Robot navigation switching methods, devices, equipment, and media based on work scenarios

Abstract

This application relates to a robot navigation switching method, apparatus, device, and medium based on a work scenario, belonging to the technical field of robot route planning. The method includes acquiring multiple images of a robot's work area to be analyzed, where the multiple images are consecutive frames; identifying the multiple images to obtain a first work scenario to be determined for each image; if multiple first work scenarios match, selecting the first work scenario as the target work scenario; selecting a target navigation map corresponding to the target work scenario from a navigation map information database; and generating control commands based on the target navigation map to enable the robot to travel based on the target navigation map. This application effectively reduces the complexity of path planning calculations by determining the target work scenario from multiple images and navigating using the target navigation map corresponding to the target work scenario, ensuring the robot quickly and accurately reaches the target location.

Description

Technical Field

[0001] This application relates to the technical field of robot route planning, and in particular to a robot navigation switching method, device, equipment and medium based on a work scenario. Background Technology

[0002] With the development of technology, robots have been widely used in many fields such as warehousing, logistics, medical care and services due to their significant advantages such as high efficiency, precision and repeatable tasks, and have become a key force in promoting the intelligent upgrading of various industries.

[0003] Typically, robots need to perform various tasks in large, complex, and finely partitioned environments. For example, in a large smart warehouse, the internal space is rationally planned into multiple functional areas such as shelving areas, picking areas, charging areas, and packaging areas. Each area has its specific task requirements and operating rules. Similarly, in a hospital environment, there are different areas such as the outpatient hall, inpatient corridors, and the perimeter of the operating room. These areas differ significantly in terms of personnel flow, spatial layout, and functional uses.

[0004] Currently, mainstream robot navigation methods mainly rely on a pre-built single global map. However, this seemingly mature navigation model has revealed many obvious shortcomings in practical applications.

[0005] On the one hand, when the environment is extremely large, the amount of data covered by a single global map grows exponentially, becoming exceptionally large. This not only puts enormous pressure on the robot's storage system but also causes the computational complexity of path planning algorithms to increase dramatically when processing such massive amounts of data. The significantly extended algorithm runtime leads to a severe decrease in the real-time performance of robot path planning, which in turn greatly affects the robot's navigation efficiency, making it difficult for it to reach the target location quickly and accurately.

[0006] On the other hand, navigation based on a single global map makes the robot lack the ability to perceive the specific functional scene in which it is located. Because it cannot accurately identify the current scene, the robot cannot flexibly adjust its navigation behavior according to the characteristics of different scenes.

[0007] Therefore, there is an urgent need in this field for a solution that enables robots to automatically, accurately, and smoothly switch between different scenarios. Summary of the Invention

[0008] To address the aforementioned technical problems, this application provides a robot navigation switching method, apparatus, device, and medium based on a work scenario.

[0009] Firstly, this application provides a robot navigation switching method based on a work scenario, employing the following technical solution:

[0010] A robot navigation switching method based on a work scenario includes:

[0011] Acquire multiple images of the robot's working area to be analyzed, wherein the multiple images to be analyzed are consecutive frames;

[0012] Multiple images to be analyzed are identified to obtain a first working scene to be determined corresponding to each image to be analyzed;

[0013] If multiple first work scenarios to be determined match, the first work scenario to be determined shall be taken as the target work scenario;

[0014] Select the target navigation map corresponding to the target work scenario from the navigation map information database;

[0015] Generate control commands based on the target navigation map to enable the robot to drive based on the target navigation map, wherein generating control commands based on the target navigation map includes: obtaining the driving segment of the robot on the target navigation map, the driving segment including obstacle areas and open areas; determining the driving speed of the robot based on the driving segment; and generating the control commands based on the driving speed.

[0016] By adopting the above technical solution, the target working scene is identified through the image to be analyzed, and the target navigation map corresponding to the target working scene is called. Since the data volume of a single scene map is much smaller than that of the global map, the pressure on the robot's storage system is greatly reduced. At the same time, the computational complexity of the path planning algorithm is reduced, the real-time performance of path planning is improved, and the robot is ensured to reach the target position quickly and accurately. Scene perception is achieved by recognizing multiple consecutive frames of the image to be analyzed, and a verification step is added to match multiple first working scenes to be determined. Compared with single image recognition, the consistency verification of consecutive frames can avoid scene misjudgment caused by image blurring or partial occlusion, ensuring the accuracy of target working scene recognition, and finally realizing the robot's automatic and accurate navigation switching between different scenes.

