Laser ground hitting detection method and device and storage medium

Abstract

The invention discloses a laser ground hitting detection method and device and a storage medium. The laser ground hitting detection method comprises the steps of obtaining target point cloud data collected by a mobile robot; performing straight line fitting processing on the target point cloud data to obtain a target fitting straight line; performing fitting evaluation processing on the target fitting straight line to obtain a fitting evaluation result of the target fitting straight line; and judging whether the mobile robot has a laser ground hitting problem or not according to a fitting evaluation result. According to the scheme, the laser ground hitting problem can be accurately detected, and data misinformation is avoided.

Description

Technical Field

[0001] The present application relates to the field of robotics technology, and in particular to a method, device and storage medium for detecting laser ground marking. Background Art

[0002] As one of the commonly used sensors for robots, lidar plays an important role in complex and changing operating environments.

[0003] For mobile robots, such as AMR (Autonomous Mobile Robot), some laser radars will be installed at the front and bottom of the AMR. The advantage of this design is that the installation height is low, and low obstacles can be identified, thus improving the safety performance of the AMR.

[0004] However, it also brings certain disadvantages. For example, the laser radar is generally fixed to the installation position by screws and other devices. However, as the robot moves and the screws loosen, the laser radar will slowly tilt, making it easy for the laser radar data to hit the ground, causing false data problems. Summary of the invention

[0005] The present application at least provides a method, device, equipment and computer-readable storage medium for detecting laser ground marking.

[0006] The first aspect of the present application provides a method for detecting laser ground penetration, comprising: acquiring target point cloud data collected by a mobile robot; performing straight line fitting processing on the target point cloud data to obtain a target fitting straight line; performing fitting evaluation processing on the target fitting straight line to obtain a fitting evaluation result of the target fitting straight line; and judging whether the mobile robot has a laser ground penetration problem based on the fitting evaluation result.

[0007] In one embodiment, the target point cloud data is subjected to straight line fitting processing to obtain a target fitting straight line, including: selecting at least one group of target point cloud data groups from the target point cloud data; the target point cloud data group includes at least two target point cloud data; performing fitting processing on the target point cloud data in each group of target point cloud data groups respectively to obtain an initial fitting straight line corresponding to each group of target point cloud data groups; determining the inner point of each initial fitting straight line according to the distance between each initial fitting straight line and the target point cloud data; and determining the target fitting straight line from each initial fitting straight line according to the number of inner points of each initial fitting straight line.

[0008] In one embodiment, the selecting at least one target point cloud data group from the target point cloud data includes: randomly selecting every two target point cloud data from the target point cloud data to form a target point cloud data group until the number of target point cloud data groups is the same as the preset number of fitting times; determining the target fitting line from each initial fitting line according to the number of inner points of each initial fitting line includes: selecting the initial fitting line with the largest number of inner points as the target fitting line.

[0009] In one embodiment, the fitting evaluation result includes a first evaluation result and a second evaluation result, and the fitting evaluation processing of the target fitting straight line to obtain the fitting evaluation result of the target fitting straight line includes: determining a distance statistic value based on the distance information between each target point cloud data and the target fitting straight line; comparing the distance statistic value with a preset distance threshold; in response to the distance statistic value being greater than the preset distance threshold, obtaining the first evaluation result; in response to the distance statistic value being less than or equal to the preset distance threshold, obtaining the second evaluation result.

[0010] In one embodiment, in response to the distance statistic being less than or equal to the preset distance threshold, after obtaining the second evaluation result, the method further includes: recording the existence time of the target fitting straight line; in response to the existence time being greater than the preset time threshold, determining that the mobile robot has a laser ground drilling problem.

[0011] In one embodiment, in response to the distance statistic being less than or equal to the preset distance threshold, after obtaining the second evaluation result, the method further includes: comparing the target fitting straight line with a priori map currently used by the mobile robot; in response to the target fitting straight line not matching the prior map, determining that the mobile robot has a laser ground drilling problem.

[0012] In one embodiment, after obtaining the second evaluation result in response to the distance statistic being less than or equal to the preset distance threshold, the method further includes: acquiring movement information of the mobile robot during the current movement process; in response to the movement information indicating that the movement direction of the mobile robot is perpendicular to the direction of gravity, determining that the mobile robot has a laser ground drilling problem.

[0013] In one embodiment, the step of acquiring target point cloud data collected by the mobile robot includes:

[0014] Acquire initial point cloud data collected by the mobile robot; perform point cloud screening processing on the initial point cloud data according to the prior map used by the mobile robot to obtain screened point cloud data; determine the target point cloud data according to the point cloud distance between the screened point cloud data.

[0015] The second aspect of the present application provides a laser ground marking detection device, including: an acquisition module, used to acquire target point cloud data collected by a mobile robot; a point cloud fitting module, used to perform straight line fitting processing on the target point cloud data to obtain a target fitting straight line; a fitting evaluation module, used to perform fitting evaluation processing on the target fitting straight line to obtain a fitting evaluation result of the target fitting straight line; a judgment module, used to judge whether the mobile robot has a laser ground marking problem based on the fitting evaluation result.

[0016] A third aspect of the present application provides an electronic device, including a memory and a processor, wherein the processor is used to execute program instructions stored in the memory to implement the above-mentioned laser ground detection method.

