Charging pile recognition method based on laser radar scanning, chip and robot

Abstract

This application discloses a charging pile identification method, chip, and robot based on lidar scanning. The charging pile identification method is used to identify a charging pile when it emits a charging guidance signal. The execution carrier of the charging pile identification method is a robot equipped with lidar. At least two signal receivers are installed on the front of the robot to receive the charging guidance signal emitted by the charging pile. The charging pile identification method includes: the robot controlling the lidar to emit laser lines to form a laser scanning area; setting at least three detection angle areas from the laser scanning area, so that there is at least one detection angle area in front of the robot and on each of its two sides; extracting the ranging data of the target laser points from each detection angle area; and identifying the charging pile based on the ranging differences between different target laser points and the reception of the charging guidance signal by the at least two signal receivers. This improves the accuracy of identifying the charging pile.

Description

Technical Field

[0001] This application relates to the technical field of charging pile positioning, specifically to a charging pile identification method, chip, and robot based on LiDAR scanning. Background Technology

[0002] Regarding the robot vacuum's recharging function, you need to first navigate the robot to a location that can recognize a charging station.

[0003] The charging guidance signal emitted by the charging pile is generally an infrared beam signal. When using infrared guidance to locate the charging pile, most robots adopt a random mode to find the infrared beam signal directly in front of the charger. Due to the poor linearity of infrared light, it is easily interfered with by strong light and is easily blocked; it also has the disadvantage of low recharge docking success rate.

[0004] Current technology uses radar to scan charging piles and performs fitting algorithms on laser point clouds to locate them. When the surface of the charging pile is black or reflective, the reflected laser point clouds are scarce and unstable due to the characteristics of the lidar sensor itself. This results in poor performance of the fitting algorithm, leading to problems such as misalignment (the located charging pile position does not match the actual location of the charging pile, and other objects are mistakenly identified as the charging pile position), and the charging pile cannot be correctly identified. Summary of the Invention

[0005] This application discloses a charging pile identification method, chip, and robot based on LiDAR scanning, and proposes the following technical solutions: A charging pile identification method based on lidar scanning is used to identify charging piles when they emit charging guidance signals. The method is executed by a robot equipped with lidar. At least two signal receivers are installed on the front of the robot to receive the charging guidance signals emitted by the charging pile. The method includes: the robot controlling the lidar to emit laser lines to form a laser scanning area; setting at least three detection angle areas within the laser scanning area, such that there is at least one detection angle area in front of the robot and on each of its two sides; extracting ranging data of target laser points from each detection angle area; and identifying the charging pile based on the ranging differences between different target laser points and the reception of the charging guidance signals by the at least two signal receivers.

[0006] In summary, this application identifies charging piles based on the ranging differences between different target laser points and the reception of the charging guidance signal by the at least two signal receivers. This allows for the identification of charging piles even when the laser point cloud is incomplete due to the charging pile's surface being made of black or reflective material, preventing misjudgments and allowing the robot to continue its recharging function using the at least two signal receivers. Thus, during the robot's initial straight-line docking recharging phase, the ranging data of the target laser points within each detection angle region and the reception of the charging guidance signal compensate for the inability of the laser point cloud to accurately fit the effective pose information of the charging pile. This optimizes the robot's recharging scenario and forms a recharging processing method that overcomes the influence of the charging pile's surface material. It overcomes the shortcomings of relying solely on laser or infrared, improving the reliability and accuracy of the robot's active recharging.

[0007] Furthermore, the method for identifying the charging pile based on the ranging differences between different target laser points and the reception of the charging guidance signal by the at least two signal receivers includes: if the ranging differences between target laser points in different detection angle areas are all within a preset distance error range, and the at least two signal receivers simultaneously receive the charging guidance signal, then it is determined that the robot has identified the charging pile, and that the charging pile exists directly in front of the robot, so as to control the robot to walk in a straight line to the front of the charging pile. Therefore, after determining that the robot has identified the charging pile, based on the ranging data of the target laser points in each detection angle area, if the charging structure is installed on the front of the robot, the robot moves in a straight line to the front of the charging pile to achieve docking of the front charging structure with the charging pile for charging; alternatively, if the charging structure is installed on the back of the robot, the robot first turns around and moves backward in a straight line to the front of the charging pile to achieve docking of the rear charging structure with the charging pile for charging.

[0008] Furthermore, if the distance difference between target laser points within the detection angle regions of different blocks is not all within the preset distance error range, then: the target laser point in the detection angle region set on one side directly in front of the robot is set as the first target laser point, the target laser point in the detection angle region set in front of the robot is set as the second target laser point, and the target laser point in the detection angle region set on the other side directly in front of the robot is set as the third target laser point; if the distance measurement data of the third target laser point and the distance measurement data of the second target laser point are both greater than the distance measurement data of the first target laser point, then during the process of the robot walking along the vertical line at the middle position in front of the charging pile, it is determined that there is an obstacle in the detection angle region on the side where the first target laser point is located, preventing the robot from walking in a straight line to the front of the charging pile, and the robot does not recognize the obstacle as the charging pile. Therefore, based on the first target laser point, the positional relationship of the obstacle relative to the robot can be determined, so that the obstacle prevents the robot from recognizing the charging pile; then the robot can adjust its posture according to the ranging data of the first target laser point to avoid the obstacle, so that the ranging difference between the target laser points in different detection angle areas of the robot is within the preset distance error range, and the at least two signal receivers simultaneously receive the charging guidance signal. Only on this basis can the straight-line recharge be completed without obstacles, and the obstacle in front of the charging pile can be distinguished without causing misjudgment.

