Robot steering wheel idling recognition method and device and electronic equipment

Abstract

The embodiment of the invention discloses a robot steering wheel idling recognition method and device and electronic equipment. The method comprises the steps that whether steering wheel idling recognition needs to be conducted or not is judged according to the current motion state of the robot; laser point clouds of the robot at different moments are obtained, the laser point clouds at the different moments are filtered, target point cloud data at the different moments are obtained, and filtering processing comprises filtering of ground point clouds, roof point clouds and dynamic obstacle point clouds; wherein the different moments comprise t1 and t2; t2 is later than t1, and the time interval between t1 and t2 is smaller than a preset time threshold value; and matching the target point clouds at different moments, calculating matching scores of the target point clouds at different moments, and determining whether the steering wheel idles or not based on the matching scores and a preset threshold value. According to the technical scheme, whether the environment information is changed or not is determined based on the point clouds at different moments, the problem of low reliability caused by detecting parameters such as the rotating speed and the torque of the steering wheel through a sensor is solved, and the steering wheel idling recognition precision is improved.

Description

Technical Field

[0001] The present invention relates to the field of robot technology, and in particular to a method, a device and an electronic device for identifying idling of a robot steering wheel. Background Art

[0002] For high-precision robots (Automated Guided Vehicle, automated guided transport vehicles or unmanned guided vehicles), it is crucial to identify the idling of the robot's steering wheel in the field of logistics and warehousing. The steering wheel is the core component of the robot's movement. If the steering wheel idling phenomenon occurs and is not identified in time, it is easy to cause the robot to lose its positioning and fail to accurately reach the designated location to complete cargo handling and storage, thus affecting the execution of the task. In addition, steering wheel idling may also increase the wear of the steering wheel, shorten its service life, and increase maintenance costs.

[0003] In the prior art, the idling identification of the robot steering wheel is performed by monitoring the rotation speed and torque and other parameters of the steering wheel through sensors. When the rotation speed increases abnormally and the torque is lower than a certain threshold, it is judged that idling may occur. However, this method has high requirements on the accuracy of the sensor. Once the sensor fails or the accuracy decreases, it may not be able to accurately identify idling, affecting the reliability of the robot. Summary of the invention

[0004] The present invention provides a method, device and equipment for identifying idling of a robot steering wheel. The method determines environmental information based on point clouds collected by a laser radar at different times, and further determines whether the steering wheel is idling in combination with the current motion state of the robot, thereby avoiding the problem of low reliability of parameters such as the rotation speed and torque of the steering wheel detected by sensors, effectively reducing the cost of identifying idling of the steering wheel, and improving the accuracy of identifying idling of the steering wheel.

[0005] According to one aspect of the present invention, a method for identifying idling of a robot steering wheel is provided, the method comprising:

[0006] Determine whether steering wheel idling identification is required based on the robot's current motion state;

[0007] Acquire laser point clouds of the robot at different times, filter the laser point clouds at different times, and obtain target point cloud data at different times, wherein the filtering process includes filtering ground point clouds, roof point clouds, and dynamic obstacle point clouds; wherein the different times include t1 and t2; t2 is later than t1, and the time interval between t1 and t2 is less than a preset time threshold;

[0008] The target point clouds at different times are matched, the matching scores of the target point clouds at different times are calculated, and whether the steering wheel idles is determined based on the matching scores and a preset threshold.

[0009] Optionally, judging whether to perform steering wheel idling detection based on the current motion state of the robot includes:

[0010] Acquiring motion data of the robot at different times, including velocity data, angular velocity data and / or acceleration data;

[0011] If the motion data meets the preset conditions, it is determined that the robot needs to perform steering wheel idling identification detection.

[0012] Optionally, after determining that the robot needs to perform steering wheel idling identification detection, the method further includes:

[0013] Constructing a grid, mapping the target point cloud at the first moment t1 into the grid to obtain a grid map at the first moment t1;

[0014] Each grid in the grid map at the first time t1 where a laser point cloud exists is determined, and a score of each grid where a laser point cloud exists is calculated according to a preset distribution function.

[0015] Optionally, calculating the grid score of each laser point cloud according to a preset distribution function includes:

[0016] Determine the central grid and the neighborhood grid of the grid map at the first time t1;

[0017] Calculating the score of each neighborhood grid relative to the central grid according to a preset distribution function;

[0018] The preset distribution function is as follows:

[0019]

[0020] Among them, S represents the score value of the neighborhood grid adjacent to the central grid Ogm(x,y) at a distance (idx_x, idx_y), x and y represent the horizontal and vertical coordinates of the central grid respectively; idx_x, idx_y represent the relative offset of the horizontal and vertical coordinates of the neighborhood grid relative to the central grid respectively, d represents the grid size, and σ is the preset spatial smoothing factor.

