Automatic guided vehicle path optimization method and device based on laser navigation

Abstract

The invention provides an automatic guided vehicle path optimization method and device based on laser navigation. The method comprises the following steps: acquiring real-time reference point scanning data and real-time environment scanning data through a laser radar; carrying out compensation processing on the real-time scanning data to obtain positioning coordinate data of an automatic guided vehicle (AGV); obtaining dynamic map data according to the real-time environment scanning data, the AGV positioning coordinate data and preset map data; obtaining an initial optimization path according to the AGV positioning coordinate data, the dynamic map data and preset target coordinate data; and smoothing the initial optimized path to obtain a target optimized path. According to the invention, the efficiency and accuracy of path optimization of the AGV can be improved.

Description

Technical Field

[0001] This invention relates to the field of path optimization technology, and also to a method and apparatus for optimizing the path of an automated guided vehicle based on laser navigation. Background Technology

[0002] Automated Guided Vehicles (AGVs) are core material handling equipment in smart warehousing, flexible manufacturing, and logistics sorting scenarios. Their navigation accuracy and path efficiency directly determine the operational efficiency of the entire automation system. Laser navigation, with its advantages of high positioning accuracy (typically ±10mm), strong environmental adaptability (unaffected by light or color), and no need for pre-set tracks (flexible layout), has become the mainstream navigation method for mid-to-high-end AGVs. However, current path optimization methods for laser-guided AGVs do not fully integrate the positioning characteristics of laser navigation, resulting in limited optimization accuracy, poor path smoothness, and inability to adapt to complex scenarios. Summary of the Invention

[0003] The technical problem to be solved by the present invention is to provide a method and apparatus for optimizing the path of an automated guided vehicle based on laser navigation, so as to improve the smoothness of the path and the accuracy of the path optimization.

[0004] To solve the above-mentioned technical problems, the technical solution of the present invention is as follows: A first aspect of the present invention provides a method for optimizing the path of an automated guided vehicle (AGV) based on laser navigation, comprising: Real-time reference point scanning data and real-time environmental scanning data are acquired using lidar. By performing compensation processing on the real-time scanning data, the positioning coordinate data of the automated guided vehicle (AGV) is obtained. Dynamic map data is obtained based on the real-time environmental scanning data, the AGV positioning coordinate data, and the preset map data; Based on the AGV positioning coordinate data, the dynamic map data, and the preset target coordinate data, an initial optimized path is obtained; The initial optimized path is smoothed to obtain the target optimized path.

[0005] Optionally, real-time reference point scanning data and real-time environmental scanning data can be acquired via LiDAR, including: Real-time reference point scanning data is acquired using a lidar; the real-time reference point scanning data includes the distance values ​​between the lidar and multiple preset reference points. Real-time environmental scanning data is acquired using lidar; the real-time environmental scanning data includes the coordinates of static obstacles, the coordinates of dynamic obstacles, and the movement speed of dynamic obstacles.

[0006] Optionally, by performing compensation processing on the real-time scanning data, the positioning coordinate data of the automated guided vehicle (AGV) is obtained, including: pass Obtain compensation data; among which, To compensate for the i-th distance value in the data, Let k be the i-th distance value in the real-time scan data, and k be the environmental correction coefficient. The distance error to the i-th reference point; The compensation data is fitted to obtain the positioning coordinate data of the automated guided vehicle (AGV).

[0007] Optionally, dynamic map data is obtained based on the real-time environmental scanning data, the AGV positioning coordinate data, and the preset map data, including: Acquire preset map data; the preset map data is a global raster map of a preset scene; The preset map data is initialized based on the AGV positioning coordinate data to obtain an initialized grid map; Obstacle marking is performed on the initial grid map based on the real-time environmental scan data to obtain dynamic map data.

[0008] Optionally, based on the AGV positioning coordinate data, the dynamic map data, and the preset target coordinate data, an initial optimized path is obtained, including: Initial node data is obtained based on the AGV positioning coordinate data, the dynamic map data, and the preset target coordinate data; The initial node data is pruned to obtain the target node data; An initial optimized path is obtained based on the target node data and the dynamic map data.

