A Long-Term SLAM Method and System for Inspection Robots Adapting to Environmental Changes

Abstract

This invention discloses a long-term SLAM method and system for inspection robots that adapt to environmental changes. The method includes the following steps: S1: Input the collected current frame laser data, and match the current frame laser data with a temporary sub-map sequence and a global map; S2: Output the location and / or point cloud of the successfully matched current frame laser data, and determine whether the current frame laser data can be inserted into the temporary sub-map sequence. If it can, update the temporary sub-map sequence; S3: Based on the updated temporary sub-map sequence, determine whether the current frame laser data can update the global map. If it can, update the global map. This application utilizes the updated temporary sub-map to slowly update the global map, maintaining the consistency and accuracy of the map and the environment. Simultaneously, it performs regular backups of the global map and employs a map loading strategy to load the map, improving the inspection robot's survivability and inspection efficiency.

Description

Technical Field

[0001] This invention relates to the field of SLAM technology, and in particular to a long-term SLAM method and system for an inspection robot that adapts to environmental changes. Background Technology

[0002] SLAM (Simultaneous Localization and Mapping) technology enables robots to move from an unknown location in an unknown environment. During the movement, the robot performs localization based on its location and a map, and simultaneously builds an incremental map based on its localization, thus achieving autonomous localization and navigation.

[0003] The robot first builds an environmental map using SLAM (Simultaneous Localization and Mapping), and then uses this map for localization and navigation, enabling it to replace humans in security, inspection, and other tasks. In existing technologies, inspection robots are typically required to work for extended periods in certain scenarios. However, due to human factors such as vehicles and buildings, or seasonal changes such as trees and weeds, the operating environment can change significantly, creating a large discrepancy with the map. This can lead to inaccurate or undetectable localization, requiring the robot to rebuild the map, thus reducing its independent survivability and inspection efficiency. Summary of the Invention

[0004] The purpose of this invention is to provide a long-term SLAM method for inspection robots that can adapt to environmental changes. This invention provides a SLAM method that can adapt to environmental changes. When the inspection environment in which the robot is located changes, this method can achieve accurate positioning without rebuilding the map, thereby improving the robot's survivability in the environment.

[0005] To achieve the above objectives, this invention provides a long-term SLAM method for inspection robots that adapt to environmental changes, the method comprising the following steps:

[0006] S1: Input the collected current frame laser data, and match the current frame laser data with a temporary sub-map sequence and a global map; the temporary sub-map sequence includes several temporary sub-maps;

[0007] S2: Output the location and / or point cloud of the successfully matched current frame laser data, determine whether the current frame laser data can be inserted into the temporary sub-map sequence, and if so, insert the current frame laser data into the temporary sub-map sequence and update the temporary sub-map sequence;

[0008] S3: Based on the updated temporary sub-map sequence, determine whether the current frame laser data can update the global map. If so, update the global map.

[0009] In one embodiment of the present invention, determining whether the current frame laser data can be inserted into the temporary sub-map sequence includes:

[0010] The current frame laser data is registered with the temporary sub-map sequence. If the registration score is greater than the first preset threshold, the robot's movement distance is compared with the time interval between the current frame laser data and the last time the temporary sub-map was inserted.

[0011] If the moving distance is greater than the second preset threshold, the current frame laser data is inserted as a temporary sub-map into the temporary sub-map sequence.

[0012] In one embodiment of the present invention, determining whether the current frame laser data can update the global map includes:

[0013] The current frame laser data is registered with the global map. If the registration score is greater than a third preset threshold, the time interval between the current frame laser data and the most recently updated frame data is compared.

[0014] If the time interval is greater than a fourth preset threshold, the current frame laser data will be used to update the global map.

[0015] In one embodiment of the present invention, updating the global map specifically includes:

[0016] S301: Find the spatial index of the current frame laser data in the global map based on the location of the current frame laser data;

[0017] S302: Determine whether laser data exists in the index space. If not, directly insert the current frame laser data into the global map; if it exists, put the laser data in the index space into the deletion queue.

[0018] S303: Compare the difference between the positioning yaw angle of the current frame laser data and the positioning yaw angle of any laser data in the deletion queue one by one, and remove the laser data in the deletion queue whose difference exceeds the fifth preset threshold.