[0017] Optional, also includes:

[0018] In cases where there are inconsistencies among multiple first-to-be-determined work scenarios, the coordinate information of the robot is obtained;

[0019] Based on the coordinate information, a second working scenario to be determined is identified;

[0020] Based on the robot's historical driving routes, one of the first and second undetermined work scenarios is selected as the target work scenario.

[0021] By adopting the above technical solution, when the image recognition results conflict, coordinate information is used as a supplement, and combined with the previous driving route, a multi-dimensional verification of image + coordinate + historical route is formed to prevent navigation switching confusion caused by a single scene recognition error.

[0022] Optionally, obtaining the robot's driving route on the target navigation map includes:

[0023] The required travel time for each travel segment is determined based on the target navigation map and the travel speed;

[0024] Obtain the initial travel time of the robot;

[0025] The switching time of the driving segment is determined based on the initial driving time and the required driving time;

[0026] When the switching time arrives, obtain the current image of the current scene;

[0027] The current image is input into the road segment recognition model to obtain the driving road segment corresponding to the current image;

[0028] If the travel segment matches the travel segment corresponding to the switching time, the step of determining the robot's travel speed based on the travel segment is executed.

[0029] By adopting the above technical solution, the road segment switching time is first determined by the required travel time plus the initial time. When the switching time is reached, the speed is not directly set according to the predicted road segment. Instead, the current image is first acquired and the actual road segment is identified. If the actual road segment is consistent with the prediction, the speed is adjusted. This prevents speed errors caused by temporary road segment changes from forming a closed loop of prediction, verification, and execution, thus minimizing the risk of road segment misjudgment.

[0030] Optionally, selecting the target navigation map corresponding to the target work scenario from the navigation map information database includes:

[0031] Obtain the first number information corresponding to the work scenario;

[0032] If a second number matching the first number exists in the navigation map information database, the navigation map corresponding to the second number will be used as the target navigation map.

[0033] If no second number information matching the first number information exists in the navigation map information database, the process of creating a new navigation map is initiated, and the updated navigation map is added to the navigation map information database.

[0034] By adopting the above technical solution, matching the second number with the first number avoids traversing all maps in the information database, shortens map selection time, and improves the overall efficiency of navigation switching.

[0035] Optionally, after the robot travels based on the target navigation, the method further includes:

[0036] Real-time acquisition of environmental change information of the robot during its navigation based on the target map;

[0037] If the difference between the environmental change information and the target navigation map exceeds a preset value, the target navigation map is locally updated based on the environmental change information to generate an optimized navigation map.

[0038] The optimized navigation map is uploaded to the navigation map information database.

[0039] Optionally, after uploading the optimized navigation map to the navigation map database, the method further includes:

[0040] Determine whether multiple optimized navigation maps are uploaded by multiple robots for the same work scenario within a preset time period;

[0041] If so, the multiple optimized navigation maps are fused and denoised to generate a globally optimized map for the work scenario;

[0042] The globally optimized map is used as the navigation map for the corresponding work scenario.

[0043] By adopting the above technical solutions, the local update of a single robot may only cover environmental changes within its perception range, while fusion denoising can integrate the update data of all robots, eliminate the perception blind spots of a single robot, and generate a more comprehensive and accurate global optimization map.

[0044] Secondly, this application provides a robot navigation switching device based on a work scenario, which adopts the following technical solution:

[0045] A robot navigation switching device based on a work scenario includes:

[0046] The acquisition module is used to acquire multiple images of the robot's working area to be analyzed, wherein the multiple images to be analyzed are consecutive frames;

[0047] The recognition module is used to recognize multiple images to be analyzed and obtain a first working scene to be determined corresponding to each image to be analyzed;

[0048] The determination module is used to identify the first work scenario to be determined as the target work scenario when multiple first work scenarios to be determined match.

[0049] The selection module is used to select the target navigation map corresponding to the target work scenario from the navigation map information database;

[0050] A generation module is used to generate control commands based on the target navigation map, so that the robot can drive based on the target navigation. The generation of control commands based on the target navigation map includes: obtaining the driving segment of the robot on the target navigation map, the driving segment including obstacle areas and open areas; determining the driving speed of the robot based on the driving segment; and generating the control commands based on the driving speed.

