Multi-robot path planning system and method based on multi-source sensor data fusion architecture

Abstract

The invention belongs to the technical field of nuclear power, and particularly relates to a multi-robot path planning system and method based on a multi-source sensor data fusion architecture. The system provided by the invention aims at nuclear power plant multi-robot tasks, encapsulates a special data fusion processing tool, generates a navigation path suitable for the multi-robot tasks in real time by analyzing and positioning related robot bodies and nuclear power plant deployment sensor data in combination with environmental map information, and provides the navigation path for engineering application after virtual environment verification. Therefore, data sources such as robot body sensors, visual monitoring in a nuclear power plant, wireless communication access points and the like are read, analyzed, converted and normalized integrated, a decentralized global scheduling mode is adopted, and preset structured map information is combined to form a practical multi-robot path planning solution. And rapid generation of the optimal navigation path of the multiple robots in the plant is realized.

Description

Technical Field

[0001] This invention belongs to the field of nuclear power technology, specifically relating to a multi-robot path planning system and method based on a multi-source sensor data fusion architecture. Background Technology

[0002] Nuclear power plants have multiple robots or autonomous mobile intelligent equipment that integrates robot body and external sensor data, combined with functions such as task planning, collision protection, and temporary networking interaction between multiple robots, to realize path planning for multi-robot operation tasks in nuclear power plant areas under complex conditions such as narrow environments, signal interference, and obstacle obstruction.

[0003] Unlike single-robot operations, multi-robot operations in nuclear power plants face various constraints, including spatial limitations, unstable network communication with the server, time windows, radiation protection, industrial safety management, and collision avoidance among robots. These constraints all affect multi-robot path planning. In addition, during power plant overhauls, some areas will experience temporary and random layout changes, making it difficult for conventional real-time positioning and online navigation technologies to fully meet the needs of multi-robot path planning. Summary of the Invention

[0004] To overcome the problems existing in related technologies, a multi-robot path planning system and method based on a multi-source sensor data fusion architecture is provided.

[0005] According to one aspect of the present disclosure, a multi-robot path planning system based on a multi-source sensor data fusion architecture is provided, the system comprising:

[0006] Multi-machine positioning data fusion tool is used to parse, assemble and fuse data from multiple functional modules within the system, and to generate, update and manage standard documents for the navigation planning data warehouse;

[0007] The navigation planning data warehouse is connected to the multi-machine positioning data fusion tool and is used to store data entity files with a standard structure generated by the multi-machine positioning data fusion tool, providing unified data services for the system.

[0008] Multiple functional modules are included, such as a navigation task planning module, a robot equipment scheduling module, a navigation environment data processing module, a multi-machine path pre-execution module, a multi-machine positioning and collision protection module, and a multi-machine route planning module; each of the functional modules interacts with the multi-machine positioning data fusion tool through a data transmission interface.

[0009] In one possible implementation, the multi-machine positioning data fusion tool includes:

[0010] The task data parsing and assembly submodule is used to obtain task information from the navigation task planning module, and generate a navigation task parsing and assembly result data set after parsing and assembling.

[0011] The equipment data parsing and assembly submodule is used to obtain equipment parameter information from the robot equipment scheduling module, and generate a set of navigation equipment parsing and assembly result data after parsing and assembly.

[0012] The map data parsing and assembly submodule is used to obtain map and environment information from the navigation environment data processing module, and generate a map parsing and assembly result data set after parsing and assembly.

[0013] The sensor data parsing and assembly submodule is used to obtain sensor data from the robot equipment scheduling module, and generate a set of sensor parsing and assembly result data after parsing and assembling.

[0014] The navigation planning data warehouse processing submodule is used to merge, match, and standardize the data sets of the above-mentioned parsing and assembly results, and to perform the generation, updating, encryption / decryption operations of the navigation planning data warehouse library files;

[0015] The navigation planning data warehouse interaction interface is used to perform real-time data interaction between the result data generated by the navigation planning data warehouse processing submodule and the generated navigation planning data warehouse result file.

[0016] The data transmission interface is used to enable data interaction between the multi-machine positioning data fusion tool and other functional modules within the system.

[0017] In one possible implementation, the structure entity file of the navigation planning data warehouse contains the following dataset:

[0018] The data warehouse index is used to record the registration information, relationships, and update status of all datasets in the navigation planning data warehouse entity file;

[0019] Task dataset, used to record information related to all loaded navigation tasks;

[0020] Robot information set, used to record all loaded navigation device-related information;

[0021] An environmental map dataset is used to record all loaded maps and related environmental information.

[0022] Sensor dataset, used to record information related to all loaded sensor acquisition data;

[0023] Navigation and positioning dataset, used to record updated location calculation results;

[0024] The navigation route information set is used to record the updated navigation strategy calculation results.

