Base station attitude detection method, device and system and storage medium

Abstract

The invention relates to the field of robots, and discloses a base station attitude detection method, device and system and a storage medium, which are used for improving the detection accuracy. The base station attitude detection method comprises the following steps: scanning a surrounding environment through a laser radar of a mobile robot to obtain environment radar data, extracting a base station radar point cloud and an identification radar point cloud corresponding to a base station from the environment radar data, the intensity of the identification radar point cloud being greater than the intensity of the base station radar point cloud, and when the distance between the fitting arc midpoint of the base station radar point cloud and the fitting arc midpoint of the identification radar point cloud is smaller than a preset distance, determining the attitude of the base station according to the base station radar point cloud or the identification radar point cloud.

Description

Technical Field

[0001] The present invention relates to the field of robots, and in particular to a method, device, system and storage medium for detecting a posture of a base station. Background Art

[0002] In the current market, mobile robots are usually used in conjunction with base stations. For example, a cleaning robot with sweeping and mopping functions is equipped with a base station for replenishing water, charging, washing mops, and other functions for the cleaning robot. The area near the ground of the base station is provided with an entrance allowing the cleaning robot to pass through, and the bottom of the base station is provided with an inner cavity allowing all or part of the cleaning robot to be parked. When the cleaning robot needs to be used in conjunction with the base station, the cleaning robot needs to plan a route to return to the base station to avoid collisions with the entrance or failure to fully enter the inner cavity of the base station.

[0003] In the prior art, the position of the base station and the inner cavity deflection angle, i.e., the base station posture, are usually calculated based on the base station radar point cloud, and the mobile robot plans the route back to the base station based on the base station posture. However, the material or color of the base station may cause the radar point cloud received by the mobile robot to be missing, and the base station cannot be accurately identified. Therefore, a strong reflective material is attached to the inner cavity surface of the base station to help identify the base station, and then the base station posture is determined based on the radar point cloud corresponding to the base station.

[0004] Since there are components distributed on the inner cavity of the base station that need to be set at different positions, such as fans or transparent windows, there are large faults in the base station radar point cloud obtained by the mobile robot. In addition, the vibration and tilt of the cleaning robot when moving will cause the position of the radar point cloud to jump, and it cannot form a complete arc corresponding to the inner cavity. In this case, the accurate base station posture cannot be obtained based on the base station radar point cloud. Summary of the invention

[0005] The main purpose of the present invention is to provide a method, device, system and storage medium for detecting the posture of a base station, so as to solve the problem that the accurate posture of the base station cannot be obtained due to faults or jumps in the radar point cloud, and to improve the accuracy of detection.

[0006] A first aspect of the present invention provides a method for detecting a base station posture, comprising: scanning the surrounding environment with a laser radar of the mobile robot to obtain environmental radar data; extracting a base station radar point cloud and an identification radar point cloud corresponding to the base station from the environmental radar data, wherein the intensity of the identification radar point cloud is greater than the intensity of the base station radar point cloud; when the distance between the midpoint of the fitted arc of the base station radar point cloud and the midpoint of the fitted arc of the identification radar point cloud is less than a preset distance, determining the base station posture according to the base station radar point cloud or the identification radar point cloud.

[0007] In a feasible implementation manner, the extracting the base station radar point cloud and identification radar point cloud corresponding to the base station from the environmental radar data includes: clustering and curve fitting the environmental radar data, and determining the fitting circle with a radius within a preset radius in the generated fitting circle as the initial base station fitting circle; performing outward diffusion clustering based on the points corresponding to the initial base station fitting circle to obtain discrete points within a preset radius around the initial base station fitting circle; aggregating the points corresponding to the initial base station fitting circle and the discrete points to obtain the base station radar point cloud; extracting a point cloud with an intensity higher than the preset intensity from the environmental radar point cloud to obtain an initial identification point cloud; clustering the initial identification point cloud, and determining the point cloud obtained after clustering as the identification radar point cloud.

[0008] In a feasible implementation, when the distance between the midpoint of the fitted arc of the base station radar point cloud and the midpoint of the fitted arc of the identification radar point cloud is less than a preset distance, determining the base station posture according to the base station radar point cloud or the identification radar point cloud, including: when the distance between the midpoint of the fitted arc of the base station radar point cloud and the midpoint of the fitted arc of the identification radar point cloud is less than a preset distance, determining whether the point corresponding to the entrance of the base station is missing; if so, determining the base station posture according to the identification radar point cloud; if not, determining the base station posture according to the base station radar point cloud.

[0009] In a feasible implementation manner, when the distance between the midpoint of the fitting arc of the base station radar point cloud and the midpoint of the fitting arc of the identification radar point cloud is less than a preset distance, determining whether the point corresponding to the entrance of the base station is missing includes: when the distance between the midpoint of the fitting arc of the base station radar point cloud and the midpoint of the fitting arc of the identification radar point cloud is less than a preset distance, determining whether the number of points in the identification radar point cloud is greater than a preset number; if so, generating a base station fitting arc according to the base station radar point cloud; determining the points at both ends of the base station fitting arc as a first endpoint and a second endpoint; calculating the arc length of the base station fitting arc according to the posture information of the first endpoint and the second endpoint, the posture information including the position and angle relative to the mobile robot; if the arc length of the base station fitting arc is within the preset arc length range, determining that the point corresponding to the entrance of the base station is not missing; if the arc length of the base station fitting arc is not within the preset arc length range, determining that the point corresponding to the entrance of the base station is missing.