[0009] Furthermore, if the at least two signal receivers do not simultaneously receive the charging guidance signal, it is determined that an obstacle exists in front of the robot, preventing it from walking in a straight line to the front of the charging station, and that the obstacle obstructs part or all of the detection view of the at least two signal receivers. Therefore, when the signal receivers do not receive the charging guidance signal, the obstacle is located based on the ranging data of the target laser points within each detection angle area. Subsequently, the obstacle's position is calculated based on the ranging data of all target laser points, and the robot avoids the obstacle or re-identifies the charging station by continuously adjusting its posture and movement orientation.

[0010] Furthermore, the target laser point within each detection angle area is the laser reflection point within that detection angle area that is closest to the center of the robot's body or the center of the lidar; wherein, the lidar is used to collect the laser reflection points generated by the reflection on the surface material to be identified and to calculate the distance data of the laser reflection points at each scanning angle, so as to represent the distance between the surface material to be identified and the robot.

[0011] In summary, by implementing the charging pile identification method disclosed in the foregoing embodiments, this application can identify charging piles even when the laser point cloud is incomplete due to the black or reflective material of the charging pile surface, without causing misjudgment. It can also continue to rely on the at least two signal receivers to complete the recharging function. Therefore, in the scenario where the robot performs the recharging function, the ranging data of the target laser points in each detection angle area and the reception of the charging guidance signal are used to compensate for the defect that the laser point cloud is too scarce to fit effective pose information, thereby optimizing the robot recharging scenario and forming a recharging processing method that can overcome the influence of the charging pile surface material.

[0012] Furthermore, the method of setting at least three detection angle regions within the laser scanning area includes: setting a target detection angle with the robot's walking direction as the bisector of the angle within the laser scanning area, and uniformly dividing the target detection angle into at least three detection angle regions with the center of the laser radar as the vertex; wherein, one detection angle region covers the vertical line at the center of the robot's front, and at least two other detection angle regions are located on either side of the robot's walking direction; wherein, the vertical line at the center of the robot's front extends along the robot's walking direction; the front of the robot is represented by the robot's walking direction. Thus, by uniformly dividing the target detection angle region with the center of the laser radar as the vertex, and with each of the three detection angle regions being symmetrical about the vertical line at the center of the robot's front, the ranging data of target laser points distributed on both sides of the robot's walking direction can be uniformly detected; thereby, charging pile and obstacle identification can be performed in different areas within a limited detection angle range and detection distance.

[0013] Furthermore, when the robot is positioned on the vertical line at the center of the front of the charging pile, the target detection angle is the maximum detection angle required for the lidar to scan the charging pile, so that the two laser lines emitted from this maximum detection angle cut through the two sides of the charging pile respectively; wherein, the charging guidance signal emitted by the charging pile covers the vertical line at the center of the front of the charging pile; the front of the charging pile is opposite to the front of the robot; the front of the charging pile is the location for installing the charging structure; wherein, the at least three detection angle regions are distributed between the robot and the charging pile; wherein, the vertical line at the center of the front of the robot is pre-configured to represent a coordinate axis of the two-dimensional coordinate system required for robot navigation. Thus, the positional relationship of the obstructing obstacle relative to the robot can be determined within the maximum detectable range in front of the charging pile, and the differences in ranging data of target laser points within each detection angle region and the simultaneous reception of signals by the robot's signal receiver can be compared within the coverage area of ​​the maximum detection angle.

[0014] Furthermore, the at least three detection angle regions are a left detection angle region, a middle detection angle region, and a right detection angle region. The middle detection angle region is located directly in front of the robot, covering the vertical line from the center of the robot's front, so that the robot can detect obstacles in the center of the charging pile or in the corresponding direction when walking along the vertical line from the center of the charging pile's front. The left detection angle region is located to the left of the robot's front, so that the robot can detect obstacles to the left of the charging pile or in the corresponding direction when walking along the vertical line from the center of the charging pile's front. The right detection angle region is located to the right of the robot's front, so that the robot can detect obstacles to the right of the charging pile or in the corresponding direction when walking along the vertical line from the center of the charging pile's front.

[0015] A chip stores a computer program, which is the program code corresponding to the charging pile identification method. The chip controls a robot to scan the surrounding environment using lidar during its movement. Based on the distance measurement differences between different target lidar points and the reception of the charging guidance signal by at least two signal receivers, the chip identifies the charging pile and distinguishes obstacles between the charging pile and the robot, improving the accuracy of charging pile identification. Furthermore, during the chip-controlled robot's straight-line docking recharging phase, the chip uses the distance measurement data of target lidar points within each detection angle area and the reception of the charging guidance signal to compensate for the deficiency of the scarce lidar point cloud, which cannot fit effective pose information. This optimizes the robot's recharging scenario and forms a recharging processing method that can overcome the influence of the charging pile's surface material.

[0016] A robot is disclosed, comprising a lidar sensor mounted on its top and at least two signal receivers horizontally positioned on either side of a vertical line centered on the front of the robot. The robot's interior houses the aforementioned chip. When the robot begins to move and directly engage with a charging station, the chip executes a computer program to implement the charging station identification method. This robot, by implementing the charging station identification method, can identify charging stations even when the lidar point cloud is incomplete due to the charging station's surface being made of black or reflective material, preventing false positives, and can continue to perform recharging functionality using the at least two signal receivers. Attached Figure Description

[0017] Figure 1A schematic diagram of three detection angle regions set in front of a robot is provided for one embodiment of this application; wherein, viewed from a clockwise direction, the three detection angle regions are, in order, the region marked "left" (-15 degrees to -5 degrees), the region marked "middle" (-5 degrees to 5 degrees), and the region marked "right" (5 degrees to 15 degrees), all of which are within the scanning area of ​​the lidar laser; 0 degrees represents the angle corresponding to the angle bisector of the angle region between -15 degrees and 15 degrees.

[0018] Figure 2 Another embodiment of this application provides a schematic diagram showing an obstacle obs in the detection angle area on the left side of the front of the robot, wherein the obstacle obs is the projection of a cylindrical obstacle onto a horizontal ground, and the horizontal ground is the walking plane of the robot.