[0025] In one possible implementation, the robot equipment scheduling module further includes a multi-machine communication interface for enabling direct data interaction between different robot devices.

[0026] In one possible implementation, the multi-machine path pre-execution module includes:

[0027] The multi-machine positioning data fusion tool loading interface is used to call the calculation results fed back by the multi-machine positioning data fusion tool;

[0028] The virtual environment generation submodule is used to generate a test virtual scene containing a map, obstacle markers, and navigation points based on the loaded data.

[0029] The virtual navigation submodule is used to perform multi-robot path operation tests in the virtual scene.

[0030] In one possible implementation, the multi-machine positioning and collision protection module includes:

[0031] The sensor data parsing submodule is used to parse obstacle avoidance related data;

[0032] The localization calculation submodule is used to return the localization results and obstacle identification results of multiple robots based on the parsed data.

[0033] In one possible implementation, the multi-machine route planning module includes:

[0034] The obstacle data parsing submodule is used to acquire and parse obstacle data from the navigation planning data warehouse;

[0035] The navigation algorithm submodule is used to combine the real-time location of the navigation device, map data, and parsed obstacle data to generate the final navigation path and detailed motion parameters for multiple robots.

[0036] In one possible implementation, the navigation planning data warehouse processing submodule fuses data from different data sources by applying sensor information fusion technology or multi-source fusion positioning methods.

[0037] According to another aspect of the present disclosure, a multi-robot path planning method based on a multi-source sensor data fusion architecture is provided. The method is implemented based on the system described above and includes:

[0038] The multi-machine positioning data fusion tool obtains navigation task data, device and sensor data, and map and environment data from the navigation task planning module, robot equipment scheduling module, and navigation environment data processing module, respectively, and parses and assembles the various types of data to generate corresponding standardized data sets.

[0039] The standardized data set is merged, matched, and normalized to generate and update a navigation planning data warehouse file that conforms to a preset standard. The navigation planning data warehouse file contains interrelated data warehouse indexes, task datasets, robot information sets, environmental map datasets, sensor datasets, navigation and positioning datasets, and navigation route information sets.

[0040] Based on the data loaded in the navigation planning data warehouse file, a test scenario containing a map, obstacle markers, and navigation points is generated in a virtual environment, and a multi-robot path operation test is executed to verify the feasibility of the navigation strategy.

[0041] During the actual operation of the robot, based on the navigation planning data warehouse file and the real-time sensor data, the robot performs real-time positioning calculations and obstacle analysis for multiple robots, and generates positioning results and obstacle avoidance feedback.

[0042] Integrating virtually validated navigation strategies, real-time positioning results, and obstacle information, it generates multi-robot navigation paths and motion parameters for actual execution.

[0043] According to another aspect of the present disclosure, a non-volatile computer-readable storage medium is provided, on which computer program instructions are stored, which, when executed by a processor, implement the above-described method.

[0044] The beneficial effects of this disclosure are as follows: The system provided here is designed for multi-robot tasks in nuclear power plants. It encapsulates a dedicated data fusion processing tool, analyzes and integrates data from robot sensors and nuclear power plant deployments related to positioning, and combines this with environmental map information to generate navigation paths suitable for multi-robot tasks in real time. After verification in a virtual environment, it is then applied in engineering. This system reads, analyzes, converts, and standardizes data sources such as robot sensors, visual monitoring within the nuclear power plant, and wireless communication access points. It employs a decentralized global scheduling approach, combined with pre-set structured map information, to form a practical multi-robot path planning solution, enabling rapid generation of optimal navigation paths for multiple robots within the plant. Furthermore, this disclosure avoids complete reliance on robot sensors and wireless networks for real-time navigation strategy generation and execution in actual tasks when individual robots cannot interact with the server in real time due to temporary communication interruptions. This allows for temporary data interaction between robots and offline autonomous navigation. Attached Figure Description

[0045] Figure 1 This is a schematic diagram illustrating a multi-robot path planning system based on a multi-source sensor data fusion architecture, according to an exemplary embodiment.

[0046] Figure 2This is a schematic diagram illustrating a multi-machine positioning data fusion tool according to an exemplary embodiment.

[0047] Figure 3 This is a schematic diagram illustrating a navigation planning data warehouse according to an exemplary embodiment.

[0048] Figure 4 This is a flowchart illustrating a multi-robot path planning method based on a multi-source sensor data fusion architecture, according to an exemplary embodiment. Detailed Implementation

[0049] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0050] In related technologies, an analysis of the working scenarios and business characteristics of multiple robots in nuclear power plants reveals that factors such as limited space in the plant or passageways, weak wireless communication signal quality, radiation protection requirements, and temporary isolation measures for passageways during major overhauls make it impossible to successfully execute multi-robot navigation path planning by relying solely on wireless communication positioning technology.