[0051] Thirdly, this application provides an electronic device that adopts the following technical solution:

[0052] An electronic device includes a processor and a memory, wherein the processor is coupled to the memory;

[0053] The processor is configured to execute a computer program stored in the memory, causing the electronic device to perform the method as described in any of the first aspects.

[0054] Fourthly, this application provides a computer-readable storage medium, which adopts the following technical solution:

[0055] A computer-readable storage medium includes a computer program or instructions that, when executed on a computer, cause the computer to perform the method as described in any of the first aspects. Attached Figure Description

[0056] Figure 1 This is a flowchart illustrating a robot navigation switching method based on a work scenario in an embodiment of this application.

[0057] Figure 2 This is a structural block diagram illustrating a robot navigation switching method based on a work scenario in the embodiments of this application.

[0058] Figure 3 This is a structural block diagram illustrating an electronic device in the embodiments of this application. Detailed Implementation

[0059] The present application will be further described in detail below with reference to the accompanying drawings.

[0060] This specific embodiment is merely an explanation of this application and is not intended to limit it. After reading this specification, those skilled in the art can make modifications to this embodiment without contributing any inventive step, but such modifications are protected by patent law as long as they fall within the scope of the claims of this application.

[0061] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0062] Furthermore, the term "and / or" in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. Additionally, the character "" in this article, unless otherwise specified, generally indicates that the preceding and following related objects have an "or" relationship.

[0063] The embodiments of this application will now be described in further detail with reference to the accompanying drawings.

[0064] This application provides a robot navigation switching method based on a work scenario. This method can be executed by an electronic device, which can be a server or a terminal device. The server can be a standalone physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud computing services. The terminal device can be a smartphone, tablet, desktop computer, etc., but is not limited to these.

[0065] like Figure 1 As shown, a robot navigation switching method based on a work scenario is described in the following main process flow (steps S101 to S105):

[0066] Step S101: Obtain multiple images of the robot's working area to be analyzed, wherein the multiple images to be analyzed are consecutive frames;

[0067] In this embodiment, the robot integrates a camera, a lidar, and a positioning module. The robot uploads images of the work site to the electronic device in real time. In this embodiment, the images of the work site are referred to as multiple images to be analyzed.

[0068] Among them, the multiple images to be analyzed are a sequence of environmental images continuously collected by the camera equipment. A binocular camera can be used to collect them at a rate of 30 frames per second to ensure the temporal continuity of scene recognition.

[0069] Step S102: Identify multiple images to be analyzed to obtain the first working scene to be determined corresponding to each image to be analyzed;

[0070] In this embodiment, multiple images to be analyzed are input into an image recognition model to obtain a first working scene to be determined corresponding to each image to be analyzed. The image recognition model is a trained neural network model.

[0071] Step S103: If multiple first work scenarios to be determined match, the first work scenario to be determined is taken as the target work scenario.

[0072] In this embodiment, when multiple first work scenarios to be determined are obtained, the multiple first work scenarios to be determined are compared. When the multiple first work scenarios to be determined are consistent, the first work scenario to be determined is directly used as the target work scenario.

[0073] The fields corresponding to the first work scenario to be determined include, but are not limited to, warehousing, logistics, medical care and libraries; the first work scenario to be determined is illustrated by the library as an example, and the first work scenario to be determined includes, but is not limited to, the lobby, social science reading area, science and technology reading area, quiet reading area and service desk;

[0074] When there are inconsistencies among multiple first-to-be-determined work scenarios, the robot's coordinate information is obtained; a second-to-be-determined work scenario is determined based on the coordinate information; and the robot's historical travel route is used to determine one of the first and second work scenarios as the target work scenario.

[0075] When the first work scenario to be determined is inconsistent, it proves that there is a contradiction among multiple first work scenarios to be determined. The coordinate information is the coordinate of the robot's current position, which is implemented using a GPS positioning system or SLAM algorithm to help determine the authenticity of the work scenario in which the robot is located. The second work scenario to be determined is a candidate scenario obtained by matching the coordinate information with the work scenario coordinate mapping relationship database. In this embodiment, the mapping query between coordinates and scenarios is realized by establishing a polygon coordinate boundary database of the work scenario area. The historical driving route refers to the trajectory data of the robot's historical navigation path, which can be realized by using the trajectory point sequence recorded by the path planning system to analyze the correlation between the robot's movement trend and the application scenario switching logic.

