Recharging method, recharging device, self-moving device, and storage medium

Abstract

The application discloses a method and device for returning to a charging base, a self-moving device and a computer storage medium. The method comprises: collecting a first environment point cloud; filtering the first environment point cloud according to a preset reflectivity condition to obtain actual point cloud data of the charging base; wherein the reflectivity of the shell of the charging base is greater than or equal to a preset reflectivity threshold; obtaining reference point cloud data corresponding to the charging base; determining a vertical direction pose constraint according to a boundary point in the actual point cloud data and a boundary point in the reference point cloud data; performing point cloud registration on the actual point cloud data and the reference point cloud data based on the vertical direction pose constraint through a preset point cloud registration algorithm to determine pose information of the self-moving device; and controlling the self-moving device to move to the charging base based on the pose information. The method can improve the returning-to-charging accuracy of the self-moving device.

Description

Technical Field

[0001] This application belongs to the field of charging technology, and in particular relates to a recharge method, a recharge device, a self-moving device, and a computer-readable storage medium. Background Technology

[0002] With the development of technology, the application of self-moving devices is becoming more and more widespread. Generally, self-moving devices are equipped with batteries. When the battery power is low, the self-moving device can be charged in order to continue to work normally.

[0003] Currently, the most common charging method for self-operated mobile devices is autonomous recharging. This means that when the self-operated mobile device detects that the battery level is below a certain threshold, it can identify the charging base, achieve autonomous positioning, and finally move into the charging base based on the positioning result to achieve autonomous charging.

[0004] However, during autonomous recharging, noise in the collected environmental point cloud can reduce the accuracy of the charging base's recognition, thus lowering the accuracy of the positioning results and preventing the self-device from recharging accurately. In other words, current recharging methods for self-devices suffer from low recharging accuracy. Summary of the Invention

[0005] This application provides a recharging method, a recharging device, a self-moving device, and a computer-readable storage medium, which can improve the recharging accuracy of self-moving devices.

[0006] Firstly, this application provides a recharge method, including:

[0007] Collect point clouds of the first environment;

[0008] The first environmental point cloud is filtered according to the preset reflectivity conditions to obtain the actual point cloud data of the charging base; wherein the reflectivity of the shell of the charging base is greater than or equal to the preset reflectivity threshold.

[0009] Obtain the reference point cloud data corresponding to the charging base;

[0010] Determine the pose constraints in the vertical direction based on the boundary points in the actual point cloud data and the boundary points in the reference point cloud data.

[0011] Based on vertical pose constraints, the actual point cloud data is registered with the reference point cloud data using a preset point cloud registration algorithm to determine the pose information of the self-moving device.

[0012] The self-moving device is controlled to move towards the charging base based on its pose information.

[0013] Secondly, this application provides a recharge device, comprising:

[0014] The acquisition module is used to acquire the point cloud of the first environment;

[0015] The filtering module is used to filter the first environmental point cloud according to the preset reflectivity conditions to obtain the actual point cloud data of the charging base; wherein the reflectivity of the shell of the charging base is greater than or equal to the preset reflectivity threshold.

[0016] The first acquisition module is used to acquire the reference point cloud data corresponding to the charging base;

[0017] The determination module is used to determine the pose constraints in the vertical direction based on the boundary points in the actual point cloud data and the boundary points in the reference point cloud data.

[0018] The registration module is used to register the actual point cloud data with the reference point cloud data based on the pose constraints in the vertical direction, and to determine the pose information of the self-moving device by using a preset point cloud registration algorithm.

[0019] The control module is used to control the movement of the self-moving device towards the charging base based on the pose information.

[0020] Thirdly, this application provides a self-moving device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps of the method described in the first aspect.

[0021] Fourthly, this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the method described in the first aspect above.

[0022] Fifthly, this application provides a computer program product comprising a computer program that, when executed by one or more processors, implements the steps of the method described in the first aspect.

[0023] The advantages of this application compared to existing technologies are as follows: When a self-moving device has a recharging requirement, a first environmental point cloud containing actual point cloud data of the charging base can be collected. To reduce the interference of noise in the environmental point cloud on the charging base identification results, a charging base with a shell reflectivity greater than a preset reflectivity threshold can be set to improve the identification accuracy of the charging base. Therefore, the actual point cloud data of the charging base can be accurately determined from the first environmental point cloud based on the preset reflectivity conditions. The actual point cloud data can accurately reflect the shape of the charging base, improving the accuracy of point cloud base identification and helping to improve the recharging accuracy of self-moving devices.