[0010] In a feasible implementation, if so, the base station posture is determined according to the identification radar point cloud, including: if the point corresponding to the entrance of the base station is missing, generating an identification fitting arc according to the identification radar point cloud; selecting the midpoint of the identification fitting arc to obtain a identification reference point; determining the direction in which the identification reference point points to the center of the identification fitting arc as the identification direction; obtaining a preset deflection direction and deflection angle, wherein the deflection direction and the deflection angle are related to the reflection intensity distribution of the scanned area in the base station; rotating the identification direction toward the deflection direction by the deflection angle, and determining the deflected identification direction and the position of the identification fitting arc as the base station posture.

[0011] In a feasible implementation, if not, determining the base station posture according to the base station radar point cloud, including: if the points corresponding to the entrance of the base station are not missing, generating a base station fitting arc according to the base station radar point cloud; selecting the midpoint of the base station fitting arc to obtain a base station reference point; determining the direction of the base station reference point pointing to the center of the base station fitting arc as the base station direction; determining the base station direction and the position of the base station fitting arc as the base station posture.

[0012] In a feasible implementation, when the distance between the base station radar point cloud and the identification radar point cloud is less than a preset distance, after determining the base station posture according to the base station radar point cloud or the identification radar point cloud, it also includes: extracting the target arc and target direction of the base station posture, the target direction being the direction from the midpoint of the target arc to the center of the target arc; determining the target return point according to the midpoint of the target arc, the target direction and the preset target distance; planning the route of the mobile robot from the current position to the target return point; when the mobile robot moves to the target return point, adjusting the forward direction of the mobile robot so that the forward direction of the mobile robot is opposite to the target direction; controlling the mobile robot to move toward the base station until the mobile robot moves to a preset position within the base station.

[0013] The second aspect of the present invention provides a detection device for a base station posture, comprising: a scanning module, used to scan the surrounding environment through the laser radar of the mobile robot to obtain environmental radar data; an extraction module, used to extract the base station radar point cloud and the identification radar point cloud corresponding to the base station from the environmental radar data, the intensity of the identification radar point cloud being greater than the intensity of the base station radar point cloud; a determination module, used to determine the base station posture according to the base station radar point cloud or the identification radar point cloud when the distance between the midpoint of the fitted arc of the base station radar point cloud and the midpoint of the fitted arc of the identification radar point cloud is less than a preset distance.

[0014] In a feasible implementation, the extraction module is specifically used to: cluster and curve fit the environmental radar data, and determine the fitting circle with a radius within a preset radius in the generated fitting circle as the initial base station fitting circle; perform outward diffusion clustering based on the points corresponding to the initial base station fitting circle to obtain discrete points within a preset radius around the initial base station fitting circle; summarize the points corresponding to the initial base station fitting circle and the discrete points to obtain a base station radar point cloud; extract point clouds with intensities higher than the preset intensity from the environmental radar point cloud to obtain an initial identification point cloud; cluster the initial identification point cloud, and determine the point cloud obtained after clustering as the identification radar point cloud.

[0015] In a feasible implementation, the determination module includes: a judgment unit, used to determine whether the point corresponding to the entrance of the base station is missing when the distance between the midpoint of the fitted arc of the base station radar point cloud and the midpoint of the fitted arc of the identification radar point cloud is less than a preset distance; a first determination unit, used to determine the base station posture according to the identification radar point cloud if so; and a second determination unit, used to determine the base station posture according to the base station radar point cloud if not.

[0016] In a feasible implementation manner, the judgment unit is specifically used to: when the distance between the midpoint of the fitting arc of the base station radar point cloud and the midpoint of the fitting arc of the identification radar point cloud is less than a preset distance, determine whether the number of midpoints in the identification radar point cloud is greater than a preset number; if so, generate a base station fitting arc based on the base station radar point cloud; determine the points at both ends of the base station fitting arc as the first endpoint and the second endpoint; calculate the arc length of the base station fitting arc based on the posture information of the first endpoint and the second endpoint, and the posture information includes the position and angle relative to the mobile robot; if the arc length of the base station fitting arc is within the preset arc length range, determine that the point corresponding to the entrance of the base station is not missing; if the arc length of the base station fitting arc is not within the preset arc length range, determine that the point corresponding to the entrance of the base station is missing.

[0017] In a feasible implementation manner, the first determination unit is specifically used to: if the point corresponding to the entrance of the base station is missing, generate a marker fitting arc according to the marker radar point cloud; select the midpoint of the marker fitting arc to obtain a marker reference point; determine the direction in which the marker reference point points to the center of the marker fitting arc as the marker direction; obtain a preset deflection direction and deflection angle, wherein the deflection direction and the deflection angle are related to the reflection intensity distribution of the scanned area in the base station; rotate the marker direction toward the deflection direction by the deflection angle, and determine the deflected marker direction and the position of the marker fitting arc as the base station posture.

[0018] In a feasible implementation manner, the second determination unit is specifically used to: if the point corresponding to the entrance of the base station is not missing, generate a base station fitting arc according to the base station radar point cloud; select the midpoint of the base station fitting arc to obtain a base station reference point; determine the direction of the base station reference point pointing to the center of the base station fitting arc as the base station direction; determine the base station direction and the position of the base station fitting arc as the base station posture.

[0019] In a feasible embodiment, the detection device of the base station posture also includes a moving module, which is specifically used to: extract the target arc and target direction of the base station posture, the target direction being the direction from the midpoint of the target arc to the center of the target arc; determine the target return point according to the midpoint of the target arc, the target direction and a preset target distance; plan the route of the mobile robot from the current position to the target return point; when the mobile robot moves to the target return point, adjust the forward direction of the mobile robot so that the forward direction of the mobile robot is opposite to the target direction; control the mobile robot to move toward the base station until the mobile robot moves to a preset position within the base station.