[0021] Optionally, after calculating the score of each neighborhood grid relative to the central grid according to a preset distribution function, the method further includes:

[0022] Acquire the target point cloud at the second time t2;

[0023] Calculate the grid coordinate set of the target point cloud at the second time t2 that falls into the grid map at the first time t1;

[0024] A score for matching the target point cloud at the second moment t2 with the target point cloud at the first moment t1 is determined according to the grid coordinate set and the corresponding grid score.

[0025] Optionally, after obtaining the target point cloud at the second time t2, the method further includes:

[0026] The target point cloud data at the second time t2 is downsampled according to the grid map resolution at the first time, to obtain a downsampled point cloud set.

[0027] Calculate the grid coordinates of the point cloud after downsampling at the second time t2 that falls into the grid map generated at the first time t1, and record it as

[0028] The formula for calculating the grid map falling into the first time t1 is as follows:

[0029]

[0030] represents the horizontal coordinate of the target point cloud at the second moment t2, represents the ordinate of the target point cloud at the second moment t2, and d represents the grid size.

[0031] According to the target grid coordinates (x i ,y i ) and the corresponding grid score, and determine the matching score between the target point cloud at the second moment t2 and the target point cloud at the first moment t1.

[0032] Optionally, based on the target grid coordinates (x i ,y i ) and the corresponding grid score, determining the matching score of the target point cloud at the second time t2 and the target point cloud at the first time t1, including:

[0033] The point cloud matching score is determined based on the following formula:

[0034]

[0035] Wherein, n is the total number of grids of the grid map generated at the second moment t2 where the target point cloud falls into at the moment t1;

[0036] If the matching score between the second moment t2 and the first moment t1 is greater than a preset threshold, it is determined that the steering wheel is idling.

[0037] Optionally, it also includes the following steps to handle false positives:

[0038] Build multi-resolution raster maps;

[0039] Perform multiple matching score calculations within a predetermined time window to determine the number of consecutive successful matches between the target point cloud at the second time t2 and the target point cloud at the first time t1;

[0040] When the number of consecutive matches exceeds a preset threshold, it is determined that the steering wheel is idling.

[0041] According to another aspect of the present invention, a robot steering wheel idling identification device is provided, comprising:

[0042] The idling identification and judgment module is used to determine whether the steering wheel idling identification is required according to the current motion state of the robot;

[0043] A laser point cloud processing module is used to obtain the laser point clouds of the robot at different times, and filter the laser point clouds at different times to obtain target point cloud data at different times, wherein the filtering process includes filtering the ground point cloud, the roof point cloud and the dynamic obstacle point cloud;

[0044] A point cloud matching calculation module, used to match the target point clouds at different times and calculate the matching scores of the target point clouds at different times;

[0045] The steering wheel idling determination module is used to determine whether the steering wheel is idling according to the target point cloud matching scores at different times combined with a preset threshold.

[0046] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising:

[0047] at least one processor; and

[0048] a memory communicatively connected to the at least one processor; wherein,

[0049] The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the robot steering wheel idling identification method described in any embodiment of the present invention.

[0050] The technical solution of the embodiment of the present invention determines whether the robot performs steering wheel idling identification detection based on the robot's motion data; obtains the robot's laser point cloud at different times, and filters the laser point cloud at different times, including filtering the ground point cloud, the roof point cloud at the top, and the dynamic obstacle point cloud, matches the target point cloud at different times, and determines whether the steering wheel is idling based on the matching score and the preset threshold. This technical solution determines whether to perform steering wheel idling identification detection based on the robot's current motion state data, filters the point cloud obtained at different times, filters out the ground, roof, and dynamic obstacle points with little feature change, ensures the reliability and accuracy of the target point cloud, performs grid point cloud matching based on the point cloud collected by the laser radar at different times to determine the matching score, and then determines whether the environment at different times has changed, avoiding the problem of low reliability caused by detecting parameters such as the rotation speed and torque of the steering wheel through sensors, effectively reducing the steering wheel idling identification cost, and improving the steering wheel idling identification accuracy.