[0024] In addition, this application optimizes the point cloud registration method to further improve the recharging accuracy of the self-moving device. Specifically, before registering the actual point cloud data with the reference point cloud data, the vertical pose constraints can be determined based on the boundary points of the actual point cloud data and the reference point cloud data; then, point cloud registration is performed based on the vertical pose. It can be understood that in this point cloud registration process, there is no need to consider point cloud registration in the vertical direction, which reduces the computational load of point cloud registration, improves the speed and accuracy of point cloud registration, and thus further enhances the recharging accuracy of the self-moving device.

[0025] It is understood that the beneficial effects of the second to fifth aspects mentioned above can be found in the relevant descriptions in the first aspect mentioned above, and will not be repeated here. Attached Figure Description

[0026] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0027] Figure 1 This is a flowchart illustrating the recharge method provided in an embodiment of this application;

[0028] Figure 2 This is a flowchart illustrating the point cloud registration method provided in an embodiment of this application;

[0029] Figure 3 This is a flowchart illustrating the method for determining the covariance matrix corresponding to each grid provided in an embodiment of this application;

[0030] Figure 4 This is a schematic flowchart of the optimization process provided in the embodiments of this application;

[0031] Figure 5 This is a schematic diagram of a point cloud acquired by a lidar according to an embodiment of this application;

[0032] Figure 6 This is a flowchart illustrating two methods for determining boundary points provided in the embodiments of this application;

[0033] Figure 7 This is a schematic diagram of the recharge device provided in the embodiments of this application;

[0034] Figure 8 This is a schematic diagram of the structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0035] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0036] In related technologies, during the recharging process of self-moving devices, noise in the collected environmental point cloud can reduce the recognition accuracy of the charging base, thereby reducing the accuracy of the positioning results and causing the self-moving device to fail to recharge accurately.

[0037] To address this issue, this application proposes a recharging method that can accurately identify the actual point cloud data of the charging base, improve the registration accuracy of the point cloud, and enable the self-moving device to recharge with more accurate pose information, thereby improving recharging accuracy. The determination method proposed in this application will be described below through specific embodiments.

[0038] It is understood that the recharging method provided in this application can be applied not only to self-moving devices, but also to other electronic devices capable of controlling self-moving devices, such as mobile phones, tablets, in-vehicle devices, augmented reality (AR) / virtual reality (VR) devices, laptops, ultra-mobile personal computers (UMPCs), netbooks, personal digital assistants (PDAs), etc. This application does not impose any limitations on the specific type of electronic device. Specifically, a communication connection channel can be established between the electronic device and the self-moving device, through which the electronic device can control the self-moving device.

[0039] The aforementioned self-moving device can be a device that includes self-movement assistance functionality. This self-movement assistance functionality can be implemented through an in-vehicle terminal, and the corresponding self-moving device can be a vehicle equipped with that in-vehicle terminal. The self-moving device can also be a semi-self-moving device or a fully autonomous device. Examples include lawnmowers, sweepers, or other robots with navigation capabilities.

[0040] To illustrate the technical solutions proposed in this application, the following description will use a self-moving device as the execution subject to illustrate various embodiments.

[0041] Figure 1 A schematic flowchart of the recharge method provided in this application is shown, the recharge method comprising:

[0042] Step 110: Collect the first environmental point cloud.

[0043] During the recharging process, the self-moving device uses the charging dock as a reference point to determine its own pose information. Clearly, accurate identification of the charging dock is crucial. Therefore, acquiring the actual point cloud data of the current charging dock can be the primary task during recharging. To obtain this actual point cloud data, the self-moving device can acquire the currently collected environmental point cloud, which can then be filtered to obtain the actual point cloud data later. To distinguish it from other subsequent environmental point clouds, this environmental point cloud can be designated as the first environmental point cloud.

[0044] The first environmental point cloud can be determined from a depth image, which can be obtained by other instruments such as a depth camera, laser scanner, or lidar mounted on a mobile device.

[0045] In some embodiments, the self-mobile device may collect the first environmental point cloud when it receives a recharging request, or when it finishes its task and needs to be recharged. This application does not limit the timing of the collection.

[0046] Step 120: Filter the first environmental point cloud according to the preset reflectivity conditions to obtain the actual point cloud data of the charging base.

[0047] To mitigate the impact of noise in the initial environmental point cloud on the identification of actual point cloud data, the charging base's outer shell can be made of a high-reflectivity material, or a high-reflectivity coating can be applied to the outer shell, making its reflectivity greater than a preset reflectivity threshold. Such a shell allows the actual point cloud data of the charging base to be more easily identifiable in the initial point cloud environment, helping to improve the accuracy of actual point cloud data identification.