[0020] The third aspect of the present invention provides a base station posture detection system, comprising: a mobile robot and a base station, the mobile robot comprising a memory and at least one processor, the memory storing instructions; the at least one processor calls the instructions in the memory so that the base station posture detection system executes the above-mentioned base station posture detection method.

[0021] A fourth aspect of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores instructions, which, when executed on a computer, enable the computer to execute the above-mentioned method for detecting the posture of a base station.

[0022] In the technical solution provided by the present invention, the surrounding environment is scanned by the laser radar of the mobile robot to obtain environmental radar data, and the base station radar point cloud and the identification radar point cloud corresponding to the base station are extracted from the environmental radar data. The intensity of the identification radar point cloud is greater than the intensity of the base station radar point cloud. When the distance between the midpoint of the fitted arc of the base station radar point cloud and the midpoint of the fitted arc of the identification radar point cloud is less than a preset distance, the base station posture is determined according to the base station radar point cloud or the identification radar point cloud. In an embodiment of the present invention, the base station radar point cloud and the identification radar point cloud are extracted from the scanned environmental radar data. When the distance between the base station radar point cloud and the identification radar point cloud is less than a preset distance, it means that the base station radar point cloud and the identification radar point cloud are valid. At this time, the base station radar point cloud or the identification radar point cloud is selected according to the integrity of the point cloud to detect the base station posture, thereby improving the accuracy of the detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 Schematic diagram of an embodiment of a base station posture detection system in an embodiment of the present invention;

[0024] Figure 2 Schematic diagram of an embodiment of a method for detecting a base station posture in an embodiment of the present invention;

[0025] Figure 3 is environmental radar data in an embodiment of the present invention;

[0026] Figure 4 Schematic diagram of another embodiment of a method for detecting a posture of a base station in an embodiment of the present invention;

[0027] Figure 5 A schematic diagram of the marking direction for marking a fitting arc in an embodiment of the present invention;

[0028] Figure 6 is a schematic diagram of the inner cavity of a base station in an embodiment of the present invention;

[0029] Figure 7 is a schematic diagram of a deflection angle and a deflection direction in an embodiment of the present invention;

[0030] Figure 8 A schematic diagram of a target return point in an embodiment of the present invention;

[0031] Fig. 9 Schematic diagram of an embodiment of a device for detecting a posture of a base station in an embodiment of the present invention;

[0032] Fig.10 Schematic diagram of another embodiment of a device for detecting a base station posture in an embodiment of the present invention. DETAILED DESCRIPTION

[0033] The embodiments of the present invention provide a method, device, system and storage medium for detecting the posture of a base station, which are used to improve the accuracy of detection.

[0034] The terms "first", "second", "third", "fourth", etc. (if any) 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 described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "including" or "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 that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0035] In order to facilitate the introduction of the detection method of the base station posture, the detection system of the base station posture is first described. Figure 1 As shown, the detection system 100 of the base station posture includes a mobile robot 110 and a base station 120, and the mobile robot 110 and the base station 120 can be used in combination. The outer contour of the top view of the mobile robot 110 is circular, and a laser radar is set on the top. The base station 120 is provided with an opening near the ground, and the opening allows the mobile robot to enter the inner cavity of the base station 120. A fixed position is set in the inner cavity for the mobile robot to park. The cross-section of the top view of the inner cavity is an arc, and a strong reflective material is pasted on the surface of the inner cavity. When the laser radar scans the base station 120, the area with the strong reflective material pasted has a greater signal intensity than the area without the strong reflective material pasted. The mobile robot 110 includes a memory 111 and at least one processor 112. The memory 111 stores instructions, and the mobile robot can call the instructions in the memory 111 through the processor 112 to execute.

[0036] The most common supporting usage scenario for mobile robots and base stations is that the charging head on the mobile robot is connected to the charging base on the base station to charge the mobile robot. In addition, on robots in some special usage scenarios, the base station also has corresponding additional functions. For example, the base station that is compatible with the mopping robot is equipped with a water tank and a sink for the mopping robot to clean the mop. Therefore, the mobile robot needs to recognize the accurate posture of the base station, which includes the position and orientation, to avoid the situation where the supporting functions cannot be used normally due to the inability to smoothly enter the fixed position set in the base station.

[0037] Those skilled in the art will understand that Figure 1 The base station posture detection system shown does not constitute a limitation on the base station posture detection system, and may include more or fewer components than shown in the figure, or different component arrangements.

[0038] For ease of understanding, the specific process of the embodiment of the present invention is described below. It can be understood that the execution subject of the present invention can be a detection device for the posture of a base station, or a mobile robot, which is not limited here. The embodiment of the present invention is described by taking a mobile robot as the execution subject as an example.

[0039] See also Figure 2 , an embodiment of a method for detecting a base station posture in an embodiment of the present invention includes:

[0040] 201. Scan the surrounding environment through the laser radar of the mobile robot to obtain environmental radar data;

[0041] The mobile robot transmits laser signals by rotating 360° through the laser radar on the top, obtains the laser signals reflected by surrounding obstacles, and obtains a frame of environmental radar data. The environmental radar data is a radar point cloud, such as Figure 3 As shown, each point contains intensity information, position information and angle information relative to the mobile robot. The laser reflected by the strong reflective material generates a black point cloud, and the laser reflected by the area other than the strong reflective material generates a gray point cloud. The intensity of the black point cloud is greater than the intensity of the gray point cloud. The mobile robot determines whether there is a point cloud generated by reflection of the reflective material in the environmental radar data. If so, it is determined that the obtained environmental radar data is valid. If not, the environmental radar data is re-acquired.