[0019] Figure 3 This application provides a schematic diagram of a linear obstacle that passes through three detection angle regions sequentially between a charging station and a robot, according to another embodiment of the present application. Implementation

[0020] The technical solutions of the embodiments of this application will be described in detail below with reference to the accompanying drawings. Unless otherwise specified, the following embodiments and features can be combined with each other. In this application, it should be understood that the terms "center," "middle position," "central axis," "upper," "lower," "left," "right," "vertical," "horizontal," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the present invention. If terms such as "first," "second," and "third" appear in the embodiments, it is for the convenience of distinguishing related features, and should not be construed as indicating or implying their relative importance, order, or number of technical features.

[0021] The lidar associated with this embodiment is a laser sensor composed of a 360-degree rotating laser emitting probe. The laser emitting probe can scan the contour features of the surrounding environment by rotating one full circle, obtaining at least one frame of laser point cloud data. A frame of laser point cloud data is a collective term for the laser data points contained within a single frame of laser data, reflecting the coordinates of each laser reflection point in the laser sensor coordinate system to represent laser ranging information. It can be understood that the laser line emitted by the laser emitting probe forms laser reflection points after being reflected by the surface of an object, thus forming a laser point cloud. A single frame of laser point cloud data contains multiple laser reflection points. Laser reflection points generated by the reflection of laser lines on black or reflective surfaces are sparse. Although the black or reflective surface of the charging pile may lead to incomplete laser point clouds, this application can determine the maximum detection angle of the laser line within a certain object and define a certain detection angle range. If the laser point cloud data is projected onto a two-dimensional laser grid plane, the maximum detection angle generated by the lidar on one side of the charging pile can be displayed from the horizontal plane, such as... Figures 1 to 3 The combination of the angled areas marked "left", "middle" and "right" in the diagram is shown; the lidar can receive reflected laser lines within this maximum detection angle and obtain the coordinates and ranging data of a batch of laser reflection points.

[0022] Because laser beams are unidirectional and have excellent directionality, and are plane waves, they have relatively low scattering, and the laser beam emitted by laser emission detectors is close to parallel light.

[0023] To address the problem of inaccurate charging pile identification in robot recharging scenarios due to the scarcity and instability of the laser point cloud received by the robot, this application provides a charging pile identification method based on lidar scanning. This method identifies the charging pile when it emits a charging guidance signal. The charging guidance signal uses an infrared guidance signal as suggested in the background technology; in some implementation scenarios, laser or ultrasonic guidance signals can also be used. The execution carrier of the charging pile identification method is a robot equipped with lidar, which scans and plans a specific detection area using laser scanning. To receive the charging guidance signal emitted by the charging pile and cooperate with the lidar scanning operation, at least two signal receivers are installed on the front side or front of the robot. Here, "front side" and "front" are defined according to the robot's walking direction, and the signal receivers are preferably infrared receiving sensors. The installation position of the charging structure can be on the front of the robot to allow the robot to move straight back to charge, or on the rear side of the robot to allow the robot to turn around and back to the charging pile. Correspondingly, the charging guidance signal can be a center docking signal emitted by an infrared emitting sensor installed in the center of the charging pile, with an emission angle of 30 to 60 degrees. Its emission direction is simplified to... Figure 2 and Figure 3The downward-pointing arrow indicates that the signal receivers mounted on the front side or front of the robot must be aligned with a vertical line (the central axis of the charging pile) directly in front of the center of the charging pile, or within a certain angle range in the vertical forward direction relative to the center of the front side of the charging pile, so that at least two signal receivers can simultaneously receive the intermediate docking signal. The front side of the charging pile can be a plane or a curved surface symmetrical about the central axis of the charging pile.

[0024] In this application, the charging pile identification method includes: a robot controlling a lidar to emit laser lines to form a laser scanning area. This laser scanning area can be an angular region with the center of the lidar (i.e., the rotation center of the laser emitting probe used to emit the laser lines) as its vertex, such as... Figures 1 to 3 The diagram shows the angled region formed by the divergence of the LiDAR laser forward. It should be noted that, to improve recognition accuracy and accelerate recharging efficiency, the charging station identification method is executed during the robot's straight-line recharging phase. This straight-line recharging phase is when the robot begins to return to the charging station along a straight line. This straight line can be a vertical line perpendicular to the center of the charging station's front, and this vertical line is pre-defined and set as an axis in the two-dimensional coordinate system required for robot navigation. The two-dimensional coordinate system required for robot navigation is preferably the coordinate system of the laser grid map.

[0025] Then, the robot sets at least three detection angle areas within the laser scanning area, so that there is at least one detection angle area in front of the robot and on each of its two sides. Specifically, a detection angle area can be divided based on the robot's central axis, and then multiple detection angle areas can be divided on the left and right sides of this detection angle area. Thus, setting at least three detection angle areas within the laser scanning area can cover at least one side of the charging pile that is connected to the robot for charging. Therefore, the at least three detection angle areas will cover the front of the robot and its two sides, as well as the front of the charging pile and its two sides.

[0026] Then, the robot extracts the ranging data of the target laser point from each detection angle region. Each detection angle region contains a target laser point and its ranging information is calculated to represent the distance between the target laser point and the center of the robot (or the center of the lidar) in the corresponding direction, or the distance between the robot and an obstacle (which can be a charging pile or a non-charging pile, and the specific distance needs to be determined by the signal reception of the at least two signal receivers) in one direction.