[0019] S304: Determine whether the update time of the laser data corresponding to the difference in the queue to be deleted that does not exceed the sixth preset threshold exceeds the preset time. If it does not exceed the preset time, abandon the insertion of the current frame laser data of the inserted frame data. If it exceeds the preset time, insert the current frame laser data into the index space and delete the remaining laser data in the queue to be deleted.

[0020] In one embodiment of the present invention, the third preset threshold is greater than the first preset threshold.

[0021] In one embodiment of the present invention, the global map is also backed up periodically. The periodic backup includes original map backup, daily backup and quarterly backup. The original map backup is backed up only once. The daily backup is backed up once a day and replaces the previous daily backup. The quarterly backup is backed up once a quarter and replaces the corresponding quarterly backup of the previous year.

[0022] In one embodiment of the present invention, the inspection robot loads a map using a map loading strategy, the map loading strategy including:

[0023] Check if the original map exists. If it does not exist, generate an error message to the user indicating that the original map is missing and map loading has failed.

[0024] If it exists, check if a daily backup exists; if it exists, load the daily backup.

[0025] If it does not exist, check if a quarterly backup exists; if it does, load the latest quarterly backup.

[0026] In one embodiment of the present invention, when updating the temporary sub-map sequence, the current frame laser data is inserted into the temporary sub-map sequence according to the first-in-first-out principle.

[0027] This invention also provides a long-term SLAM system for inspection robots that adapt to environmental changes, characterized in that it includes:

[0028] At least one processor;

[0029] At least one memory for storing at least one program;

[0030] When the at least one program is executed by the at least one processor, the at least one processor implements the long-term SLAM method for inspection robots that adapts to environmental changes as described above.

[0031] Compared with existing technologies, this invention registers and updates the collected current frame laser data with a temporary sub-map sequence, and then selectively updates the global map using the updated temporary sub-map. This achieves differentiated updates between the temporary sub-map and the global map, ensuring both timely updates of the temporary sub-map and the accuracy of the updated global map data. This prevents permanent environmental changes from affecting the robot's positioning, maintaining consistency and accuracy between the map and the environment. Furthermore, regular backups are performed after the global map is updated, and a map loading strategy is used to load the map, improving the robot's survivability and inspection efficiency. Attached Figure Description

[0032] Figure 1This is a flowchart of a long-term SLAM method for an inspection robot that adapts to environmental changes according to an embodiment of the present invention.

[0033] Figure 2 This is a flowchart of a method for inserting a temporary submap according to an embodiment of the present invention;

[0034] Figure 3 This is a flowchart of a method for updating a global map according to an embodiment of the present invention;

[0035] Figure 4 This is a flowchart of a loading strategy according to an embodiment of the present invention. Detailed Implementation

[0036] The specific embodiments of the present invention will now be described in detail with reference to the accompanying drawings, but it should be understood that the scope of protection of the present invention is not limited to the specific embodiments.

[0037] Unless otherwise expressly stated, throughout the specification and claims, the term "comprising" or its variations such as "including" or "comprises" shall be understood to include the stated elements or components without excluding other elements or other components.

[0038] In the description of this invention, it should be noted that the terms "center," "upper," "lower," "left," "right," "vertical," "horizontal," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are used only for the convenience of describing the invention and for simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the invention. The terms "first," "second," and "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance. Furthermore, unless otherwise explicitly specified and limited, the terms "installed," "connected," and "linked" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0039] The global map described in this application is stored in an octree manner. In the laser data collected by the inspection robot, a frame of laser data includes the location of the frame of laser data and the point cloud of the frame of laser data. When inserting the map, the frame of laser data is inserted into the corresponding spatial position in the octree map according to the location of the laser data. If a point cloud map is needed, the point cloud in the laser data can be projected onto the world coordinate system through its location.

[0040] like Figure 1 As shown, a long-term SLAM method for an inspection robot adapted to environmental changes, according to a preferred embodiment of the present invention, includes the following steps:

[0041] S1: Input the collected current frame laser data, and match the current frame laser data with the temporary sub-map sequence and the global map; the temporary sub-map sequence includes several temporary sub-maps.

[0042] In this embodiment, the inspection robot detects the current environment to obtain the current frame laser data. The inspection robot divides the pre-constructed map into a global map and several temporary sub-maps. The robot matches the collected current frame laser data with the sequence of temporary sub-maps and the global map to determine whether the image formed by the collected current frame laser data is in the global map and the sequence of temporary sub-maps.