[0042] The way to re-acquire environmental radar data is as follows:

[0043] The mobile robot obtains a preset semantic map, reads the position of the base station in the preset semantic map and its own position. If the distance between the mobile robot and the base station is greater than the set value, the mobile robot is controlled to move a first distance toward the base station, and then moves a second distance clockwise or counterclockwise with the base station as the origin, and then obtains a frame of environmental radar data through the lidar, and determines whether the obtained environmental radar data is valid. If it is still invalid, it continues to move a second distance in the same direction as the last time, and then obtains a frame of environmental radar data through the lidar and determines the validity of the environmental radar data until the obtained environmental radar data is valid.

[0044] 202. Extracting a base station radar point cloud and an identification radar point cloud corresponding to the base station from the environmental radar data, wherein the intensity of the identification radar point cloud is greater than the intensity of the base station radar point cloud;

[0045] The environmental radar data is the radar point cloud. The mobile robot clusters and curve fits the radar point cloud to generate multiple fitting circles, from which the fitting circle with a radius matching the base station inner cavity radius is selected as the base station fitting circle. To avoid inaccurate clustering caused by point cloud faults, the mobile robot searches for points within a preset radius near the base station fitting circle, summarizes the searched points and the points corresponding to the base station fitting circle to obtain the base station radar point cloud, summarizes and clusters the points in the environmental radar point cloud with intensities greater than the preset intensity to obtain the identification radar point cloud, which is a point cloud obtained by laser reflection from a strong reflective material.

[0046] 203. When the distance between the midpoint of the fitted arc of the base station radar point cloud and the midpoint of the fitted arc of the identification radar point cloud is less than a preset distance, determine the base station posture according to the base station radar point cloud or the identification radar point cloud.

[0047] In order to filter out the erroneous data that appear during the scanning process, the mobile robot calculates the distance between the midpoint of the fitted arc of the base station radar point cloud and the midpoint of the fitted arc of the identification radar point cloud. When the distance is less than the preset distance, the mobile robot determines that the base station radar point cloud and the identification radar point cloud data are valid. At this time, the integrity of the base station radar point cloud is detected. The integrity means that the point clouds on both sides of the base station opening are not missing. When the base station radar point cloud is complete, the base station posture is determined according to the base station radar point cloud. When the base station radar point cloud is incomplete, the base station posture is determined according to the identification radar point cloud.

[0048] In an embodiment of the present invention, a base station radar point cloud and a marker radar point cloud are extracted from the scanned environmental radar data. When the distance between the base station radar point cloud and the marker radar point cloud is less than a preset distance, it means that the base station radar point cloud and the marker radar point cloud are valid. At this time, the base station radar point cloud or the marker radar point cloud is selected according to the integrity of the point cloud to detect the base station posture, thereby improving the accuracy of the detection.

[0049] See also Figure 4 Another embodiment of the method for detecting the posture of a base station in the embodiment of the present invention includes:

[0050] 401. Scan the surrounding environment through the laser radar of the mobile robot to obtain environmental radar data;

[0051] In this embodiment, step 401 is the same as step 201 and will not be described in detail here.

[0052] 402. Extracting a base station radar point cloud and an identification radar point cloud corresponding to the base station from the environmental radar data, wherein the intensity of the identification radar point cloud is greater than the intensity of the base station radar point cloud;

[0053] The mobile robot clusters and curve fits the environmental radar data, and determines the fitting circle with a radius within a preset radius range in the generated fitting circle as the initial base station fitting circle, and the preset radius range is set based on the base station inner cavity.

[0054] For example, the radius of the arc formed on the inner surface of the base station is 16 cm. An error range of 1 cm is added to 16 cm, that is, 15 cm-17 cm is determined as the preset radius range. The mobile robot determines the fitting circle with a radius of 15 cm-17 cm in the fitting circle as the initial base station fitting circle.

[0055] If there are two or more fitting circles with a radius within the preset radius range, the mobile robot calculates the number of points with an intensity greater than the preset intensity in the coverage area of ​​each of the two or more fitting circles, and determines the fitting circle with the largest number as the initial base station fitting circle.

[0056] To avoid inaccurate clustering caused by point cloud faults, the mobile robot searches for discrete points within a preset radius based on the center of the initial base station fitting circle, and summarizes the points corresponding to the initial base station fitting circle and the discrete points to obtain the base station radar point cloud. The discrete points include fault points and outliers around the initial base station fitting circle.

[0057] Point clouds with intensities higher than a preset intensity are extracted from the environmental radar point cloud to obtain an initial identification point cloud, the initial identification point cloud is clustered, and the point cloud obtained after clustering is determined as the identification radar point cloud.

[0058] Furthermore, due to the limited width of the strongly reflective material, the number of acquired identification radar point clouds is small and relatively dense. In order to avoid misjudgment caused by noise, the points corresponding to the class with fewer points after clustering are judged as noise. The class with a point number greater than the set threshold is selected to filter out interference. The points corresponding to the class with a point number greater than the set threshold are summarized to obtain a purer identification radar point cloud, thereby improving the accuracy of obtaining the identification radar point cloud.

[0059] Furthermore, the mobile robot determines the current position as the basic position, rotates clockwise at a preset angle based on the center of the initial base station fitting circle at the basic position to reach the first position, obtains the first environmental radar data at the first position, rotates counterclockwise at a preset angle based on the center of the initial base station fitting circle at the basic position to reach the second position, obtains the second environmental radar data at the second position, splices the first environmental radar data, the second environmental radar data and the point cloud corresponding to the initial base station fitting circle to obtain a spliced ​​radar point cloud, re-clustering and curve fitting the spliced ​​radar point cloud, selects a secondary base station fitting circle with a radius within a preset radius, and determines the secondary base station fitting circle as a new initial base station fitting circle. When there are two or more secondary base station fitting circles, the number of points with an intensity greater than a preset intensity in the coverage area of ​​each secondary base station fitting circle in the two or more secondary base station fitting circles is calculated, and the secondary base station fitting circle with the largest number value is determined as the new initial base station fitting circle.