[0051] This disclosure fully utilizes multiple robot bodies and sensor data sources such as vision, distance, and wireless network access points in a nuclear power plant environment. Dedicated tools are used to load and parse the data required for multi-robot navigation path planning from various data sources, integrate them into a unified dedicated data warehouse through comprehensive calculations, generate the required multi-robot navigation path, and export and distribute the path navigation strategy to the actual robots after verification by the pre-execution functional module. In the event of poor network communication quality or complete offline status, each robot autonomously completes navigation and obstacle avoidance according to the distributed path. After establishing communication with the server or other robots, the current status is immediately reported, and the navigation strategy is updated through the server.

[0052] See Figures 1 to 3 To ensure that the data is easily called and calculated by functional modules, the system disclosed herein is configured with a multi-bots navigation data fusion utility (MNDF) and a supporting navigation planning data warehouse (NPD) to aggregate relevant work scenario data information necessary for navigation positioning and path planning. At the same time, it is designed to cover all business operations of multi-robot path planning calculation in nuclear power plants, including navigation task planning module, robot equipment scheduling module, navigation environment data processing module, multi-bot path pre-execution module, multi-bot positioning and collision protection module, and multi-bot route planning module. The following is a detailed description of the technical solution.

[0053] The Multi-Machine Positioning Data Fusion Tool (MNDF) is used to read and parse data related to navigation tasks, robot devices, sensors, and navigation maps. It performs data normalization through corresponding sub-modules, and generates, edits, updates, and manages permissions for standard Navigation Planning Data Warehouse (NPD) files. The MNDF tool includes sub-modules for task data parsing and assembly, device data parsing and assembly, map data parsing and assembly, sensor data parsing and assembly, NPD processing, NPD interaction, and data transmission.

[0054] The task data parsing and assembly submodule retrieves information from the task details database and navigation key information database of the navigation task planning module through the data transmission interface. After information parsing and data assembly, it generates the navigation task parsing and assembly result data set RDC. n-t ;

[0055] The equipment data parsing and assembly submodule retrieves information from the detailed equipment parameter library of the robot equipment scheduling module via a data transmission interface. After information parsing and data assembly, it generates a navigation equipment parsing and assembly result data set RDC. n-d ;

[0056] The map data parsing and assembly submodule retrieves data from the navigation point library and obstacle marker library of the navigation environment data management module via a data transmission interface, and generates a map parsing and assembly result data set RDC after parsing and assembly. i-m ;

[0057] The sensor data parsing and assembly submodule obtains information from the sensor data interface inside the robot equipment scheduling module through the data transmission interface, and generates a sensor parsing and assembly result data set RDC after information parsing and data assembly. s ;

[0058] The NPD data security submodule performs data merging, matching, and normalization calculations on the result matrix generated by the above submodules, executes NPD library file generation, updating, and encryption / decryption processing, and generates an intermediate calculation result data set RDC. i-r ;

[0059] NPD Interaction Interface: Used to process the result data set RDC generated by the NPD processing submodule. i-r Real-time data interaction with the generated NPD result files;

[0060] Data transmission interface: Used for data interaction with other functional modules within the system.

[0061] The Navigation Planning Data Warehouse (NPD) records the basic information required for performing measurements and related information for subsequent calculations and verifications. Each NPD standard structure entity file contains one or more records; a single MNDF structure entity contains: Data Warehouse Index I s-i Task Dataset C t Robot Information Set C d Environmental map dataset C m Sensor Dataset C s Navigation point dataset C l Navigation route information set C n See structure Figure 2 The explanations for each part are as follows:

[0062] Data Warehouse Index I s-i : Records all dataset registration information in the current NPD entity file, including the entry number subset SC n Data Relationship Subset SC r Key Field Subset SC k Data update status subset SC s The structure is I s-i <SC n |SC r |SC k |SC s >;

[0063] Task Dataset C t Records all navigation task information loaded in the current NPD entity file, including the task type subset ST. c Related object subset ST o Task information subset ST i Task state subset ST s The structure is C t <ST c ST o ST i ST s >;

[0064] Robot Information Set C d Records all navigation device information loaded into the current NPD entity file, including the device type subset SD. c Related task type subset SD t Device status subset SD s The structure is C d <SD c |SD t |SD s >;

[0065] Environmental Map Dataset C mRecords all loaded map and environment-related information in the current NPD entity file, including the map summary information subset SM. a SM, a subset of identification information i Obstacle information subset SM h The structure is C m <SM a SM i SM h >;

[0066] Sensor Dataset C s Records information related to all loaded sensor data in the current NPD entity file, including the location information subset SS. l Type information subset SS c SS, a subset of collected data information d The structure is C s <SS l |SS c |SS d >;

[0067] Navigation and positioning dataset C l Records information related to the updated location calculation results in the current NPD entity file, including the location base data subset SL. i Location-related task data subset SL t Location-related device data subset SL d The structure is C l <SL i |SL t |SL d >;

