Robot control method, system and electronic device

Abstract

This application provides a robot control method, system, and electronic device. The method includes: acquiring initial point cloud data from a scanning device, wherein the initial point cloud data is data obtained by the scanning device moving and scanning within a target scene, and the initial point cloud data corresponds to the acquisition coordinate system of the scanning device; performing coordinate transformation on the initial point cloud data to obtain deployment point cloud data; wherein the deployment point cloud data corresponds to the deployment coordinate system of the mobile robot. This separates the scanning deployment task from the formal work task, allowing the scanning device to independently acquire and process the initial point cloud data to obtain the deployment point cloud data, thereby completing the initial deployment of the mobile robot. It eliminates the need for remote control or manual pushing of the mobile robot to scan the working environment, making the operation more flexible, especially suitable for complex environments, and effectively improving deployment efficiency.

Description

Technical Field

[0001] This application relates to the field of robot deployment technology, and in particular to a robot control method, system and electronic device. Background Technology

[0002] In recent years, with the continuous development of science and technology and the need for industrial upgrading, more and more intelligent mobile robots have entered the fields of industry, agriculture and animal husbandry, freeing people from dangerous, heavy and repetitive work, and reducing labor costs and improving production efficiency for enterprises.

[0003] Before a mobile robot can be put into use, it needs to be deployed. The typical deployment method involves remotely controlling or pushing the mobile robot to move within the work environment. The robot scans its surroundings and uses technologies such as SLAM (Simultaneous Localization and Mapping) to acquire a point cloud map of the work environment. However, this method is relatively inefficient. Summary of the Invention

[0004] In view of the above, it is necessary to provide a robot control method, system, and electronic device that can improve deployment efficiency.

[0005] The first aspect of this application provides a robot control method, the method comprising: The process involves acquiring initial point cloud data from a scanning device. This initial point cloud data is obtained by the scanning device moving and scanning within the target scene, and it corresponds to the scanning device's acquisition coordinate system. A map is then deployed using the initial point cloud data to obtain deployed point cloud data, which corresponds to the deployment coordinate system of the mobile robot. As the mobile robot moves within the target scene, the deployed point cloud data is used to match and locate the target robot against the real-time point cloud data obtained from its scanning, thus achieving the mobile robot's target localization result.

[0006] In some embodiments, before deploying the initial point cloud data to obtain the deployed point cloud data, the method further includes: filtering points in the initial point cloud data other than the target area, where the target area is used to characterize the movement range of the mobile robot in the target scene; and / or performing point cloud filtering on the initial point cloud data; and / or performing downsampling on the initial point cloud data.

[0007] In some embodiments, the acquisition coordinate system has a first reference point, and the deployment coordinate system has a second reference point. The first reference point and the second reference point are mapped to the same position in the target scene.

[0008] In some embodiments, map deployment is performed on the initial point cloud data to obtain deployed point cloud data, including: determining the transformation parameters between the acquisition coordinate system and the deployment coordinate system, and transforming the initial point cloud data to the deployment coordinate system based on the transformation parameters to obtain deployed point cloud data; wherein, the transformation parameters include rotation matrix and translation amount.

[0009] In some embodiments, matching localization includes: obtaining a grid map based on deployed point cloud data, the grid map having a multi-layer resolution structure; determining a target point cloud and a candidate set, wherein the target point cloud includes at least a portion of the points in the real-time point cloud data; the candidate set includes multiple elements, each element being used to simulate the localization result of the mobile robot; calculating the matching score of all elements in the candidate set based on the target point cloud and the grid map, and outputting a search localization result based on the element with the highest matching score; and obtaining the target localization result based on the search localization result.

[0010] In some embodiments, obtaining a raster map based on deployed point cloud data includes: constructing a raster map based on multiple points in the deployed point cloud data located within a preset target height region; determining a target point cloud includes: using multiple points in real-time point cloud data located within the target height region as the target point cloud.

[0011] In some embodiments, obtaining target location results based on search location results includes: iteratively optimizing the search location results to obtain target location results.

[0012] In some embodiments, the method further includes: matching and locating the mobile robot based on deployed point cloud data and real-time point cloud data to obtain the target localization result of the mobile robot; acquiring the operation information of the mobile robot; and mapping the operation information to a pre-built visualization model based on the target localization result of the mobile robot.

