Target positioning method and system for AGV

Abstract

The invention discloses a target positioning method and system for an AGV, and relates to the technical field of target positioning, and the method comprises the steps: obtaining point cloud data and a polarization angle; decomposing the polarization angle to generate a dynamic polarization fingerprint, and matching the point cloud data and the dynamic polarization fingerprint with an environmental object polarization feature point of an SMT production line environmental map to generate a successful pairing point set; a Beidou-laser radar cooperative positioning network is established by using the successful pairing point set, AGV indoor and outdoor positioning information is fused, and Beidou-laser radar cooperative positioning coordinates are generated; and simulating an action line of the AGV along the AGV navigation path points in the SMT production line environment map according to the Beidou-laser radar cooperative positioning coordinates and the successful AGV positioning pairing point set. According to the invention, seamless positioning fusion of the AGV in indoor and outdoor environments is realized by establishing the Beidou-laser radar cooperative positioning network.

Description

Technical Field

[0001] The present invention relates to the field of target positioning technology, and in particular to a target positioning method and system for AGV. Background Art

[0002] Automated Guided Vehicles (AGVs) are a crucial component of modern manufacturing and logistics systems. With the continuous advancement of industrial automation, the requirements for AGV positioning accuracy and environmental adaptability are increasing. In their early development, AGVs primarily relied on fixed paths such as magnetic strips or ribbons for navigation, and later evolved to utilize QR codes and laser reflectors for increased flexibility. In recent years, the use of LiDAR scanning to generate point cloud data combined with environmental feature matching has significantly improved positioning accuracy. At the same time, polarization detection technology has begun to be applied to the field of optical sensing, enhancing target recognition capabilities by analyzing the polarization characteristics of light reflected from an object's surface. However, relying solely on geometric information for position alignment faces challenges in complex and changing manufacturing environments.

[0003] Although existing technologies have made certain progress, they are still insufficient when dealing with dynamically changing industrial scenarios, especially in high-precision manufacturing such as SMT production lines. For example, when performing position matching based on point cloud data obtained by a single lidar, the lack of in-depth consideration of the optical properties of the object surface leads to a high mismatch rate when lighting conditions change or there is occlusion. In addition, the lack of an effective fusion mechanism between different indoor and outdoor positioning systems makes it difficult for AGVs to maintain continuous and stable positioning results when crossing indoor and outdoor transition areas, affecting the overall task execution efficiency and safety. Summary of the Invention

[0004] In view of the above existing problems, the inventors proposed the present invention.

[0005] Therefore, the present invention provides a target positioning method for AGV to solve the problem of lack of in-depth consideration of the optical properties of the object surface.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0007] In the first aspect, the present invention provides a target positioning method for AGV, which includes obtaining point cloud data and polarization angles; decomposing the polarization angles to generate dynamic polarization fingerprints, matching the point cloud data and the dynamic polarization fingerprints with the polarization feature points of environmental objects in the SMT production line environment map, generating a set of successful pairing points, and determining the AGV position coordinates; using the successful pairing point set to establish a Beidou-LiDAR collaborative positioning network, integrating the AGV indoor and outdoor positioning information, and generating Beidou-LiDAR collaborative positioning coordinates; based on the Beidou-LiDAR collaborative positioning coordinates and the successful AGV positioning pairing point set, simulating the AGV's movement route along the AGV navigation path points in the SMT production line environment map, generating an optimized navigation path, and guiding the AGV to perform material handling tasks.

[0008] As a preferred solution of the target positioning method for AGV of the present invention, wherein: the polarization angle is decomposed to generate a dynamic polarization fingerprint, specifically as follows:

[0009] Calculate the total polarized reflection energy corresponding to the polarization angle;

[0010] The total polarization reflection energy is decomposed into primary polarization energy and secondary polarization energy through layered coding technology. After normalization, the proportion of primary polarization energy and secondary polarization energy in the total polarization reflection energy is determined.

[0011] The total polarization reflection energy, the main polarization energy ratio, and the secondary polarization energy ratio are combined to generate a dynamic polarization fingerprint.

[0012] As a preferred solution of the target positioning method for AGV described in the present invention, the matching of point cloud data and dynamic polarization fingerprints with polarization feature points of environmental objects in the SMT production line environment map refers to setting a pairing threshold, comparing the point cloud data and dynamic polarization fingerprints with the polarization feature points of environmental objects in the SMT production line environment map, and screening AGV positioning pairing points that meet the pairing threshold.

[0013] As a preferred solution of the target positioning method for AGV described in the present invention, the AGV positioning pairing point is a correspondence between the polarization feature points of environmental objects in the point cloud data and dynamic polarization fingerprint and the polarization feature points of environmental objects in the SMT production line environment map.