[0076] In this embodiment, when multiple first work scenarios to be determined are inconsistent, the current coordinate information collected by the positioning module is obtained, and the current coordinate information is matched with the pre-stored regional coordinate boundary data of the work scenario to obtain the first work scenario to be determined; the robot's most recent complete navigation path data is obtained, and the correlation between its movement direction and the current coordinates is analyzed. If the historical driving path shows that the robot is moving towards the region of the first work scenario to be determined, the corresponding first work scenario to be determined is selected as the target work scenario; if the coordinate matching result shows that the current position is already within the boundary of the second work scenario to be determined, the second work scenario to be determined is selected as the target work scenario.

[0077] Traditional solutions typically employ repeated scanning or manual intervention when identifying conflicts in work scenarios. However, in the final solution of this invention, the work scenario is automatically determined through the collaborative analysis of coordinate positioning and path history data. Existing technologies that rely solely on visual recognition are prone to misjudgment in areas with similar environmental features. This solution, by introducing a physical coordinate verification mechanism, effectively improves the reliability of scenario determination.

[0078] Step S104: Select the target navigation map corresponding to the target work scenario from the navigation map information database;

[0079] Specifically, the process involves obtaining the first identification number corresponding to the work scenario; if a second identification number matching the first identification number exists in the navigation map information database, the navigation map corresponding to the second identification number is used as the target navigation map; if no second identification number matching the first identification number exists in the navigation map information database, the process of creating a new navigation map is initiated, and the updated navigation map is added to the navigation map information database.

[0080] Before the robot is put into use, a navigation map information database is built. This database, or file indexing system, is used to store and manage navigation resources for different work scenarios. The construction of the navigation map information database includes:

[0081] For each known work area, the robot is driven to perform SLAM mapping within that work area. Specifically, the robot uses its onboard LiDAR and camera equipment to collect environmental data and generate a high-precision two-dimensional or three-dimensional grid map, which serves as the navigation map for that area.

[0082] While constructing the navigation map, the robot collects multiple representative scene images at several key locations within the work area. These multiple scene images should fully reflect the visual characteristics of the work area, with each work area constituting a work scene. A unique secondary identification number is generated for each work scene. The primary identification number is not merely a simple ID but a structured data object, containing at least: a unique scene identifier such as "Room_A101" and a scene category such as "Meeting Room No. 1". The secondary identification number is obtained by hashing the unique scene identifier and scene category and is then stored in the navigation map information database.

[0083] A target navigation map is a navigation map that is bound to a specific work scenario. Each work scenario map only contains key waypoints and obstacle distribution information for that work area.

[0084] The first identifier refers to a unique identifier generated by encoding the features of the work scene using image recognition technology. This identifier can be quickly matched and queried against the navigation map database using the image to be analyzed. The second identifier refers to a scene identifier pre-stored in the navigation map database. The process of creating a new navigation map refers to the process of constructing a 3D map of the scene using LiDAR scanning combined with visual SLAM technology. In this embodiment, a real-time positioning and map building algorithm can be used to dynamically expand the coverage of the navigation map database.

[0085] In this embodiment, when the robot recognizes the current work scene, it first calculates the second number information of the work scene. By comparing the first number information with the second number information in the navigation map information database, if there is a second number information that matches the first number information, the navigation map corresponding to the second number information is directly obtained. If there is no second number information that matches the first number information, the map building process is triggered. The LiDAR node runs the SLAM node to collect the contour information of the work area, converts the distance of obstacles detected by the radar into the actual distance between the robot and the obstacles, and generates a new map after scanning the surrounding environment. For example, in a warehouse scenario, when the robot enters an unrecorded packaging area for the first time, a high-precision navigation map of the area is built and a corresponding number is generated. When other robots enter the same area later, they can directly call the stored map.

[0086] Traditional navigation systems rely on fixed preset map libraries, which cannot cope with new or changed scenarios, causing navigation failures when the robot encounters unrecorded areas. In this embodiment, a dynamic map building mechanism is used to enable navigation to have self-expansion capabilities, effectively solving the problem of incomplete map library coverage.

[0087] Step S105: Generate control commands based on the target navigation map so that the robot can drive based on the target navigation.

[0088] Specifically, the robot's travel route on the target navigation map is obtained, including obstacle areas and open areas; the robot's travel speed is determined based on the travel route; and control commands are generated based on the travel speed.

[0089] The navigation path is divided into different attribute regions within the robot's navigation path. This can be achieved using image recognition technology combined with map coordinate information. The target navigation map identifies obstacle zones and open zones. Obstacle zones refer to areas with obstacles, while open zones are areas without significant obstacles and with good passage conditions. The driving speed is the robot's dynamically adjusted movement rate based on the road segment attributes. This can be achieved by using a preset speed threshold combined with real-time environmental perception data, optimizing navigation efficiency by matching the passage requirements of different areas.