[0051] It should be understood that the contents described in this section are not intended to identify the key or important features of the embodiments of the present invention, nor are they intended to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0053] Figure 1-1 is a flow chart of a method for identifying idling of a robot steering wheel provided according to Embodiment 1 of the present invention;

[0054] Figure 1-2 is a flow chart of a method for identifying idling of a robot steering wheel provided according to Embodiment 1 of the present invention;

[0055] Figure 2 is a flow chart of a method for identifying idling of a robot steering wheel provided according to a second embodiment of the present invention;

[0056] Figure 3 is a schematic structural diagram of a robot steering wheel idling identification device provided according to Embodiment 3 of the present invention;

[0057] Figure 4 The present invention is a schematic diagram of the structure of an electronic device for realizing a method for identifying self-idling of a robot steering wheel according to an embodiment of the present invention. DETAILED DESCRIPTION

[0058] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0059] It should be noted that the terms "first", "second", "target", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0060] Embodiment 1

[0061] Figure 1-1 This is a flow chart of a method for identifying idling of a robot steering wheel provided in the first embodiment of the present invention. This embodiment is applicable to the case of automatically identifying idling of a robot steering wheel. The method can be executed by a robot steering wheel idling identification device. The robot steering wheel idling identification device can be implemented in the form of hardware and / or software. The robot steering wheel idling identification device can be configured in an electronic device with data processing capabilities. Figure 1-1 and Figure 1-2 As shown, the method includes:

[0062] S110, judging whether it is necessary to perform steering wheel idling identification according to the current motion state of the robot.

[0063] Among them, a steering wheel is deployed in the robot, which integrates an integrated mechanical structure such as a drive motor, a steering motor, and a reducer, and integrates product, walking, traction and steering functions, and can load and haul heavier goods. The current motion state of the robot may refer to the robot's motion data including speed data, angular velocity data and / or acceleration data. Among them, the speed information can be obtained by detecting a wheel speed meter, and the angular velocity and acceleration data can be obtained by detecting an inertial sensor.

[0064] S120: If it is detected that the current speed data, angular velocity data and / or acceleration data of the robot meets a preset condition, it is determined that the robot needs to perform a steering wheel idling identification test.

[0065] The current motion data of the robot reflects the current motion state of the robot. In the specific implementation of this embodiment, if the robot speed data, angular velocity data and acceleration data detected by the sensor are all 0, it means that the robot is in a stopped state and there is no need to perform steering wheel idling identification detection; if any one of the robot's speed data, angular velocity data or acceleration data is not 0, it is determined that steering wheel idling identification detection is required.

[0066] S130, obtaining laser point clouds of the robot at different times, filtering the laser point clouds at different times, and obtaining target point cloud data at different times.

[0067] Specifically, after determining that the robot needs to perform steering wheel idling identification detection, the laser point cloud of the robot at different times is obtained, wherein the different times include a first time t1 and a second time t2; t2 is later than t1, and the time interval between t1 and t2 is less than a preset time threshold.

[0068] When the robot's motion data at the current moment meets the preset conditions, that is, the data is not 0 or close to 0, it is determined that the steering wheel idling identification detection needs to be performed. The laser point cloud data at different moments are obtained, and the time intervals between t1 and t2 at different moments can be adjusted according to the actual situation. For example, when the robot is running in a scene with relatively large environmental changes, the time interval between t1 and t2 can be selected as 150ms; for scenes with little environmental changes, such as open outdoor spaces, the time interval between t1 and t2 can be selected to be greater than 250ms. The time interval selected when implementing this technical solution is 200ms.

[0069] S140, filtering the laser point clouds at different times to obtain target point cloud data at different times, wherein the filtering includes filtering ground point clouds, roof point clouds, and dynamic obstacle point clouds.

[0070] When the robot is running in an indoor scene, the point cloud collected by the laser radar includes ground points, top laser point clouds such as roofs or ceilings, and dynamic and static obstacle point clouds. In order to improve the accuracy of point cloud perception of environmental changes, this embodiment uses the RANSAC algorithm to perform plane fitting on the ground and top of the scene. When fitting, the plane normal vector threshold can be set to ±5°, and the distance fitting plane threshold can be set to 0.1m. According to the fitted plane and the set distance threshold, the ground and top point clouds such as roofs or ceilings are filtered.