[0017] A fourth aspect of the present application provides a computer-readable storage medium having program instructions stored thereon, which implement the above-mentioned laser ground detection method when the program instructions are executed by a processor.

[0018] The above scheme, by acquiring the target point cloud data collected by the mobile robot and performing straight line fitting processing on the target point cloud data, can obtain the target fitting straight line for judging whether the target point cloud data collected by the mobile robot is arranged in an approximate straight line shape; the target fitting straight line is subjected to fitting evaluation processing to obtain the fitting evaluation result of the target fitting straight line, and the fitting effect of the target fitting straight line can be judged according to the fitting evaluation result, thereby judging whether the mobile robot has the problem of laser ground marking according to the fitting evaluation result, thereby realizing efficient detection of the problem of laser ground marking of the mobile robot.

[0019] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] The drawings herein are incorporated into the specification and constitute a part of the specification. These drawings illustrate embodiments consistent with the present application and are used together with the specification to illustrate the technical solution of the present application.

[0021] Figure 1 It is a simplified top view of the movable robot in the laser ground detection method of the present application;

[0022] Figure 2 It is a simplified side view of the movable robot in the laser ground detection method of the present application;

[0023] Figure 3 It is a flow chart of an exemplary embodiment of the laser ground detection method of the present application;

[0024] Figure 4It is an exemplary point cloud data schematic diagram in the laser ground detection method of the present application;

[0025] Figure 5 It is an exemplary straight line fitting schematic diagram in the laser ground detection method of the present application;

[0026] Figure 6 is a block diagram of a laser ground detection device shown in an exemplary embodiment of the present application;

[0027] Figure 7 It is a structural schematic diagram of an embodiment of the electronic device of the present application;

[0028] Figure 8 It is a structural diagram of an embodiment of a computer-readable storage medium of the present application. DETAILED DESCRIPTION

[0029] The scheme of the embodiment of the present application is described in detail below in conjunction with the drawings of the specification.

[0030] In the following description, for the purpose of explanation rather than limitation, specific details such as specific system structures, interfaces, and technologies are provided to facilitate a thorough understanding of the present application.

[0031] The term "and / or" in this article is only a description of the association relationship of associated objects, indicating that there may be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article generally indicates that the objects associated before and after are in an "or" relationship. In addition, "many" in this article means two or more than two. In addition, the term "at least one" in this article means any combination of at least two of any one or more of a plurality of, for example, including at least one of A, B, and C, can mean including any one or more elements selected from the set consisting of A, B, and C.

[0032] For ease of understanding, one of the applicable scenarios of the present application is now exemplarily described. In the field of robotics, laser radar, as one of the commonly used sensors of robots, plays an important role in complex and changeable operating environments.

[0033] For mobile robots (such as AMR, AGV, etc.), taking AMR as an example, reference can be made to the Figure 1 As shown, Figure 1 This is a simplified top view of the movable robot in the laser ground detection method of this application. Two laser radars (or laser sensors) can be installed below the left and right sides of the robot head (in the forward direction). The laser radar can be installed 10 cm above the ground and is mainly used to identify low obstacles to improve the safety of AMR operation.

[0034] However, it also brings certain disadvantages. Figure 2 As shown, Figure 2 This is a simplified side view of the movable robot in the laser ground detection method of the present application. For example, the laser radar is generally fixed to the installation position by screws and other devices. However, as the robot moves and the screws loosen, the laser radar will slowly tilt. For example, the pitch angle of the laser radar changes, making it easy for the laser radar data to hit the ground, and failing to correctly collect the point cloud data in front of the robot, resulting in data misreporting.

[0035] See also Figure 3 , Figure 3 This is a flowchart of an exemplary embodiment of the laser ground detection method of the present application. The detection method of the present application can be applied to a mobile robot and / or to other devices that are in communication connection with the mobile robot. This is not limited here and can be adaptively adjusted and implemented according to specific needs. Specifically, it can include the following steps:

[0036] Step S110, acquiring target point cloud data collected by the mobile robot.

[0037] The target point cloud data refers to the point cloud data (or laser data) used to determine whether the mobile robot has a laser grounding problem. The target point cloud data can be data obtained after data processing or data obtained without data processing, which is not limited here.

[0038] Exemplarily, in this step, the target point cloud data collected by the mobile robot in real time may be obtained, or the target point cloud data collected by the mobile robot before may be obtained.

[0039] Step S120, performing straight line fitting processing on the target point cloud data to obtain a target fitting straight line.

[0040] Line fitting processing refers to fitting the target point cloud data according to the shape of a straight line to obtain a target fitting straight line. The target fitting straight line is equivalent to characterizing the distribution of the target point cloud data to a certain extent.

[0041] For example, in the present application, the target point cloud data may be subjected to a linear fitting process according to a preset linear fitting algorithm. The linear fitting algorithms available in the present application may include, but are not limited to, RANSAC algorithm (Random Sample Consensus), Flitline algorithm, least squares method, weighted fitting linear equation method, gradient descent method, etc.

[0042] Step S130, performing fitting evaluation processing on the target fitting straight line to obtain a fitting evaluation result of the target fitting straight line.