[0027] Based on the distance measurement differences between different target laser points and the reception status of the charging guidance signal by the at least two signal receivers, the charging pile is identified, thereby distinguishing the charging pile from other obstacles in the robot's environment. In this application, the distance measurement differences between different target laser points are understood as the distance difference between target laser points within any two detection angle regions. This requires calculating multiple distance differences, and then threshold comparison can be used to eliminate the problem of poor fitting effect caused by the scarcity of laser point clouds. Furthermore, the reception status of the charging guidance signal by the at least two signal receivers emphasizes the consistency of reception by all signal receivers, including at least simultaneous reception of the charging guidance signal. Due to the interference problem of the charging guidance signal, the signal strength or signal incident direction received by each signal receiver may not be the same, but it can be used to confirm whether the target laser point originates from the surface of the charging pile. Thus, the robot disclosed in this application accurately identifies the charging pile by using the scanning data of the lidar and the reception status of the charging guidance signal.

[0028] Regardless of whether the surface material to be identified by the charging pile identification method is non-reflective or reflective, during the scanning of the charging pile / robot environment by the LiDAR, the robot executes the charging pile identification method to identify the charging pile based on the ranging differences between different target laser points and the reception of the charging guidance signal by the at least two signal receivers. This allows for the identification of the charging pile even when the laser point cloud is incomplete due to the charging pile surface being black or reflective, preventing misjudgments and allowing the robot to continue its recharging function using the at least two signal receivers. Furthermore, during the robot's linear docking recharging phase, the ranging data of the target laser points within each detection angle region and the reception of the charging guidance signal compensate for the inability of the scarce laser point cloud to fit the effective pose information of the charging pile, optimizing the robot's recharging scenario and forming a recharging processing method that overcomes the influence of the charging pile surface material. This overcomes the shortcomings of relying solely on laser or infrared, improving the reliability and accuracy of the robot's active recharging.

[0029] This application, by implementing the charging pile identification method, identifies the charging pile and eliminates interference from obstacles based on the ranging differences between different target laser points and the reception of the charging guidance signal by the at least two signal receivers. Thus, during the robot's straight-line docking and recharging phase, the charging pile is accurately identified by combining the ranging data extracted by the laser radar in different areas and the charging guidance signal received by the signal receivers, reducing misjudgment of obstacles, overcoming the shortcomings of relying solely on laser or infrared, and improving the reliability and accuracy of the robot's active recharging.

[0030] In the above embodiments, the method for identifying the charging pile based on the ranging differences between different target laser points and the reception of the charging guidance signal by the at least two signal receivers includes: if the ranging differences between target laser points in different detection angle areas are all within a preset distance error range, and the at least two signal receivers simultaneously receive the charging guidance signal, then it is determined that the robot has identified the charging pile, and that the charging pile exists directly in front of the robot, so as to control the robot to walk in a straight line to the front of the charging pile; therefore, after determining that the robot has identified the charging pile, based on the ranging data of the target laser points in each detection angle area, if the charging structure is installed on the front of the robot, the robot moves in a straight line to the front of the charging pile to achieve docking of the front charging structure with the charging pile for charging; alternatively, if the charging structure is installed on the back of the robot, the robot first turns around and moves in a straight line to the front of the charging pile to achieve docking of the rear charging structure with the charging pile for charging.

[0031] In some recharging scenarios, it can be determined that the robot is walking along a vertical line in front of the charging pile (the robot begins to recharge and dock in a straight line) and recognizes the presence of the charging pile in front of it. In this scenario, there is an open area between the charging pile and the robot, and the robot has entered the docking stage of returning to the charging pile in a straight line. The signal receivers installed on the front of the robot (especially the two signal receivers installed on both sides of the center position on the front) can receive the signal normally and can combine the distance measurement differences between different target laser points to eliminate the influence of factors such as obstacles, environment, and reflective materials.

[0032] In some embodiments, if the distance difference between target laser points within the detection angle regions of different blocks is not all within a preset distance error range, and / or the at least two signal receivers do not simultaneously receive the charging guidance signal, it is determined that the robot has not identified the charging pile. This can be accurately understood as the robot not identifying the charging pile while walking along the vertical line from the center of the charging pile's front, i.e., the robot does not identify the charging pile directly in front of it, thus failing to meet the straight-line recharge condition. Alternatively, it can be understood that the robot may not be walking along the vertical line from the center of the charging pile's front but instead moving to the left or right of the charging pile, equivalent to the front side of the charging pile being tilted relative to the robot's front. Or, the robot may be unable to identify the charging pile while walking along the vertical line from the center of the charging pile's front due to obstruction by an obstacle, only recognizing an obstacle on one side of the charging pile's front or the charging guidance signal being blocked by an obstacle, preventing the robot from recharging in a straight line. Therefore, the robot needs to adjust its pose based on the ranging data of the target laser points in the detection angle area of ​​each block, so that the ranging difference between the target laser points in the detection angle areas of different blocks is within the preset distance error range, and the at least two signal receivers simultaneously receive the charging guidance signal.

[0033] The ranging difference between target laser points in different detection angle regions includes the ranging difference between a target laser point in one of the at least three detection angle regions and a target laser point in any of the other detection angle regions.

[0034] In this embodiment, the preset distance error range is related to the system ranging error of the lidar, but does not exceed the ranging difference caused by obstacle obstruction. The preset distance error range can be understood as the distance range delineated based on the ranging difference between target laser points within different detection angle areas during the process of the robot walking along the vertical line at the center of the front of the charging pile (when the robot starts to move straight back to dock) without obstacles obstructing the charging pile. Moreover, the at least two signal receivers can simultaneously receive the charging guidance signal.

[0035] When the front side of the charging pile is tilted relative to the front of the robot, the distance difference between the target laser points in different detection angle areas may not all be within the preset distance error range. In this case, if the robot's charging structure is installed in front of the robot, the robot cannot return to the charging pile in a straight line for docking and charging. At the same time, because the at least two signal receivers are in front of the robot and the charging pile emits charging guidance signals at a limited angle, the at least two signal receivers cannot receive the charging guidance signal simultaneously. Therefore, in this embodiment, when determining whether the robot has recognized the charging pile, the distance difference between different target laser points and the reception status of the charging guidance signal by the at least two signal receivers are considered together.