[0048] Based on the housing configuration of the charging base, the self-moving device can filter the first environmental point cloud according to a preset reflectivity condition to obtain the actual point cloud data of the charging base. Based on the aforementioned example, it is known that the reflectivity of the housing is greater than a reflectivity threshold. Considering that reflectivity loss is inevitable during point cloud data acquisition, the reflectivity condition could be: environmental points with reflectivity greater than or equal to the reflectivity threshold are also identified as actual point cloud data. This reflectivity threshold can be slightly less than the reflectivity of the housing to improve the accuracy and comprehensiveness of the actual point cloud data determination.

[0049] Specifically, the process of filtering the first environmental point cloud by the self-moving device can be as follows: determining the reflectance of each environmental point in the first environmental point cloud; comparing the reflectance of each environmental point with a preset reflectance threshold; and determining environmental points with reflectance greater than or equal to the reflectance threshold as actual point cloud data.

[0050] In some embodiments, to improve the reflectivity of the charging base, the housing of the charging base may be made of a specified material; a coating made of the specified material may also be applied to the housing of the charging base; or a thin film made of the specified material may be attached to the housing of the charging base. Wherein, the reflectivity of the specified material is greater than a reflectivity threshold.

[0051] The selection of designated materials can be considered in conjunction with their price and stability, but no restrictions are imposed in this application.

[0052] Step 130: Obtain the reference point cloud data corresponding to the charging base.

[0053] To determine the pose information of the mobile device based on actual point cloud data, reference point cloud data of the charging base can be acquired. This reference point cloud data can be determined from the second environment point cloud. Knowing the reference pose information of the mobile device when acquiring the second environment point cloud, the pose information of the mobile device can be determined by registering the actual point cloud data and the reference point cloud data.

[0054] In some embodiments, after the charging base is placed, the self-moving device can acquire a second environmental point cloud at a preset reference point with a reference pose; after obtaining the second environmental point cloud, the self-moving device can determine the reference point cloud data from the second environmental point cloud using the same determination method as in step 120 above.

[0055] Generally, there's no need to strictly limit the location of the self-moving device when acquiring the first environmental point cloud, as long as it can obtain point cloud data that includes the actual point cloud data of the charging base. However, the shape and size of the actual point cloud data acquired from different locations may vary, which can easily reduce the registration accuracy of the point cloud. To improve the registration accuracy of the point cloud, the self-moving device can limit its location when acquiring the first environmental point cloud data, thereby reducing the difference between the actual point cloud data and the reference point cloud data. Specifically, when the self-moving device needs to return to its charging base, it can first control itself to move to the reference point and acquire the first environmental point cloud at the reference point.

[0056] Step 140: Determine the pose constraints in the vertical direction based on the boundary points in the actual point cloud data and the boundary points in the reference point cloud data.

[0057] Point cloud data is three-dimensional data. If the actual point cloud data is directly registered with the reference point cloud data, then for each data point in the actual point cloud data, it is necessary to register with each data point in the reference point cloud data in multiple degrees of freedom (such as multiple degrees of freedom in the horizontal axis, vertical axis, vertical direction, rotation angle, pitch angle, and yaw angle). The computational load is large, which can easily lead to low registration efficiency of point clouds.

[0058] To improve point cloud registration efficiency, the self-moving device can first determine the pose constraints of the actual point cloud data and the reference point cloud data in the vertical direction. Specifically, these pose constraints can be determined based on the boundary points in the actual point cloud data and the boundary points in the reference point cloud data.

[0059] Step 150: Based on the pose constraints in the vertical direction, the actual point cloud data and the reference point cloud data are registered using a preset point cloud registration algorithm to determine the pose information of the self-moving device.

[0060] With vertical pose constraints, during point cloud registration, each data point in the actual point cloud data only needs to be registered with data points in the reference point cloud data at the same height in a few additional degrees of freedom. In other words, the registration dimensionality of the point cloud is reduced. Clearly, with vertical pose constraints, the computational load of point cloud registration can be significantly reduced, helping to improve registration efficiency. After the actual point cloud data and the reference point cloud data are registered, the pose information of the mobile device can be obtained.

[0061] Step 160: Control the self-moving device to move towards the charging base based on the pose information.

[0062] Knowing the location information of the charging base, after determining the pose information, the self-moving device can control itself to move towards the charging base based on the location and pose information of the charging base, so as to improve the accuracy of recharging, avoid collision with the charging base due to inaccurate positioning during the recharging process, and improve the reliability of the recharging method.