[0014] As a preferred solution of the target positioning method for AGV described in the present invention, the establishment of a Beidou-LiDAR collaborative positioning network refers to establishing a Beidou-LiDAR collaborative positioning network by sharing the dynamic polarization fingerprint in the successful AGV positioning pairing point set and the local coordinates of the SMT production line environment map, so as to complete the collaborative positioning of multiple AGVs in indoor and outdoor environments.

[0015] As a preferred embodiment of the target positioning method for AGV of the present invention, the action route of the simulated AGV along the AGV navigation path point in the SMT production line environment map is as follows:

[0016] Initialize the virtual simulation environment, load the AGV navigation path points and polarization feature points of environmental objects in the SMT production line environment map, and form a virtual 3D scene;

[0017] Load the point cloud data and dynamic polarization fingerprints in the successful AGV positioning paired point set into the virtual 3D scene;

[0018] The coordinates of the AGV navigation path points are used as the AGV's action route, and the Beidou-LiDAR collaborative positioning coordinates are set as the starting position of the virtual three-dimensional scene. The AGV is simulated to move along the action route from the Beidou-LiDAR collaborative positioning coordinates until it reaches the target site coordinates.

[0019] As a preferred solution of the target positioning method for AGV of the present invention, wherein: the SMT production line environment map is pre-stored in the central control computer;

[0020] The AGV indoor and outdoor positioning information is the position of the AGV in indoor and outdoor environments.

[0021] In a second aspect, the present invention provides a target positioning system for AGV, comprising:

[0022] Acquisition module, used to obtain point cloud data and polarization angle;

[0023] The matching module is used to decompose the polarization angle, generate a dynamic polarization fingerprint, match the point cloud data and dynamic polarization fingerprint with the polarization feature points of environmental objects in the SMT production line environment map, generate a set of successful matching points, and determine the AGV position coordinates;

[0024] The collaborative module is used to establish a BeiDou-LiDAR collaborative positioning network using a set of successfully paired points, integrate the AGV's indoor and outdoor positioning information, and generate BeiDou-LiDAR collaborative positioning coordinates;

[0025] The simulation module is used to simulate the AGV's movement route along the AGV navigation path points in the SMT production line environment map based on the Beidou-LiDAR collaborative positioning coordinates and the set of successful AGV positioning pairing points, generate an optimized navigation path, and guide the AGV to perform material handling tasks.

[0026] In a third aspect, the present invention provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program is executed by the processor, any step of the target positioning method for AGV as described in the first aspect of the present invention is implemented.

[0027] In a fourth aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, any step of the target positioning method for AGV as described in the first aspect of the present invention is implemented.

[0028] The beneficial effects of the present invention are as follows: the present invention realizes seamless positioning fusion of AGV in indoor and outdoor environments by establishing a Beidou-LiDAR collaborative positioning network. Based on the dynamic polarization fingerprint in the successful pairing point set and the local coordinates of the SMT production line environment map, combined with Beidou satellite positioning information, the coordinate transformation matrix is used to uniformly map the positioning data of different references to the same spatial coordinate system; by performing time synchronization, deviation correction and coordinate fusion on multi-source positioning data, high-precision and highly continuous collaborative positioning coordinates are generated; effectively solving the problems of positioning discontinuity and coordinate offset that exist in the traditional AGV during the indoor and outdoor switching process, improving the positioning stability and operation continuity of the AGV in complex industrial scenarios, and ensuring the autonomous navigation capability and task execution efficiency in mixed environments. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0030] Figure 1 Flowchart of the target positioning method for AGV.

[0031] Figure 2 Schematic diagram of the target positioning system for AGV.

[0032] Figure 3 Flowchart for dynamic polarization fingerprint generation.

[0033] Figure 4 Build a flow chart for the co-location network. DETAILED DESCRIPTION

[0034] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0035] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0036] Secondly, the term "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in various places throughout this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive of other embodiments.

[0037] Reference Figures 1 to 4 , is an embodiment of the present invention, which provides a target positioning method for an AGV, comprising the following steps:

[0038] S1. Obtain point cloud data and polarization angle.

[0039] Furthermore, AGVs (automated guided vehicles) were launched on the SMT production line, equipped with lidar and polarization encoding equipment;

[0040] The LiDAR begins operating, emitting multi-band polarized laser light to illuminate objects within the production area of the SMT production line, such as wall corners, AGV navigation path points, production equipment, and ground markers. The multi-band polarized laser light includes different polarization angles, such as 0 degrees, 45 degrees, and 90 degrees, to generate unique polarization signatures.