[0043] Fitting evaluation processing refers to evaluating the target fitting straight line to determine the fitting effect of the target fitting straight line, that is, the fitting evaluation result. It can be understood that the distance between the straight line with good fitting effect and the data points participating in the straight line fitting is close, while the distance between the straight line with poor fitting effect and the data points participating in the straight line fitting is far away.

[0044] For example, reference may be made to Figure 4 As shown, Figure 4 This is an exemplary point cloud data diagram in the laser ground detection method of the present application. Figure 2 and Figure 4 ,It is understandable that when the robot has a laser grounding problem (e.g. Figure 4 The laser radar on the left side of the robot has a problem with hitting the ground). The laser emitted by the laser radar will be reflected by hitting the ground, so the laser point cloud data determined by the robot based on the received reflected laser will form a nearly straight line. In fact, this area may be an open flat land with no obstacles, such as Figure 4 The point cloud data detected by the laser radar on the right side of the robot is shown in the figure.

[0045] Therefore, in the present application, it is possible to determine whether the robot has a laser grounding problem by using the straight line fitting effect (fitting evaluation result) of the target fitting straight line.

[0046] Step S140, judging whether the mobile robot has a laser ground-piercing problem according to the fitting evaluation result.

[0047] Combined with the above steps, if the fitting effect of the target fitting line is good, it means that the target point cloud data currently acquired is almost arranged in a straight line, and it can be determined that the robot has a laser ground problem. On the contrary, if the fitting effect of the target fitting line is poor, it means that the target point cloud data currently acquired is not arranged in a straight line, which may be because the laser radar does scan the existence of an object in front, so it can be determined that the robot does not have a laser ground problem.

[0048] It can be seen that the present application obtains the target point cloud data collected by the mobile robot, performs straight line fitting processing on the target point cloud data, and can obtain a target fitting straight line for judging whether the target point cloud data collected by the mobile robot is arranged in an approximate straight line shape; performs fitting evaluation processing on the target fitting straight line to obtain a fitting evaluation result of the target fitting straight line, and can judge the fitting effect of the target fitting straight line according to the fitting evaluation result, thereby judging whether the mobile robot has a laser ground marking problem according to the fitting evaluation result, thereby realizing efficient detection of the mobile robot's laser ground marking problem.

[0049] Based on the above embodiments, the embodiment of the present application exemplarily illustrates the steps of performing straight line fitting processing on the target point cloud data to obtain the target fitting straight line. Specifically, the method of this embodiment includes the following steps:

[0050] At least one target point cloud data group is selected from the target point cloud data; the target point cloud data group includes at least two target point cloud data; fitting processing is performed on the target point cloud data in each target point cloud data group to obtain an initial fitting straight line corresponding to each target point cloud data group; the inner point of each initial fitting straight line is determined according to the distance between each initial fitting straight line and the target point cloud data; and the target fitting straight line is determined from each initial fitting straight line according to the number of inner points of each initial fitting straight line.

[0051] In combination with the above-mentioned embodiments, this embodiment mainly illustrates the straight line fitting process of the present application.

[0052] Specifically, in the straight line fitting process of the present application, reference can be made to the following Figure 5 As shown, Figure 5 This is an exemplary straight line fitting schematic diagram in the laser ground detection method of the present application. At least one group of target point cloud data is randomly selected from the target point cloud data or selected according to a set rule; each group of target point cloud data includes at least two target point cloud data, so that the target point cloud data in each group of target point cloud data can generate a corresponding fitting straight line after the straight line fitting processing.

[0053] Then, the target point cloud data in each target point cloud data group is fitted to obtain the initial fitting straight line corresponding to each target point cloud data group. If there is only one initial fitting straight line, the initial fitting straight line can be directly used as the target fitting straight line; if there are at least two initial fitting straight lines, the initial fitting straight lines need to be screened, and the target fitting straight line is determined from each initial fitting straight line. Among them, the initial fitting straight lines can be screened according to the number of inner points of the initial fitting straight line. The inner point refers to the data point (point cloud data) within a certain distance from the fitting straight line. The distance between the inner point and the straight line can be preset or flexibly adjusted, which is not limited here. In summary, the target fitting straight line can be determined from each initial fitting straight line according to the number of inner points of each initial fitting straight line.

[0054] Exemplarily, each group of target point cloud data includes two target point cloud data. In the implementation process of this embodiment, every two target point cloud data can be randomly selected from the target point cloud data as a target point cloud data group for straight line fitting, until the initial fitting straight line obtained reaches a specified number, or the number of randomly selected target point clouds (which can be deduplicated) relative to the total number of point clouds reaches a specified threshold (for example, half of the point cloud data in the target point cloud data has been selected, or all the point cloud data in the target point cloud data has been selected, etc., which is not limited here). It should be noted that in this embodiment, the process of selecting target point cloud data for straight line fitting can be performed synchronously and concurrently, or it can be performed step by step (for example, iterating point cloud selection and straight line fitting 100 times to obtain 100 initial fitting straight lines), which is not limited here.

[0055] For each initial fitting line, the distance from all or part of the target point cloud data to the initial fitting line can be calculated, and then the number of target point cloud data corresponding to each initial fitting line that satisfies the distance between the point cloud and the line (for example, the straight line distance) is less than the set point-line threshold can be counted, and the point cloud data that meets the above conditions can be determined as the inner point of the initial fitting line.