[0068] Navigation route information set C n Records information related to the updated navigation strategy calculation results in the current NPD entity file, including the strategy basic information subset SN. i Strategy-related task information subset SN t Policy-related device information subset SN d Task status information subset SN t-s The structure is C n <SN i |SN t |SN d |SN t-s >

[0069] The Navigation Task Planning (NTP) module is used to read and parse detailed task information from an external task data source (nuclear power plant production management platform). It stores the read task data in a dedicated task detail information library and a navigation key information library, and interacts with other functional modules through a data transmission interface. Simultaneously, it writes data entries to the navigation task data vector in the positioning and navigation combined data structure entity file. The Navigation Task Planning module includes: a task data loading interface, a task detail information library, a navigation key information library, and a data transmission interface. (Structure diagram shown) Figure 1 Part (1);

[0070] The Robot Equipment Scheduling Module (BDP) is used to read and parse detailed data from robots and other data source devices. It stores the acquired basic parameters, motion control information, and real-time status information of the robots and other data source devices in a detailed device parameter library. It can acquire real-time data such as images and measurements collected by registered vision sensors, distance sensors, and displacement sensors in the task environment through sensor data interfaces. This module can store and acquire data and perform data interaction with other functional modules through a data transmission interface. This module also provides a dedicated multi-machine interaction interface to allow data interaction between robots. The Robot Equipment Scheduling Module includes: an equipment data loading interface, a detailed device parameter library, a sensor data interface, a multi-machine communication interaction interface, and a data transmission interface. (Structure diagram shown) Figure 1 Part (2);

[0071] The Navigation Environment Data Processing Module (NEDM) is used to read and parse detailed map information of the area to be navigated from the nuclear power plant's authorized map data source. It stores the acquired navigation points and obstacle markers in the corresponding databases and interacts with other functional modules through a data transmission interface. The NEDM includes: a map data loading interface, a navigation point database, an obstacle marker database, and a data transmission interface. (Structure diagram shown) Figure 1 Part (3);

[0072] Multi-robot Path Pre-Execution Module (MPF): This module executes path execution in a virtual environment by calling the calculation results fed back by the Multi-robot Localization Data Fusion Tool (MNDF). It acquires data by calling the MNDF loading interface, calls the virtual environment generation submodule to generate a test virtual scene (including a map, obstacle markers, and navigation points), calls the virtual navigation submodule to execute multi-robot path operation, and can also interact with other functional modules through a data transmission interface. The Multi-robot Path Pre-Execution Module includes: the MNDF loading interface, the virtual environment generation submodule, a process information database, a virtual navigation submodule, and a data transmission interface; structural diagram shown. Figure 1 Part (6);

[0073] Multi-robot Localization and Collision Protection Module (MLCP): This module performs localization and collision protection during multi-robot path execution. It acquires data by calling the MNDF loading interface and completes obstacle avoidance-related data parsing through the sensor data parsing submodule. Finally, the localization calculation submodule returns the multi-robot localization and obstacle identification results. The MLCP module includes: an MNDF loading interface, a sensor data parsing submodule, a process information database, a localization calculation submodule, and a data transmission interface; structural diagram shown. Figure 1 Part (7);

[0074] Multi-robot Route Planning Module (MPP): This module generates multi-robot navigation paths verified in a virtual environment. It uses the MNDF loading interface to parse NPD structure entity files, acquiring navigation task data, navigation map data, and positioning and labeling data. It then performs obstacle analysis and, combined with the real-time position of the navigation devices, generates the multi-robot navigation path and detailed motion parameters. Simultaneously, it can interact with other functional modules via a data transmission interface. The MPP module includes an MNDF loading interface, an obstacle data parsing submodule, a process information database, a navigation algorithm submodule, and a data transmission interface. (Structure diagram shown) Figure 1 Part (8).

[0075] This disclosure enables the unified and standardized integration of feedback data from external and internal navigation and positioning sensors of the navigation device system, and real-time comprehensive calculation. By comparing the calculation results with preset map information of the navigation environment, the real-time positioning calculation and navigation strategy generation of the navigation device can be realized during the implementation of the navigation task.

[0076] This disclosure integrates navigation task information, basic navigation equipment information, and environmental map information necessary for multi-robot navigation path planning into a corresponding data structure, and integrates them into a navigation planning data warehouse through an encapsulated multi-robot positioning data fusion tool. The system obtains information such as navigation task content and a list of objects to be navigated through the navigation task planning module, obtains detailed parameters of all robot devices, onboard sensor devices, and access interfaces of various visual acquisition and distance measurement sensors through the robot device scheduling module, and loads a map through the navigation environment data processing module to obtain navigation point information and obstacle markers. The integration of these modules forms a dedicated entity file for each dedicated functional module to perform data parsing and calculation, virtual environment pre-execution, and finally obtains real-time positioning information and navigation strategy information of the navigation devices.