[0090] In this embodiment, during the generation of control commands, the distribution of obstacle areas and free areas is first identified by analyzing the path features in the target navigation map. For example, in a warehouse environment, densely packed shelving areas are marked as obstacle areas, while aisle areas are classified as free areas. Based on the classification results, a lower driving speed is used in obstacle areas to improve obstacle avoidance capabilities, while a higher driving speed is used in free areas to improve task execution efficiency. By integrating speed parameters and path coordinate information, a comprehensive control signal containing speed commands and steering commands is formed to drive the robot to complete the navigation task.

[0091] Traditional navigation solutions typically use fixed speed parameters or simple segmented speed adjustments, which cannot be dynamically adjusted according to real-time road characteristics. In the technical solution of this application, by dividing obstacle zones and idle zones and establishing a speed matching mechanism, navigation efficiency is significantly improved while ensuring safety. For example, existing technologies require frequent starts and stops when encountering temporary obstacles, while this solution can reduce speed in advance by pre-setting a low-speed strategy for obstacle zones, thus reducing the number of emergency stops.

[0092] The process of obtaining the robot's travel route on the target navigation map includes: determining the required travel time for each travel route based on the target navigation map and the travel speed; obtaining the robot's initial travel time; determining the switching time of the travel route based on the initial travel time and the required travel time; obtaining the current image of the current scene when the switching time is reached; inputting the current image into the road segment recognition model to obtain the travel route corresponding to the current image; and, if the travel route matches the travel route corresponding to the switching time, performing the step of determining the robot's travel speed based on the travel route.

[0093] The required travel time refers to the theoretical time for the robot to traverse a specific road segment at its current speed, which can be calculated by the ratio of path length to speed. The initial travel time refers to the time when the robot begins to execute the current navigation task. The road segment recognition model refers to a trained deep learning model, such as a convolutional neural network, used to identify the type of road segment the robot is currently in from an image.

[0094] In this embodiment, during navigation, the theoretical travel time of each road segment is first predicted based on the planned route and real-time speed. The expected switching time of each road segment is calculated by combining the initial time point. When the time reaches the preset switching time point, the camera device is triggered to capture the current environmental image and input the current environmental image into the pre-trained road segment classification model for real-time recognition. If the recognition result is consistent with the expected road segment, the driving speed is adjusted according to the attributes of the road segment, such as reducing the speed in the obstacle area to avoid collision. If the recognition result is inconsistent with the expectation, the speed adjustment is paused and the scene is reconfirmed.

[0095] Traditional methods rely solely on static time planning for road segment switching, without considering the impact of speed fluctuations or environmental changes during actual driving, which can easily lead to deviations in switching timing. In contrast, the solution in this application uses dynamic calculation of switching time combined with real-time image verification to effectively eliminate accumulated time errors and ensure that road segment switching is synchronized with the actual environmental conditions.

[0096] After the robot travels based on the target navigation, the process also includes: acquiring environmental change information in real time during the robot's travel based on the target navigation map; when the difference between the environmental change information and the target navigation map exceeds a preset value, locally updating the target navigation map based on the environmental change information to generate an optimized navigation map; and uploading the optimized navigation map to the navigation map information database.

[0097] Among them, environmental change information refers to real-time environmental data such as the distribution of dynamic obstacles, changes in scene layout, or adjustments in equipment position detected during robot movement. This can be achieved by fusing LiDAR scanning data with depth information collected by visual sensors, and is used to reflect the degree of deviation between the actual environment and the preset map. Local update refers to correcting map data only for local areas where environmental changes have occurred. This can be achieved through an incremental update algorithm of obstacle coordinates in the grid map, avoiding the computational resource consumption caused by full map reconstruction. The preset value refers to the pre-set environmental difference threshold, which can be set as a quantitative indicator that the offset of obstacle position in the map coordinate system exceeds 0.5 meters or the area occupied by newly added obstacles exceeds 1 square meter, and is used to trigger the map update mechanism.