[0013] A second aspect of this application provides a robot control system, comprising: a data acquisition module for acquiring initial point cloud data from a scanning device, wherein the initial point cloud data is data obtained by the scanning device moving and scanning within a target scene, and the initial point cloud data corresponds to the acquisition coordinate system of the scanning device; and a map deployment module for deploying the initial point cloud data to obtain deployed point cloud data, wherein the deployed point cloud data corresponds to the deployment coordinate system of the mobile robot; and when the mobile robot moves within the target scene, the deployed point cloud data is used to match and locate with the real-time point cloud data obtained by the mobile robot scanning, so as to obtain the target positioning result of the mobile robot.

[0014] A third aspect of this application provides an electronic device, comprising: a processor for implementing one or more instructions; and a memory for storing multiple instructions adapted to be loaded by the processor and executed as in the first aspect.

[0015] The robot control method, system, and electronic equipment provided in this application separate the scanning deployment task from the formal work task. The scanning device independently obtains initial point cloud data and processes it to obtain deployment point cloud data to complete the initial deployment of the mobile robot. There is no need to remotely control or manually push the mobile robot to scan the working environment, making the operation more flexible and especially suitable for complex environments, which can effectively improve deployment efficiency. Attached Figure Description

[0016] Figure 1 This is a schematic diagram illustrating an application scenario of the robot control method provided in this application.

[0017] Figure 2 This is a first flowchart illustrating the robot control method provided in this application.

[0018] Figure 3 This is a second flowchart illustrating the robot control method provided in this application.

[0019] Figure 4 This is a schematic diagram of the third process of the robot control method provided in this application.

[0020] Figure 5 A schematic diagram of step S401 of the robot control method provided in this application.

[0021] Figure 6 A schematic diagram of the modules of the robot control system provided in this application.

[0022] Figure 7 A schematic diagram of the modules of the electronic device provided in this application. Detailed Implementation

[0023] In the description of the embodiments in this application, the words "exemplary," "or," and "for example" are used to indicate examples, illustrations, or explanations. Any embodiment or design that is described as "exemplary" or "for example" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or design options. Specifically, the use of the words "exemplary," "or," and "for example" is intended to present the relevant concepts in a specific manner.

[0024] The methods disclosed in the embodiments of this application or the methods shown in the flowcharts include one or more steps for implementing the method. Without departing from the scope of the claims, the execution order of the multiple steps may be interchanged, and some steps may be deleted.

[0025] Currently, in the actual use of mobile robots, there is usually a deployment step. Before the mobile robot is officially used, the working environment of the mobile robot is scanned using technologies such as SLAM (Simultaneous Localization and Mapping) to obtain its point cloud map. When the mobile robot is actually working, the scanned point cloud of the mobile robot's built-in 3D LiDAR is matched with the deployed point cloud map using technologies such as map matching to obtain the precise position and orientation of the mobile robot.

[0026] When deploying mobile robots, it is usually necessary to remotely control or manually push the mobile robot to scan the working environment. However, in some special scenarios, such as when the ground environment is complex (e.g., uneven ground, many obstacles, etc.) which is not conducive to the movement of the mobile robot, when the movement speed of the mobile robot is too fast or too slow, or when the mobile robot is not conducive to movement due to its own reasons (e.g., large size, heavy weight, inflexible turning, etc.), the above deployment method is less efficient and more time-consuming.

[0027] Furthermore, during the movement of mobile robots, operational information such as sensor data, operating status, decision-making and planning information, and fault information is usually stored in the form of text or tables, which is not convenient to display to users and affects the user experience.

[0028] Therefore, embodiments of this application provide a robot control method, system, and electronic device that can improve deployment efficiency and enhance user experience.

[0029] This application first provides a robot control method.

[0030] Figure 1 This is a schematic diagram illustrating an application scenario of the robot control method provided in this application. For example... Figure 1 As shown, the robot control method uses the mobile robot's motion environment as the target scene. The scanning device 10 scans the target scene to obtain point cloud data, and then uses this point cloud data to deploy a point cloud map for the mobile robot 20, accurately modeling the target scene and constructing a high-precision visualization model. During the movement of the mobile robot 20, the scanning results and the deployed point cloud map are matched for localization to obtain the real-time position and attitude of the mobile robot 20. Furthermore, the operational information of the mobile robot 20 can be mapped onto the pre-built visualization model for display, allowing users to intuitively understand the operating environment of the mobile robot 20 or quickly troubleshoot its malfunctions.