[0060] The base station is scanned by moving left and right to avoid the problem of blind spots in the inner cavity caused by the mobile robot's offset position, ensuring the comprehensiveness of the scan, increasing the amount of radar data, and helping to improve the accuracy of detection.

[0061] 403. When the distance between the midpoint of the fitted arc of the base station radar point cloud and the midpoint of the fitted arc of the marker radar point cloud is less than a preset distance, determine whether the point corresponding to the entrance of the base station is missing;

[0062] In order to filter out the erroneous data that appear during the scanning process, the mobile robot calculates the distance between the midpoint of the fitting arc of the base station radar point cloud and the midpoint of the fitting arc of the identification radar point cloud. When the distance between the base station radar point cloud and the identification radar point cloud is less than the preset distance, the mobile robot determines that the base station radar point cloud and the identification radar point cloud data are valid. At this time, the mobile robot determines whether the number of midpoints in the identification radar point cloud is greater than the preset number. If so, a base station fitting arc is generated according to the base station radar point cloud. If not, it is confirmed that the number of point clouds obtained from the strong reflection area is insufficient. The main reason is that the mobile robot is too far away from the base station, the laser radar accuracy is limited, the width of the reflective material is limited by the inner cavity structure of the base station, and the scanned width is small. The corresponding number of base station radar point clouds is also small. In order to increase the accuracy of the calculation, the mobile robot moves the first distance in the direction close to the base station and re-executes steps 401-403.

[0063] The mobile robot determines the points at both ends of the generated base station fitting arc as the first endpoint and the second endpoint, and calculates the arc length of the base station fitting arc according to the posture information of the first endpoint and the second endpoint. The posture information includes the position and angle relative to the mobile robot. If the arc length of the base station fitting arc is within the preset arc length range, the mobile robot determines that the points corresponding to the entrance of the base station are not missing. If the arc length of the base station fitting arc is not within the preset arc length range, the mobile robot determines that the points corresponding to the entrance of the base station are missing. The preset arc length range is obtained based on the actual base station inner cavity parameters, specifically by obtaining the actual arc length of the arc corresponding to the inner cavity cross section, and adding a tolerance based on the actual arc length.

[0064] The identification radar point cloud is a part of the base station radar point cloud. Under normal circumstances, the distance between the identification radar point cloud and the base station radar point cloud is very close. The correctness of the data is verified by distance calculation. When the distance between the two exceeds the preset distance, it means that the data is wrong and cannot be used for subsequent calculations.

[0065] 404. If yes, determine the base station posture according to the identification radar point cloud;

[0066] If the point corresponding to the entrance of the base station is missing, it means that the base station radar point cloud is incomplete. At this time, the mobile robot generates a logo fitting arc based on the logo radar point cloud, selects the midpoint of the logo fitting arc, obtains the logo reference point, and determines the direction of the logo reference point pointing to the center of the logo fitting arc as the logo direction.

[0067] Specifically, the mobile robot generates a target identification fitting circle according to the identification radar point cloud, and intercepts an identification fitting arc from the target identification fitting circle. The identification radar point cloud is distributed on the identification fitting arc, and the two endpoints of the identification fitting arc are points in the identification radar point cloud. The mobile robot extracts the center and radius of the target identification fitting circle and the position information of the two end points of the identification fitting arc, determines the midpoint of the identification fitting arc according to the center and radius of the target identification fitting circle and the position information of the two end points of the identification fitting arc, obtains the identification reference point, and determines the direction of the identification reference point pointing to the center of the target identification fitting circle as the identification direction, and the center of the target identification fitting circle is the same as the center of the identification fitting arc.

[0068] like Figure 5 As shown in the figure, the gray circle is the target identification fitting circle, point O is the center of the target identification fitting circle and the identification fitting arc, the black dot is the identification radar point cloud, the black arc is the identification fitting arc, D1 and D2 are the two endpoints of the identification fitting arc, D3 is the midpoint of the identification fitting arc, and the direction from D3 to point O is the identification direction.

[0069] Since the identification radar point cloud is obtained by reflecting laser from strongly reflective materials, the position distribution of the identification radar point cloud in the base station radar point cloud can reflect the distribution of strongly reflective materials in the inner cavity. When the distribution of strongly reflective materials in the inner cavity is not symmetrical based on the central axis of the base station, there is a certain angle between the identification direction and the central axis of the base station. The central axis of the base station is the axis of symmetry of the outer contour arc of the cross-section of the inner cavity top view. At this time, the mobile robot obtains the preset deflection direction and deflection angle, and the deflection direction and deflection angle are related to the reflection intensity distribution of the scanned area in the base station.

[0070] Figure 6 (a) is an image of the base station inner cavity taken from the base station opening. The white square area is the area where the strong reflective material is pasted, and the white oval area is the window or fan set on the inner cavity surface. It can be seen that due to the structural design of the base station, the strong reflective material is not symmetrically pasted on the inner cavity surface.

[0071] Figure 6 (b) Figure 6 (a) The inner cavity structure corresponding to the image. Figure 6 (b) also shows the opening side door of the base station, from which it is easy to see that the position where the strong reflective material is pasted is generally biased towards one side of the base station. At this time, the obtained identification radar point cloud position is also distributed on one side of the base station radar point cloud position. According to the pasting position of the strong reflective material on the inner cavity surface, the deflection direction and deflection angle between the midpoint of the strong reflection area and the midpoint of the inner cavity can be calculated. The deflection direction and deflection angle are the preset deflection direction and deflection angle.