[0077] Figure 4 This is a flowchart illustrating a multi-robot path planning method based on a multi-source sensor data fusion architecture according to an exemplary embodiment, such as... Figure 4 As shown, the method includes:

[0078] Step 1, Task Parsing, involves the batch acquisition and parsing of navigation-related data required from external task information. Each step includes data loading, information filling, information filtering, and information transmission, including:

[0079] Step 1.1 Data Loading: This step retrieves raw task information from external systems in batches; it calls the task data loading interface in the Navigation Task Planning (NTP) module to perform task data reading operations.

[0080] Step 1.2 Information Population: This step is used to perform information preprocessing; it performs structured matching on the read task data and batch transfers it to the task details library of the navigation task planning module.

[0081] Step 1.3 Information Filtering: Used to perform key task information matching, retrieval and extraction; based on the current task requirements, extract key information such as task number, associated area code, compatible robot equipment and working time / cycle window according to task priority requirements, and transfer it to the navigation task planning module navigation key information database;

[0082] Step 1.4 NPD Information Transfer: This step is used to update and transfer data to the NPD structure entity file in the navigation planning data warehouse. The Multi-Machine Positioning Data Fusion Tool (MNDF) is called via the data transfer interface to perform data transfer operations from the navigation task planning module's navigation key information database to the NPD structure entity file in the navigation planning data warehouse. After the data transfer is completed, the structure index vector of the NPD structure entity file is updated using the Multi-Machine Positioning Data Fusion Tool (MNDF).

[0083] Step 2, Robot Equipment and Environmental Information Analysis, involves the batch acquisition and analysis of navigation-related data required from the robot equipment and environmental information. Each sub-step includes, in sequence, data loading, information filling / robot parameter filtering, real-time measurement value reading, point information comparison, obstacle information comparison, and MNDF information transmission, including:

[0084] Step 2.1 Data Loading: This step is used to obtain raw device and map information from outside the system in batches; it calls the device data loading interface of the Robot Device Scheduling Module (BDP) and the map data loading interface of the Navigation Environment Data Processing Module to perform data reading operations.

[0085] Step 2.2 Information Filling / Robot Parameter Filtering: This step is used to perform matching, retrieval, and extraction of key equipment and map information; to perform structured matching on the read equipment data and batch transfer it to the detailed equipment parameter library of the robot equipment scheduling module; and to perform structured matching on the read map data and batch transfer it to the navigation point information library and obstacle marker information library.

[0086] Step 2.3 Real-time measurement value reading: Used to acquire sensor data in real time; call the sensor data interface of the robot equipment scheduling module to open the real-time measurement value data channel of the equipment;

[0087] Step 2.4 Point Information Comparison: This step is used to perform point information differential and update processing. Through the sensor data interface of the robot equipment scheduling module, the visual images and distance measurement data of key navigation points are compared with the navigation point information database of the navigation environment data processing module. After the comparison is completed, the navigation point information database of the navigation environment data processing module is updated.

[0088] Step 2.5 Obstacle Information Comparison: This step is used to perform map obstacle information differential and update processing; through the sensor data interface of the robot equipment scheduling module, the visual images and distance measurement data of key navigation points are compared with the obstacle identification information database of the navigation environment data processing module, and the obstacle identification information database of the navigation environment data processing module is updated after the comparison is completed.

[0089] Step 2.6 Multi-machine network communication: Used to perform network communication connections and data interaction between robots; by calling the multi-machine transmission and interaction interface, a network connection between robot devices is established using a self-organizing network method, allowing data information to be uploaded and received through other robot devices in this mode even when the current robot's communication with the server is restricted.

[0090] Step 2.7 MNDF Information Transmission: The Multi-Machine Positioning Data Fusion Tool (MNDF) is called through the data transmission interface to perform data transmission between the robot equipment scheduling module's detailed parameter library, the navigation environment data processing module's navigation point information library, and the NPD structure entity file; after the data transmission is completed, the structure index vector of the NPD structure entity file is updated using the Multi-Machine Positioning Data Fusion Tool (MNDF).

[0091] Step 3, data fusion processing, involves using the Multi-Machine Positioning Data Fusion Tool (MNDF) to perform operations related to the generation, editing, and encryption of NPD structure entity files; each sub-step includes, in sequence, task information filling, equipment and environment information filling, data permission processing, and MNDF index information registration.