[0031] Specifically, the robot control method can be applied to a robot system, which includes a scanning device 10 and a mobile robot 20. The scanning device 10 can be a 3D laser scanner or a walking device equipped with a 3D laser scanner. The walking device, compared to the mobile robot 20, is more suitable for moving and walking in a specified scenario (such as an uneven ground surface or numerous obstacles). When the scanning device 10 is a 3D laser scanner, the user can hold the scanning device 10 and walk within the specified scenario to allow the scanning device 10 to collect point cloud data of the specified scenario. When the scanning device 10 is a walking device, the walking device can walk autonomously within the specified scenario and collect point cloud data of the specified scenario. The point cloud data collected by the scanning device 10 can be used to deploy a point cloud map of the working scenario of the mobile robot 20.

[0032] In this embodiment, the mobile robot 20 is an intelligent walking device equipped with a scanning module, which can be a LiDAR. During movement, the mobile robot 20 can collect point cloud data through the scanning module. The point cloud data collected by the mobile robot 20 can be used to match and locate itself against a deployed point cloud map, obtaining the current location result of the mobile robot 20. The location result includes the position information and attitude information of the mobile robot 20.

[0033] The following describes the process steps of the robot control method provided in this application.

[0034] Figure 2 This is a first flowchart illustrating the robot control method provided in this application. Please refer to it as well. Figure 2 The robot control method includes the following steps.

[0035] S201. Acquire the initial point cloud data of the scanning device 10.

[0036] The initial point cloud data is the data obtained by the scanning device 10 moving and scanning within the target scene, and is the point cloud map obtained by the scanning device 10 scanning the target scene. The target scene refers to the environment in which the mobile robot 20 is working. The target scene can be an indoor environment or an outdoor environment, which can be set according to actual needs.

[0037] The initial point cloud data corresponds to the acquisition coordinate system of the scanning device 10, which is a local or global reference system established based on the initial pose of the scanning device 10 or a fixed reference point. For example, the scanning device 10 can be moved within the target scene by being held by a user or carried while walking, and point cloud data is acquired during the movement to obtain the initial point cloud data.

[0038] The initial point cloud data is a set of raw environmental points collected by the scanning device 10 during its movement and scanning process. It includes multiple points and their coordinate values ​​in the acquisition coordinate system. These multiple points can represent objects in the target scene (such as the ground, walls, obstacle surfaces, etc.).

[0039] S202. Deploy the map on the initial point cloud data to obtain the deployed point cloud data.

[0040] The deployment point cloud data corresponds to the deployment coordinate system of the mobile robot 20. The deployment coordinate system is the world coordinate system or task coordinate system in the working environment of the mobile robot 20, and it is established with reference points such as the map origin and the robot's starting point. The deployment point cloud data is a point cloud map that has already been deployed.

[0041] By deploying the initial point cloud data into a map, the initial point data bound to the acquisition coordinate system is transformed into the deployment coordinate system, so as to convert the point cloud map of the scanning device 10 into the point cloud map of the deployable mobile robot 20, and thus obtain the deployment point cloud data.

[0042] When the mobile robot 20 moves within the target scene, point cloud data is deployed to match and locate the target with the real-time point cloud data scanned by the mobile robot 20, so as to obtain the target location result of the mobile robot 20.

[0043] Understandably, steps S201-S202 are used to deploy the mobile robot 20 before it begins operation. After deployment, the mobile robot 20 can move within the target scene during its operation. During movement, the mobile robot 20 can match and locate itself using real-time point cloud data and deployed point cloud data, thereby obtaining its position and attitude information to maintain normal operation.

[0044] In this embodiment, the deployed point cloud data can also be used to construct a visualization model of the target scene. During the movement of the mobile robot 20, the operation information of the mobile robot 20 can be mapped onto the visualization model based on the positioning results of the mobile robot 20. The visualization model can be a three-dimensional model, and both the visualization model and the corresponding mapping content can be displayed on the user's terminal, allowing the user to quickly and intuitively understand the operating environment of the mobile robot 20 and to quickly troubleshoot anomalies in case of unexpected situations.