[0009] Optionally, initial node data is obtained based on the AGV positioning coordinate data and preset target coordinate data, including: Candidate nodes are obtained by randomly sampling based on the dynamic map data; Based on the candidate nodes and the preset search radius, the surrounding nodes are obtained; pass Obtain the total path cost; where Cost is the total path cost. , , All are dynamic weighting factors, where L is the total path length. The average steering angle of the path, For obstacle avoidance safety redundancy; The initial node data is obtained based on the total path cost and the surrounding nodes.

[0010] Optionally, the initial optimized path is smoothed to obtain the target optimized path, including: pass The fitted coordinates are obtained; where, To fit the coordinates, n is the number of nodes in the initial optimized path minus 1. The node coordinates are for the initial optimized path. The basis functions are 3rd order B-spline functions; The target optimized path is obtained based on the fitted coordinates.

[0011] A second aspect of the present invention provides a laser-guided vehicle path optimization device, comprising: The acquisition module is used to acquire real-time reference point scanning data and real-time environmental scanning data through LiDAR; The processing module is used to obtain AGV positioning coordinate data by compensating the real-time scanning data; to obtain dynamic map data based on the real-time environmental scanning data, the AGV positioning coordinate data, and preset map data; to obtain an initial optimized path based on the AGV positioning coordinate data, the dynamic map data, and preset target coordinate data; and to obtain a target optimized path by smoothing the initial optimized path.

[0012] A third aspect of the present invention provides a computing device, comprising: a processor and a memory storing a computer program, wherein the computer program, when executed by the processor, performs the method described in the first aspect.

[0013] A fourth aspect of the present invention provides a computer-readable storage medium storing instructions that, when executed on a computer, cause the computer to perform the method as described in the first aspect.

[0014] The above-described solution of the present invention has at least the following beneficial effects: The above-described solution of the present invention acquires real-time reference point scanning data and real-time environmental scanning data using a lidar system. By compensating the real-time scanning data, it obtains AGV positioning coordinate data. Based on the real-time environmental scanning data, the AGV positioning coordinate data, and preset map data, it obtains dynamic map data. Based on the AGV positioning coordinate data, the dynamic map data, and preset target coordinate data, it obtains an initial optimized path. Finally, it smooths the initial optimized path to obtain the target optimized path, thereby improving the efficiency and accuracy of AGV path optimization. Attached Figure Description

[0015] Figure 1 This is a flowchart illustrating the laser navigation-based automated guided vehicle path optimization method in an embodiment of the present invention. Figure 2 This is a schematic diagram of the structure of the laser-guided vehicle path optimization device in an embodiment of the present invention. Detailed Implementation

[0016] Exemplary embodiments of the invention will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the invention are shown in the drawings, it should be understood that the invention may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this invention will be thorough and complete, and will fully convey the scope of the invention to those skilled in the art.

[0017] like Figure 1 As shown, an embodiment of the present invention proposes a laser-guided vehicle path optimization method, comprising the following steps: Step 101: Acquire real-time reference point scanning data and real-time environmental scanning data using lidar; Step 102: By performing compensation processing on the real-time scanning data, the positioning coordinate data of the automated guided vehicle (AGV) is obtained. Step 103: Obtain dynamic map data based on the real-time environmental scanning data, the AGV positioning coordinate data, and the preset map data; Step 104: Obtain the initial optimized path based on the AGV positioning coordinate data, the dynamic map data, and the preset target coordinate data; Step 105: Smooth the initial optimized path to obtain the target optimized path.

[0018] The laser-guided vehicle path optimization method of this invention acquires real-time reference point scanning data and real-time environmental scanning data using a laser radar. By compensating the real-time scanning data, AGV positioning coordinate data is obtained. Based on the real-time environmental scanning data, the AGV positioning coordinate data, and preset map data, dynamic map data is obtained. Based on the AGV positioning coordinate data, the dynamic map data, and preset target coordinate data, an initial optimized path is obtained. Finally, the initial optimized path is smoothed to obtain the target optimized path. This method can improve the efficiency and accuracy of AGV path optimization.