[0036] The following is for reference. Figures 2 to 3 The description of the charging pile, robot, and laser scanning area disclosed in this application is shown on a horizontal plane. This horizontal plane can be a grid plane formed by establishing a grid at the same height in the three-dimensional environment where the robot and the charging pile are located; it can also be understood as the robot's walking plane.

[0037] As one embodiment, if the distance difference between the target laser points in the detection angle regions of the different blocks is not all within the preset distance error range, then: the at least two signal receivers may receive the charging guidance signal simultaneously, or only some signal receivers may receive the charging guidance signal or may not receive the charging guidance signal at all; then the target laser point in the detection angle region set on one side directly in front of the robot is set as the first target laser point, the target laser point in the detection angle region set on the front of the robot is set as the second target laser point, and the target laser point in the detection angle region set on the other side directly in front of the robot is set as the third target laser point. If the ranging data of the third target laser point and the second target laser point are both greater than the ranging data of the first target laser point, then it is determined that there is an obstacle in the detection angle area on the side of the first target laser point, preventing the robot from walking in a straight line to the front of the charging pile. Furthermore, the robot does not recognize the obstacle as a charging pile. Specifically, it can be determined that while the robot is walking along the vertical line (the central axis of the charging pile) in front of the middle position, it does not recognize the charging pile directly in front of it. Instead, it detects an obstacle in the area on the side where it receives the first target laser point. Based on the first target laser point, the positional relationship of the obstacle relative to the robot can be determined, causing the obstacle to prevent the robot from recognizing the charging pile. Subsequently, the robot can adjust its posture based on the ranging data of the first target laser point to avoid the obstacle. This ensures that the ranging differences between the target laser points in different detection angle areas are all within a preset distance error range, and that at least two signal receivers simultaneously receive the charging guidance signal. Only under these conditions can the robot achieve unobstructed straight-line recharging and distinguish obstacles in front of the charging pile, preventing misjudgments.

[0038] Indicatively, such as Figure 1 and Figure 2 As shown, the laser reflection point closest to the robot within the -15° to -5° area marked "left" can be defined as the first target laser point, representing the target laser point within the detection angle area set on the left side directly in front of the robot; correspondingly, the laser reflection point closest to the robot within the 5° to 15° area marked "right" can be defined as the third target laser point, representing the target laser point within the detection angle area set on the right side directly in front of the robot; and the laser reflection point closest to the robot within the -5° to 5° area marked "center" can be defined as the second target laser point, representing the target laser point within the detection angle area set directly in front of the robot. (Combined...) Figure 2It can be seen that during the robot's journey towards the charging pile along the vertical line from the center of its front, an obstacle (obs) exists in the area marked "left". Since the obs is a cylindrical obstacle projected onto the horizontal ground and occupies a small area, the distance measurements of the third and second target laser points are both greater than the distance measurement of the first target laser point. This means that the distance measurements measured on the robot's right front and front sides are significantly greater than those measured on its left front. Therefore, when the difference in distance measurements between the first, second, and third target laser points exceeds a certain threshold, the charging pile and the obstacle can be distinguished, increasing the accuracy and reliability of the robot's charging pile identification.

[0039] As one embodiment, if the at least two signal receivers do not receive the charging guidance signal simultaneously, it is determined that an obstacle exists in front of the robot, and this obstacle obstructs part or all of the detection view of the at least two signal receivers, preventing them from receiving the charging guidance signal simultaneously. More precisely, it can be determined that while the robot is walking along a vertical line (the central axis of the charging pile) directly in front of the charging pile, it fails to identify the charging pile directly in front of it and instead detects an obstacle obstructing the charging guidance signal, preventing at least one signal receiver from receiving the signal. Therefore, when the signal receivers do not receive the charging guidance signal, the obstacle is located based on the ranging data of the target laser points within each detection angle area. Subsequently, the obstacle's position is calculated based on the ranging data of all target laser points, and the robot avoids the obstacle or re-identifies the charging pile by continuously adjusting its posture and movement.

[0040] In this embodiment, the ranging difference between target laser points within the detection angle regions of different blocks can all be within a preset distance error range, as illustrated below. Figure 3As shown in the diagram, the linear obstacle obstacle is located on the side directly in front of the robot. The distance measurement data reflected back from the linear obstacle obstacle ensures that the distance difference between the target laser points in different detection angle regions is within a preset distance error range. Although the linear obstacle obstacle blocks the charging guidance signal emitted by the charging pile, it also blocks the entire detection angle of the charging pile from at least two signal receivers when passing through the three detection angle regions in sequence. Under this obstacle distribution, the robot in the diagram cannot walk in a straight line along a predetermined direction (e.g., along the direction of the vertical line (the central axis of the charging pile) extending from the middle position directly in front of the charging pile) to the front of the charging pile. Therefore, it can be determined that the robot cannot identify the charging pile directly in front of it, but the general distribution position of the obstacles can be determined. Since the linear obstacle obstacle runs through the area between the robot and the charging pile, the target laser points extracted in different detection angle regions are different from the target laser points in the previous embodiment, and the distance measurement data of the target laser points extracted in different detection angle regions are smaller. Therefore, obstacles can be identified in front of the charging station based on the reception of the charging guidance signal by the at least two signal receivers.

[0041] In the aforementioned embodiments, the target laser point within each detection angle region is the laser reflection point within that region that is closest to the robot's body center or the center of the lidar. Considering the error caused by laser scattering, it needs to be stated that the carrier of the laser reflection is a surface of a certain material. The lidar is used to collect the laser reflection points generated by the reflections on the surface material to be identified and to calculate the distance measurement data of the laser reflection points at each scanning angle, representing the distance between the surface material to be identified and the robot. The surface material to be identified can be the material of one side of the charging pile or the surface material of obstacles distributed near the charging pile. Based on the distance measurement data of the laser reflection points at each scanning angle, the coordinates of the reflection location are calculated and converted to global coordinates. This coordinates can be marked on a grid map and reserved for subsequent steps for recharging navigation.