[0045] The operational information that the visualization model can map includes, but is not limited to: position, attitude, sensor data, operational status, decision-making and planning information, and fault information. User terminals include, but are not limited to, smartphones and tablets.

[0046] In some embodiments, the acquisition coordinate system has a first reference point preset, and the deployment coordinate system has a second reference point preset. The first reference point and the second reference point are mapped to the same location in the target scene.

[0047] The first reference point is a specified coordinate position in the acquisition coordinate system, such as the origin of the acquisition coordinate system. The second reference point is a specified coordinate position in the deployment coordinate system, such as the origin of the deployment coordinate system. By binding the first and second reference points to the same position in the target scene, the first and second reference points can be made to coincide, thus spatially aligning the acquisition and deployment coordinate systems to facilitate map deployment in subsequent step Sx02.

[0048] For example, the first reference point can be the origin of the point cloud map in the initial point cloud data. When the scanning device 10 performs movement and scanning in the target scene, the scanning starting point of the scanning device 10 is mapped to the first reference point in the acquisition coordinate system. The second reference point can be the origin of the deployment coordinate system. When the mobile robot 20 is in operation, the starting point of the mobile robot 20 is preset to the second reference point in the deployment coordinate system.

[0049] Specifically, before the scanning operation, the scanning starting point of the scanning device 10 can be set at the moving starting point of the mobile robot 20 to establish a correspondence between the acquisition coordinate system of the scanning device 10 and the deployment coordinate system of the mobile robot 20, so as to achieve initial alignment of the two coordinate systems, improve the efficiency of subsequent map deployment, and thus improve deployment efficiency.

[0050] In some embodiments, step S202 specifically includes: Determine the transformation parameters between the acquisition coordinate system and the deployment coordinate system, and based on the transformation parameters, transform the initial point cloud data to the deployment coordinate system to obtain the deployment point cloud data.

[0051] The transformation parameters include a rotation matrix and a translation amount. The rotation matrix is ​​used to compensate for the directional deviation between the acquisition coordinate system and the deployment coordinate system, and the translation amount is used to compensate for the positional deviation between the acquisition coordinate system and the deployment coordinate system. The transformation parameters can be calculated based on the structural parameters of the scanning device 10 and the mobile robot 20.

[0052] For example, the initial point cloud data and the deployed point cloud data can be used for map deployment according to formula (1).

[0053] (1) in, To deploy the coordinates of points in the coordinate system, To collect the coordinates of points in the coordinate system, R is the rotation matrix and t is the translation amount.

[0054] Thus, by using a rotation matrix to align the attitude of the acquisition coordinate system and the deployment coordinate system, and by using translation to compensate for the spatial distance between the acquisition and deployment coordinate systems, a map of multiple points in the initial point cloud data can be deployed to the deployment coordinate system. Furthermore, by aligning the first reference point of the acquisition coordinate system with the second reference point of the deployment coordinate system, the spatial deviation between the two coordinate systems can be reduced, thereby minimizing the errors caused by spatial deviation, simplifying the calculation of translation, and improving the efficiency and accuracy of map deployment.

[0055] The robot control method provided in this application separates the scanning deployment task from the formal work task. The scanning device 10 independently obtains initial point cloud data and processes it to obtain deployment point cloud data to complete the preliminary deployment of the mobile robot 20. It does not require remote control or manual pushing of the mobile robot 20 to scan the working environment, making the operation more flexible and especially suitable for complex environments, which can effectively improve deployment efficiency.

[0056] Figure 3 This is a second flowchart illustrating the robot control method provided in this application. Please refer to it as well. Figure 3 In some embodiments, prior to step S202, the robot control method further includes optimizing the initial point cloud data. In step S202, the optimized initial point cloud data is used for map deployment to obtain deployed point cloud data. The optimization processing methods include at least one of the following: point cloud cropping, point cloud filtering, and downsampling.

[0057] Specifically, prior to step S202, the robot control method further includes: S301. Perform point cloud cropping on the initial point cloud data. Specifically, filter out points in the initial point cloud data that are not in the target area.

[0058] The target area is used to characterize the range of movement of the mobile robot 20 within the target scene.

[0059] Understandably, by setting filtering rules for the initial point cloud data within the target area, points located outside the target area are removed from the initial point cloud data. This effectively trims the entire point cloud map, deleting data from points that do not affect the movement of the mobile robot 20. In this way, the initial point cloud data can be focused on the actual activity area of ​​the mobile robot 20, reducing the point cloud volume, simplifying the point cloud to save memory, and reducing the amount of invalid point cloud data processing in subsequent steps, thus improving deployment efficiency.