[0041] Multi-band polarized lasers illuminate objects within the production workshop area and are reflected back to the lidar;

[0042] The LiDAR receives the reflected signals of multi-band polarized lasers and generates point cloud data and polarization angles. The point cloud data records three-dimensional coordinates, which are the horizontal coordinate, the vertical coordinate, and the height coordinate. The polarization angle records the polarization characteristics of the reflected signals of the multi-band polarized lasers, ranging from 0 to 180 degrees. The point cloud data and polarization angle are used to describe the spatial position and optical characteristics for AGV positioning and map matching.

[0043] The lidar collects point cloud data and polarization angles of multiple points (e.g., 1,000) within a radius (e.g., 10 meters) to ensure coverage of key objects around the AGV, such as AGV navigation path points and wall corners.

[0044] The point cloud data and polarization angle are preliminarily sorted by the internal processor of the lidar to form an unstored temporary data set; the temporary data set contains the horizontal coordinate, vertical coordinate, height coordinate and polarization angle of each AGV point; among them, the preliminary sorting is specifically as follows: first, the internal processor of the lidar processes the reflected signal, extracts the three-dimensional coordinates and polarization angle of each point, and corrects the three-dimensional coordinates based on the installation position and posture of the lidar to ensure that the three-dimensional coordinates are consistent with the actual spatial position; secondly, the three-dimensional coordinates and polarization angles are denoised (such as mean filtering) to eliminate abnormal points due to light interference; then, the polarization angle is normalized to the range of 0 to 180 degrees; finally, the three-dimensional coordinates and polarization angle of each point are associated and organized into a temporary data set.

[0045] The AGV local processor receives the temporary data set and organizes the point cloud data and polarization angles into an environmental feature data set. The environmental feature data set is stored in a table format, with each row recording the horizontal coordinate, vertical coordinate, height coordinate, and polarization angle of a point. The points represent the points in the point cloud data, indicating the location and characteristics of objects in the production workshop area of the SMT production line.

[0046] After the AGV local processor completes the sorting, it saves the environmental feature dataset in the local memory and prepares for transmission. The environmental feature dataset is transmitted from the local memory to the central control computer via the wireless network. The wireless network must ensure that the AP signal is stable before transmission to avoid transmission interruption.

[0047] After receiving the environmental feature data set, the central control computer stores the environmental feature data set in a database on the Windows platform. After storage is completed, the environmental feature data set can be called by the central control computer for subsequent map matching.

[0048] S2. Decompose the polarization angle to generate a dynamic polarization fingerprint. Match the point cloud data and dynamic polarization fingerprint with the polarization feature points of environmental objects in the SMT production line environment map to generate a set of successful matching points and determine the AGV position coordinates.

[0049] Furthermore, the central control computer receives the environmental feature data set via the wireless network and verifies the integrity of the environmental feature data set (e.g., checking the number of points, format, and transmission consistency of the environmental feature data set);

[0050] The central control computer calls the pre-stored SMT production line environment map; the SMT production line environment map is generated through lidar scanning. The scanning process is: collecting point cloud data and polarization angles of all objects in the production workshop area of the SMT production line to form an environmental feature data set, and then extracting key points of the objects (such as wall corners, AGV navigation path points and ground markers) through manual annotation. At the same time, a reference point is determined (for example, the workshop entrance, the three-dimensional coordinates of the workshop entrance are calibrated as the reference point of the SMT production line environment map coordinate system), and the polarization feature points of environmental objects in the SMT production line environment map are generated, including AGV navigation path points and AGV station coordinates within the SMT production line, covering the SMT production line production workshop area.

[0051] The polarization angle is extracted from the environmental feature dataset and decomposed into primary polarization energy and secondary polarization energy using a layered encoding technique to generate a dynamic polarization fingerprint. The details are as follows:

[0052] Calculate the total polarized reflected energy corresponding to the polarization angle of the reflected light from objects in the SMT production line environment and decompose it into primary polarization energy and secondary polarization energy. The expression is:

[0053] E t =I r ·cos 2 (φ);

[0054] E m =β·E t ;

[0055] E s =(1-β)·E t ;

[0056] Among them, E t I is the total energy of polarized reflection corresponding to the polarization angle, in joules (J), that is, the total energy of the polarized light signal reflected back after the multi-band polarized laser emitted by the lidar illuminates the object. r The intensity of light reflected from the surface of objects in the SMT production line environment, expressed in watts per square meter (W / m 2 ), φ is the polarization angle in degrees (°), cos 2 E is the cosine square, which describes the energy contribution of polarized light and has no unit. m It is the dominant polarization energy in the reflected light from objects in the SMT production line environment, reflecting the main polarization direction characteristics of the polarization angle. The unit is joule (J). β is the proportion of the main polarization energy. It has no unit. For example, 0.8 (80%) indicates the proportion of the main polarization energy to the total polarization reflection energy. E s The secondary polarization energy in the light reflected from objects in the SMT production line environment captures the secondary polarization characteristics of the polarization angle (such as scattering), and the unit is joule (J);

[0057] Normalize the primary polarization energy and the secondary polarization energy, and calculate the ratio of the primary polarization energy and the secondary polarization energy to the total polarization reflection energy, generating the expression:

[0058]

[0059] Where ∝ is the fraction of secondary polarization energy, which has no unit. For example, 0.2 means 20%, which indicates the ratio of secondary polarization energy to the total polarized reflection energy.