[0071] Then, the DBSCAN algorithm is used to cluster the point cloud. The point cloud search radius value can be selected according to the obstacle type in the scene. When there are no large obstacles in the scene, the search radius value can be selected from 30cm to 50cm, and the minimum number of clustering points is preferably 8 to 12 points. The point cloud of the dynamic obstacles obtained by clustering is filtered to obtain the target laser point cloud at time t1 and the target laser point cloud at time t2.

[0072] S150, rasterizing the filtered point cloud.

[0073] Specifically, the laser point cloud collected by the robot at the first moment t1 is obtained, and the target laser point cloud is obtained after filtering. Construct a grid, and map the target point cloud at the first moment t1 into the grid to obtain a grid map at the first moment t1.

[0074] Specifically, a grid is created, and the grid size can be adjusted according to the point cloud density, preferably 5cm-15cm. In this embodiment, a grid size of 10cm*10cm is constructed, and the target point cloud at the first moment t1 is rasterized. For each point cloud in the point cloud set, the grid coordinates where the point cloud falls can be calculated as follows:

[0075]

[0076] represents the horizontal coordinate of the point cloud, It represents the ordinate of the point cloud, d represents the grid size. In this embodiment, the size of d is 10 cm. The point cloud falls into the grid Ogm(x,y), and the grid coordinates are (x,y).

[0077] like Figure 1-2 As shown:

[0078] S160, determining each grid having a laser point cloud in the grid map at the first time t1, and calculating the grid score according to a preset distribution function.

[0079] Specifically, the target point cloud acquired at the first time t1 is mapped to the grid map, the grid coordinates of each point cloud in the grid map are determined, and the central grid and the neighborhood grid of the grid map at the first time t1 are determined; according to the coordinates of the point cloud and the preset grid size, the corresponding grid coordinates of each point cloud in the grid map are determined. For each grid with a point cloud, its coordinate information is recorded.

[0080] Next, determine the central grid of the grid map at the first time t1. The central grid can be determined by calculating the geometric center of the grid map or according to specific scene requirements. After the central grid is determined, its neighborhood grid is determined. The neighborhood grid can be determined according to a preset neighborhood range. For example, the grid within a certain distance range around the central grid is the neighborhood grid. The score values ​​of the central grid and the neighborhood grid of the grid are calculated using a preset distribution function.

[0081] In this embodiment, the Gaussian function is used to calculate the scores of the central grid and the neighborhood grid. The Gaussian function can effectively smooth the values ​​of the grids, so that the values ​​of the central grid and the neighborhood grid more accurately reflect the distribution of the point cloud in the area.

[0082] The preset Gaussian distribution function formula is as follows:

[0083]

[0084] Among them, S represents the score value of the neighborhood grid adjacent to the central grid Ogm(x,y) at a distance of (idx_x, idx_y), x and y represent the horizontal and vertical coordinates of the central grid respectively; idx_x, idx_y represent the relative offset of the horizontal and vertical coordinates of the neighborhood grid relative to the central grid respectively, d represents the grid size, σ is a preset spatial smoothing factor, and σ is preferably 0.3-0.7 when taking a specific value. If σ is less than 0.3, the selection range of the neighborhood grid is small, and the point cloud features are easily affected by noise; if σ is greater than 0.7, it is easy to cause the neighborhood grid range to be too large and the point cloud features are not obvious. In this embodiment, σ is taken as 0.5.

[0085] S170, after determining the neighborhood grid score, obtaining the target point cloud at the second time t2.

[0086] After the scores of the central grid and each neighborhood grid are determined by the above calculation, the target point cloud at the second time t2 is obtained; the target laser point cloud data obtained after filtering at time t2 is downsampled according to the resolution of the grid map, and the downsampled laser point cloud set is recorded as Calculate the grid coordinates of the downsampled laser point cloud that fall into the grid map generated at time t1, and the grid coordinates are recorded as

[0087] Specifically, at time t2, the acquired laser point cloud data is first filtered to remove ground point clouds, top point clouds such as roofs or ceilings, and dynamic obstacle point clouds. The filtering method is the same as the laser point cloud filtering method at time t1, and will not be repeated here.

[0088] The filtered laser point cloud data at time t2 is downsampled according to the resolution of the grid map generated at time t1. The specific method of downsampling can be random sampling, uniform sampling, or sampling according to specific rules to reduce the amount of point cloud data and maintain the point cloud features.