[0063] In this embodiment, the reflectivity of the charging base's shell is greater than a reflectivity threshold, which improves the recognition accuracy of the actual point cloud data of the charging base in the first environmental point cloud. After obtaining the first environmental point cloud, the self-moving device can accurately determine the actual point cloud data based on the reflectivity of each environmental point. After obtaining the actual point cloud data, the self-moving device can perform point cloud registration based on the acquired reference point cloud data to obtain its own pose information. However, to improve the point cloud registration efficiency, the actual point cloud data is not directly registered with the reference point cloud data. Instead, the pose constraints of the actual point cloud data and the reference point cloud data in the vertical direction are first determined; and based on the pose constraints in the vertical direction, the actual point cloud data and the reference point cloud data are registered using a point cloud registration algorithm, which can efficiently and accurately determine the pose information of the self-moving device. Finally, the pose information is used to control the self-moving device to move towards the charging base, which can improve the accuracy of the self-moving device's recharging.

[0064] In some embodiments, the point cloud registration algorithm may include the Normal Distributions Transform (NDT) algorithm, the Iterative Closest Point (ICP) algorithm, and other derivative algorithms based on the ICP algorithm. Different algorithms may also be combined. The specific registration algorithm is not limited in this application.

[0065] In some embodiments, the point cloud registration algorithm is the NDT algorithm. Unlike the ICP algorithm, the NDT algorithm does not require feature matching, thus avoiding problems that arise in feature matching, such as the influence of point cloud noise, object movement, or point cloud overlap. It uses a normal distribution probability density function to describe the local properties of the point cloud, and all derivatives can be solved analytically, resulting in accurate and relatively fast calculations.

[0066] The basic idea of ​​the NDT algorithm includes: first, converting the reference point cloud data into a normally distributed reference frame with multidimensional variables; then, transforming the real point cloud data to the normally distributed reference frame using pose transformation parameters to obtain the corresponding transformed points; if the real point cloud data and the reference point cloud data are highly matched, the sum of probabilities of each transformed point in the normally distributed reference frame will be relatively large. Therefore, the pose transformation parameters that maximize this probability sum can be determined through optimization, thereby accurately determining the pose information. For details, see [link to relevant documentation]. Figure 2 The point cloud registration method includes:

[0067] Step 210: Rasterize the reference point cloud data to obtain multiple grids.

[0068] A normal distribution reference system can be constructed based on a raster map; therefore, a self-moving device can rasterize the reference point cloud to obtain multiple grids.

[0069] Step 220: Calculate the multidimensional normal distribution parameters for each grid based on the reference point cloud data falling into each grid.

[0070] For each grid cell, the mobile device can collect the baseline point cloud data falling within that grid cell to calculate the multidimensional normal distribution parameters for each grid cell. These multidimensional normal distribution parameters can be used to describe the distribution characteristics of the baseline point cloud data within each grid cell.

[0071] Step 230: Project each point cloud data in the actual point cloud data into the grid based on the pose transformation parameters, and determine the probability of each point cloud data falling into the corresponding grid based on the multidimensional normal distribution parameters.

[0072] After the mesh is constructed, the self-moving device can predetermine the pose transformation parameters and project each point cloud data in the actual point cloud data into the mesh using these pose transformation parameters. For each point cloud data falling into the mesh, the probability corresponding to each point cloud data can be determined based on the multidimensional normal distribution parameters.

[0073] It is understandable that each data point falls into a different grid, corresponding to different normal distribution parameters, and consequently, different probabilities. Which grid a data point falls into is determined by the pose transformation parameters. Therefore, it can be considered that the pose transformation parameters determine the probability of each point cloud data point, meaning that the probability is related to the pose transformation parameters.

[0074] The initial parameter values ​​corresponding to the pose transformation parameters can be determined based on the prior pose. Specifically, the self-moving device can first determine the prior pose of the self-moving device based on the odometer or the motion model corresponding to the self-moving device, and then determine the initial parameter values ​​by comparing the prior pose with the specified pose corresponding to the reference point cloud data.

[0075] Step 240: Based on the pose constraints in the vertical direction, optimize the probability corresponding to all point cloud data to determine the maximum value of the probability corresponding to all point cloud data.

[0076] The registration accuracy between the reference point cloud data and the actual point cloud data can be determined by the probabilities corresponding to all point cloud data. As shown in the preceding steps, the pose constraints in the vertical direction between the actual point cloud data and the reference point cloud data have been obtained. Based on this, to improve the registration efficiency of the point cloud, when the self-moving device optimizes the probabilities corresponding to all point cloud data, it can directly substitute the pose constraints in the vertical direction, optimizing only the pose of each point cloud data in other degrees of freedom, thereby increasing the optimization rate and improving the registration efficiency of the point cloud.

[0077] Optionally, the algorithm used for optimization may include gradient descent, derivative algorithms of gradient descent, Newton's method, and quasi-Newton methods. The specific optimization algorithm used is not limited in this application.

[0078] Step 250: When determining the maximum value, obtain the target parameter value used for the pose transformation parameters.