[0072] For example. Figure 7It is a schematic diagram of the cross-section of the top view of the base station. The black arc is the intersection line of the area with strong reflective material and the cross-section. M is the midpoint of the inner cavity of the base station. The two black arcs are regarded as an integral arc. According to the positions of the two endpoints P1 and P2 of the integral arc on the inner cavity arc, the position of the midpoint N of the integral arc is determined. Point N is the midpoint of the strong reflection area. The size of ∠NOM is calculated to obtain the preset deflection angle. When the angle of ON deflecting ∠NOM clockwise can reach the OM position, the clockwise deflection is determined as the preset deflection direction. When ON deflects ∠NOM counterclockwise to reach the OM position, the counterclockwise deflection is determined as the preset deflection direction.

[0073] The mobile robot rotates the marker fitting arc in the deflection direction by a deflection angle, and determines the marker direction and the position of the marker fitting arc after deflection as the base station posture. For the convenience of calculation, the position of the center of the marker fitting arc can be used as the position of the marker fitting arc.

[0074] When the base station radar point cloud is incomplete, the association between the lidar point cloud and the base station radar point cloud is established through the position distribution characteristics of the lidar point cloud, and the base station posture is detected through the lidar point cloud to improve the detection accuracy.

[0075] 405. If not, determine the base station posture according to the base station radar point cloud;

[0076] If the points corresponding to the entrance of the base station are not missing, it means that the base station radar point cloud is complete. At this time, the mobile robot generates a base station fitting arc based on the base station radar point cloud, selects the midpoint of the base station fitting arc, obtains the base station reference point, and determines the direction in which the base station reference point points to the center of the base station fitting arc as the base station direction. The base station direction and the position of the base station fitting arc are determined as the base station posture. The method of selecting the focus of the base station fitting arc is similar to the method of selecting the midpoint of the identified fitting arc in step 404, which will not be repeated here.

[0077] When the base station radar point cloud is complete, the accurate base station posture can be obtained directly based on the base station radar point cloud, which is convenient for calculation.

[0078] 406. Extract the target arc and target direction of the base station posture, where the target direction is the direction from the midpoint of the target arc to the center of the target arc;

[0079] like Figure 8 As shown, the black arc is the target arc, the target arc is the base station fitting arc or the marker fitting arc after deflection, O is the center of the target arc, M is the midpoint of the target arc, and the direction from point M to point O is the target direction.

[0080] 407. Determine the target return point according to the midpoint of the target arc, the target direction and the preset target distance;

[0081] Still like Figure 8 As shown, assuming that the preset target distance is L, the mobile robot calculates the position of the target return point B with a distance of L from point M along the target direction.

[0082] 408. Plan a route for the mobile robot to move from the current position to the target return point;

[0083] 409. When the mobile robot moves to the target return point, adjust the forward direction of the mobile robot so that the forward direction of the mobile robot is opposite to the target direction;

[0084] 410. Control the mobile robot to move toward the base station until the mobile robot moves to a preset position in the base station.

[0085] After the mobile robot reaches the target return point, it can move toward the base station according to the guidance direction, or it can enter the base station through infrared guidance. When the mobile robot enters the preset position in the inner cavity of the base station, an electronic signal is triggered. After receiving the electronic signal, the base station provides corresponding supporting operations to the mobile robot.

[0086] In an embodiment of the present invention, a base station radar point cloud and a marking radar point cloud are obtained from scanned environmental radar data. When the distance between the base station radar point cloud and the marking radar point cloud is less than a preset distance, it means that the base station radar point cloud and the marking radar point cloud are valid. At this time, the integrity of the base station radar point cloud is judged. If the base station radar point cloud is complete, the base station posture is detected according to the base station radar point cloud, and the calculation method is simple and accurate. Otherwise, the base station posture is detected according to the marking radar point cloud to calculate the accurate base station posture, so as to select more accurate point cloud features to determine the posture of the base station in different situations, improve the accuracy of detection, and guide the mobile robot to accurately enter the base station according to the detected base station posture.

[0087] The above describes the detection method of the base station posture in the embodiment of the present invention. The following describes the detection device of the base station posture in the embodiment of the present invention. Fig. 9 , an embodiment of a detection device for a base station posture in an embodiment of the present invention includes:

[0088] The scanning module 901 is used to scan the surrounding environment through the laser radar of the mobile robot to obtain environmental radar data;

[0089] An extraction module 902 is used to extract a base station radar point cloud and an identification radar point cloud corresponding to the base station from the environmental radar data, wherein the intensity of the identification radar point cloud is greater than the intensity of the base station radar point cloud;

[0090] The determination module 903 is used to determine the base station posture according to the base station radar point cloud or the identification radar point cloud when the distance between the midpoint of the fitted arc of the base station radar point cloud and the midpoint of the fitted arc of the identification radar point cloud is less than a preset distance.

[0091] In an embodiment of the present invention, a base station radar point cloud and a marker radar point cloud are extracted from the scanned environmental radar data. When the distance between the base station radar point cloud and the marker radar point cloud is less than a preset distance, it means that the base station radar point cloud and the marker radar point cloud are valid. At this time, the base station radar point cloud or the marker radar point cloud is selected according to the integrity of the point cloud to detect the base station posture, thereby improving the accuracy of the detection.