[0089] Calculate the grid coordinates of the downsampled point cloud data in the grid map generated at time t1. Determine the laser point cloud at the second time t2 based on the coordinates of the point cloud and the resolution of the grid map. The corresponding grid coordinate set of each point cloud in the grid map

[0090]

[0091] Calculate the laser point cloud at the second moment t2 The way to fall into the first moment t1 grid map is as follows

[0092]

[0093] represents the horizontal coordinate of the target point cloud at the second moment t2, represents the ordinate of the target point cloud at the second time t2, and d represents the size of the grid.

[0094] For point clouds that fall on the grid boundary, they can be allocated according to specific rules, such as being allocated to the nearest grid or to adjacent grids in a certain ratio.

[0095] S180, determining a score for matching the target point cloud at the second moment t2 with the target point cloud at the first moment t1.

[0096] According to the grid coordinate set and the corresponding grid score, the score of the target point cloud at the second time t2 matching the target point cloud at the first time t1 is determined. The point cloud matching score is determined based on the following formula:

[0097]

[0098] Where n is the total number of grids of the grid map generated at time t1 where the target point cloud at the second time t2 falls into;

[0099] S190: If the matching score between the second moment t2 and the first moment t1 is greater than a preset threshold, it is determined that the steering wheel is idling.

[0100] In the scheme of this embodiment, firstly, it is determined whether the robot needs to perform steering wheel idling identification detection according to the robot's motion data. When the robot's motion data meets the preset conditions, the robot starts the steering wheel idling identification detection program. The laser point cloud data at different times are collected by the laser radar sensor configured on the robot, and the collected laser point cloud is filtered to obtain the target point cloud at different times t1 and t2. The target laser point cloud at time t1 is rasterized, and the scores of each grid where the point cloud exists are calculated using the preset Gaussian distribution function; the target point cloud at the second time t2 is mapped to the grid map at time t1, and the target point cloud score at the second time t2 is calculated according to the grid score. When the target point cloud score at the second time exceeds the preset threshold, it is determined that the steering wheel is idling. The grid point cloud matching is performed based on the point cloud collected at different times by the laser radar to determine the matching score, and then determine whether the environment at different times has changed. Avoid the problem of low reliability caused by detecting parameters such as the rotation speed and torque of the steering wheel through the sensor, effectively reduce the cost of steering wheel idling identification, and improve the accuracy of steering wheel idling identification.

[0101] Embodiment 2

[0102] Figure 2 This is a flow chart of a method for identifying idling of a robot steering wheel provided in the second embodiment of the present invention. This embodiment is optimized based on the above embodiment. The specific optimization is: it also includes a misjudgment processing step.

[0103] In order to avoid the situation where the robot does not change much in the scene, such as in an open scene, the technical solution disclosed in the present invention may also include a misjudgment processing method when it is implemented. Multiple matching score calculations are performed within a predetermined time window to determine the number of consecutive successful matches; when the number of consecutive matches exceeds a preset threshold, it is determined that the steering wheel is idling. The specific steps include:

[0104] S210, construct a multi-resolution raster map.

[0105] When the technical solution is implemented, a high-resolution grid can be established in an area less than 5 meters away from the robot, and the size d of the grid can be 0.05 meters; for an area greater than 5 meters away from the robot, a low-resolution grid can be established, such as a grid size d of 0.2 meters. The calculation method of the Gaussian distribution function is consistent with the method disclosed in the first embodiment, and will not be repeated in this embodiment.

[0106] When downsampling the point cloud at the second moment t2, the grid size can also be adaptively adjusted; for point clouds with a distance less than 5 meters, the original density during laser acquisition is maintained; for points with a distance greater than or equal to 5 meters, voxel downsampling is used, and the voxel size increases with the distance.

[0107] S220, multiple laser point cloud matching score calculations within a predetermined time window

[0108] First, the length of the time window can be set according to the application scenario and requirements during specific implementation. For example, it can be a few seconds, a few minutes or even longer. The starting point and the end point of the time window are determined, and the laser point cloud is processed and analyzed within this time length. Multiple groups of time windows can also be set as needed. For example, the time window is selected to be 1.5 seconds to 2 seconds, the laser radar acquisition frequency is 20HZ, the laser radar point cloud in the environment collected by the robot is obtained, and the laser radar point cloud at different times is calculated for multiple matching scores. According to the results of the laser point cloud matching scores at different times, it is determined whether the match is successful.

[0109] S230, determining the number of consecutive successful matches of the laser point cloud at different times within the preset time window. Based on the collected point clouds at different times, perform point cloud filtering according to the point cloud processing method in the first embodiment, calculate the point cloud grid score value, calculate the laser point cloud matching value at different times, and determine whether the laser point cloud matching at different times is successful.