[0092] Step 3.1 Task Information Population: Call the task data parsing and assembly submodule of the Multi-Machine Positioning Data Fusion Tool (MNDF) to perform structured matching on the loaded task data and batch transfer it to the navigation task data vector of the NPD structure entity file;

[0093] Step 3.2 Device and Environment Information Population: Call the device data parsing and assembly submodule, sensor data parsing and assembly submodule, and map data parsing and assembly submodule of the Multi-device Positioning Data Fusion Tool (MNDF) to perform structured matching on the loaded navigation device, sensor, and map data (if it is the first time to use it, a brand new NPD structure entity file will be generated), and batch transfer the navigation device data vector, sensor data vector, and navigation map data vector to the NPD structure entity file;

[0094] Step 3.3 Data permission processing: Invoke the MNDF data processing submodule of the Multi-Machine Positioning Data Fusion Tool (MNDF) to perform encryption / decryption operations on the NPD structure entity file;

[0095] Step 3.4 NPD Index Information Registration: Update the structure index vector of the NPD structure entity file, including the number of entries, data relationships, key fields, data update status, etc.

[0096] Step 4, Multi-robot Path Pre-execution (Virtual Scene Multi-robot Navigation Path Verification): Use the Multi-robot Path Pre-execution Module (MPF) and call the Multi-robot Localization Data Fusion Tool (MNDF) to perform a trial run verification operation of the multi-robot navigation path in the virtual scene. The sub-steps include virtual scene data loading, NPD data parsing, virtual scene position calculation, virtual path generation, virtual scene multi-robot navigation, and verification result feedback.

[0097] Step 4.1 Virtual Scene Data Loading: The multi-machine path pre-execution module calls the multi-machine positioning data fusion tool (MNDF) through the MNDF loading interface. Through the NPD data processing submodule, it performs data decryption and reading operations on the NPD structure entity file; through the virtual environment generation submodule, it realizes the map generation, identification setting and other operations of the current nuclear power plant working environment. After the construction is completed, it is updated to the process information database of the multi-machine path pre-execution module.

[0098] Step 4.2 NPD Data Parsing: Call the MNDF parsing submodule and obstacle parsing submodule to perform calculations such as category filtering, key point information matching and identifier feature comparison on the navigation task data, robot data, navigation map and sensor data vector in the NPD structure entity file. Among them, the navigation map data and the virtual map perform identifier comparison and identifier incremental update (updating to the virtual map). The calculation results are stored in the process information database of the multi-machine path pre-execution module.

[0099] Step 4.3 Virtual Scene Position Calculation: Read the process information database, perform global coarse positioning and local precise positioning calculations for multiple robots in the virtual environment, and optionally transfer the calculation results to the positioning information vector of the NPD structure entity file through the multi-robot positioning data fusion tool (MNDF) via the data transmission interface. Users can choose whether to perform data encryption operations according to their options and data management requirements.

[0100] Step 4.4 Virtual Path Generation: In the virtual environment, the multi-robot navigation submodule reads the process information database, performs inflection point calculation, local path distance optimization, and multi-robot path cross detection, and generates a navigation path by combining the state information of each robot in the virtual environment, including detailed path information and path execution plan;

[0101] Step 4.5 Multi-robot navigation in virtual scene: Run the navigation path fed back in step 4.4, summarize and record the multi-robot operation information of the whole process after the operation, and batch transfer it to the positioning dataset and navigation route information set of the NPD structure entity file. You can choose whether to perform data encryption operation according to data management requirements.

[0102] Step 4.6 Verification Result Feedback: Based on the multi-machine path test in the virtual environment, update the structure index vector of the NPD structure entity file, including the number of entries, internal data relationships, key fields, data update status, etc.

[0103] Step 5, Real-time Positioning Calculation and Navigation Strategy Generation: Using the Multi-Machine Positioning and Collision Protection Module (MLCP), the Multi-Machine Route Planning Module (MPP), and the Multi-Machine Positioning Data Fusion Tool (MNDF), the robot's real-time position calculation and navigation path data generation operations are performed. Each sub-step includes data loading, data parsing, position calculation, path generation, information filling, and NPD index update.

[0104] Step 5.1 Data Loading: The multi-machine positioning and collision protection module and the multi-machine route planning module respectively call the multi-machine positioning data fusion tool (MNDF) through the data permission processing submodule to perform data decryption and reading operations on the NPD structure entity file;

[0105] Step 5.2 Data parsing: The multi-machine positioning and collision protection module and the multi-machine route planning module respectively call the MNDF parsing submodule and the obstacle parsing submodule to perform calculations such as category filtering, key point information matching and identification feature comparison on the navigation task data vector, navigation device data vector, navigation map vector and sensor data vector in the NPD structure entity file. The calculation results are stored in the process information database of the multi-machine route planning module.

[0106] Step 5.3 Position Calculation: The positioning calculation submodule of the multi-machine positioning and collision protection module reads the process information database, performs global coarse positioning and local precise positioning calculations for each robot in real-time multi-machine tasks, and transmits the calculation results to the positioning information vector of the NPD structure entity file through the multi-machine positioning data fusion tool (MNDF) via the data transmission interface. Data encryption can be performed according to data management requirements.