[0098] In this embodiment, during the robot's task execution, environmental data is continuously collected through multi-sensor fusion. When changes in the scene are detected, such as shelf displacement or temporary stacking of materials, the current environmental features are compared with the coordinates of the navigation map. If the difference exceeds a preset threshold, the geographic coordinate range of the changed area is locked, and the map topology of the working area is updated through point cloud registration technology. After the updated local map is verified, it is uploaded to the map database in the form of differential data for subsequent navigation tasks. For example, in a warehouse scenario, when the handling robot detects temporary stacking of materials in a certain aisle, only the data of the occupied area of ​​that aisle is updated, while the maps of other areas remain unchanged.

[0099] Traditional navigation systems typically need to rebuild the global map when detecting environmental changes, resulting in high computational resource consumption and significant response delays. However, in this neural network's technical solution, local incremental updates reduce the computational load of map maintenance and shorten the response time to environmental changes to the second level, effectively solving the real-time navigation problem in dynamic environments.

[0100] After uploading the optimized navigation map to the navigation map information database, the process also includes: determining whether there are multiple optimized navigation maps uploaded by multiple robots for the same work scenario within a preset time period; if so, merging and denoising the multiple optimized navigation maps to generate a global optimized map for the work scenario; and using the global optimized map as the navigation map for the corresponding work scenario.

[0101] The preset time period can be within 24 hours of the acquisition time of the first optimized navigation map. The optimized navigation map refers to the map data after the robot makes local adjustments to the original navigation map according to real-time environmental changes. It can be achieved by collecting information on obstacle position changes or path dynamic adjustments through sensors. Its function is to reflect the latest state of the current scene. Fusion and denoising refers to the integration and noise filtering of multiple optimized maps uploaded within the preset time period. It can be achieved by using a weighted average algorithm or a confidence-based data fusion method. Its function is to eliminate the perception error of a single robot and improve map accuracy. The global optimized map refers to the unified scene map formed after multi-source data fusion processing. It can be generated by spatial coordinate system alignment and feature point matching technology. Its function is to provide more complete environmental information for robots that subsequently enter the scene.

[0102] In this embodiment, when multiple robots are operating in the same work environment, each robot continuously converts its perceived environmental changes into map update data and uploads it to a shared navigation map database. This update data includes, for example, the new coordinates of a moved shelf or the location information of temporary obstacles. Subsequently, the update data submitted by multiple robots undergoes spatial alignment processing, and noise reduction is performed by removing duplicate data points and abnormal offset values, ultimately forming a global map with wider coverage and more accurate details. When a new robot enters the work environment, it directly uses the global map as a navigation basis, eliminating the need to repeat the environmental exploration process.

[0103] Traditional single-robot map update methods are prone to inconsistencies due to individual perception errors, while multi-robot collaborative update mechanisms effectively improve the comprehensiveness of environmental perception through data complementarity. Existing technologies do not consider noise interference during multi-robot data fusion, but the technical solution in this application significantly improves the reliability of map data through denoising processing.

[0104] Figure 2 This application provides a structural block diagram of a robot navigation switching device 200 based on a work scenario. Figure 2 As shown, the robot navigation switching device 200 based on the work scenario may include:

[0105] The acquisition module 201 is used to acquire multiple images to be analyzed of the robot's working area, wherein the multiple images to be analyzed are consecutive frames;

[0106] The recognition module 202 is used to recognize multiple images to be analyzed and obtain the first working scene to be determined corresponding to each image to be analyzed;

[0107] The determination module 203 is used to determine the first work scenario to be determined as the target work scenario when multiple first work scenarios to be determined match.

[0108] The selection module 204 is used to select the target navigation map corresponding to the target work scenario from the navigation map information database;

[0109] The generation module 205 is used to generate control commands based on the target navigation map so that the robot can drive based on the target navigation map. The generation of control commands based on the target navigation map includes: obtaining the robot's driving segment on the target navigation map, the driving segment including obstacle area and open area; determining the robot's driving speed based on the driving segment; and generating control commands based on the driving speed.

[0110] As an optional implementation of this embodiment, the robot navigation switching device 200 based on the work scenario further includes:

[0111] The environment acquisition module is used to acquire information on environmental changes in real time as the robot moves along the target navigation map after it has started moving based on the target navigation.

[0112] The update generation module is used to locally update the target navigation map based on the environmental change information and generate an optimized navigation map when the difference between the detected environmental change information and the target navigation map exceeds a preset value.

[0113] The upload module is used to upload the optimized navigation map to the navigation map information database.