[0110] S240: If the number of successful matches exceeds a preset threshold, it is determined that the steering wheel of the current robot is idling; otherwise, it is determined that the steering wheel of the current robot is not idling.

[0111] Within the predetermined time window, the number of consecutive successful matches is recorded. If the number of consecutive successful matches of the laser point cloud exceeds the preset threshold, it can be determined that the steering wheel has idling. The preset threshold setting can be adjusted according to the type of robot, operating environment, etc. to ensure that the steering wheel idling can be accurately identified in different situations.

[0112] The technical solution of the embodiment of the present invention adds a misjudgment processing step on the basis of the first embodiment. A multi-resolution grid is adaptively constructed according to the environmental characteristics, a high-resolution grid is used for the area close to the robot, and a low-resolution grid is used for the area far from the robot, so as to increase the reliability of the point cloud and prevent the misjudgment of the steering wheel idling.

[0113] Perform multiple laser point cloud continuous matching score calculations within a predetermined time window to determine the number of continuous matching successes; when the number of continuous matching times exceeds a preset threshold, determine that the steering wheel is idling. Avoid misjudgment of steering wheel idling due to relying solely on a single laser point cloud match. After adding the misjudgment processing step, by performing multiple matching score calculations within a predetermined time window, it is possible to make judgments based on data from multiple time points. Only when the number of continuous matching successes exceeds a preset threshold is it determined that the steering wheel is idling, which greatly reduces the probability of misjudgment caused by accidental factors, thereby significantly improving the accuracy of steering wheel idling identification. The steering wheel idling can still be accurately identified in complex environments, enhancing the reliability and accuracy of steering wheel idling identification, improving the stability of the detection system, and extending the service life of the robot.

[0114] Embodiment 3

[0115] Figure 3 This is a schematic diagram of the structure of a robot steering wheel idling identification device provided in the third embodiment of the present invention. The device can execute the robot steering wheel idling identification method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method. Figure 3 As shown, the device comprises:

[0116] The idling identification and judgment module 310 is used to determine whether the steering wheel idling identification is required according to the current motion state of the robot;

[0117] The laser point cloud processing module 320 is used to obtain the laser point clouds of the robot at different times, and filter the laser point clouds at different times to obtain target point cloud data at different times, wherein the filtering process includes filtering the ground point cloud, the roof point cloud and the dynamic obstacle point cloud; wherein the different times include a first time t1 and a second time t2; t2 is later than t1, and the time interval between t1 and t2 is less than a preset time threshold;

[0118] A point cloud matching calculation module 330 is used to match the target point clouds at different times and calculate the matching scores of the target point clouds at different times;

[0119] The steering wheel idling determination module 340 is used to determine whether the steering wheel is idling according to the target point cloud matching scores at different times and a preset threshold.

[0120] Optionally, the idling identification and judgment module 310 is further used to:

[0121] Acquire motion data of the robot at different times, including velocity data, angular velocity data and / or acceleration data;

[0122] If the motion data meets the preset conditions, it is determined that the robot needs to perform steering wheel idling identification detection.

[0123] Optionally, the laser point cloud processing module 320 is further used for:

[0124] Constructing a grid, mapping the target point cloud at the first moment t1 into the grid to obtain a grid map at the first moment t1;

[0125] Each grid in the grid map at the first time t1 where a laser point cloud exists is determined, and the grid score is calculated according to a preset distribution function.

[0126] Optionally, the device further comprises:

[0127] A grid score determination module, used to determine the central grid and the neighborhood grid of the grid map at the first time t1;

[0128] Calculating the score of each neighborhood grid relative to the central grid according to a preset distribution function;

[0129] The preset distribution function is as follows:

[0130]

[0131] Among them, S represents the score value of the neighborhood grid adjacent to the central grid Ogm(x,y) at a distance (idx_x, idx_y), x and y represent the horizontal and vertical coordinates of the central grid respectively; idx_x, idx_y represent the relative offset of the horizontal and vertical coordinates of the neighborhood grid relative to the central grid respectively, d represents the grid size, and σ is the preset spatial smoothing factor.

[0132] A point cloud matching score determination module, used to obtain the target point cloud at the second time t2;

[0133] Calculate the grid coordinate set of the target point cloud at the second time t2 that falls into the grid map at the first time t1

[0134] A score for matching the target point cloud at the second moment t2 with the target point cloud at the first moment t1 is determined according to the grid coordinate set and the corresponding grid score.

