.

6. An artificial intelligence-based RFID data collection system, characterized in that: The artificial intelligence-based RFID data collection system is used to implement the artificial intelligence-based RFID data collection method described in any one of claims 1 to 5.

7. The artificial intelligence-based RFID data collection method according to claim 6, characterized in that: A neural network model is set in the artificial intelligence server.

8. An artificial intelligence-based RFID data collection device, characterized in that: The artificial intelligence-based RFID device is used to implement the artificial intelligence-based RFID data collection method described in any one of claims 1 to 5.

9. An artificial intelligence-based RFID data collection platform, characterized in that: The artificial intelligence-based RFID platform includes an artificial intelligence-based RFID data acquisition module; the module is used to implement the artificial intelligence-based RFID data acquisition method described in any one of claims 1 to 5.

10. A computer-readable storage medium, characterized in that: The invention comprises a program which, when running on a computer, enables the computer to execute the RFID data collection method based on artificial intelligence as described in any one of claims 1 to 5.

Images