[0019] In an optional embodiment of the present invention, step 101, acquiring real-time reference point scanning data and real-time environmental scanning data via lidar, may include: Step 1011: Acquire real-time reference point scanning data using a lidar; the real-time reference point scanning data includes the distance values ​​between the lidar and multiple preset reference points; Specifically, the original distance values ​​between the laser radar and multiple (at least 3, to ensure positioning accuracy) preset reference points (reflectors / natural features) are obtained; the real-time reference point scanning data can also include the preset coordinates of the preset reference points (such as reflectors) in the AGV global map, the angle between the laser radar and each reference point, etc.

[0020] Step 1012: Acquire real-time environmental scanning data using lidar; the real-time environmental scanning data includes static obstacle coordinates, dynamic obstacle coordinates, and the movement speed of dynamic obstacles.

[0021] Specifically, LiDAR is used to obtain the coordinates of static obstacles (such as shelves and columns), the coordinates of dynamic obstacles (such as people and temporary goods), and the movement speed of dynamic obstacles, providing a data foundation for subsequent path optimization.

[0022] In an optional embodiment of the present invention, step 102, obtaining the AGV positioning coordinate data by compensating the real-time scanning data, may include: Step 1021, through Obtain compensation data; among which, To compensate for the i-th distance value in the data, Let k be the i-th distance value in the real-time scan data, and k be the environmental correction coefficient. The distance error to the i-th reference point; Specifically, the environmental correction coefficient k can be determined according to the specific application scenario, and its value ranges from 1 to 1.05; the distance error of the i-th reference point The value range is from -0.002 to 0.002m.

[0023] Step 1022: Fit the compensation data to obtain the positioning coordinate data of the automated guided vehicle (AGV).

[0024] Specifically, it can be done by... The solution is performed to obtain the high-precision positioning coordinates of the AGV, i.e., the AGV positioning coordinate data. Among them, Provide the AGV's own location coordinates. Let i be the preset coordinates of the i-th reference point (e.g., a reflector). The distance value is the i-th distance value in the real-time scan data, and n is the number of preset reference points. The more preset reference points there are, the higher the positioning accuracy. For example, for a warehouse scenario, the value can be 4, 5, or 6.

[0025] In an optional embodiment of the present invention, step 103, obtaining dynamic map data based on the real-time environmental scanning data, the AGV positioning coordinate data, and the preset map data, may include: Step 1031: Obtain preset map data; the preset map data is a global raster map of a preset scene; Specifically, the preset scene can be a warehousing or manufacturing scene, and the size of the grid map is 0.1m×0.1m to fit the size of the AGV.

[0026] Step 1032: Initialize the preset map data according to the AGV positioning coordinate data to obtain an initialized grid map; Specifically, based on preset map data and combined with AGV positioning coordinate data, the initial grid position of the AGV in the preset map data is determined. (The mapping relationship between raster coordinates and Cartesian coordinates is:) , Where x and y are Cartesian coordinates, and i and j are the raster row and column numbers (rounded down). Simultaneously, raster attributes are initialized, with all rasters marked as "unrecognized rasters" by default. Three core attribute identifiers are reserved: free rasters (drivable), obstacle rasters (non-drivable), and dynamic obstacle prediction rasters (the movement range of dynamic obstacles within the next 1 second), resulting in an initialized raster map.

[0027] Step 1033: Mark obstacles on the initialized grid map based on the real-time environmental scan data to obtain dynamic map data.

[0028] Specifically, static obstacles (such as shelves, columns, and objects whose position remains unchanged for 300ms in a LiDAR scan, and whose speed...) are filtered out from real-time environmental scanning data. (target), its global coordinates Convert to raster coordinates The grid and its surrounding grid (for redundancy and to prevent AGV from scratching) are marked as "obstacle grids". At the same time, the grid range of static obstacles is recorded to form a set of static obstacle grids. No remarking is required when no new static obstacles appear.