[0135] Optionally, the point cloud matching score determination module is further used to:

[0136] The target point cloud data at the second time t2 is downsampled according to the grid map resolution at the first time, to obtain a downsampled point cloud set.

[0137] Calculate the grid coordinates of the point cloud after downsampling at the second time t2 that falls into the grid map generated at the first time t1, and record it as

[0138] The formula for calculating the grid map falling into the first time t1 is as follows:

[0139]

[0140] represents the horizontal coordinate of the target point cloud at the second moment t2, represents the ordinate of the target point cloud at the second time t2, and d represents the resolution of the grid.

[0141] According to the target grid coordinates (x i ,y i) and the corresponding grid score, and determine the matching score between the target point cloud at the second moment t2 and the target point cloud at the first moment t1.

[0142] Optionally, the point cloud matching score determination module further includes:

[0143] The point cloud matching score is determined based on the following method:

[0144]

[0145] Wherein, n is the total number of grids of the grid map generated at time t1 that the target point cloud at the second time t2 falls into.

[0146] If the matching score between the second moment t2 and the first moment t1 is greater than a preset threshold, it is determined that the steering wheel is idling.

[0147] Optionally, the device further includes a misjudgment processing module, configured to:

[0148] Perform multiple matching score calculations within a predetermined time window to determine the number of consecutive successful matches;

[0149] When the number of consecutive matches exceeds a preset threshold, it is determined that the steering wheel is idling.

[0150] A robot steering wheel idling identification device provided in an embodiment of the present invention can execute a robot steering wheel idling identification method provided in any embodiment of the present invention, and has functional modules and beneficial effects corresponding to the execution method.

[0151] Embodiment 4

[0152] Figure 4 A schematic diagram of the structure of an electronic device 10 that can be used to implement an embodiment of the present invention is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or required herein.

[0153] like Figure 4As shown, the electronic device 10 includes at least one processor 11, and a memory connected to the at least one processor 11, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc., wherein the memory stores a computer program that can be executed by at least one processor, and the processor 11 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 12 or the computer program loaded from the storage unit 18 to the random access memory (RAM) 13. In the RAM 13, various programs and data required for the operation of the electronic device 10 can also be stored. The processor 11, the ROM 12, and the RAM 13 are connected to each other through a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0154] A number of components in the electronic device 10 are connected to the I / O interface 15, including: an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a disk, an optical disk, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0155] The processor 11 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The processor 11 executes the various methods and processes described above, such as a robot steering wheel idling identification method.

[0156] In some embodiments, the robot steering wheel idling identification method can be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as a storage unit 18. In some embodiments, part or all of the computer program can be loaded and / or installed on the electronic device 10 via the ROM 12 and / or the communication unit 19. When the computer program is loaded into the RAM 13 and executed by the processor 11, one or more steps of the robot steering wheel idling identification method described above can be performed. Alternatively, in other embodiments, the processor 11 can be configured to execute the robot steering wheel idling identification method in any other appropriate manner (for example, by means of firmware).

[0157] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0158] Computer programs for implementing the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that when the computer program is executed by the processor, the functions / operations specified in the flow chart and / or block diagram are implemented. The computer program may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0159] In the context of the present invention, a computer-readable storage medium may be a tangible medium that may contain or store a computer program for use by or in combination with an instruction execution system, device or equipment. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices or equipment, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0160] To provide interaction with a user, the systems and techniques described herein may be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or trackball) through which the user can provide input to the electronic device. Other types of devices may also be used to provide interaction with the user; for example, the feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form (including acoustic input, voice input, or tactile input).

[0161] The systems and techniques described herein may be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0162] A computing system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The client and server relationship is generated by computer programs running on the corresponding computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services.

[0163] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution of the present invention can be achieved, and this document does not limit this.

[0164] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for identifying idling of a robot steering wheel, characterized in that: The method comprises: Determine whether steering wheel idling detection is required based on the robot's current motion state; Acquire laser point clouds of the robot at different times, filter the laser point clouds at different times, and obtain target point cloud data at different times, wherein the filtering process includes filtering ground point clouds, roof point clouds, and dynamic obstacle point clouds; wherein the different times include a first time t1 and a second time t2; t2 is later than t1, and the time interval between t1 and t2 is less than a preset time threshold; The target point clouds at different times are matched, the matching scores of the target point clouds at different times are calculated, and whether the steering wheel idles is determined based on the matching scores and a preset threshold.