[0043] If the match is successful, proceed to step S2 and subsequent steps; if the match is unsuccessful, delete the current frame laser data and input new laser data to perform the match again.

[0044] S2: Output the location and / or point cloud of the current frame laser data that has been successfully matched, determine whether the current frame laser data can be inserted into the temporary sub-map sequence, and if so, insert the current frame laser data into the temporary sub-map sequence and update the temporary sub-map sequence.

[0045] In this embodiment, the positioning location and / or point cloud of the current frame laser data are output. If a point cloud map is needed, the point cloud in the current frame laser data is projected onto the world coordinate system using its positioning location. It is determined whether the current frame laser data can be inserted into a temporary sub-map sequence. If it can, the temporary sub-map is updated to improve the inspection robot's independent survival capability in the field.

[0046] Preferably, when updating the temporary sub-map sequence, the current frame laser data is inserted into the temporary sub-map sequence according to the first-in-first-out principle.

[0047] In this embodiment, the temporary sub-map periodically stores laser data over a certain period. Each time the laser data of the current frame is acquired, it is registered with both the temporary sub-map and the global map. The successfully registered laser data of the current frame is added to the temporary sub-map sequence to provide the latest environmental data during registration, addressing rapid environmental changes and preventing matching failures due to the map's inability to provide effective reference data when environmental changes are too drastic. The temporary sub-map is stored using a sliding window-like method. When new laser data is inserted, the oldest data in the temporary sub-map sequence is deleted, thus maintaining the size of the temporary sub-map.

[0048] S3: Based on the updated temporary sub-map sequence, determine whether the current frame laser data can update the global map. If so, update the global map.

[0049] In this embodiment of the application, when the collected current frame laser data is inserted into the temporary sub-map sequence, it will only affect the current short-term positioning, while when it is inserted into the global map, it will be saved for a long time. Therefore, in order to ensure the accuracy and consistency of the global map, the update frequency of the global map is lower than that of the temporary sub-map.

[0050] In a preferred embodiment of this application, the ratio of the current frame laser data updating temporary sub-map data to the data updating the global map is 1:5.

[0051] like Figure 2 As shown, determining whether the laser data can be inserted into the temporary sub-map sequence includes:

[0052] The current frame laser data is registered with the sub-map sequence. If the registration score is greater than the first preset threshold, the robot's movement distance is compared with the interval between the current frame laser data and the most recent insertion of the temporary sub-map.

[0053] If the moving distance is greater than the second preset threshold, the current frame laser data is inserted as a temporary sub-map into the temporary sub-map sequence.

[0054] By setting a first preset threshold, it is possible to prevent the insertion of poorly registered laser data into the temporary sub-map sequence, which could lead to map errors. Preferably, the first preset threshold can be changed according to user needs, and the second preset threshold can be determined based on the resolution of the temporary sub-map and the global map.

[0055] The step of determining whether the current frame laser data can update the global map includes registering the current frame laser data with the global map; if the registration score is greater than a third preset threshold, then comparing the time interval between the current frame laser data and the most recent updated frame data.

[0056] If the time interval is greater than a fourth preset threshold, the current frame laser data will be used to update the global map.

[0057] In this embodiment, the global map update uses the current frame laser data interval to insert or replace old data, slowly updating the map. This aims to address the positioning problem when the environment undergoes permanent changes, maintaining consistency between the map and the environment. By setting a third preset threshold greater than the first preset threshold, the registration score during global map updates can be improved, ensuring the accuracy and slowness of the global map update, while reducing the probability of errors and improving the reliability of the inspection robot's positioning and navigation. Preferably, the third preset threshold is greater than the first preset threshold, and the fourth preset threshold is greater than 0.5 hours.

[0058] Furthermore, such as Figure 3 As shown, updating the global map specifically includes:

[0059] S301: Find the spatial index of the current frame laser data in the global map based on the location of the current frame laser data;

[0060] S302: Determine whether laser data exists in the index space. If not, directly insert the current frame laser data into the global map; if it exists, put the laser data in the index space into the deletion queue.

[0061] S303: Compare the difference between the positioning yaw angle of the current frame laser data and the positioning yaw angle of any laser data in the deletion queue one by one, and remove the laser data in the deletion queue whose difference exceeds the fifth preset threshold.

[0062] S304: Determine whether the update time of the laser data corresponding to the difference in the queue to be deleted that does not exceed the sixth preset threshold exceeds the preset time. If it does not exceed the preset time, then abandon the insertion of the current frame laser data; if it exceeds the preset time, then insert the current frame laser data into the index space and delete the remaining laser data in the queue to be deleted.