[0056] There are many methods for calculating the distance between a point and a straight line (for details, please refer to the prior art and will not be described in detail here). This application may exemplarily provide one of the calculation methods, for example:

[0057]

[0058] Where (x, y) represents the coordinates of the input target point cloud data, k and b represent the slope and intercept of the initial fitting line, respectively. In this embodiment, the point line threshold for judging the inliers can be set to 6 cm, and the number of iterations can be set to 100 times. That is, the target point cloud data within 6 cm of an initial fitting line is regarded as the inliers of the initial fitting line, and the above straight line fitting processing method can be iterated 100 times to obtain 100 initial fitting lines.

[0059] Based on the above embodiment, the embodiment of the present application describes the steps of selecting at least one set of target point cloud data from the target point cloud data and determining the target fitting line from each initial fitting line according to the number of inner points of each initial fitting line. Specifically, the method of this embodiment includes the following steps:

[0060] Randomly select every two target point cloud data from the target point cloud data to form a target point cloud data group until the number of target point cloud data groups is the same as the preset fitting number; select the initial fitting straight line with the largest number of internal points as the target fitting straight line.

[0061] In conjunction with the foregoing embodiment, when selecting target point cloud data to form a target point cloud data group in this embodiment, a preset number of fitting times can be set in advance. The preset number of fitting times represents the number of fitting processes expected to be performed during the straight line fitting process of this application. Since in some scenarios, the number of target point cloud data may be very large, when the number of target point cloud data groups selected for straight line fitting processing in this application is the same as the preset number of fitting times, or when the current number of straight line fitting processes is the same as the preset number of fitting times, the straight line fitting process can be paused or stopped, and the target point cloud data in each target point cloud data group that has been selected can be subjected to straight line fitting processing respectively, and it is not necessary to perform straight line fitting processing on all target point cloud data.

[0062] Furthermore, after obtaining at least two initial fitting lines, the target fitting line can be determined from the initial fitting lines according to the number of inliers of each initial fitting line. It can be understood that the optimal fitting line under an ideal state should be that all data points are on the same straight line. Therefore, when determining the target fitting line, the initial fitting line with the largest number of inliers can be selected as the target fitting line; it is not limited to selecting the initial fitting line with the second largest number of inliers, or the initial fitting line with the number of inliers reaching a certain threshold as the target fitting line, which is not limited here.

[0063] Optionally, in combination with the above-mentioned embodiment, the present application may also provide a method that determines the number of inliers of an initial fitting line each time an initial fitting line is obtained. In response to the ratio of the number of inliers of the currently obtained initial fitting line to the total number of all target point cloud data being greater than a preset inlier ratio threshold, the currently obtained initial fitting line may be directly determined as the target fitting line without continuing the subsequent target point cloud selection and straight line fitting process until the preset number of fitting times is reached, thereby significantly improving processing efficiency.

[0064] Based on the above embodiment, the embodiment of the present application exemplarily illustrates the steps of performing fitting evaluation processing on the target fitting straight line to obtain the fitting evaluation result of the target fitting straight line. The fitting evaluation result includes a first evaluation result and a second evaluation result. Specifically, the method of this embodiment includes the following steps:

[0065] The distance statistics are determined according to the distance information between each target point cloud data and the target fitting straight line; the distance statistics are compared with the preset distance threshold; in response to the distance statistics being greater than the preset distance threshold, a first evaluation result is obtained; in response to the distance statistics being less than or equal to the preset distance threshold, a second evaluation result is obtained.

[0066] In conjunction with the above-mentioned embodiments, the fitting evaluation results in the present application are mainly used to evaluate the fitting effect of the target fitting line. If the distance between each target point cloud data and the target fitting line is far, the fitting effect of the target fitting line is poor; if the distance between each target point cloud data and the target fitting line is close, the fitting effect of the target fitting line is good. Among them, the fitting evaluation results include a first evaluation result and a second evaluation result. The first evaluation result is used to characterize that the fitting effect of the target fitting line is poor, and the second evaluation result is used to characterize that the fitting effect of the target fitting line is good.

[0067] Specifically, the distance statistics are determined according to the distance information between each target point cloud data and the target fitting line. The distance statistics may include but are not limited to the average value, covariance, etc. of the distance information, which are not limited here. This embodiment is mainly described by taking the average value as an example.

[0068] According to the distance information between each target point cloud data and the target fitting line, determine the distance average of the distance information. Compare the distance average with the preset distance threshold. Among them, the preset distance threshold is used to judge the fitting effect of the target fitting line. The preset distance threshold can be the same as or different from the point line threshold used to judge the inner point in the aforementioned embodiment, and is not limited here. For example, the preset distance threshold of the present application can be 6cm. If the average value of the distance from all target point cloud data to the target fitting line is greater than 6cm, a first evaluation result is obtained, indicating that the fitting effect of the target fitting line is poor; if the average value of the distance from all target point cloud data to the target fitting line is less than or equal to 6cm, a second evaluation result is obtained, indicating that the fitting effect of the target fitting line is good.

[0069] If the second evaluation result is obtained, it can be determined that the mobile robot has a laser ground-piercing problem.