[0060] The total polarized reflection energy, the primary polarization energy ratio, and the secondary polarization energy ratio corresponding to the polarization angle are directly combined into a polarization feature vector and used as a dynamic polarization fingerprint. The primary polarization energy is the main identification information of the dynamic polarization fingerprint, which is used to improve the accuracy of polarization feature points of environmental objects in the SMT production line environmental map matching, especially under normal lighting conditions. The secondary polarization energy is used to enhance the robustness of the dynamic polarization fingerprint, especially to improve the discrimination of polarization feature points of environmental objects under extreme lighting or dynamic interference.

[0061] The dynamic polarization fingerprint is stored in the environmental feature dataset and associated with the corresponding polarization feature points of the environmental objects. The polarization feature points of the environmental objects refer to the spatial position and optical characteristics of specific objects in the SMT production line environment (such as wall corners, AGV navigation path points, and ground markers). These feature points are described by the point cloud data collected by the lidar and the polarization feature vector.

[0062] The point cloud data and dynamic polarization fingerprints in the environmental feature dataset are compared one by one with the polarization feature points of environmental objects in the SMT production line environment map to determine the position of the AGV in the production workshop area of the SMT production line, as follows:

[0063] Set the pairing threshold based on the LiDAR performance and the SMT production line environment characteristics;

[0064] The Euclidean distance difference of three-dimensional coordinates is less than 0.1 meters;

[0065] The difference in the main polarization energy ratio is less than 0.03 (i.e. 3%);

[0066] The difference in the proportion of sub-polarization energy is less than 0.03 (i.e. 3%);

[0067] Based on the pairing threshold, the three-dimensional coordinate Euclidean distance difference and dynamic polarization fingerprint of each AGV positioning pairing point are compared; if all conditions of the pairing threshold are met, the match is successful, and the AGV positioning pairing point is included in the successful AGV positioning pairing point set; if any condition of the pairing threshold is not met, the match fails, and the operator is prompted to manually calibrate the AGV position through the operation interface, and the reason for the failure (such as "abnormal lighting" or "obstacle occlusion") is recorded; Among them, the Euclidean distance difference refers to the straight-line distance between two points in multidimensional space, which is usually used to measure the spatial proximity between two points. The Euclidean distance difference is an existing and conventional calculation method, widely used in mathematics, computer science and engineering fields; the AGV positioning pairing point is composed of a polarization feature point of an environmental object in the environmental feature dataset and a polarization feature point of an environmental object in the SMT production line environment map, indicating a possible correspondence (for example, the two may point to the same object in the SMT production line production workshop area, such as an AGV navigation path point);

[0068] During the matching process, the environmental feature dataset and the SMT production line environment map generate multiple AGV positioning pairing points. Based on the LiDAR scanning accuracy (e.g., point cloud resolution of approximately 0.01 meters) and the resolution requirements for the SMT production line environmental object feature points (e.g., AGV navigation path point spacing of approximately 0.5 meters), AGV positioning pairing points that meet the matching threshold and have a high matching degree are screened and combined to generate a set of successful AGV positioning pairing points. If multiple AGV positioning pairing points meet the matching threshold, the AGV positioning pairing point with the smallest absolute difference in total polarized reflection energy is prioritized to improve matching reliability. The matching degree is a comprehensive score of the AGV positioning pairing points, ranging from 0 to 1. A higher score indicates that the AGV positioning pairing point is more likely to correspond to the same object in the SMT production line workshop area.

[0069] The expression of the total polarized reflection energy difference is:

[0070] ΔE t =|E t,d -E t,o |;

[0071] Where ΔE t E is the absolute difference in the total polarization reflection energy of the two polarization feature points of the environmental objects in the AGV positioning pairing point. t,d is the total polarization reflection energy of the polarization feature points of environmental objects in the environmental feature dataset, E t,o is the total polarization reflection energy of the polarization feature points of environmental objects in the SMT production line environment map, and d is the environmental feature dataset;

[0072] The matching expression is:

[0073]

[0074] Where S is the matching score, D is the Euclidean distance difference (unit: meter), w1 is the unitless weight of the Euclidean distance difference, w2 is the unitless weight of the primary polarization energy ratio difference, which indicates the contribution of the primary polarization energy ratio difference in the dynamic polarization fingerprint to the matching score, and w3 is the unitless weight of the secondary polarization energy ratio difference, which indicates the contribution of the secondary polarization energy ratio difference in the dynamic polarization fingerprint to the matching score.