[0092] See also Fig.10 Another embodiment of the detection device of the base station posture in the embodiment of the present invention includes:

[0093] The scanning module 901 is used to scan the surrounding environment through the laser radar of the mobile robot to obtain environmental radar data;

[0094] An extraction module 902 is used to extract a base station radar point cloud and an identification radar point cloud corresponding to the base station from the environmental radar data, wherein the intensity of the identification radar point cloud is greater than the intensity of the base station radar point cloud;

[0095] The determination module 903 is used to determine the base station posture according to the base station radar point cloud or the identification radar point cloud when the distance between the midpoint of the fitted arc of the base station radar point cloud and the midpoint of the fitted arc of the identification radar point cloud is less than a preset distance.

[0096] Optionally, the extraction module 902 is specifically used for:

[0097] The environmental radar data are clustered and curve fitted, and the fitting circle with a radius within a preset radius in the generated fitting circle is determined as the initial base station fitting circle; based on the points corresponding to the initial base station fitting circle, diffusion clustering is performed outward to obtain discrete points within a preset radius around the initial base station fitting circle; the points corresponding to the initial base station fitting circle and the discrete points are summarized to obtain the base station radar point cloud; the point cloud with an intensity higher than the preset intensity is extracted from the environmental radar point cloud to obtain the initial identification point cloud; the initial identification point cloud is clustered, and the point cloud obtained after clustering is determined as the identification radar point cloud.

[0098] Optionally, the determining module 903 includes:

[0099] The judging unit 9031 is used to judge whether the point corresponding to the entrance of the base station is missing when the distance between the midpoint of the fitting arc of the base station radar point cloud and the midpoint of the fitting arc of the identification radar point cloud is less than a preset distance;

[0100] The first determining unit 9032 is used to determine the base station posture according to the identification radar point cloud if yes;

[0101] The second determination unit 9033 is used to determine the base station posture according to the base station radar point cloud if no.

[0102] Optionally, the judging unit 9031 is specifically configured to:

[0103] When the distance between the midpoint of the fitting arc of the base station radar point cloud and the midpoint of the fitting arc of the identification radar point cloud is less than the preset distance, determine whether the number of midpoints in the identification radar point cloud is greater than the preset number; if so, generate a base station fitting arc based on the base station radar point cloud; determine the points at both ends of the base station fitting arc as the first endpoint and the second endpoint; calculate the arc length of the base station fitting arc based on the posture information of the first endpoint and the second endpoint, the posture information includes the position and angle relative to the mobile robot; if the arc length of the base station fitting arc is within the preset arc length range, determine that the point corresponding to the entrance of the base station is not missing; if the arc length of the base station fitting arc is not within the preset arc length range, determine that the point corresponding to the entrance of the base station is missing.

[0104] Optionally, the first determining unit 9032 is specifically configured to:

[0105] If the point corresponding to the entrance of the base station is missing, generate a marker fitting arc based on the marker radar point cloud; select the midpoint of the marker fitting arc to obtain the marker reference point; determine the direction in which the marker reference point points to the center of the marker fitting arc as the marker direction; obtain the preset deflection direction and deflection angle, which are related to the reflection intensity distribution of the scanned area in the base station; rotate the marker direction toward the deflection direction by the deflection angle, and determine the deflected marker direction and the position of the marker fitting arc as the base station posture.

[0106] Optionally, the second determining unit 9033 is specifically configured to:

[0107] If the points corresponding to the entrance of the base station are not missing, a base station fitting arc is generated according to the base station radar point cloud; the midpoint of the base station fitting arc is selected to obtain the base station reference point; the direction in which the base station reference point points to the center of the base station fitting arc is determined as the base station direction; the base station direction and the position of the base station fitting arc are determined as the base station posture.

[0108] Optionally, the detection of the base station posture further includes a moving module 904, and the moving module 904 is specifically used for:

[0109] Extract the target arc and target center corresponding to the base station posture; determine the direction from the target center to the midpoint of the target arc as the guidance direction; determine the target return point according to the midpoint of the target arc, the guidance direction and the preset target distance; plan the route of the mobile robot from the current position to the target return point; when the mobile robot moves to the target return point, adjust the forward direction of the mobile robot so that the forward direction of the mobile robot is the same as the guidance direction; control the mobile robot to move toward the base station until the mobile robot moves to the preset position in the base station.

[0110] In an embodiment of the present invention, a base station radar point cloud and a marking radar point cloud are obtained from the scanned environmental radar data. When the distance between the base station radar point cloud and the marking radar point cloud is less than a preset distance, it means that the base station radar point cloud and the marking radar point cloud are valid. At this time, the integrity of the base station radar point cloud is judged. If the base station radar point cloud is complete, the base station posture is detected according to the base station radar point cloud, and the calculation method is simple and accurate. Otherwise, the base station posture is detected according to the marking radar point cloud to calculate the accurate base station posture, so as to select more accurate point cloud features to determine the posture of the base station in different situations, improve the accuracy of detection, and guide the mobile robot to accurately enter the base station according to the detected base station posture. The present invention also provides a computer-readable storage medium, which can be a non-volatile computer-readable storage medium, or a volatile computer-readable storage medium. The computer-readable storage medium stores instructions, and when the instructions are executed on a computer, the computer executes the steps of the base station posture detection method.

[0111] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0112] 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 invention 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. The computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk and other media that can store program codes.

[0113] As described above, the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features thereof may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for detecting the posture of a base station, for a mobile robot used in conjunction with a base station, characterized in that: The base station posture detection method comprises: Scan the surrounding environment by using the laser radar of the mobile robot to obtain environmental radar data; Extracting a base station radar point cloud and an identification radar point cloud corresponding to the base station from the environmental radar data, wherein the intensity of the identification radar point cloud is greater than the intensity of the base station radar point cloud; When the distance between the midpoint of the fitted arc of the base station radar point cloud and the midpoint of the fitted arc of the identification radar point cloud is less than a preset distance, the base station posture is determined according to the base station radar point cloud or the identification radar point cloud.