[0042] The charging pile identification method requires the identification of a non-reflective or reflective surface material. This is because, during the scanning process of the charging pile / robot environment by the LiDAR, the laser lines emitted by the LiDAR reflect off the surface material, resulting in a sparse distribution of laser points. This can lead to incomplete laser point clouds due to the charging pile surface being black or reflective. Therefore, by implementing the charging pile identification method disclosed in the aforementioned embodiments, this application can identify the charging pile even when the laser point cloud is incomplete due to the charging pile surface being black or reflective, without misjudgment. It can also continue to rely on the at least two signal receivers to complete the recharging function. Therefore, in scenarios where the robot performs the recharging function, the ranging data of the target laser points within each detection angle area and the reception of the charging guidance signal are used to compensate for the deficiency of the laser point cloud in fitting effective pose information, thereby optimizing the robot's recharging scenario and forming a recharging processing method that can overcome the influence of the charging pile surface material.

[0043] As one embodiment, the method of setting at least three detection angle regions within the laser scanning area includes: setting a target detection angle with the robot's walking direction as the angle bisector within the laser scanning area, and uniformly dividing the target detection angle into at least three detection angle regions with the center of the lidar as the vertex; wherein, one detection angle region covers the vertical line to the center of the robot's front, and at least two other detection angle regions are located on either side of the robot's walking direction; to improve the recognition efficiency of the charging pile, three detection angle regions are set within the laser scanning area. Figure 1 This application provides a schematic diagram of three detection angle regions set in front of a robot in one embodiment. Viewed clockwise, these three detection angle regions are, in order, the region marked "left" (-15° to -5°), the region marked "middle" (-5° to 5°), and the region marked "right" (5° to 15°), all within the scanning area of ​​the LiDAR laser. 0° represents the angle corresponding to the angle bisector of the angle region between -15° and 15°. Thus, with the center of the LiDAR as the vertex, three detection angle regions are evenly divided within the target detection angle, and these three regions are symmetrical about the vertical line at the center of the robot's front, uniformly detecting the ranging data of target laser points distributed on both sides of the robot's walking direction. This allows for the identification of charging stations and obstacles within a limited detection angle and detection distance.

[0044] It should be noted that the vertical line at the center of the robot's front extends along the robot's walking direction; the front of the robot is indicated by its walking direction; the front of the robot is used to install at least two signal receivers. For example, a binocular receiving mold can be set on the front of the robot, and two signal receivers can be installed inside the binocular receiving mold. The receiving probe receives the charging guidance signal emitted by the charging pile through the binocular receiving mold. In addition, the position in the robot used to install the charging structure (including electrodes) can also be on the front of the robot, so as to cooperate with the front of the robot to dock with the charging pile for electrode docking and charging.

[0045] Preferably, when the robot is positioned on the vertical line at the center of the front of the charging pile, the target detection angle is the maximum detection angle required for the lidar to scan the charging pile. Two laser lines emitted from this maximum detection angle cut across the two sides of the charging pile, representing the detection angle range scanned by the lidar between two moments. The straight edges of the detection angle range are tangent to the two sides of the charging pile. At least three detection angle regions are distributed between the robot and the charging pile. The charging guidance signal emitted by the charging pile covers the vertical line at the center of the front of the charging pile, which also serves as the bisector of the angle of the maximum detection angle. The front of the charging pile is opposite to the front of the robot, representing the stage where the robot walks in a straight line to the front of the charging pile (straight-line docking and recharging stage). The established directional relationship, and the vertical line at the center of the robot's front is pre-configured. This vertical line represents one axis of the two-dimensional coordinate system required for robot navigation. When the robot begins docking for charging, it can walk along this vertical line while executing the charging pile identification method. Based on this, the maximum detection angle coverage area, the robot, and the charging pile can be transformed into a two-dimensional coordinate system. The positional relationship between the charging pile and the robot can be calculated. The positional relationship of the obstructing obstacle relative to the robot can also be indirectly determined within the maximum detectable range directly in front of the charging pile. Furthermore, the differences in the ranging data of the target laser points in each detection angle area within the maximum detection angle coverage area and the simultaneous reception of signals by the robot's signal receiver can be compared.

[0046] As one embodiment, the at least three detection angle regions are respectively the left detection angle region, the middle detection angle region, and the right detection angle region, combined with Figures 1 to 3As can be seen, the detection angle area set directly in front of the robot is the central detection angle area. The area marked "center" in the diagram is the central detection angle area. The central detection angle area covers the vertical line from the center of the robot's front, allowing the robot to detect the center of the charging pile's front when walking along this vertical line. This enables the robot to move in a straight line to contact the charging pile, or to detect obstacles in the corresponding direction. These obstacles include... Figure 3 The linear obstacle shown is a straight line. Figure 3 The straight-line obstacle shown can pass through the central detection angle area. It should be noted that the vertical line at the center of the robot's front and the vertical line at the center of the charging pile's front can both be understood as axes extending along the front-rear direction of their respective device models. The detection angle area located on the left side of the robot's front is the left detection angle area. The area marked "left" in the diagram is the left detection angle area. This area is located to the left of the robot's walking direction, allowing the robot to detect obstacles to the left of the charging station or in the corresponding direction when walking along a vertical line from the center of the charging station's front. Obstacles in the corresponding direction include... Figure 2 The obstacle obs shown is covered by the left detection corner area.