[0070] Based on the above embodiment, the embodiment of the present application describes the steps after obtaining the second evaluation result in response to the distance statistic value being less than or equal to the preset distance threshold. Specifically, the method of this embodiment includes the following steps:

[0071] The existence time of the target fitting straight line is recorded; in response to the existence time being greater than a preset time threshold, it is determined that the mobile robot has a laser ground-piercing problem.

[0072] In conjunction with the above-mentioned embodiment, it should be further explained that in order to improve the accuracy of the laser ground detection of the present application, after obtaining the second evaluation result of the target fitting straight line, further judgment can be made to avoid the problem of misjudgment.

[0073] Exemplarily, the existence time of the target fitting straight line can be recorded. The existence time of the target fitting straight line is equivalent to the duration of the target point cloud data collected by the mobile robot being distributed approximately in a straight line. It is understandable that if there is an obstacle on the robot's driving path, causing the obstacle point cloud data collected by the robot to be distributed approximately in a straight line, then after the robot passes the obstacle, the target fitting straight line formed by the obstacle point cloud data will no longer be detected. Therefore, if the target point cloud data collected by the robot is distributed approximately in a straight line for a long time, the characterization may be caused by the existence of a laser grounding problem.

[0074] Therefore, in response to the existence time being greater than the preset time threshold, it can be determined that the mobile robot has a laser grounding problem. If the target fitting straight line corresponding to the target point cloud data changes within the preset time threshold, it can be determined that the mobile robot does not have a laser grounding problem.

[0075] Based on the above embodiment, the embodiment of the present application describes the steps after obtaining the second evaluation result in response to the distance statistic value being less than or equal to the preset distance threshold. Specifically, the method of this embodiment includes the following steps:

[0076] The target fitting straight line is compared with a priori map currently used by the mobile robot; in response to the target fitting straight line not matching the priori map, it is determined that the mobile robot has a laser grounding problem.

[0077] With reference to the foregoing embodiments, in addition to further judging whether the robot has a laser ground-hitting problem based on the existence time of the target fitting straight line, the present application can also compare the target point cloud data with the prior map currently used by the robot to judge whether the robot has a laser ground-hitting problem.

[0078] The prior map is map data based on prior knowledge, which is equivalent to map data constructed based on the historical point cloud data collected by the mobile robot during its previous movement. Therefore, based on the prior map, it is known whether there are obstacles in the current moving scene of the mobile robot, that is, whether the target point cloud data where the target fitting line is located is generated by obstacles.

[0079] Specifically, the target fitting line is compared with the prior map currently used by the mobile robot; in response to the target fitting line not matching the prior map (the target point cloud data corresponding to the target fitting line does not exist in the prior map), it is determined that the mobile robot has a laser ground-hitting problem. In response to the target fitting line matching the prior map (the target point cloud data corresponding to the target fitting line exists in the prior map, which may be generated by the laser radar irradiating an obstacle), it can be determined that the mobile robot does not have a laser ground-hitting problem.

[0080] It should be noted that the judgment method of this embodiment may include but is not limited to being implemented alone after obtaining the second evaluation result; it may also be combined with the method of judging based on the existence time in the aforementioned embodiment, which will not be elaborated here.

[0081] Based on the above embodiment, the embodiment of the present application describes the steps after obtaining the second evaluation result in response to the distance statistic value being less than or equal to the preset distance threshold. Specifically, the method of this embodiment includes the following steps:

[0082] The movement information of the mobile robot in the current movement process is obtained; in response to the movement information indicating that the movement direction of the mobile robot is perpendicular to the gravity direction, it is determined that the mobile robot has a laser grounding problem.

[0083] In combination with the above-mentioned embodiment, in addition to being able to further determine whether the robot has a laser ground hitting problem based on the existence time of the target fitting straight line, the target point cloud data is compared with the prior map currently used by the robot to determine whether the robot has a laser ground hitting problem. This embodiment also provides a method for determining whether the robot has a laser ground hitting problem based on the robot's movement information.

[0084] The movement information of the robot may include but is not limited to the movement direction of the robot. The movement direction can be obtained by sensors such as gyroscopes, accelerometers, and levels provided on the robot, which are not limited here. It is understandable that in some special scenarios (such as sloped roads), the robot may still have laser grounding problems when moving on a sloped road, but such problems are caused by road factors, not by the tilt of the robot's laser radar in this application. Therefore, when judging whether the robot has a laser grounding problem, such special conditions can be excluded.

[0085] Therefore, when there is a target fitting line with a good fitting effect, if the movement information indicates that the moving direction of the mobile robot is perpendicular to the gravity direction, that is, the mobile robot moves on the horizontal plane, it can be determined that the mobile robot has a laser grounding problem. If the movement information indicates that the moving direction of the mobile robot is not perpendicular to the gravity direction, that is, the mobile robot does not move on the horizontal plane (for example, it moves on a slope plane), it can be determined that the mobile robot does not have a laser grounding problem.

[0086] For example, in the specific implementation process of this embodiment, the state of the level of the mobile robot during the current movement process can be directly obtained. If the level state is horizontal, it means that the moving direction of the mobile robot is perpendicular to the gravity direction, and it is determined that the mobile robot has a laser grounding problem. On the contrary, if the level state is non-horizontal, it means that the moving direction of the mobile robot is not perpendicular to the gravity direction, and it is determined that the mobile robot does not have a laser grounding problem.