[0063] In this embodiment, the global map is stored using an octree map. An index is built at each node of the octree. When the registered current frame laser data meets the conditions for updating the global map, the spatial index of the current frame laser data in the octree map is found based on the location of the current frame laser data. Laser data within the index space is then found. If there is no laser data within the index space, the current frame laser data is directly inserted into the octree map to update the global map. If there is laser data within the index space, the laser data within the index space is placed in a deletion queue. The yaw angle of the current frame laser data is compared one by one with the yaw angle of any laser data in the deletion queue. The difference in yaw angles is used to determine whether the laser data overlaps. The larger the difference in yaw angles, the smaller the overlap between the current frame laser data and the laser data in the deletion queue; conversely, the smaller the difference in yaw angles, the greater the overlap.

[0064] If the difference exceeds a fifth preset threshold, meaning the overlap between the current frame laser data and the laser data in the deletion queue is small, the laser data in the deletion queue is deleted. If the difference does not exceed the fifth preset threshold, meaning the overlap between the current frame laser data and the laser data in the deletion queue is large, then it is determined whether the update time of the laser data in the deletion queue whose difference does not exceed a sixth preset threshold exceeds a preset time. If it does not exceed the preset time, it indicates that the spatial index has been recently updated, and the insertion of the current frame laser data is abandoned. If it exceeds the preset time, it indicates that there is no recently updated data in the spatial index, and the current frame laser data is inserted into the index space of the octree map, the global map is updated, and the remaining laser data in the deletion queue is deleted. Preferably, the fifth preset threshold is less than the sixth preset threshold, which further improves the standard for judging the overlap between the current frame laser data and the laser data in the deletion queue, ensuring the accuracy of the global map update and reducing the probability of update errors.

[0065] In other embodiments of this application, the fifth preset threshold is equal to the sixth preset threshold.

[0066] In a preferred embodiment of this application, the preset time is set within one month.

[0067] Furthermore, the global map is also backed up regularly, including original map backup, daily backup and quarterly backup. The original map backup is backed up only once, the daily backup is backed up every day and replaces the previous daily backup, and the quarterly backup is backed up once every quarter and replaces the corresponding quarterly backup of the previous year.

[0068] In this application embodiment, regular backup is performed to prevent the need to rebuild the map when the inspection environment of the inspection robot changes, thereby improving the inspection robot's ability to survive independently and its inspection efficiency. Preferably, this application uses three-level backup to save map data, namely, original map backup, daily backup and quarterly backup.

[0069] The first level of backup involves backing up the original map only once, ensuring the global map remains unchanged at all times. In the event of a long-term SLAM error, the system can gradually restore to the original map file without affecting it and necessitating re-mapping, thus improving the inspection efficiency of the robot. The second level of backup involves daily backups, replacing the previous daily backup file each day. Only one backup is created to prevent excessive backups from impacting the robot's efficiency. The third level of backup involves quarterly backups, replacing the corresponding quarterly backup file from the previous year. A total of four quarterly backup files are generated annually, with subsequent backups continuously overwriting the existing quarterly files.

[0070] When the inspection robot has been charging for more than 5 minutes and the last backup time is more than 24 hours or three months, the inspection robot will be successfully triggered to perform daily or quarterly map backups. At this time, the inspection robot will be speed-limited and its threads will be locked, stopping global map data modification. It can only be restored after the backup is completed.

[0071] In this embodiment of the application, the robot needs to load a map when it is powered on or restarted. The map loading strategy is as follows: Figure 4 As shown, when the robot needs to load a map, it first checks if the original map file exists in the map folder. If the original map file does not exist, an error message is generated indicating that the original map file is missing, and the robot fails to load the map. If the original map file exists, it checks if a daily backup file exists. If it exists, the daily backup map file is loaded. If the daily backup file does not exist, it checks if a quarterly backup file exists. If a quarterly backup file exists, the latest quarterly backup file is loaded. If no quarterly backup file exists, the original map file is loaded. The robot can check backup files step-by-step and load the map in this manner to prevent map errors during loading or positioning, ensuring the reliability of the inspection robot's positioning and navigation, thereby improving the inspection robot's survivability in the field.