[0107] Step 5.4 Path Generation: The navigation algorithm submodule of the multi-robot route planning module reads the process information database, performs inflection point calculation, local path distance optimization, obstacle avoidance marking based on the obstacle marker set obtained in Step 5.3, and multi-robot path cross detection, and generates a navigation strategy, including path information and path execution plan, by combining the current operating status information of the navigation device; the robot body can autonomously perform path navigation in an offline environment;

[0108] Step 5.5 Information Population: Perform structured matching on the calculated location information and navigation strategy information, and batch transfer the location information vector and navigation strategy information vector to the NPD structure entity file through the Multi-Machine Positioning Data Fusion Tool (MNDF). Data encryption can be performed according to data management requirements.

[0109] Step 5.6 NPD Index Information Update: Update the structure index vector of the NPD structure entity file, including the number of entries, data relationships, key fields (all data / information sets), data update status, etc.

[0110] This disclosure can be a system, method, and / or computer program product. A computer program product may include a computer-readable storage medium having computer-readable program instructions loaded thereon for causing a processor to implement various aspects of this disclosure.

[0111] Computer-readable storage media can be tangible devices capable of holding and storing instructions for use by an instruction execution device. Computer-readable storage media can be, for example—but not limited to—electrical storage devices, magnetic storage devices, optical storage devices, electromagnetic storage devices, semiconductor storage devices, or any suitable combination thereof. More specific examples (a non-exhaustive list) of computer-readable storage media include: portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), static random access memory (SRAM), portable compact disc read-only memory (CD-ROM), digital multifunction disc (DVD), memory sticks, floppy disks, mechanical encoding devices, such as punch cards or recessed protrusions storing instructions thereon, and any suitable combination thereof. The computer-readable storage media used herein are not to be construed as transient signals themselves, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through waveguides or other transmission media (e.g., light pulses through fiber optic cables), or electrical signals transmitted through wires.

[0112] The computer-readable program instructions described herein can be downloaded from computer-readable storage media to various computing / processing devices, or downloaded via a network, such as the Internet, local area network, wide area network, and / or wireless network, to an external computer or external storage device. The network may include copper transmission cables, fiber optic transmission, wireless transmission, routers, firewalls, switches, gateway computers, and / or edge servers. A network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards them to the computer-readable storage media in the respective computing / processing device.

[0113] Computer program instructions used to perform the operations of this disclosure may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, status setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Smalltalk, C++, etc., and conventional procedural programming languages ​​such as the "C" language or similar programming languages. The computer-readable program instructions may execute entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or may be connected to an external computer (e.g., via the Internet using an Internet service provider). In some embodiments, electronic circuitry, such as programmable logic circuitry, field-programmable gate arrays (FPGAs), or programmable logic arrays (PLAs), is personalized by utilizing the status information of the computer-readable program instructions to implement various aspects of this disclosure.

[0114] Various aspects of this disclosure are described herein with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this disclosure. It should 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-readable program instructions.

[0115] These computer-readable program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus to produce a machine such that, when executed by the processor of the computer or other programmable data processing apparatus, they create means for implementing the functions / actions specified in one or more blocks of the flowchart and / or block diagram. These computer-readable program instructions can also be stored in a computer-readable storage medium that causes a computer, programmable data processing apparatus, and / or other device to operate in a particular manner; thus, the computer-readable medium storing the instructions comprises an article of manufacture that includes instructions for implementing aspects of the functions / actions specified in one or more blocks of the flowchart and / or block diagram.

[0116] Computer-readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device to cause a series of operational steps to be performed on the computer, other programmable data processing apparatus, or other device to produce a computer-implemented process, thereby causing the instructions executed on the computer, other programmable data processing apparatus, or other device to perform the functions / actions specified in one or more boxes of a flowchart and / or block diagram.

[0117] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of an instruction containing one or more executable instructions for implementing a specified logical function. In some alternative implementations, the functions marked in the blocks may occur in a different order than those shown in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, may be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions.

[0118] The various embodiments of this disclosure have been described above. These descriptions are exemplary and not exhaustive, nor are they limited to the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described embodiments. The terminology used herein is chosen to best explain the principles, practical application, or improvement of the technology in the market, or to enable others skilled in the art to understand the embodiments disclosed herein.

Claims

1. A multi-robot path planning system based on a multi-source sensor data fusion architecture, characterized in that, The system includes: Multi-machine positioning data fusion tool is used to parse, assemble and fuse data from multiple functional modules within the system, and to generate, update and manage standard documents for the navigation planning data warehouse; The navigation planning data warehouse is connected to the multi-machine positioning data fusion tool and is used to store data entity files with a standard structure generated by the multi-machine positioning data fusion tool, providing unified data services for the system. Multiple functional modules are included, such as a navigation task planning module, a robot equipment scheduling module, a navigation environment data processing module, a multi-machine path pre-execution module, a multi-machine positioning and collision protection module, and a multi-machine route planning module; each of the functional modules interacts with the multi-machine positioning data fusion tool through a data transmission interface.