[0060] S302. Perform point cloud filtering processing on the initial point cloud data.

[0061] Point cloud filtering can remove randomly distributed noise and isolated points from the initial point cloud data. Randomly distributed noise refers to discrete points generated during the scanning process due to environmental interference or equipment errors. Isolated points refer to outliers in space that are far from the main structure and have few surrounding neighboring points; the local density of isolated points is below a preset threshold. For example, point cloud filtering can be at least one of statistical filtering and radius filtering.

[0062] Understandably, by filtering the initial point cloud data, scanning errors and local anomalies can be reduced or eliminated, data quality can be improved, effective environmental structure can be preserved, and the reliability of point cloud maps can be enhanced.

[0063] S303. Downsample the initial point cloud data.

[0064] Among them, voxel filtering is used to downsample the initial point cloud data, reducing the number and density of points in the point cloud map, thereby reducing memory usage, improving running efficiency, and accelerating subsequent positioning and matching.

[0065] It is understood that, in the examples of this application, the optimization process includes point cloud cropping, point cloud filtering, and downsampling, and steps S301, S302, and S303 can be executed sequentially. In other embodiments, the optimization process can be added, reduced, and adjusted according to actual needs, and this application does not impose any restrictions on this.

[0066] Figure 4 This is a third flowchart illustrating the robot control method provided in this application. Please refer to it as well. Figure 4 In some embodiments, the control method further includes: S401. Based on the matching and positioning of the deployed point cloud data and the real-time point cloud data, the target positioning result of the mobile robot 20 is obtained.

[0067] The real-time point cloud data consists of point clouds from scanned frames obtained by the mobile robot 20 during its movement. The target localization results include the current position and orientation of the mobile robot 20.

[0068] Understandably, when the mobile robot 20 is in operation, during its movement, the mobile robot 20 scans the target scene through the scanning module to obtain the scan frame point cloud. The scan frame point cloud is then matched and positioned with the deployed point cloud map to obtain the current position and posture of the mobile robot 20.

[0069] Figure 5 This is a schematic diagram of the sub-process of step S401 of the robot control method provided in this application. Please refer to it as well. Figure 5 Specifically, step S401 includes: S501. Based on the deployed point cloud data, a raster map is obtained.

[0070] Among them, the raster map has a multi-resolution structure.

[0071] A raster map is a two-dimensional planar representation that discretizes a continuous three-dimensional space into grid cells, using multiple grates to record environmental information within its internal space. A raster map consists of multiple raster layers, each containing multiple grates. The resolution of each raster layer is N times the resolution of the layer below it (N > 0). Resolution characterizes the physical size of a single grates in the real world. Higher resolution raster layers represent smaller physical sizes and a greater number of grates within them.

[0072] For example, the total number of raster layers can be 3-5, and the resolution multiplier between adjacent raster layers can be 2. In this way, the raster map adopts a structure where the resolution is halved layer by layer, so that the upper raster contains the aggregated information of the corresponding area in the lower layer. It can be understood that the number of layers and the resolution multiplier can be configured according to actual needs.

[0073] The elements of a raster map include raster coordinates and a hash table. Raster coordinates are generated by dividing the coordinates of a point by the resolution of the current raster layer and rounding down. The hash table is the data storage format for the raster map, and its value represents the number of points that fall within the raster.

[0074] In this embodiment, step S501 can be: using multiple points located in a preset target height area in the deployed point cloud data as key point clouds, and constructing a raster map based on the key point clouds.

[0075] The target height region is the area that retains the effective characteristics of the mobile robot 20 during normal operation, and can be configured according to the application environment of the mobile robot 20. For example, the target height region is a certain range of area obtained with the height of the scanning device 10 or scanning module as a reference point.

[0076] Understandably, by setting the target height area, points outside the target height area can be excluded from subsequent grid map construction, eliminating interference from points in non-working areas (such as ground point clouds and roof point clouds), focusing the grid map within the working range of the mobile robot 20, and improving computational efficiency and positioning accuracy.