[0114] As an optional implementation of this embodiment, the robot navigation switching device 200 based on the work scenario further includes:

[0115] The coordinate acquisition module is used to obtain the robot's coordinate information when there are inconsistencies among multiple first-to-be-determined working scenarios;

[0116] The first scenario determination module is used to determine the second working scenario to be determined based on coordinate information;

[0117] The second scenario determination module is used to determine one of the first and second working scenarios as the target working scenario based on the robot's historical driving route.

[0118] As an optional implementation of this embodiment, the generation module 205 is specifically used for:

[0119] The required travel time for each travel segment is determined based on the target navigation map and travel speed; the initial travel time of the robot is obtained; the switching time of the travel segment is determined based on the initial travel time and the required travel time; when the switching time is reached, the current image of the current scene is obtained; the current image is input into the road segment recognition model to obtain the travel segment corresponding to the current image; if the travel segment is consistent with the travel segment corresponding to the switching time, the step of determining the robot's travel speed based on the travel segment is executed.

[0120] As an optional implementation of this embodiment, the selection module 204 is specifically used for:

[0121] Obtain the first ID information corresponding to the work scenario; if a second ID information matching the first ID information exists in the navigation map information database, use the navigation map corresponding to the second ID information as the target navigation map; if no second ID information matching the first ID information exists in the navigation map information database, start the process of creating a new navigation map and add the updated navigation map to the navigation map information database.

[0122] As an optional implementation of this embodiment, the robot navigation switching device 200 based on the work scenario further includes:

[0123] The judgment module is used to determine whether there are multiple optimized navigation maps uploaded by multiple robots in the same work scene within a preset time period after the optimized navigation map is uploaded to the navigation map information database; if so, the multiple optimized navigation maps are fused and denoised to generate a global optimized map of the work scene.

[0124] The map module is used to use the globally optimized map as a navigation map for the corresponding work scenario.

[0125] The functional modules in the embodiments of this application can be integrated together to form an independent part, or each module can exist independently, or two or more modules can be integrated to form an independent part. If the function is implemented as a software functional module and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause an electronic device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of a robot navigation switching method based on a work scenario according to various embodiments of this application.

[0126] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and modules described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0127] Figure 3 This is a structural block diagram of an electronic device 300 provided in an embodiment of this application. (See diagram below.) Figure 3 As shown, the electronic device 300 includes a memory 301, a processor 302, and a communication bus 303; the memory 301 and the processor 302 are connected via the communication bus 303. The memory 301 stores a robot navigation switching method based on a work scenario, which can be loaded and executed by the processor 302, as provided in the above embodiment.

[0128] The memory 301 can be used to store instructions, programs, code, code sets, or instruction sets. The memory 301 may include a program storage area and a data storage area. The program storage area may store instructions for implementing an operating system, instructions for at least one function, and instructions for implementing the robot navigation switching method based on a work scenario provided in the above embodiments. The data storage area may store data involved in the robot navigation switching method based on a work scenario provided in the above embodiments.

[0129] Processor 302 may include one or more processing cores. Processor 302 executes instructions, programs, code sets, or instruction sets stored in memory 301, and calls data stored in memory 301 to perform various functions and process data as described in this application. Processor 302 may be at least one of the following: Application-Specific Integrated Circuit (ASIC), Digital Signal Processor (DSP), Digital Signal Processing Device (DSPD), Programmable Logic Device (PLD), Field Programmable Gate Array (FPGA), Central Processing Unit (CPU), controller, microcontroller, and microprocessor. It is understood that, for different devices, the electronic devices used to implement the functions of processor 302 may also be other types, and this application embodiment does not specifically limit the specific implementation.

[0130] The communication bus 303 may include a path for transmitting information between the aforementioned components. The communication bus 303 may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus, etc. The communication bus 303 can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 3 The symbol is represented by a single double arrow, but this does not mean that there is only one bus or one type of bus.

[0131] This application provides a computer-readable storage medium storing a computer program that can be loaded by a processor and executed as described in the above embodiments, which is a robot navigation switching method based on a work scenario.

[0132] In this embodiment, the computer-readable storage medium can be a tangible device that holds and stores instructions used by an instruction execution device. The computer-readable storage medium can be, but is not limited to, an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any combination thereof. Specifically, the computer-readable storage medium can be a portable computer disk, a hard disk, a USB flash drive, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), staging random access memory (SRAM), portable compact disc read-only memory (CD-ROM), digital multifunction disc (DVD), memory stick, floppy disk, optical disk, magnetic disk, mechanical encoding device, or any combination thereof.