[0075] The 3D coordinates of the polarization feature points of the environmental objects in the SMT production line environment map in the set of successful AGV positioning pairing points are calculated by geometric weighted average to determine the AGV center position as the AGV position coordinates;

[0076] The AGV position coordinates are displayed through the interface for the operator to view in the form of two-dimensional or three-dimensional coordinate points. The AGV position coordinates are then stored in the Windows database as the final output of the laser polarization coding map matching.

[0077] S3. Use the set of successful paired points to establish a Beidou-LiDAR collaborative positioning network, integrate the AGV indoor and outdoor positioning information, and generate Beidou-LiDAR collaborative positioning coordinates.

[0078] Furthermore, the central control computer obtains the AGV position coordinates from the Windows database as the starting coordinates for AGV path planning;

[0079] The AGV receives Beidou satellite signals through Beidou positioning equipment, generates Beidou positioning coordinates, and transmits them to the central control computer via a wireless network;

[0080] The central control computer verifies the integrity of the Beidou positioning coordinates, checks their format, timestamp, and signal strength; if the check passes, it is stored in the Windows database; if the check fails, it records the reason (such as "Beidou satellite signal is weak" or "data format error") and prompts the user interface to re-acquire the Beidou satellite signal;

[0081] The central control computer converts the Beidou positioning coordinates (latitude, longitude and altitude) into Beidou positioning projection coordinates (abscissa, ordinate and altitude coordinates, in meters) through the Mercator projection method. Specifically, based on the earth coordinate system, the units of latitude and longitude are first converted from degrees to radians. Then, based on the origin of the earth coordinate system and the equatorial plane, the radian value of the latitude is mapped to the ordinate on the Beidou positioning coordinate projection plane, and the surface position of the earth corresponding to the latitude is expanded into a plane position to form the ordinate value (in meters). The radian value of the longitude is mapped to the abscissa on the Beidou positioning coordinate projection plane, and the surface position of the earth corresponding to the longitude is expanded into a plane position to form the abscissa value (in meters). The altitude remains unchanged and is directly used as the altitude coordinate value (in meters).

[0082] Based on the reference points of the SMT production line environment map, the preset coordinate conversion matrix is used to convert the Beidou positioning projection coordinates into the local coordinates of the SMT production line environment map;

[0083] The preset coordinate transformation matrix is expressed as:

[0084]

[0085] Among them, x is the horizontal coordinate in the local coordinate system after transformation (in meters), corresponding to the X-axis position of the SMT production line environment map, y is the vertical coordinate in the local coordinate system after transformation (in meters), corresponding to the Y-axis position of the SMT production line environment map, z is the height coordinate in the local coordinate system after transformation (in meters), corresponding to the Z-axis position of the SMT production line environment map, 1 is the constant term of the homogeneous coordinate, which is used for the standardized form of coordinate transformation matrix operation, so that the rotation and translation transformation can be completed by a single coordinate transformation matrix. 11 , R 12 , R 13 is the first row component of the rotation matrix, R 21 , R 22 , R 23 is the second row component of the rotation matrix, R 31 , R 32 , R 33 It is the third row component of the rotation matrix. R is a 3x3 rotation matrix, which represents the direction alignment transformation from the Beidou positioning projection coordinates to the local coordinate system of the SMT production line environment map. T x is the translation component in the horizontal direction, T y is the translation component in the ordinate direction, T z The translation component in the height coordinate direction is determined by the reference point of the SMT production line environment map and is used to adjust the origin of the Beidou positioning projection coordinate and align the origin of the local coordinate system of the SMT production line environment map. pis the horizontal coordinate of Beidou positioning projection coordinate (in meters), p is the vertical coordinate of Beidou positioning projection coordinate (in meters), z p Beidou positioning projection coordinate height coordinate (in meters);

[0086] The central control computer obtains the set of successful AGV positioning pairing points from the Windows database;

[0087] Based on the AGV position coordinates, the local coordinates of the SMT production line environment map, and the set of successful AGV positioning pairing points, a BeiDou-LiDAR collaborative positioning network is established to generate the BeiDou-LiDAR collaborative positioning coordinates, as follows:

[0088] In indoor environments, AGV position coordinates are preferred. The positioning deviation of the AGV is corrected by the dynamic polarization fingerprint of the positioning pairing point set to maintain high accuracy of the AGV position coordinates.