2. The system according to claim 1, characterized in that, The multi-machine positioning data fusion tool includes: The task data parsing and assembly submodule is used to obtain task information from the navigation task planning module, and generate a navigation task parsing and assembly result data set after parsing and assembling. The equipment data parsing and assembly submodule is used to obtain equipment parameter information from the robot equipment scheduling module, and generate a set of navigation equipment parsing and assembly result data after parsing and assembly. The map data parsing and assembly submodule is used to obtain map and environment information from the navigation environment data processing module, and generate a map parsing and assembly result data set after parsing and assembly. The sensor data parsing and assembly submodule is used to obtain sensor data from the robot equipment scheduling module, and generate a set of sensor parsing and assembly result data after parsing and assembling. The navigation planning data warehouse processing submodule is used to merge, match, and standardize the data sets of the above-mentioned parsing and assembly results, and to perform the generation, updating, encryption / decryption operations of the navigation planning data warehouse library files; The navigation planning data warehouse interaction interface is used to perform real-time data interaction between the result data generated by the navigation planning data warehouse processing submodule and the generated navigation planning data warehouse result file. The data transmission interface is used to enable data interaction between the multi-machine positioning data fusion tool and other functional modules within the system.

3. The system according to claim 1, characterized in that, The structured entity file of the navigation planning data warehouse contains the following datasets: The data warehouse index is used to record the registration information, relationships, and update status of all datasets in the navigation planning data warehouse entity file; Task dataset, used to record information related to all loaded navigation tasks; Robot information set, used to record all loaded navigation device-related information; An environmental map dataset is used to record all loaded maps and related environmental information. Sensor dataset, used to record information related to all loaded sensor acquisition data; Navigation and positioning dataset, used to record updated location calculation results; The navigation route information set is used to record the updated navigation strategy calculation results.

4. The system according to claim 1, characterized in that, The robot equipment scheduling module also includes a multi-machine communication interface for enabling direct data interaction between different robot devices.

5. The system according to claim 1, characterized in that, The multi-machine path pre-execution module includes: The multi-machine positioning data fusion tool loading interface is used to call the calculation results fed back by the multi-machine positioning data fusion tool; The virtual environment generation submodule is used to generate a test virtual scene containing a map, obstacle markers, and navigation points based on the loaded data. The virtual navigation submodule is used to perform multi-robot path operation tests in the virtual scene.

6. The system according to claim 1, characterized in that, The multi-machine positioning and collision protection module includes: The sensor data parsing submodule is used to parse obstacle avoidance related data; The localization calculation submodule is used to return the localization results and obstacle identification results of multiple robots based on the parsed data.

7. The system according to claim 1, characterized in that, The multi-machine route planning module includes: The obstacle data parsing submodule is used to acquire and parse obstacle data from the navigation planning data warehouse; The navigation algorithm submodule is used to combine the real-time location of the navigation device, map data, and parsed obstacle data to generate the final navigation path and detailed motion parameters for multiple robots.

8. The system according to claim 2, characterized in that, The navigation planning data warehouse processing submodule fuses data from different data sources by applying sensor information fusion technology or multi-source fusion positioning methods.

9. A multi-robot path planning method based on a multi-source sensor data fusion architecture, characterized in that, The method is implemented based on the system according to any one of claims 1-8, and the method includes: The multi-machine positioning data fusion tool obtains navigation task data, device and sensor data, and map and environment data from the navigation task planning module, robot equipment scheduling module, and navigation environment data processing module, respectively, and parses and assembles the various types of data to generate corresponding standardized data sets. The standardized data set is merged, matched, and normalized to generate and update a navigation planning data warehouse file that conforms to a preset standard. The navigation planning data warehouse file contains interrelated data warehouse indexes, task datasets, robot information sets, environmental map datasets, sensor datasets, navigation and positioning datasets, and navigation route information sets. Based on the data loaded in the navigation planning data warehouse file, a test scenario containing a map, obstacle markers and navigation points is generated in a virtual environment, and a multi-robot path operation test is executed to verify the feasibility of the navigation strategy. During the actual operation of the robot, based on the navigation planning data warehouse file and the real-time sensor data, the robot performs real-time positioning calculations and obstacle analysis for multiple robots, and generates positioning results and obstacle avoidance feedback. Integrating virtually validated navigation strategies, real-time positioning results, and obstacle information, it generates multi-robot navigation paths and motion parameters for actual execution.

10. A non-volatile computer-readable storage medium storing computer program instructions thereon, characterized in that, When the computer program instructions are executed by the processor, they implement the method of claim 9.

Images