[0047] The detection angle area located on the right side of the robot's front is the right detection angle area. The area marked "right" in the diagram is the right detection angle area. This area is located to the right of the robot's walking direction, allowing the robot to detect obstacles to the right of the charging pile or in the corresponding direction when walking along a vertical line from the center of the charging pile's front. Obstacles in the corresponding direction include... Figure 3 The obstacle shown passes through the right detection corner area.

[0048] Preferably, the included angle of each detection angle region is 10 degrees; the receiving range of one of the signal receivers can overlap with one of the detection angle regions. Generally, the signal receiver can only receive the charging guidance signal when the charging pile is within the receiving range of the robot's signal receiver and there are no obstacles obstructing it. Within each 10-degree region, the robot extracts a target laser point, representing the laser point closest to the robot within that 10-degree region, used for distance data to reflect the lateral distance information of obstacles or the charging pile in the corresponding direction.

[0049] Combination Figure 2It can be seen that during the process of the robot walking towards the charging pile along the vertical line in the middle of the front of the charging pile, a first target laser point is detected from the left detection angle area, a second target laser point is detected from the middle detection angle area, and a third target laser point is detected from the right detection angle area; when the ranging data of the third target laser point and the ranging data of the second target laser point are both greater than the ranging data of the first target laser point, it is determined that there is an obstacle OBS in the left detection angle area, and it is located on the corresponding side in front of the charging pile.

[0050] For example Figure 3 As shown, the ranging data reflected back from the straight-line obstacle located directly in front of the robot ensures that the ranging differences between target laser points within different detection angle regions are all within a preset distance error range. Although the straight-line obstacle blocks the charging guidance signal emitted by the charging pile, it also blocks the entire detection angle of at least two signal receivers on the charging pile when passing through the left detection angle region, the middle detection angle region, and the right detection angle region in sequence. Under this obstacle distribution, Figure 3 The robot shown cannot walk in a straight line along a predetermined direction (e.g., along the vertical line extending from the center of the charging pile's front) to the front of the charging pile. Therefore, it can be determined that the robot cannot identify the charging pile directly in front of it. However, the general distribution of obstacles can be determined. Since the straight obstacle traverses the area between the robot and the charging pile, the target laser points extracted from different detection angle areas are different. Figure 2 The target laser point in the embodiment shown is smaller, and the ranging data of the target laser point extracted from the detection angle region of different blocks is smaller.

[0051] In summary, by utilizing the ranging differences between different target laser points and the reception status of the charging guidance signal by the at least two signal receivers, the robot can effectively distinguish the distribution of various obstacles in front of the charging pile, which facilitates the subsequent adaptive adjustment of the robot's return route based on the ranging data of the target laser points.

[0052] This application also discloses a chip storing a computer program, which is the program code corresponding to the charging pile identification method. Based on the foregoing embodiments, the chip controls the robot to scan the surrounding environment using lidar during its movement. Based on the ranging differences between different target laser points and the reception of the charging guidance signal by at least two signal receivers, the chip identifies the charging pile and distinguishes obstacles between the charging pile and the robot, improving the accuracy of charging pile identification. Therefore, during the chip-controlled robot's straight-line docking recharging phase, the ranging data of target laser points within each detection angle area and the reception of the charging guidance signal are used to compensate for the deficiency of the scarce laser point cloud, which cannot fit effective pose information. This optimizes the robot's recharging scenario and forms a recharging processing method that can overcome the influence of the charging pile's surface material.

[0053] When the robot begins walking and directly aligns with the charging pile, the surface material to be identified by the charging pile identification method is either non-reflective or reflective. This results in a sparse distribution of laser points generated by the laser lines emitted by the LiDAR reflecting off the surface material during the scanning process of the charging pile / robot's walking environment. Consequently, the laser point cloud becomes incomplete due to the charging pile's black or reflective surface, especially when the robot and the charging pile are far apart. Therefore, the chip disclosed in this embodiment, by controlling the robot to execute the charging pile identification method disclosed in the aforementioned embodiment, can identify the charging pile even when the laser point cloud is incomplete due to the charging pile's black or reflective surface, preventing misjudgment. It can also continue to complete the recharging function using the at least two signal receivers. Therefore, in the scenario where the robot performs the recharging function, the ranging data of the target laser points in each detection angle area and the reception of the charging guidance signal are used to compensate for the deficiency of the laser point cloud in fitting effective pose information, optimizing the robot's recharging scenario and forming a recharging processing method that overcomes the influence of the charging pile's surface material.

[0054] Based on the foregoing embodiments, this application also discloses a robot. A lidar is mounted on the top of the robot's body, and at least two signal receivers are horizontally distributed on either side of a vertical line at the center of the robot's front. The chip is installed inside the robot. When the robot begins to walk and directly connect with a charging station, the chip executes the computer program to implement the charging station identification method, thereby controlling the robot to identify charging stations and obstacles during its movement, including but not limited to walking to... Figure 2 or Figure 3The scenario depicts the distribution of charging piles and obstacles. Optionally, during the entire recharging process, the robot can construct a grid map using LiDAR to obtain the robot's current pose and the locations of identified obstacles. The center of the charging pile is preset as the origin of the grid map's coordinate system. Correspondingly, the direction of the vertical axis of the grid map's coordinate system can be considered as the direction along which the robot moves during the execution of the charging pile identification method. The robot disclosed in this application, by executing the charging pile identification method, identifies the charging pile and eliminates interference from obstacles based on the ranging differences between different target laser points and the reception of the charging guidance signal by the at least two signal receivers. Thus, during the robot's initial straight-line docking recharging phase, it accurately identifies the charging pile by combining the ranging data extracted by the LiDAR from different areas and the charging guidance signal received by the signal receivers. This allows for the identification of the charging pile even when the laser point cloud is incomplete due to the charging pile's surface being made of black or reflective material, preventing misjudgment. The robot can continue to rely on the at least two signal receivers to complete the recharging function, reducing obstacle misjudgment. Furthermore, it overcomes the shortcomings of relying solely on laser or infrared, improving the reliability and accuracy of the robot's active recharging.