[0089] In outdoor or indoor-outdoor transition areas, local coordinates of the SMT production line environment map are preferred, and the longitude and latitude information of the Beidou positioning coordinates are used to expand the positioning coverage of the AGV;

[0090] Multiple AGVs share the dynamic polarization fingerprint of the successful AGV positioning pairing point set and the local coordinates of the SMT production line environment map through the wireless network. Together with the timestamp of each AGV and each AGV identification (AGV unique number, such as "AGV001"), a real-time positioning cloud is formed to dynamically reflect the positioning status of all AGVs and SMT production line environment information;

[0091] The central control computer processes the real-time positioning cloud shared by multiple AGVs, and generates Beidou-LiDAR collaborative positioning coordinates through synchronization, fusion, correction and aggregation; specifically: first, according to the timestamp of each AGV, the dynamic polarization fingerprint and the local coordinates of the SMT production line environment map in the real-time positioning cloud are synchronized to ensure that the dynamic polarization fingerprint and the local coordinates of the SMT production line environment map of each AGV reflect the latest positioning status of each AGV and the latest environmental information of the SMT production line production workshop area; then, each AGV applies the AGV position coordinates and the local coordinates of the SMT production line environment map, selects the main coordinate source according to the environment where the AGV is located (indoor or outdoor), and generates preliminary positioning coordinates; then, the dynamic polarization fingerprint of the successful AGV positioning pairing point set is used to match the polarization feature points of the environmental objects in the SMT production line environment map to correct the deviation of the preliminary positioning coordinates, such as the coordinate offset caused by weak Beidou signal or LiDAR occlusion; finally, the corrected preliminary positioning coordinates of all AGVs are summarized to generate Beidou-LiDAR collaborative positioning coordinates and store them in the Windows database.

[0092] S4. Based on the Beidou-LiDAR collaborative positioning coordinates and the set of successful AGV positioning pairing points, simulate the AGV's movement route along the AGV navigation path points in the SMT production line environment map, generate an optimized navigation path, and guide the AGV to perform material handling tasks.

[0093] Furthermore, the BeiDou-LiDAR collaborative positioning coordinates, the set of successful AGV positioning pairing points, the AGV navigation path points in the SMT production line environment map, and the AGV station coordinates are read from the Windows database;

[0094] By integrating point cloud data, dynamic polarization fingerprints, and BeiDou-LiDAR collaborative positioning coordinates through virtual simulation technology, the AGV's movement path along the AGV navigation path points is simulated. The specific operations are as follows:

[0095] The central control computer initializes the virtual simulation environment based on the SMT production line environment map, loads the polarization feature points of environmental objects, AGV navigation path points, and AGV station coordinates within the SMT production line into the virtual simulation environment, and forms a virtual 3D scene that includes the fixed object layout of the SMT production workshop area.

[0096] The point cloud data from the successful AGV positioning paired point set is loaded into the virtual 3D scene. The point cloud data reflects the position information of objects in the SMT production line workshop area. The central control computer reconstructs the position of objects in the SMT production line workshop area in the virtual 3D scene based on the point cloud data to form the distribution of environmental objects.

[0097] The dynamic polarization fingerprint from the set of successful AGV positioning pairing points is applied to the distributed environmental objects. The dynamic polarization fingerprint describes the optical characteristics of the environmental objects. The central control computer marks the object optical characteristic information for each environmental object, such as the reflectance characteristics of the AGV navigation path points, for subsequent path interference prediction.

[0098] The AGV navigation path points in the virtual 3D scene are used as the AGV's action route. The action route represents the AGV's planned movement trajectory from the Beidou-LiDAR collaborative positioning coordinates to the AGV's target station. The AGV's target station is derived from the AGV station coordinates within the SMT production line. The planned movement trajectory is generated through virtual simulation based on the AGV navigation path points.

[0099] Set the BeiDou-LiDAR collaborative positioning coordinates to the starting position of the AGV in the virtual 3D scene to ensure that the simulation starts from the current position of the AGV;

[0100] Set the virtual 3D scene parameters based on the SMT production line environment map. The virtual 3D scene parameters include the boundary range of the SMT production line workshop area and the positions of fixed objects to ensure that the virtual 3D scene is consistent with the actual SMT production line workshop area. The boundary range is, for example, from the coordinates of the origin of the SMT production line workshop entrance (southwest corner) to the coordinates of the farthest end of the SMT production line workshop area (northeast corner).

[0101] According to the dynamic simulation requirements, dynamic change factors in the virtual 3D scene are set. The dynamic simulation requirements come from the operating characteristics of the SMT production line production workshop area and the AGV operating rules. The operating characteristics include the possible presence of mobile personnel or temporarily stacked materials in the SMT production line production workshop. The AGV operating rules include restrictions on the occupation of AGV navigation path points. For example, when an AGV navigation path point is occupied by another AGV, it must wait. Dynamic change factors include the possible presence of mobile personnel, temporarily stacked materials, and restrictions on the occupation of AGV navigation path points in the SMT production line production workshop area.