[0077] Specifically, according to the Z-axis value corresponding to the target height region, key point clouds located in the target height region are extracted from the deployed point cloud data. The two-dimensional coordinate values ​​(X-axis and Y-axis) of the key point clouds and the grid resolution are used to calculate the grid coordinates of the grid map. The grid coordinate values ​​are obtained by dividing the two-dimensional coordinate values ​​of the points by the grid resolution and then rounding. In this embodiment, the data structure of the grid map is a hash table. The key of the hash table is the hash value obtained by calculating the grid coordinates. The hash value can be calculated according to formula (2).

[0078] (2) Where h is the hash value, x is the value of the raster coordinate x-axis, and y is the value of the raster coordinate y-axis.

[0079] S502, Determine the target point cloud and candidate set.

[0080] The target point cloud includes at least a portion of the points in the real-time point cloud data. In this embodiment, multiple points located within the target height region in the real-time point cloud data are used as the target point cloud, so that the spatial height of the target point cloud corresponds to the spatial height of the key point cloud.

[0081] The candidate set is the position and orientation simulation set of the mobile robot 20. The candidate set includes multiple elements, each of which is used to simulate the localization result of the mobile robot 20. The simulated localization result includes position and orientation.

[0082] S503. Based on the target point cloud and the raster map, calculate the matching score of all elements in the candidate set, and output the search and positioning results based on the element with the highest matching score.

[0083] The matching score of an element is used to evaluate the degree of matching between the actual position and pose of the mobile robot 20 and the position and pose simulated by the element. The higher the matching score of an element, the higher the probability that the position and pose simulated by the element is the actual position and pose of the mobile robot 20.

[0084] The specific implementation of step S503 is as follows: a threshold score is set, and the initial value of the threshold score can be 0. The threshold score is updated based on the highest matching score currently calculated.

[0085] The search proceeds sequentially through each raster layer, from lowest to highest resolution. The matching score for each element in the candidate set is calculated, and these scores are then stored in a stack in descending order. The matching score is calculated as the ratio of the number of points falling into a non-empty raster (i.e., a raster with at least one point) to the total number of points in the current keypoint cloud.

[0086] During the search process, the current key point cloud is rotated according to the preset search angle step size, the raster map is reconstructed, and the matching score of all elements in the candidate set is calculated. The search angle step size can be calculated according to formula (3).

[0087] (3) in, The search angle step size, The resolution of the current layer raster map. This is the maximum distance that the scanning module can detect.

[0088] During the search cycle of each grid layer, first check if the stack is empty, and pop the top element (the element with the highest matching score) as the current element. If the matching score of the current element is not higher than the current threshold score, skip the current element. If the current element's matching score is higher than the threshold score, use the simulated localization result of the current element as a candidate localization result.

[0089] Furthermore, if the current element's raster is not the highest resolution, the search proceeds to the next lower resolution raster, recalculating the matching score for each element and storing the matching scores in descending order onto a stack. This ensures that the element at the top of the stack always has the highest matching score. The search loop ends when the stack is empty, yielding the element with the highest final matching score. The position and pose simulated by this element are then used as the search position and pose.

[0090] Understandably, in this embodiment, an environmental space is established through a grid map, and spatial indexing is performed based on hash values. The data structure can be optimized by using hash tables, skipping empty grids in a large area of ​​the scene, reducing storage memory waste, improving retrieval speed and accuracy, and ensuring real-time positioning performance.

[0091] S504. Based on the search and positioning results, the target positioning results are obtained.

[0092] In this process, the search and location results are iteratively optimized to obtain the target location results.

[0093] Specifically, the search positioning result from step S503 is used as the initial positioning, and fine matching is performed using methods including but not limited to ICP (Iterative Closest Point) and NDT (Normal Distributions Transform) to obtain the target positioning result.

[0094] In some embodiments, the robot control method further includes: S402. Obtain the operation information of the mobile robot 20, and based on the target positioning results of the mobile robot 20, map the operation information to a visualization model corresponding to the deployment coordinate system.

[0095] The operational information includes, but is not limited to: position, attitude, sensor data, operational status, decision-making and planning information, and fault information. The visualization model can be a high-precision 3D model obtained by structuring point cloud data.

[0096] The target localization result can determine the current position and posture of the mobile robot 20. Based on the target localization result, the mapping position of the mobile robot 20 in the visualization model can be calculated, and the operation information of the mobile robot 20 can be presented in the mapping position of the visualization model, so that users can quickly and intuitively understand the operating environment of the mobile robot 20 and quickly troubleshoot anomalies in case of unexpected situations.