[0072] In this embodiment, the robot loads a map during startup or restart and performs navigation and positioning based on the map. The temporary sub-map sequence includes several temporary sub-maps. The robot matches the current frame laser data collected by the robot with the temporary sub-map sequence and the global map. It determines whether the current frame laser data is in the temporary sub-map sequence and the global map. The location and / or point cloud of the successfully matched current frame laser data are output. It is also determined whether the current frame laser data can be inserted into the sub-map sequence. If the registration score between the current frame laser data and the temporary sub-map is greater than a first preset threshold and the robot's movement distance within the interval between the current frame laser data and the most recent insertion into the temporary sub-map is greater than a second preset threshold, then the current frame laser data is inserted into the temporary sub-map sequence to update the sub-map. It is also determined whether the current frame laser data can update the global map. If it can, then the global map is updated. By selectively updating the global map using updated temporary sub-maps, differentiated updates between the temporary and global maps are achieved. This ensures both timely updates of the temporary sub-maps and the accuracy of the updated global map data, preventing permanent environmental changes from affecting the robot's localization and maintaining consistency and accuracy between the map and the environment. Additionally, the global map is backed up regularly to prevent long-term SLAM errors from affecting the original map and requiring remapping. This improves the robot's survivability and inspection efficiency.

[0073] A preferred embodiment of the present invention also provides a long-term SLAM system for an inspection robot that adapts to environmental changes, comprising:

[0074] At least one processor;

[0075] At least one memory for storing at least one program;

[0076] When the at least one program is executed by the at least one processor, the at least one processor implements the long-term SLAM method for inspection robots that adapts to environmental changes as described above.

[0077] Another embodiment of the present invention provides a long-term SLAM system for inspection robots that adapt to environmental changes, comprising:

[0078] The acquisition unit is used by the inspection robot to acquire the laser data of the current frame;

[0079] A matching unit is used to match the current frame laser data with the temporary sub-map sequence and the global map to determine whether the laser data is in the temporary sub-map and the global map; the temporary sub-map sequence includes several temporary sub-maps;

[0080] The first registration unit is used to output the location and / or point cloud of the successfully matched current frame laser data, register the current frame laser data with the temporary sub-map, and determine whether the current frame laser data can be inserted into the temporary sub-map sequence.

[0081] The first update module is used to insert the current frame laser data that can be inserted into the temporary sub-map sequence into the temporary sub-map sequence, thereby updating the temporary sub-map;

[0082] The second registration unit is used to register the current frame laser data with the global map according to the updated temporary sub-map sequence, and to determine whether the current frame laser data can update the global map.

[0083] The second update module is used to update the global map with the current frame laser data that meets the conditions for updating the global map;

[0084] The backup module is used to perform regular backups after a global map update; the regular backups include original map backups, daily backups, and quarterly backups.

[0085] The loading module is used to reload the map according to the loading strategy when the inspection robot is powered on or restarted.

[0086] This invention acquires current frame laser data through an acquisition unit. The current frame laser data is then matched with a temporary sub-map sequence and a global map in a matching unit. The system determines whether the current frame laser data is present in either the temporary sub-map or the global map. Successfully matched current frame laser data is registered with the temporary sub-map sequence through a first registration unit, which then determines whether the current frame laser data can be inserted into the temporary sub-map sequence. A first update template is used to insert the current frame laser data that can be inserted into the temporary sub-map sequence, updating the temporary sub-map. A second registration unit determines whether the current frame laser data can update the global map based on the updated temporary sub-map and selectively updates the global map using a second update template. The global map is periodically backed up using a backup module to prevent long-term SLAM errors from affecting the original map and requiring remapping. When the inspection robot is powered on or restarted, a loading strategy is used to load the map, ensuring the reliability of the inspection robot's positioning and navigation.

[0087] This disclosure provides a non-volatile computer storage medium storing computer-executable instructions that can perform the method steps described in the above embodiments.

[0088] It should be noted that the computer-readable medium described in this disclosure can be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. A computer-readable storage medium can be, for example,—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this disclosure, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in connection with an instruction execution system, apparatus, or device. In this disclosure, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium can be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to: wires, optical fibers, RF (radio frequency), etc., or any suitable combination thereof.

[0089] The aforementioned computer-readable medium may be included in the aforementioned electronic device; or it may exist independently and not assembled into the electronic device.