[0102] The outdoor scene parameters (outdoor boundary range, fixed object position, terrain features, AGV navigation path points and dynamic environmental factors) are set according to the Beidou positioning coordinates. The Beidou positioning coordinates reflect the positioning information of the AGV in the outdoor or transition area. The central control computer marks the scene range of the outdoor area in the virtual 3D scene to ensure that the simulation adapts to the mixed indoor and outdoor environment.

[0103] The central control computer starts virtual simulation and sets a simulation time (such as 1 hour). It gradually advances the AGV's movement trajectory according to the set simulation time and generates the AGV's dynamic movement path.

[0104] The central control computer analyzes potential problems in the AGV's dynamic movement path and generates simulation results. The potential problems include: detecting the positions of mobile personnel and temporarily stacked materials in the SMT production line workshop area based on point cloud data, determining whether the mobile personnel and temporarily stacked materials in the SMT production line workshop area will block the AGV navigation path points, identifying changes in the reflective characteristics of the AGV navigation path points based on changes in dynamic polarization fingerprints, and determining whether the AGV navigation path points are unavailable due to occlusion, which includes AGV navigation path points being covered by temporarily stacked materials. Analyzing the movement trajectory of the outdoor area based on Beidou positioning coordinates to determine whether the path in the outdoor area is blocked by the impact of stacked objects.

[0105] The central control computer adjusts the coordinates of the AGV navigation path points based on the simulation results, avoiding moving personnel, temporarily stacked materials and stacked objects in the SMT production line production workshop area, and generates an optimized navigation path to ensure the efficient and safe movement of the AGV from the Beidou-LiDAR collaborative positioning coordinates to the AGV target site coordinates. The optimized navigation path is stored in the Windows database to complete the virtual simulation.

[0106] The virtual simulation AGV target site coordinates are transmitted to the AGV local processor via a wireless network. The AGV moves to the AGV target site coordinates according to the optimized navigation path, performs the material handling task, and provides real-time feedback on the material handling task status and Beidou-LiDAR collaborative positioning coordinates. The material handling task status and Beidou-LiDAR collaborative positioning coordinates are transmitted to the central control computer and stored in the Windows database for real-time monitoring and subsequent query by the operator;

[0107] The operation interface displays the material handling task status, Beidou-LiDAR collaborative positioning coordinates, navigation path status and AGV status; the material handling task status includes task in progress and task completed; the navigation path status includes the current AGV navigation path point and the distance to the AGV target site; the AGV status includes idle, material handling task in progress and charging. If the material handling task fails, the operator is prompted to intervene, the failure reason is recorded and stored in the Windows database; if the AGV deviates from the navigation path, the operator switches to manual mode through the manual operation interface, controls the AGV to return to the AGV navigation path point, and restores automatic mode.

[0108] This embodiment also provides a target positioning system for an AGV, including:

[0109] Get point cloud data and polarization angle;

[0110] The polarization angle is decomposed to generate a dynamic polarization fingerprint. The point cloud data and dynamic polarization fingerprint are matched with the polarization feature points of environmental objects in the SMT production line environment map to generate a set of successful matching points and determine the AGV position coordinates.

[0111] A BeiDou-LiDAR collaborative positioning network is established using the set of successfully paired points, integrating the AGV's indoor and outdoor positioning information to generate BeiDou-LiDAR collaborative positioning coordinates.

[0112] Based on the Beidou-LiDAR collaborative positioning coordinates and the set of successful AGV positioning pairing points, the AGV's movement route along the AGV navigation path points in the SMT production line environment map is simulated to generate an optimized navigation path to guide the AGV to perform material handling tasks.

[0113] This embodiment also provides a computer device suitable for the target positioning method of AGV, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute computer-executable instructions to implement the target positioning method for AGV proposed in the above embodiment.

[0114] The computer device may be a terminal, comprising a processor, a memory, a communication interface, a display screen and an input device connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device comprises a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner may be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. The display screen of the computer device may be a liquid crystal display or an electronic ink display screen, and the input device of the computer device may be a touch layer covering the display screen, or a button, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse.

[0115] This embodiment also provides a storage medium having a computer program stored thereon, which, when executed by a processor, implements the target positioning method for AGV proposed in the above embodiment; the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, disk or optical disk.

[0116] In summary, the present invention achieves seamless positioning fusion of AGVs in indoor and outdoor environments by establishing a Beidou-LiDAR collaborative positioning network. Based on the dynamic polarization fingerprint in the successful pairing point set and the local coordinates of the SMT production line environment map, combined with Beidou satellite positioning information, the coordinate transformation matrix is used to uniformly map the positioning data of different references to the same spatial coordinate system; by performing time synchronization, deviation correction and coordinate fusion on multi-source positioning data, high-precision and highly continuous collaborative positioning coordinates are generated; effectively solving the problems of positioning interruption and coordinate offset that exist in the traditional AGV during the indoor and outdoor switching process, improving the positioning stability and operation continuity of AGVs in complex industrial scenarios, and ensuring autonomous navigation capabilities and task execution efficiency in mixed environments.