[0079] When the probability corresponding to each point cloud data is at its maximum value, it indicates that the registration effect between the projected actual point cloud data and the reference point cloud data is relatively high. At this point, optimization can be stopped, and the parameter values ​​used for pose transformation can be obtained. To distinguish them from the initial parameter values ​​mentioned above, they can be denoted as target parameter values.

[0080] Step 260: Determine the pose information of the self-moving device based on the target parameter values.

[0081] After determining the target parameter value, the self-moving device can accurately determine its pose information based on that parameter value. Specifically, the pose information can be obtained by transforming a specified pose using the target parameter value.

[0082] In this embodiment, by rasterizing the reference point cloud data, the multidimensional normal distribution parameters of the reference point cloud data in each grid can be determined. These multidimensional normal distribution parameters can be used to describe the local properties of the reference point cloud data. The actual point cloud data is projected onto the grid according to the pose transformation parameters, and the probability corresponding to each point cloud data can be calculated. Under vertical pose constraints, the pose transformation parameters are adjusted through an optimization algorithm to maximize the sum of probabilities corresponding to each point cloud data. At this point, the local properties of the actual point cloud data in each grid tend to be consistent with the local properties of the corresponding reference point cloud, achieving high-precision matching of the two point cloud data. It can be understood that under vertical pose constraints, the optimization efficiency of the optimization process can be improved, thereby improving the matching efficiency of the point clouds.

[0083] In some embodiments, the multidimensional normal distribution parameters include the covariance matrix. Specifically, see [link to relevant documentation]. Figure 3 The covariance matrix corresponding to each grid can be determined through the following steps:

[0084] Step 310: Determine the set of reference point cloud data that falls within the grid.

[0085] The covariance matrix can be calculated from the mean. Therefore, the mobile device can first determine the mean to facilitate the calculation of the covariance matrix. The mean describes the location parameter of the normal distribution, characterizing the central tendency of the reference point cloud data within the grid. To determine the mean, the reference point cloud data within each grid can be statistically analyzed and treated as a set.

[0086] Step 320: Determine the mean of the set.

[0087] Once the set is determined, the mobile device can determine the mean of the set. The probability rule for the mean is: the closer the reference point cloud data is to the mean, the higher the probability; the farther the reference point cloud data is from the mean, the lower the probability.

[0088] Specifically, assuming the set contains n point cloud data points, each point cloud data point can be denoted as x. i Where i = 1, 2, 3, ..., n. Then the formula for calculating the mean q can be as follows:

[0089]

[0090] Step 330: Calculate the covariance matrix of the set within the grid based on the mean.

[0091] Once the mean of the set is determined, the covariance matrix of the set within the grid can be further calculated based on the mean.

[0092] The covariance matrix describes the dispersion of a normally distributed baseline point cloud data set. The general rule for the covariance matrix is: the larger the covariance matrix, the more dispersed the baseline point cloud data; conversely, the smaller the covariance matrix, the more concentrated the baseline point cloud data.

[0093] In some embodiments, the probability model of any point falling into the corresponding grid can be determined based on the formula for calculating the covariance matrix.

[0094] Based on the probability model, the probability of each point cloud data falling into the corresponding grid can be determined.

[0095] In some embodiments, it can be understood that if a single point cloud data point matches well, then the probability determined based on the covariance matrix is ​​greater. To determine the matching accuracy between the actual point cloud data and the reference point cloud data, it can be determined by the probabilities corresponding to all point cloud data points.

[0096] See Figure 4 The above step 240 specifically includes:

[0097] Step 410: Sum the probabilities corresponding to all point cloud data to obtain the probability sum.

[0098] To measure the matching accuracy between actual point cloud data and benchmark point cloud data, the probability sum can be determined for all point cloud data. This involves summing the probabilities corresponding to each point cloud data point.

[0099] The magnitude of this probability sum reflects the matching accuracy between the actual point cloud data and the reference point cloud data. A larger probability sum indicates a better matching effect and higher matching accuracy between the actual point cloud data and the reference point cloud data; a smaller probability sum indicates a worse matching effect and lower matching accuracy between the actual point cloud data and the reference point cloud data.

[0100] Step 420: Based on the pose constraints in the vertical direction, optimize the probability sum to determine the maximum value of the probability sum.

[0101] As the core idea of ​​the NDT algorithm suggests, the better the match between the actual point cloud data and the reference point cloud data, the greater the probability sum. Therefore, the probability sum can be optimized based on pose constraints in the numerical direction to obtain the maximum probability sum.

[0102] In some embodiments, Newton's method can be used to optimize the probability sum to maximize it. In other embodiments, other optimization algorithms may also be used, and this application does not limit the scope of these algorithms.