2. The method according to claim 1, characterized in that Determine whether steering wheel idling detection is required based on the robot's current motion state, including: Acquiring motion data of the robot at different times, including velocity data, angular velocity data and / or acceleration data; If the motion data meets the preset conditions, it is determined that the robot needs to perform steering wheel idling identification detection.

3. The method according to claim 2, characterized in that After determining that the robot needs to perform steering wheel idling identification detection, the method further includes: Constructing a grid, mapping the target point cloud at the first moment t1 into the grid to obtain a grid map at the first moment t1; Each grid in the grid map at the first time t1 where a laser point cloud exists is determined, and a score of each grid where a laser point cloud exists is calculated according to a preset distribution function.

4. The method according to claim 3, characterized in that Calculating the grid score of each laser point cloud according to a preset distribution function includes: Determine the central grid and the neighborhood grid of the grid map at the first time t1; Calculating the score of each neighborhood grid relative to the central grid according to a preset distribution function; The preset distribution function is as follows: Among them, S represents the score value of the neighborhood grid adjacent to the central grid Ogm(x,y) at a distance (idx_x, idx_y), x and y represent the horizontal and vertical coordinates of the central grid respectively; idx_x, idx_y represent the relative offset of the horizontal and vertical coordinates of the neighborhood grid relative to the central grid respectively, d represents the grid size, and σ is the preset spatial smoothing factor.

5. The method according to claim 4, characterized in that After calculating the score of each neighborhood grid relative to the central grid according to the preset distribution function, the method further includes: Acquire the target point cloud at the second time t2; Calculate the grid coordinate set of the target point cloud at the second time t2 that falls into the grid map at the first time t1; A score for matching the target point cloud at the second moment t2 with the target point cloud at the first moment t1 is determined according to the grid coordinate set and the corresponding grid score.

6. The method according to claim 5, characterized in that After obtaining the target point cloud at the second time t2, the method further includes: The target point cloud data at the second time t2 is downsampled according to the grid map resolution at the first time, to obtain a downsampled point cloud set. Calculate the grid coordinates of the point cloud after downsampling at the second time t2 that falls into the grid map generated at the first time t1, and record it as The grid coordinates falling into the grid map at the first time t1 are determined as follows: represents the horizontal coordinate of the target point cloud at the second moment t2, represents the ordinate of the target point cloud at the second moment t2, and d represents the grid size. According to the target grid coordinates (x i ,y i ) and the corresponding grid score, and determine the matching score between the target point cloud at the second moment t2 and the target point cloud at the first moment t1.

7. The method according to claim 6, characterized in that According to the target grid coordinates (x i ,y i ) and the corresponding grid score, determining the matching score of the target point cloud at the second time t2 and the target point cloud at the first time t1, including: The point cloud matching score is determined based on the following method: Wherein, n is the total number of grids of the grid map generated at time t1 that the target point cloud at the second time t2 falls into; If the matching score between the second moment t2 and the first moment t1 is greater than a preset threshold, it is determined that the steering wheel is idling.

8. The method according to any one of claims 1 to 7, characterized in that: It also includes the following steps to handle false positives: Build multi-resolution raster maps; Perform multiple matching score calculations within a predetermined time window to determine the number of consecutive successful matches between the target point cloud at the second time t2 and the target point cloud at the first time t1; When the number of consecutive matches exceeds a preset threshold, it is determined that the steering wheel is idling.

9. A robot steering wheel idling identification device, characterized in that: The device comprises: The idling identification and judgment module is used to determine whether the steering wheel idling identification is required according to the current motion state of the robot; A laser point cloud processing module is used to obtain laser point clouds of the robot at different times, and filter the laser point clouds at different times to obtain target point cloud data at different times, wherein the filtering process includes filtering ground point clouds, roof point clouds, and dynamic obstacle point clouds; wherein the different times include a first time t1 and a second time t2; t2 is later than t1, and the time interval between t1 and t2 is less than a preset time threshold; A point cloud matching calculation module, used to match the target point clouds at different times and calculate the matching scores of the target point clouds at different times; The steering wheel idling determination module is used to determine whether the steering wheel is idling based on the matching score and a preset threshold.

10. An electronic device, characterized in that: The electronic device comprises: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the robot steering wheel idling identification method according to any one of claims 1-8.

Images