2. The method for detecting the posture of a base station according to claim 1, characterized in that: The step of extracting the base station radar point cloud and the identification radar point cloud corresponding to the base station from the environmental radar data includes: Clustering and curve fitting are performed on the environmental radar data, and a fitting circle with a radius within a preset radius in the generated fitting circle is determined as an initial base station fitting circle; Based on the points corresponding to the initial base station fitting circle, diffuse clustering is performed outward to obtain discrete points within a preset radius around the initial base station fitting circle; Summarize the points corresponding to the initial base station fitting circle and the discrete points to obtain a base station radar point cloud; Extracting point clouds with intensities higher than the preset intensity from the environmental radar point clouds to obtain initial identification point clouds; The initial identification point cloud is clustered, and the point cloud obtained after clustering is determined as the identification radar point cloud.

3. The method for detecting the posture of a base station according to claim 1, characterized in that: When the distance between the midpoint of the fitted arc of the base station radar point cloud and the midpoint of the fitted arc of the identification radar point cloud is less than a preset distance, determining the base station posture according to the base station radar point cloud or the identification radar point cloud includes: When the distance between the midpoint of the fitted arc of the base station radar point cloud and the midpoint of the fitted arc of the identification radar point cloud is less than a preset distance, determining whether the point corresponding to the entrance of the base station is missing; If yes, determining the base station posture according to the identified radar point cloud; If not, the base station posture is determined according to the base station radar point cloud.

4. The method for detecting the posture of a base station according to claim 3, characterized in that: When the distance between the midpoint of the fitted arc of the base station radar point cloud and the midpoint of the fitted arc of the identification radar point cloud is less than a preset distance, determining whether the point corresponding to the entrance of the base station is missing includes: When the distance between the midpoint of the fitted arc of the base station radar point cloud and the midpoint of the fitted arc of the identification radar point cloud is less than a preset distance, determining whether the number of midpoints of the identification radar point cloud is greater than a preset number; If yes, generating a base station fitting arc according to the base station radar point cloud; Determine the points at both ends of the base station fitting arc as the first endpoint and the second endpoint; Calculate the arc length of the base station fitting arc according to the posture information of the first endpoint and the second endpoint, wherein the posture information includes a position and an angle relative to the mobile robot; If the arc length of the base station fitting arc is within a preset arc length range, it is determined that the point corresponding to the entrance of the base station is not missing; If the arc length of the base station fitting arc is not within the preset arc length range, it is determined that the point corresponding to the entrance of the base station is missing.

5. The method for detecting the posture of a base station according to claim 3, characterized in that: If so, determining the base station posture according to the identification radar point cloud includes: If the point corresponding to the entrance of the base station is missing, generating a marker fitting arc according to the marker radar point cloud; Selecting the midpoint of the marked fitting arc to obtain a marked reference point; The direction in which the marked reference point points to the center of the marked fitting arc is determined as the marked direction; Acquire a preset deflection direction and a deflection angle, wherein the deflection direction and the deflection angle are related to a reflection intensity distribution of a scanned area in the base station; The marker fitting arc is rotated toward the deflection direction by the deflection angle, and the marker direction and the position of the marker fitting arc after the deflection are determined as the base station posture.

6. The method for detecting the posture of a base station according to claim 3, characterized in that: If not, determining the base station posture according to the base station radar point cloud, including: If the points corresponding to the entrance of the base station are not missing, generating a base station fitting arc according to the base station radar point cloud; Selecting the midpoint of the base station fitting arc to obtain a base station reference point; Determine the direction in which the base station reference point points to the center of the base station fitting arc as the base station direction; The base station direction and the position of the base station fitting arc are determined as the base station posture.

7. The method for detecting the posture of a base station according to any one of claims 1 to 6, characterized in that: After determining the base station posture according to the base station radar point cloud or the marker radar point cloud when the distance between the base station radar point cloud and the marker radar point cloud is less than a preset distance, the method further includes: Extracting a target arc and a target direction of the base station posture, wherein the target direction is a direction from the midpoint of the target arc to the center of the target arc; Determining a target return point according to the midpoint of the target arc, the target direction and a preset target distance; Planning a route for the mobile robot to move from a current position to a target return point; When the mobile robot moves to the target return point, adjusting the forward direction of the mobile robot so that the forward direction of the mobile robot is opposite to the target direction; The mobile robot is controlled to move toward the base station until the mobile robot moves to a preset position in the base station.

8. A device for detecting a base station posture, characterized in that: The detection device of the base station posture comprises: A scanning module, used to scan the surrounding environment through the laser radar of the mobile robot to obtain environmental radar data; An extraction module, used to extract a base station radar point cloud and an identification radar point cloud corresponding to the base station from the environmental radar data, wherein the intensity of the identification radar point cloud is greater than the intensity of the base station radar point cloud; A determination module is used to determine the base station posture according to the base station radar point cloud or the identification radar point cloud when the distance between the midpoint of the fitted arc of the base station radar point cloud and the midpoint of the fitted arc of the identification radar point cloud is less than a preset distance.

9. A base station posture detection system, characterized in that: The base station posture detection system comprises: a mobile robot and a base station, the mobile robot comprises a memory and at least one processor, and the memory stores instructions; The at least one processor calls the instructions in the memory so that the base station posture detection system executes the base station posture detection method as described in any one of claims 1-7.

10. A computer-readable storage medium having instructions stored thereon, characterized in that: When the instruction is executed by the processor, the method for detecting the posture of the base station as described in any one of claims 1 to 7 is implemented.

Images