[0087] In some special scenarios, when the robot is traveling on a horizontal road, the laser radar will scan the sloped road ahead, resulting in point cloud data with approximate straight line distribution. Therefore, the robot's horizontal driving state can be determined when the target fitting straight line of the second evaluation result is detected, and / or when the robot passes through the path segment where the target fitting straight line is located, and / or after passing through the path segment where the target fitting straight line is located, the robot's horizontal driving state can be determined, which is not limited here.

[0088] It should be noted that before or during the operation of the robot, if it can be determined that the operating scene of the robot is a horizontal ground based on pre-set operating scene information, prior map information, etc., then there is no need to make the judgment on horizontal driving in the above embodiment. When the target fitting straight line of the second evaluation result is detected, it can be directly determined that there is a laser ground hitting problem.

[0089] It should also be noted that in the present application, in the process of making secondary judgments based on the existence time of the target fitting line, the matching result between the target fitting line and the prior map, and the horizontal driving state of the robot, at least one of the above judgment conditions can be arbitrarily selected, and no limitation is made here.

[0090] In addition, it is also possible to refer to the various judgment conditions provided in the aforementioned embodiments, and in the application scenario where at least two judgment conditions are selected for judgment, a corresponding judgment weight is set for each selected judgment condition. According to the judgment weight, the judgment result of each judgment condition selected for application is weighted and summed, and then a comprehensive judgment result obtained according to multiple judgment conditions is obtained, thereby improving the detection accuracy of laser grounding problems.

[0091] On the basis of the above embodiments, the embodiments of the present application also need to explain that in the laser ground detection method of the present application, in addition to the above embodiments, after determining the target fitting straight line of the second evaluation result, it can be determined whether the robot is in horizontal driving, and then determine whether there is a laser ground problem. It can also be that first determine whether the robot is in horizontal driving, if it is in horizontal driving, then judge whether the robot has a laser ground problem according to the fitting effect of the target fitting straight line according to the steps in the above embodiments (such as steps S110 to S140); if it is not in horizontal driving, the implementation of the steps of the above embodiments can be paused or stopped until the robot is in a horizontal driving state. This can significantly save relevant computing resources.

[0092] Based on the above embodiments, the present embodiment of the present application describes the steps of obtaining the target point cloud data collected by the mobile robot. Specifically, the method of this embodiment includes the following steps:

[0093] Acquire initial point cloud data collected by the mobile robot; perform point cloud screening on the initial point cloud data according to a priori map used by the mobile robot to obtain screened point cloud data; determine target point cloud data according to the point cloud distance between the screened point cloud data.

[0094] With reference to the foregoing embodiments, the target point cloud data in the present application may be obtained through certain data processing based on the initial point cloud data collected by the robot.

[0095] Specifically, initial point cloud data collected by the mobile robot is acquired; point cloud filtering is performed on the initial point cloud data according to a priori maps used by the mobile robot to obtain filtered point cloud data; and target point cloud data is determined according to the point cloud distance between the filtered point cloud data.

[0096] Exemplarily, the map route of the prior map may pre-store set road information (such as road width information, road slope information, road length information, etc., which are not limited here). Therefore, based on the coordinate information of the initial point cloud data, the initial point cloud data that is not on the current driving road of the robot may be eliminated, thereby obtaining filtered point cloud data.

[0097] Furthermore, usually, since the laser radar data emits a laser beam at a fixed angle within the viewing angle, in order to improve the accuracy of judging the laser grounding problem, it is necessary to extract continuous and valid point cloud data. For example, the point cloud distance value between adjacent laser point clouds cannot exceed a certain point cloud distance threshold (for example, this solution can be set to 15cm), which means that it is the same set of continuous point cloud data. At the same time, if the number of point cloud data in a point cloud data set within a certain area (such as a point cloud data set formed by adjacent laser point clouds) exceeds the point cloud quantity threshold (for example, 20), then these initial point cloud data can be considered to be valid point clouds, and can be determined as target point cloud data.

[0098] It can also be explained that, generally speaking, the coordinate information of the point cloud data is defined relative to the vehicle body coordinate system. In order to facilitate the data processing process of the present application, the point cloud coordinate information in the vehicle body coordinate system can be converted into the point cloud coordinate information in the world coordinate system.

[0099] For example: Get the angle offset of the laser radar origin in the vehicle coordinate system as θ L , the translation amount is T L =(x l ,y l ) T , the coordinates of the laser point cloud in the laser coordinate system are p0 = (x0, y0) T , then the coordinate p1 of the laser point cloud in the vehicle coordinate system is:

[0100] p1=R(θ L )p0+T L

[0101] in, Represents a rotation matrix.

[0102] Similarly, the angular offset of the center of the vehicle in the world coordinate system is known to be θ c , the translation amount is T c =(x c ,y c ) T , then the coordinate p2 of the laser point cloud in the world coordinate system is:

[0103] p2=R(θ c )p1+T c

[0104] Therefore, the coordinate information of all laser point clouds can be converted into coordinates in the world coordinate system according to the above formula.

[0105] Based on the above embodiment, the embodiment of the present application describes the steps after determining whether the mobile robot has a laser ground-piercing problem based on the fitting evaluation result.