[0117] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A target positioning method for AGV, characterized by: include, Get point cloud data and polarization angle; The polarization angle is decomposed to generate a dynamic polarization fingerprint. The point cloud data and dynamic polarization fingerprint are matched with the polarization feature points of environmental objects in the SMT production line environment map to generate a set of successful matching points and determine the AGV position coordinates. A BeiDou-LiDAR collaborative positioning network is established using the set of successfully paired points, integrating the AGV's indoor and outdoor positioning information to generate BeiDou-LiDAR collaborative positioning coordinates. Based on the Beidou-LiDAR collaborative positioning coordinates and the set of successful AGV positioning pairing points, the AGV's movement route along the AGV navigation path points in the SMT production line environment map is simulated to generate an optimized navigation path to guide the AGV to perform material handling tasks.

2. The target positioning method for AGV according to claim 1, characterized in that: The polarization angle is decomposed to generate a dynamic polarization fingerprint, as follows: Calculate the total polarized reflection energy corresponding to the polarization angle; The total polarization reflection energy is decomposed into primary polarization energy and secondary polarization energy through layered coding technology. After normalization, the proportion of primary polarization energy and secondary polarization energy in the total polarization reflection energy is determined. The total polarization reflection energy, the main polarization energy ratio, and the secondary polarization energy ratio are combined to generate a dynamic polarization fingerprint.

3. The target positioning method for AGV according to claim 1, characterized in that: Matching the point cloud data and dynamic polarization fingerprint with the polarization feature points of environmental objects in the SMT production line environment map refers to setting a pairing threshold, comparing the point cloud data and dynamic polarization fingerprint with the polarization feature points of environmental objects in the SMT production line environment map, and screening AGV positioning pairing points that meet the pairing threshold.

4. The target positioning method for AGV according to claim 3, characterized in that: The AGV positioning pairing point is a correspondence between the polarization feature points of the environmental objects in the point cloud data and the dynamic polarization fingerprint and the polarization feature points of the environmental objects in the SMT production line environment map.

5. The target positioning method for AGV according to claim 1, characterized in that: The establishment of the Beidou-LiDAR collaborative positioning network refers to establishing a Beidou-LiDAR collaborative positioning network by sharing the dynamic polarization fingerprint in the successful AGV positioning pairing point set and the local coordinates of the SMT production line environment map, and completing the collaborative positioning of multiple AGVs in indoor and outdoor environments.

6. The target positioning method for AGV according to claim 1, characterized in that: The simulated AGV's action route along the AGV navigation path points in the SMT production line environment map is as follows: Initialize the virtual simulation environment, load the AGV navigation path points and polarization feature points of environmental objects in the SMT production line environment map, and form a virtual 3D scene; Load the point cloud data and dynamic polarization fingerprints in the successful AGV positioning paired point set into the virtual 3D scene; The coordinates of the AGV navigation path points are used as the AGV's action route, and the Beidou-LiDAR collaborative positioning coordinates are set as the starting position of the virtual three-dimensional scene. The AGV is simulated to move along the action route from the Beidou-LiDAR collaborative positioning coordinates until it reaches the target site coordinates.

7. The target positioning method for AGV according to claim 1, characterized in that: The SMT production line environment map is pre-stored in the central control computer; The AGV indoor and outdoor positioning information is the position of the AGV in indoor and outdoor environments.

8. A target positioning system for an AGV, based on the target positioning method for an AGV according to any one of claims 1 to 7, characterized in that: include, Acquisition module, used to obtain point cloud data and polarization angle; The matching module is used to decompose the polarization angle, generate a dynamic polarization fingerprint, match the point cloud data and dynamic polarization fingerprint with the polarization feature points of environmental objects in the SMT production line environment map, generate a set of successful matching points, and determine the AGV position coordinates; The collaborative module is used to establish a BeiDou-LiDAR collaborative positioning network using a set of successfully paired points, integrate the AGV's indoor and outdoor positioning information, and generate BeiDou-LiDAR collaborative positioning coordinates; The simulation module is used to simulate the AGV's movement route along the AGV navigation path points in the SMT production line environment map based on the Beidou-LiDAR collaborative positioning coordinates and the set of successful AGV positioning pairing points, generate an optimized navigation path, and guide the AGV to perform material handling tasks.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the target positioning method for AGV according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the target positioning method for AGV according to any one of claims 1 to 7 are implemented.

Images