[0103] It is understandable that the optimization process is iterative. If the probability sum with the maximum value is not determined after this optimization, the process can return to step 230 and subsequent steps until the maximum probability sum is obtained, so as to determine the target parameter values ​​to be used for the pose transformation parameters.

[0104] In some embodiments, for a point cloud acquisition device mounted on a self-moving device, assuming a lidar, each laser beam emitted can be assigned a beam number. See also Figure 5 , Figure 5 A schematic diagram is shown of point cloud data of a charging base acquired by a LiDAR system, with each data point corresponding to a specific line bundle number. Data points with the same bundle number are considered to be on the same line. It is understood that when a self-moving device acquires point cloud data of the same object in different postures, the bundle numbers corresponding to data points at the same location on that object will differ. Based on this, the pose constraints in the vertical direction of the actual point cloud data and the reference point cloud data can be determined according to the bundle numbers. The actual point cloud data of the self-moving device also includes the actual bundle number, and the reference point cloud data also includes the reference bundle number. Therefore, see [reference needed]. Figure 6 The boundary points in the actual point cloud data and the boundary points in the reference point cloud data can be determined through the following steps:

[0105] Step 610: Obtain the actual wire bundle number of each data point in the actual point cloud data, and determine the data point corresponding to the actual wire bundle number of the top layer in the vertical direction as the boundary point in the actual point cloud data.

[0106] To determine the pose constraints in the vertical direction between the actual point cloud data and the reference point cloud data, a location can be first determined. Then, the difference in the numbering of this location in the two point cloud datasets can be identified. The pose constraints in the vertical direction can then be determined based on this difference. This location could be the top of the charging base. After obtaining the actual wiring harness numbers of each data point in the actual point cloud data, the self-moving device can determine the data point corresponding to the number of the topmost actual wiring harness in the vertical direction as the boundary point in the actual point cloud data.

[0107] Step 620: Obtain the baseline bundle number of each data point in the baseline point cloud data, and determine the data point corresponding to the baseline bundle number of the top layer in the vertical direction as the boundary point in the baseline point cloud data.

[0108] Similar to the process of determining boundary points in actual point cloud data, the self-moving device can obtain the actual wire harness number of each data point in the reference point cloud data, and can determine the data point corresponding to the topmost reference wire harness number in the vertical direction as the boundary point in the reference point cloud data.

[0109] In some embodiments, based on the determination of the boundary points of the actual point cloud data and the boundary points of the reference point cloud data, the pose difference between the two types of boundary points in the vertical direction can be determined as the pose difference between the actual point cloud data and the reference point cloud data in the vertical direction, and the pose constraint between the actual point cloud data and the reference point cloud data in the vertical direction can be determined by the pose difference.

[0110] Corresponding to the recharge method in the above embodiments, Figure 7 A structural block diagram of the recharge device 1 provided in an embodiment of this application is shown. For ease of explanation, only the parts related to the embodiment of this application are shown.

[0111] Reference Figure 7 The recharge device 1 includes:

[0112] Acquisition module 10 is used to acquire the first environmental point cloud;

[0113] The filtering module 11 is used to filter the first environmental point cloud according to the preset reflectivity conditions to obtain the actual point cloud data of the charging base; wherein the reflectivity of the shell of the charging base is greater than or equal to the preset reflectivity threshold.

[0114] The first acquisition module 12 is used to acquire the reference point cloud data corresponding to the charging base;

[0115] The determination module 13 is used to determine the pose constraints in the vertical direction based on the boundary points in the actual point cloud data and the boundary points in the reference point cloud data.

[0116] The registration module 14 is used to perform point cloud registration between the actual point cloud data and the reference point cloud data based on the pose constraints in the vertical direction, and to determine the pose information of the self-moving device by using a preset point cloud registration algorithm.

[0117] The control module 15 is used to control the movement of the self-moving device toward the charging base based on the pose information.

[0118] Optionally, the point cloud registration algorithm is a normal distribution transformation algorithm, and the registration module 14 may include:

[0119] The preprocessing unit is used to rasterize the baseline point cloud data to obtain multiple grids.

[0120] The computing unit is used to calculate the multidimensional normal distribution parameters of each grid based on the reference point cloud data falling into each grid.

[0121] The first determining unit is used to project each point cloud data in the actual point cloud data into the grid based on the pose transformation parameters, and to determine the probability that each point cloud data falls into the corresponding grid based on the multidimensional normal distribution parameters; the probability is related to the pose transformation parameters.

[0122] The first optimization unit is used to optimize the probability corresponding to all point cloud data based on the pose constraint in the vertical direction, so as to determine the maximum value of the probability corresponding to all point cloud data.