[0133] The terms “comprising,” “including,” or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0134] The above description is merely a preferred embodiment of this application and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of this application is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the foregoing application concept. For example, technical solutions formed by substituting the above features with (but not limited to) technical features with similar functions claimed in this application.

Claims

1. A robot navigation switching method based on a work scenario, characterized in that, include: Acquire multiple images of the robot's working area to be analyzed, wherein the multiple images to be analyzed are consecutive frames; Multiple images to be analyzed are identified to obtain a first working scene to be determined corresponding to each image to be analyzed; If multiple first work scenarios to be determined match, the first work scenario to be determined shall be taken as the target work scenario; Select the target navigation map corresponding to the target work scenario from the navigation map information database; Control commands are generated based on the target navigation map to enable the robot to drive based on the target navigation map; The step of generating control commands based on the target navigation map includes: obtaining the robot's travel segment on the target navigation map, the travel segment including obstacle areas and open areas; determining the robot's travel speed based on the travel segment; and generating the control commands based on the travel speed. In cases where there are inconsistencies among multiple first work scenarios to be determined, the coordinate information of the robot is obtained; Based on the coordinate information, a second working scenario to be determined is identified; Based on the robot's historical driving routes, one of the first and second undetermined work scenarios is selected as the target work scenario.

2. The method according to claim 1, characterized in that, The step of obtaining the robot's driving route on the target navigation map includes: The required travel time for each travel segment is determined based on the target navigation map and the travel speed; Obtain the initial travel time of the robot; The switching time of the driving segment is determined based on the initial driving time and the required driving time; When the switching time arrives, obtain the current image of the current scene; The current image is input into the road segment recognition model to obtain the driving road segment corresponding to the current image; If the travel segment matches the travel segment corresponding to the switching time, the step of determining the robot's travel speed based on the travel segment is executed.

3. The method according to claim 1, characterized in that, Selecting the target navigation map corresponding to the target work scenario from the navigation map information database includes: Obtain the first number information corresponding to the work scenario; If a second number matching the first number exists in the navigation map information database, the navigation map corresponding to the second number will be used as the target navigation map. If no second number information matching the first number information exists in the navigation map information database, the process of creating a new navigation map is initiated, and the updated navigation map is added to the navigation map information database.

4. The method according to claim 1, characterized in that, After the robot travels based on the target navigation map, the following is also included: Real-time acquisition of environmental change information of the robot during its navigation based on the target map; If the difference between the environmental change information and the target navigation map exceeds a preset value, the target navigation map is locally updated based on the environmental change information to generate an optimized navigation map. The optimized navigation map is uploaded to the navigation map information database.

5. The method according to claim 4, characterized in that, After uploading the optimized navigation map to the navigation map database, the process further includes: Determine whether multiple optimized navigation maps are uploaded by multiple robots for the same work scenario within a preset time period; If so, the multiple optimized navigation maps are fused and denoised to generate a globally optimized map for the work scenario; The globally optimized map is used as the navigation map for the corresponding work scenario.

6. A robot navigation switching device based on a work scenario, characterized in that, include: The acquisition module is used to acquire multiple images of the robot's working area to be analyzed, wherein the multiple images to be analyzed are consecutive frames; The recognition module is used to recognize multiple images to be analyzed and obtain a first working scene to be determined corresponding to each image to be analyzed; The determination module is used to select the first work scenario to be determined as the target work scenario when multiple first work scenarios to be determined match. The selection module is used to select the target navigation map corresponding to the target work scenario from the navigation map information database; A generation module is used to generate control commands based on the target navigation map, so that the robot can drive based on the target navigation map. The generation of control commands based on the target navigation map includes: obtaining the robot's driving segment on the target navigation map, the driving segment including obstacle areas and open areas; determining the robot's driving speed based on the driving segment; and generating the control commands based on the driving speed. The coordinate acquisition module is used to obtain the robot's coordinate information when there are inconsistencies among multiple first-to-be-determined working scenarios; The first scenario determination module is used to determine the second working scenario to be determined based on coordinate information; The second scenario determination module is used to determine one of the first and second working scenarios as the target working scenario based on the robot's historical driving route.

7. An electronic device, characterized in that, It includes a processor and a memory, wherein the processor is coupled to the memory; The processor is configured to execute a computer program stored in the memory, causing the electronic device to perform the method as described in any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that, It includes a computer program or instructions that, when run on a computer, cause the computer to perform the method as described in any one of claims 1 to 5.

Images