[0055] Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications can still be made to the specific embodiments of the present invention or equivalent substitutions can be made to some technical features without departing from the spirit of the technical solution of the present invention, and all such modifications and substitutions should be covered within the scope of the technical solution claimed in the present invention.

Claims

1. A charging pile recognition method based on a laser radar scan, the charging pile recognition method being used to recognize a charging pile when the charging pile sends out a charging guide signal; characterized in that, The charging pile identification method is executed by a robot equipped with a lidar; at least two signal receivers are installed on the front of the robot to receive charging guidance signals emitted by the charging pile. The charging pile identification method includes: The robot controls the lidar to emit laser lines to form a laser scanning area; Set at least three detection angle areas within the laser scanning area, so that there is at least one detection angle area in front of the robot and on each of its two sides; The ranging data of the target laser points are extracted from each detection angle area; then, based on the ranging differences between different target laser points and the reception of the charging guidance signal by the at least two signal receivers, the charging pile is identified. The method for identifying the charging pile based on the ranging differences between different target laser points and the reception of the charging guidance signal by the at least two signal receivers includes: If the distance difference between target laser points within the detection angle areas of different blocks is within a preset distance error range, and the at least two signal receivers simultaneously receive the charging guidance signal, then the robot is determined to have identified the charging pile and to have the charging pile in front of it, so as to control the robot to walk in a straight line to the front of the charging pile.

2. The charging pile identification method according to claim 1, characterized in that, When the ranging differences between target laser points within the detection angle regions of different blocks are not all within the preset distance error range, the following situation exists: The target laser point in the detection angle area set on one side directly in front of the robot is set as the first target laser point, the target laser point in the detection angle area set on the front of the robot is set as the second target laser point, and the target laser point in the detection angle area set on the other side directly in front of the robot is set as the third target laser point. If the ranging data of the third target laser point and the ranging data of the second target laser point are both greater than the ranging data of the first target laser point, then during the process of the robot walking along the vertical line at the middle position in front of the charging pile, it is determined that there is an obstacle in the detection angle area on the side where the first target laser point is located, so that the robot cannot walk in a straight line to the front of the charging pile, and the robot does not recognize the obstacle as the charging pile.

3. The charging pile identification method according to claim 1, characterized in that, If the at least two signal receivers do not receive the charging guidance signal simultaneously, it is determined that there is an obstacle in front of the robot that prevents the robot from walking in a straight line to the front of the charging pile, and the obstacle blocks part or all of the detection view of the at least two signal receivers.

4. The charging pile identification method according to claim 1, characterized in that, The target laser point within each of the aforementioned detection angle regions is the laser reflection point within that region that is closest to the center of the robot's body or the center of the lidar. Among them, the lidar is used to collect the laser reflection points generated by the reflection on the surface material to be identified and to calculate the distance data of the laser reflection points at each scanning angle, so as to indicate the distance between the surface material to be identified and the robot.

5. The charging pile identification method according to claim 1, characterized in that, The method of setting at least three detection angle regions from the laser scanning area includes: Within the laser scanning area, the target detection angle is set with the direction of the robot's walking direction as the bisector, and at least three detection angle regions are evenly divided within the target detection angle with the center of the laser radar as the vertex; among them, there is one detection angle region that covers the vertical line of the middle position of the front of the robot, and there are at least two other detection angle regions located on both sides of the robot's walking direction. The vertical line in the center of the robot's front extends along the robot's walking direction; the front of the robot is indicated by the robot's walking direction.

6. The charging pile identification method according to claim 5, characterized in that, When the robot is positioned on a vertical line at the center of the front of the charging pile, the target detection angle is the maximum detection angle required for the lidar to scan the charging pile, so that the two laser lines emitted at this maximum detection angle cut through the two sides of the charging pile respectively; wherein, the charging guidance signal emitted by the charging pile covers the vertical line at the center of the front of the charging pile; the front of the charging pile is opposite to the front of the robot; the front of the charging pile is the location for installing the charging structure; The at least three detection angle regions are distributed between the robot and the charging station; The vertical line in the center of the front of the robot is pre-configured to represent one axis of the two-dimensional coordinate system required for robot navigation.

7. The charging pile identification method according to claim 5, characterized in that, The at least three detection angle regions are the left detection angle region, the middle detection angle region, and the right detection angle region; The detection angle area set in front of the robot is the middle detection angle area, which covers the vertical line of the middle position of the front of the robot, so that when the robot walks along the vertical line of the middle position of the front of the charging pile, it can detect obstacles in the middle position of the front of the charging pile or in the corresponding direction. The detection angle area set on the left side of the front of the robot is the left detection angle area. The left detection angle area is located on the left side of the robot's walking direction, so that when the robot walks along the vertical line of the middle position in front of the charging pile, it can detect obstacles on the left side or in the corresponding direction of the charging pile. The detection angle area set on the right side of the robot's front is the right detection angle area. The right detection angle area is located on the right side of the robot's walking direction, so that when the robot walks along the vertical line of the middle position in front of the charging pile, it can detect obstacles on the right side or in the corresponding direction of the charging pile.

8. A chip storing a computer program, characterized in that, The computer program is the program code corresponding to the charging pile identification method according to any one of claims 1 to 7.

9. A robot, characterized in that, The top of the robot is equipped with a lidar, and at least two signal receivers are installed on the front of the robot, which are horizontally distributed on both sides of the vertical line in the middle of the front of the robot. The robot's internal assembly includes the chip described in claim 8; When the robot begins to walk in order to make a straight-line docking with the charging pile, the chip executes the computer program to implement the charging pile identification method according to any one of claims 1 to 7.

Images