[0123] The first acquisition unit is used to acquire the target parameter value used for pose transformation parameters when the maximum value is determined;

[0124] The second determining unit is used to determine the pose information of the self-moving device based on the target parameter values.

[0125] Optionally, the multidimensional normal distribution parameters include the covariance matrix, and the computational unit is specifically used for:

[0126] For each grid:

[0127] Determine the set of reference point cloud data that falls within the grid;

[0128] Determine the mean of the set;

[0129] The covariance matrix of the set within the grid is calculated based on the mean.

[0130] Optionally, the registration module 14 may include:

[0131] The summation unit is used to sum the probabilities corresponding to all point cloud data to obtain the probability sum;

[0132] The second optimization unit is used to optimize the probability sum based on the pose constraint in the vertical direction in order to determine the maximum value of the probability sum.

[0133] Optionally, the filtering module 11 may include:

[0134] The second acquisition unit is used to acquire the reflectance corresponding to each first environmental point cloud;

[0135] The third determining unit is used to determine the first environmental point cloud with a reflectivity greater than or equal to a preset reflectivity threshold as the actual point cloud data of the charging base.

[0136] Optionally, the actual point cloud data also includes the actual harness number, and the reference point cloud data also includes the reference harness number; the recharging device 1 may further include:

[0137] The second acquisition module is used to acquire the real wire bundle number of each data point in the actual point cloud data, and to determine the data point corresponding to the real wire bundle number of the top layer in the vertical direction as the boundary point in the actual point cloud data.

[0138] The third acquisition module is used to acquire the baseline bundle number of each data point in the baseline point cloud data, and to determine the data point corresponding to the baseline bundle number of the top layer in the vertical direction as the boundary point in the baseline point cloud data.

[0139] Optionally, the control module 15 may include:

[0140] Planning unit, used to plan recharge path based on pose information;

[0141] The control unit is used to control the movement of the self-moving device along the recharge path to the charging base.

[0142] It should be noted that the information interaction and execution process between the above-mentioned devices / units are based on the same concept as the method embodiments of this application. For details on their specific functions and technical effects, please refer to the method embodiments section, and they will not be repeated here.

[0143] Figure 8 This is a schematic diagram of the physical layer structure of a self-moving device provided in an embodiment of this application. For example... Figure 8 As shown, the self-moving device 2 of this embodiment includes: at least one processor 20 ( Figure 8 Only one is shown in the diagram), memory 21, and computer program 22 stored in memory 21 and executable on at least one processor 20. When processor 20 executes computer program 22, it implements the steps in any of the above-described recharge method embodiments, for example... Figure 1 Steps 110-160 are shown.

[0144] The processor 20 may be a Central Processing Unit (CPU), or it may be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor.

[0145] In some embodiments, memory 21 may be an internal storage unit of the self-moving device 2, such as a hard disk or memory of the self-moving device 2. In other embodiments, memory 21 may also be an external storage device of the self-moving device 2, such as a plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, etc., equipped on the self-moving device 2.

[0146] Furthermore, the memory 21 may include both internal storage units and external storage devices of the self-moving device 2. The memory 21 is used to store operating devices, application programs, bootloaders, data, and other programs, such as program code for computer programs. The memory 21 can also be used to temporarily store data that has been output or will be output.

[0147] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is merely an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the above device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. Furthermore, the specific names of the functional units and modules are only for easy differentiation and are not intended to limit the scope of protection of this application. The specific working process of the units and modules in the above system can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0148] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps described in the various method embodiments above.

[0149] This application provides a computer program product that, when run on a mobile terminal, enables the mobile terminal to implement the steps described in the above-described method embodiments.

[0150] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include at least: any entity or device capable of carrying the computer program code to a photographic device / electronic device, a recording medium, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium, such as a USB flash drive, a portable hard drive, a magnetic disk, or an optical disk.

[0151] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0152] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0153] In the embodiments provided in this application, it should be understood that the disclosed apparatus / network devices and methods can be implemented in other ways. For example, the apparatus / network device embodiments described above are merely illustrative. For instance, the division of modules or units described above is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0154] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0155] The above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A recharge method, characterized in that, include: Collect point clouds of the first environment; The first environmental point cloud is filtered according to a preset reflectivity condition to obtain the actual point cloud data of the charging base; wherein the reflectivity of the shell of the charging base is greater than or equal to a preset reflectivity threshold. Obtain the reference point cloud data corresponding to the charging base; The pose constraint in the vertical direction is determined based on the boundary points in the actual point cloud data and the boundary points in the reference point cloud data. Based on the pose constraints in the vertical direction, the actual point cloud data and the reference point cloud data are registered using a preset point cloud registration algorithm to determine the pose information of the self-moving device. Based on the pose information, control the self-moving device to move toward the charging base; The actual point cloud data also includes the actual wire harness number, and the reference point cloud data also includes the reference wire harness number; Before determining the vertical pose constraint based on the boundary points in the actual point cloud data and the boundary points in the reference point cloud data, the method further includes: Obtain the actual wire bundle number of each data point in the actual point cloud data, and determine the data point corresponding to the topmost actual wire bundle number in the vertical direction as the boundary point in the actual point cloud data. Obtain the baseline bundle number of each data point in the baseline point cloud data, and determine the data point corresponding to the baseline bundle number of the top layer in the vertical direction as the boundary point in the baseline point cloud data.