[0106] Specifically, after determining that the mobile robot has a laser ground-marking problem, the relevant information of the point cloud data can be saved, and a prompt message of the laser ground-marking problem can be generated and sent to the corresponding platform so that the staff can inspect and maintain the robot in a timely manner.

[0107] Among them, in order to avoid false alarms of laser grounding problems, you can also set the risk factor to control whether to generate prompt information for sending. For example, the risk factor can be positively correlated with the existence time of the target fitting line. The longer the existence time of the target fitting line, the higher the risk factor of the robot is adjusted. Until the risk factor is higher than the preset risk threshold, you can choose to generate a prompt message and send it. If the risk factor is always lower than or equal to the preset risk threshold, you do not need to generate a prompt message for sending, or generate a prompt message but do not send it for the time being.

[0108] It should be further explained that the execution subject of the laser ground detection method may be a laser ground detection device, for example, the laser ground detection method may be executed by a terminal device or a server or other processing device, wherein the terminal device may be a user equipment (UE), a computer, a mobile device, a user terminal, a terminal, a cellular phone, a cordless phone, a personal digital assistant (PDA), a handheld device, a computing device, a vehicle-mounted device, a wearable device, etc. In some possible implementations, the laser ground detection method may be implemented by a processor calling a computer-readable instruction stored in a memory.

[0109] Figure 6 FIG. 1 is a block diagram of a laser ground detection device shown in an exemplary embodiment of the present application. Figure 6 As shown, the exemplary laser ground detection device 600 includes: an acquisition module 610, a point cloud fitting module 620, a fitting evaluation module 630 and a judgment module 640. Specifically:

[0110] The acquisition module 610 is used to acquire the target point cloud data collected by the mobile robot.

[0111] The point cloud fitting module 620 is used to perform straight line fitting processing on the target point cloud data to obtain a target fitting straight line.

[0112] The fitting evaluation module 630 is used to perform fitting evaluation processing on the target fitting straight line to obtain a fitting evaluation result of the target fitting straight line.

[0113] The judgment module 640 is used to judge whether the mobile robot has a laser ground-piercing problem according to the fitting evaluation result.

[0114] In this exemplary laser ground marking detection device, by acquiring the target point cloud data collected by the mobile robot and performing straight line fitting processing on the target point cloud data, a target fitting straight line can be obtained to determine whether the target point cloud data collected by the mobile robot is arranged in an approximate straight line shape; a fitting evaluation process is performed on the target fitting straight line to obtain a fitting evaluation result of the target fitting straight line, and the fitting effect of the target fitting straight line can be judged according to the fitting evaluation result. Thus, it is possible to judge whether the mobile robot has a laser ground marking problem according to the fitting evaluation result, thereby achieving efficient detection of the mobile robot's laser ground marking problem.

[0115] It should be noted that the device provided in the above embodiment and the method provided in the above embodiment belong to the same concept, wherein the specific manner in which each module and unit performs the operation has been described in detail in the method embodiment, and will not be repeated here. In practical applications, the device provided in the above embodiment can distribute the above functions to different functional modules as needed, that is, divide the internal structure of the device into different functional modules to complete all or part of the functions described above, and this is not limited here.

[0116] The functions of each module can be found in the embodiment of the detection method for laser grounding, which will not be described in detail here.

[0117] See also Figure 7 , Figure 7 1 is a schematic diagram of the structure of an embodiment of an electronic device of the present application. The electronic device 100 includes a memory 101 and a processor 102, and the processor 102 is used to execute program instructions stored in the memory 101 to implement the steps in any of the above-mentioned laser ground detection method embodiments. In a specific implementation scenario, the electronic device 100 may include but is not limited to: a microcomputer, a server, and in addition, the electronic device 100 may also include a mobile device such as a laptop computer and a tablet computer, which is not limited here.

[0118] Specifically, the processor 102 is used to control itself and the memory 101 to implement the steps in any of the above-mentioned laser ground detection method embodiments. The processor 102 can also be called a CPU (Central Processing Unit). The processor 102 may be an integrated circuit chip with signal processing capabilities. The processor 102 can also be a general-purpose processor, a digital signal processor (Digital Signal Processor, DSP), an application-specific integrated circuit (Application Specific Integrated Circuit, ASIC), a field-programmable gate array (Field-Programmable Gate Array, FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc. In addition, the processor 102 can be implemented by an integrated circuit chip.

[0119] In this exemplary electronic device, by acquiring the target point cloud data collected by the mobile robot and performing straight line fitting processing on the target point cloud data, a target fitting straight line can be obtained to determine whether the target point cloud data collected by the mobile robot is arranged in an approximate straight line shape; a fitting evaluation process is performed on the target fitting straight line to obtain a fitting evaluation result of the target fitting straight line, and the fitting effect of the target fitting straight line can be judged according to the fitting evaluation result. Thus, it can be judged whether the mobile robot has a laser ground marking problem according to the fitting evaluation result, thereby achieving efficient detection of the mobile robot's laser ground marking problem.

[0120] See also Figure 8 , Figure 8 The computer-readable storage medium 110 stores program instructions 111 that can be executed by a processor, and the program instructions 111 are used to implement the steps in any of the above-mentioned laser ground detection method embodiments.