[0097] For example, the mapping position can be calculated using formula (4).

[0098] (4) in, For mapping position, This is the rotation matrix representing the current target localization result of the mobile robot. This refers to the location of sensors, operating status indicators, key decision-making and planning points, or fault occurrence points in the robot's own coordinate system. This represents the current location of the target for the mobile robot.

[0099] It is understood that the robot control method provided in this application can construct a digital twin system of mobile robot 20 through scanning device 10, achieve efficient deployment by using independent scanning, achieve real-time high-precision positioning and synchronization by using a combination of coarse matching and fine matching, and realize the mapping of operation information in the three-dimensional model, realizing a real-time interactive closed loop between the physical world and the virtual model, so that mobile robot 20 can be deployed faster, positioned more accurately, and operated and maintained more intelligently in complex scenarios.

[0100] It is worth noting that some or all of the steps in the above-described robot control method can be executed by the control module built into the scanning device 10, the control system built into the mobile robot 20, or a control terminal. The control terminal can be a cloud server or similar entity that is communicatively connected to the scanning device 10 and the mobile robot 20. For example, all steps of the above-described robot control method can be executed by the control terminal. In other embodiments, some steps of the robot control method can be executed by the control module built into the scanning device 10 or the control system built into the mobile robot 20. For example, the scanning device 10 performs the map deployment step, and the mobile robot 20 performs the matching and positioning step. This application does not impose any limitations on this.

[0101] Figure 6 This is a schematic diagram of the robot control system provided in this application. Please refer to it as well. Figure 6 This application also provides a robot control system 100. The robot control system 100 may include multiple functional sub-modules, and the functions of each functional sub-module are as follows.

[0102] The data acquisition module 110 is used to acquire the initial point cloud data of the scanning device 10. The initial point cloud data is the data obtained by the scanning device 10 moving and scanning within the target scene. The initial point cloud data corresponds to the acquisition coordinate system of the scanning device 10.

[0103] The map deployment module 120 is used to deploy the initial point cloud data to obtain deployed point cloud data. The deployed point cloud data corresponds to the deployment coordinate system of the mobile robot 20.

[0104] When the mobile robot 20 moves within the target scene, point cloud data is deployed to match and locate the target with the real-time point cloud data scanned by the mobile robot 20, so as to obtain the target location result of the mobile robot 20.

[0105] The robot control system 100 may further include a matching and positioning module 130 and an information mapping module 140. The matching and positioning module 130 is used to perform matching and positioning based on the deployed point cloud data to obtain the target positioning result of the mobile robot 20. The information mapping module 140 is used to acquire the operation information of the mobile robot 20 and, based on the target positioning result of the mobile robot 20, map the operation information to a visualization model corresponding to the deployed point cloud data.

[0106] It is understood that the module division described above is a logical functional division, and there may be other division methods in actual implementation. Furthermore, the functional modules in the various embodiments of this application can be integrated into the same processing unit, or each module can exist physically separately, or two or more modules can be integrated into the same unit. The integrated modules described above can be implemented in hardware or in a combination of hardware and software functional modules.

[0107] It is worth noting that the robot control system 100 described above can be a control module built into the scanning device 10, a control system built into the mobile robot 20, or a control terminal. The control terminal can be a cloud server or similar entity that communicates with the scanning device 10 and the mobile robot 20. For example, the robot control system 100 is deployed on a control terminal. In other embodiments, the robot control system 100 can be a combination of a control module built into the scanning device 10 and a control system built into the mobile robot 20. For example, the data acquisition module 110 and the map deployment module 120 are integrated into the scanning device 10, and the matching and positioning module 130 and the information mapping module 140 are integrated into the mobile robot 20. This application does not impose any limitations on this.

[0108] Figure 7 A schematic diagram of the electronic device provided in this application. Please refer to it as well. Figure 7The electronic device 200 includes a processor 210 and a memory 220. The processor 210 is used to implement one or more instructions. The memory 220 is used to store multiple instructions, which are adapted to be loaded by the processor 210 and executed in the robot control method described above.

[0109] This application also provides a computer storage medium. The computer storage medium stores program instructions that, when executed on the electronic device 200, cause the electronic device 200 to perform the robot control method described above.