[0090] Computer program code for performing the operations of this disclosure can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, and conventional procedural programming languages ​​such as the "C" language or similar programming languages. The program code can be executed 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 remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (AN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0091] The foregoing description of specific exemplary embodiments of the invention is for illustrative and explanatory purposes. These descriptions are not intended to limit the invention to the precise forms disclosed, and it will be apparent that many changes and variations can be made in accordance with the foregoing teachings. The exemplary embodiments were chosen and described in order to explain the specific principles of the invention and its practical application, thereby enabling those skilled in the art to implement and utilize various different exemplary embodiments of the invention, as well as various different choices and variations. The scope of the invention is intended to be defined by the claims and their equivalents.

Claims

1. A long-term SLAM method for an inspection robot that adapts to environmental changes, characterized in that, The method includes the following steps: S1: Input the collected current frame laser data, and match the current frame laser data with a temporary sub-map sequence and a global map; the temporary sub-map sequence includes several temporary sub-maps; S2: Output the location and / or point cloud of the successfully matched current frame laser data, determine whether the current frame laser data can be inserted into the temporary sub-map sequence, and if so, insert the current frame laser data into the temporary sub-map sequence and update the temporary sub-map sequence; S3: Based on the updated temporary sub-map sequence, determine whether the current frame laser data can update the global map; if so, update the global map. The determination of whether the current frame laser data can be inserted into the temporary sub-map sequence includes: The current frame laser data is registered with the temporary sub-map sequence. If the registration score is greater than the first preset threshold, the robot's movement distance is compared with the time interval between the current frame laser data and the last time the temporary sub-map was inserted. If the moving distance is greater than the second preset threshold, the current frame laser data is inserted as a temporary sub-map into the temporary sub-map sequence.

2. The long-term SLAM method for an inspection robot adapting to environmental changes as described in claim 1, characterized in that, The step of determining whether the current frame laser data can update the global map includes: The current frame laser data is registered with the global map. If the registration score is greater than a third preset threshold, the time interval between the current frame laser data and the most recently updated frame data is compared. If the time interval is greater than a fourth preset threshold, the current frame laser data will be used to update the global map.

3. The long-term SLAM method for an inspection robot adapting to environmental changes as described in claim 2, characterized in that, Updating the global map specifically includes: S301: Based on the location of the current frame laser data, find the index space corresponding to the current frame laser data in the global map; S302: Determine whether laser data exists in the index space. If not, directly insert the current frame laser data into the global map; if it exists, put the laser data in the index space into the deletion queue. S303: Compare the difference between the positioning yaw angle of the current frame laser data and the positioning yaw angle of any laser data in the deletion queue one by one, and remove the laser data in the deletion queue whose difference exceeds the fifth preset threshold. S304: Determine whether the update time of the laser data corresponding to the difference in the queue to be deleted that does not exceed the sixth preset threshold exceeds the preset time. If it does not exceed the preset time, then abandon the insertion of the current frame laser data; if it exceeds the preset time, then insert the current frame laser data into the index space and delete the remaining laser data in the queue to be deleted.

4. The long-term SLAM method for an inspection robot adapting to environmental changes as described in claim 2, characterized in that, The third preset threshold is greater than the first preset threshold.

5. The long-term SLAM method for an inspection robot adapting to environmental changes as described in claim 1, characterized in that, The global map is also backed up regularly, including original map backup, daily backup and quarterly backup. The original map backup is backed up only once. The daily backup is backed up once a day and replaces the previous daily backup. The quarterly backup is backed up once a quarter and replaces the corresponding quarterly backup of the previous year.

6. The long-term SLAM method for an inspection robot adapting to environmental changes as described in claim 1, characterized in that, The inspection robot loads a map using a map loading strategy, which includes: Check if the original map exists. If it does not exist, generate an error message to the user indicating that the original map is missing and map loading has failed. If it exists, check if a daily backup exists; if it exists, load the daily backup. If it does not exist, check if a quarterly backup exists; if it does, load the latest quarterly backup.

7. The long-term SLAM method for an inspection robot adapting to environmental changes as described in claim 1, characterized in that, When updating the temporary sub-map sequence, the current frame laser data is inserted into the temporary sub-map sequence according to the first-in-first-out principle.

8. A long-term SLAM system for an inspection robot that adapts to environmental changes, characterized in that, include: At least one processor; At least one memory for storing at least one program; When the at least one program is executed by the at least one processor, the at least one processor implements the long-term SLAM method for an inspection robot that adapts to environmental changes as described in any one of claims 1-7.

Images