[0121] In this exemplary storage medium, by running the program instructions in the storage medium, the target point cloud data collected by the mobile robot is obtained, and a straight line fitting process is performed on the target point cloud data, so that a target fitting straight line can be obtained for judging whether the target point cloud data collected by the mobile robot is arranged in an approximate straight line shape; a fitting evaluation process is performed on the target fitting straight line to obtain a fitting evaluation result of the target fitting straight line, and the fitting effect of the target fitting straight line can be judged according to the fitting evaluation result, thereby judging whether the mobile robot has a laser ground marking problem according to the fitting evaluation result, thereby achieving efficient detection of the laser ground marking problem of the mobile robot.

[0122] In some embodiments, the functions or modules included in the device provided by the embodiments of the present disclosure can be used to execute the method described in the above method embodiments. The specific implementation can refer to the description of the above method embodiments, and for the sake of brevity, it will not be repeated here.

[0123] The above description of various embodiments tends to emphasize the differences between the various embodiments. The same or similar aspects can be referenced to each other, and for the sake of brevity, they will not be repeated herein.

[0124] In the several embodiments provided in the present application, it should be understood that the disclosed methods and devices can be implemented in other ways. For example, the device implementation described above is only schematic. For example, the division of modules or units is only a logical function division. There may be other division methods in actual implementation, such as units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, and the indirect coupling or communication connection of devices or units can be electrical, mechanical or other forms.

[0125] In addition, each functional unit in each embodiment of the present application can be integrated into a processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of a software functional unit. If the integrated unit is implemented in the form of 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, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) or a processor (processor) to perform all or part of the steps of each implementation method of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program code.

Claims

1. A method for detecting laser grounding, characterized in that: The method comprises: Obtain target point cloud data collected by the mobile robot; Performing straight line fitting processing on the target point cloud data to obtain a target fitting straight line; Performing a fitting evaluation process on the target fitting straight line to obtain a fitting evaluation result of the target fitting straight line; It is determined whether the mobile robot has a laser ground-penetrating problem according to the fitting evaluation result.

2. The method according to claim 1, characterized in that The performing straight line fitting processing on the target point cloud data to obtain a target fitting straight line includes: Selecting at least one target point cloud data group from the target point cloud data; the target point cloud data group includes at least two target point cloud data; Perform fitting processing on the target point cloud data in each target point cloud data group respectively to obtain an initial fitting straight line corresponding to each target point cloud data group; Determine the inner point of each initial fitting straight line according to the distance between each initial fitting straight line and the target point cloud data; The target fitting straight line is determined from each initial fitting straight line according to the number of interior points of each initial fitting straight line.

3. The method according to claim 2, characterized in that The selecting at least one target point cloud data set from the target point cloud data comprises: Randomly selecting every two target point cloud data from the target point cloud data to form a target point cloud data group, until the number of the target point cloud data groups is the same as the preset fitting number; The step of determining the target fitting straight line from each initial fitting straight line according to the number of interior points of each initial fitting straight line comprises: The initial fitting straight line with the largest number of inliers is selected as the target fitting straight line.

4. The method according to claim 1, characterized in that: The fitting evaluation result includes a first evaluation result and a second evaluation result. The fitting evaluation process is performed on the target fitting straight line to obtain the fitting evaluation result of the target fitting straight line, including: Determine a distance statistic value according to the distance information between each target point cloud data and the target fitting straight line; Comparing the distance statistics with a preset distance threshold; In response to the distance statistic being greater than the preset distance threshold, obtaining the first evaluation result; In response to the distance statistic being less than or equal to the preset distance threshold, the second evaluation result is obtained.

5. The method according to claim 4, characterized in that In response to the distance statistic being less than or equal to the preset distance threshold, after obtaining the second evaluation result, the method further includes: Recording the existence time of the target fitting straight line; In response to the existence time being greater than a preset time threshold, it is determined that the mobile robot has a laser ground marking problem.

6. The method according to claim 4, characterized in that In response to the distance statistic being less than or equal to the preset distance threshold, after obtaining the second evaluation result, the method further includes: Comparing the target fitting straight line with a priori map currently used by the mobile robot; In response to the target fitting straight line not matching the priori map, it is determined that the mobile robot has a laser ground marking problem.

7. The method according to claim 4, characterized in that In response to the distance statistic being less than or equal to the preset distance threshold, after obtaining the second evaluation result, the method further includes: Acquiring movement information of the mobile robot during the current movement process; In response to the movement information indicating that the movement direction of the mobile robot is perpendicular to the direction of gravity, it is determined that the mobile robot has a laser ground marking problem.

8. The method according to claim 1, characterized in that The step of obtaining target point cloud data collected by the mobile robot includes: Acquiring initial point cloud data collected by the mobile robot; Performing point cloud screening processing on the initial point cloud data according to the prior map used by the mobile robot to obtain screened point cloud data; The target point cloud data is determined according to the point cloud distance between the filtered point cloud data.

9. An electronic device, characterized in that: The method comprises a memory and a processor, wherein the processor is used to execute program instructions stored in the memory to implement the method according to any one of claims 1 to 8.

10. A computer-readable storage medium having program instructions stored thereon, characterized in that: When the program instructions are executed by a processor, the method according to any one of claims 1 to 8 is implemented.

Images