[0110] The implementation principles of the robot control system 100 and electronic device 200 provided in this application embodiment can be found in the relevant descriptions in the above-mentioned robot control method, and will not be repeated here.

[0111] It will be apparent to those skilled in the art that this application is not limited to the details of the exemplary embodiments described above, and that this application can be implemented in other specific forms without departing from the spirit or essential characteristics of this application. Therefore, the embodiments described above should be considered exemplary and non-limiting in all respects, and the scope of this application is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within this application.

Claims

1. A robot control method, characterized in that, The method includes: Acquire initial point cloud data of the scanning device, wherein the initial point cloud data is data obtained by the scanning device moving and scanning within the target scene, and the initial point cloud data corresponds to the acquisition coordinate system of the scanning device; The initial point cloud data is used to deploy a map to obtain deployed point cloud data; wherein, the deployed point cloud data corresponds to the deployment coordinate system of the mobile robot; When the mobile robot moves within the target scene, the deployed point cloud data is used to match and locate the target with the real-time point cloud data scanned by the mobile robot, so as to obtain the target location result of the mobile robot.

2. The robot control method according to claim 1, characterized in that, Before deploying the map on the initial point cloud data to obtain the deployed point cloud data, the method further includes: The points in the initial point cloud data, excluding the target region, are filtered out; the target region is used to characterize the movement range of the mobile robot within the target scene; and / or, Perform point cloud filtering processing on the initial point cloud data; and / or, The initial point cloud data is downsampled.

3. The robot control method according to claim 1, characterized in that, The acquisition coordinate system has a first reference point, and the deployment coordinate system has a second reference point. The first reference point and the second reference point are mapped to the same position in the target scene.

4. The robot control method according to claim 1, characterized in that, The process of deploying a map from the initial point cloud data to obtain deployed point cloud data includes: Determine the transformation parameters between the acquisition coordinate system and the deployment coordinate system, and based on the transformation parameters, transform the initial point cloud data to the deployment coordinate system to obtain the deployment point cloud data; wherein, the transformation parameters include a rotation matrix and a translation amount.

5. The robot control method according to any one of claims 1 to 4, characterized in that, The matching and positioning includes: Based on the deployed point cloud data, a raster map is obtained, wherein the raster map has a multi-layer resolution structure; A target point cloud and a candidate set are determined, wherein the target point cloud includes at least a portion of the points in the real-time point cloud data; the candidate set includes multiple elements, each of which is used to simulate the localization result of the mobile robot. Based on the target point cloud and the grid map, calculate the matching score of all elements in the candidate set, and output the search and location result based on the element with the highest matching score; Based on the search and location results, the target location results are obtained.

6. The robot control method according to claim 5, characterized in that, The process of obtaining a raster map based on the deployed point cloud data includes: The grid map is constructed based on multiple points located in the preset target height region in the deployed point cloud data; The determination of the target point cloud includes: Multiple points located within the target height region in the real-time point cloud data are used as the target point cloud.

7. The robot control method according to claim 5, characterized in that, The process of obtaining the target location result based on the search and location results includes: The search and location results are iteratively optimized to obtain the target location result.

8. The robot control method according to claim 1, characterized in that, The method further includes: The target positioning result of the mobile robot is obtained by matching and positioning the deployed point cloud data with the real-time point cloud data. Obtain the operational information of the mobile robot; Based on the target localization results of the mobile robot, the operational information is mapped onto a pre-built visualization model.

9. A robot control system, characterized in that, include: The data acquisition module is used to acquire the initial point cloud data of the scanning device, wherein the initial point cloud data is the data obtained by the scanning device moving and scanning within the target scene, and the initial point cloud data corresponds to the acquisition coordinate system of the scanning device; The map deployment module is used to deploy the initial point cloud data to obtain deployed point cloud data; wherein, the deployed point cloud data corresponds to the deployment coordinate system of the mobile robot; When the mobile robot moves within the target scene, the deployed point cloud data is used to match and locate the target with the real-time point cloud data scanned by the mobile robot, so as to obtain the target location result of the mobile robot.

10. An electronic device, characterized in that, include: A processor is used to implement one or more instructions; and, A memory for storing a plurality of instructions adapted to be loaded by a processor and executed as described in any one of claims 1 to 8.

Images