2. The recharge method as described in claim 1, characterized in that, The point cloud registration algorithm is a normal distribution transformation algorithm. The step of registering the actual point cloud data with the reference point cloud data using a preset point cloud registration algorithm to determine the pose information of the self-moving device includes: The reference point cloud data is rasterized to obtain multiple grids; Calculate the multidimensional normal distribution parameters for each grid based on the reference point cloud data falling into each grid. Based on the pose transformation parameters, each point cloud data in the actual point cloud data is projected onto the grid, and the probability of each point cloud data falling into the corresponding grid is determined based on the multidimensional normal distribution parameters; the probability is related to the pose transformation parameters. Based on the pose constraints in the vertical direction, the probabilities corresponding to all the point cloud data are optimized to determine the maximum value of the probabilities corresponding to all the point cloud data. When determining the maximum value, the target parameter value used for the pose transformation parameters is obtained; The pose information of the self-moving device is determined based on the target parameter value.

3. The recharge method as described in claim 2, characterized in that, The multidimensional normal distribution parameters include the covariance matrix. The calculation of the multidimensional normal distribution parameters for each grid cell based on the reference point cloud data falling into each grid cell includes: For each of the aforementioned grids: Determine the set of reference point cloud data that fall within the grid; Determine the mean of the set; The covariance matrix of the set within the grid is calculated based on the mean.

4. The recharge method as described in claim 2, characterized in that, The optimization process based on the pose constraints in the vertical direction, which optimizes the probabilities corresponding to all the point cloud data to determine the maximum value of the probabilities corresponding to all the point cloud data, includes: The probabilities corresponding to all the point cloud data are summed to obtain the probability sum; Based on the pose constraints in the vertical direction, the sum of probabilities is optimized to determine the maximum value of the sum of probabilities.

5. The recharge method as described in claim 1, characterized in that, The step of filtering the first environmental point cloud according to a preset reflectivity condition to obtain the actual point cloud data of the charging base includes: Obtain the reflectance corresponding to each of the first environmental point clouds; The first environmental point cloud with a reflectivity greater than or equal to a preset reflectivity threshold is determined as the actual point cloud data of the charging base.

6. The recharge method according to any one of claims 1-5, characterized in that, The step of controlling the self-moving device to move toward the charging base based on the pose information includes: Plan the recharge path based on the pose information; Control the self-moving device to move along the recharge path toward the charging base.

7. A rechargeable device, characterized in that, include: The acquisition module is used to acquire the point cloud of the first environment; A filtering module is used to filter the first environmental point cloud according to a preset reflectivity condition to obtain the actual point cloud data of the charging base; wherein the reflectivity of the shell of the charging base is greater than or equal to a preset reflectivity threshold. The first acquisition module is used to acquire the reference point cloud data corresponding to the charging base; The determination module is used to determine the pose constraints in the vertical direction based on the boundary points in the actual point cloud data and the boundary points in the reference point cloud data. The registration module is used to perform point cloud registration between the actual point cloud data and the reference point cloud data based on the pose constraints in the vertical direction and through a preset point cloud registration algorithm, so as to determine the pose information of the self-moving device. The control module is used to control the self-moving device to move toward the charging base based on the pose information; The actual point cloud data also includes the actual wire harness number, and the reference point cloud data also includes the reference wire harness number; The recharge device also includes: The second acquisition module is used to acquire the real wire bundle number of each data point in the actual point cloud data before determining the pose constraint in the vertical direction based on the boundary points in the actual point cloud data and the boundary points in the reference point cloud data, and to determine the data point corresponding to the real wire bundle number of the top layer in the vertical direction as the boundary point in the actual point cloud data. The third acquisition module is used to acquire the baseline bundle number of each data point in the baseline point cloud data, and to determine the data point corresponding to the baseline bundle number of the top layer in the vertical direction as the boundary point in the baseline point cloud data.

8. A self-moving device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the recharge method as described in any one of claims 1 to 6.

9. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the recharge method as described in any one of claims 1 to 6.

Images