[0029] Filtering dynamic obstacles (such as people, temporary cargo, speed) in real-time environmental scan data For targets whose positions change within 300ms, first convert their current global coordinates to grid coordinates and mark them as "obstacle grids"; then, based on the current velocity of the dynamic obstacles... The system predicts the direction of movement and the range of movement within the next second, and marks all grids corresponding to the predicted points as "dynamic obstacle prediction grids" to avoid AGV driving conflicts in advance.

[0030] After excluding unidentified grids, obstacle grids, and dynamic obstacle prediction grids, the remaining grids are marked as "free grids". Free grids in narrow passage areas are marked with special emphasis (when the passage width is ≤ AGV vehicle width + 0.2m, they are marked as "narrow free grids" and should be avoided or slowed down during subsequent path planning).

[0031] By marking static obstacles, dynamic obstacles, and free grids as described above, dynamic map data is obtained.

[0032] In an optional embodiment of the present invention, step 104, obtaining an initial optimized path based on the AGV positioning coordinate data, the dynamic map data, and the preset target coordinate data, may include: Step 1041: Obtain initial node data based on the AGV positioning coordinate data, the dynamic map data, and the preset target coordinate data; In an optional embodiment of the present invention, step 1041 includes: Step 10411: Randomly sample the dynamic map data to obtain candidate nodes; Specifically, within the free grid range of the dynamic map data, a candidate node is randomly generated and converted into grid coordinates. Then, the grid attributes corresponding to the candidate node are determined. If it is a free grid and meets the obstacle avoidance constraint (the distance to the surrounding obstacle grid and the dynamic obstacle prediction grid is ≥ the safe distance), it is determined as a candidate node. If it is an obstacle grid, a dynamic obstacle prediction grid, or does not meet the obstacle avoidance constraint, the candidate node is discarded and resampling is performed until a candidate node is generated (the upper limit of the number of samplings is set to 50 times to avoid infinite sampling affecting real-time performance and to adapt to the requirement of AGV path replanning ≤200ms).

[0033] Step 10412: Obtain surrounding nodes based on the candidate nodes and the preset search radius; Specifically, the surrounding nodes are found using a preset search radius.

[0034] Step 10413, through Obtain the total path cost; where Cost is the total path cost. , , All are dynamic weighting factors, where L is the total path length. The average steering angle of the path, For obstacle avoidance safety redundancy; Specifically, in high-frequency handling scenarios in warehousing, The value is 0.5 (path length has the highest weight). The value is 0.3 (smoothness). The value is 0.2 (safety); in narrow passage scenarios, The value is 0.3. The value is 0.2. A value of 0.5 maximizes the security weight. The total path length is determined by... The average steering angle of the path was obtained. This reflects the smoothness of the path; through Where m is the number of nodes in the path (by searching for surrounding nodes, multiple path branches from the AGV to the target node can be found, and each path branch has multiple nodes). Let be the coordinates of the k-th path node. Let be the coordinates of the (k+1)th path node. Let the driving direction angle be the one at the k-th path node. Let the driving direction angle be the (k+1)th path node. The straight-line distance between the center of the AGV and the center of the obstacle. Safety distance for AGVs. Obstacle avoidance safety redundancy. The larger the value, the higher the security; therefore, taking its reciprocal ensures that higher security comes at a lower cost. Step 10414: Obtain initial node data based on the total path cost and the surrounding nodes.

[0035] Specifically, by calculating the total path cost of each surrounding node, the surrounding node with the smallest total path cost is taken as the parent node. The sampling-search-update process is repeated until a path to the target coordinates is generated. All parent nodes obtained in this process are used as initial node data.

[0036] Step 1042: The initial node data is pruned to obtain the target node data; Specifically, if three consecutive nodes in the initial node data satisfy... and If intermediate nodes are removed, the number of path nodes is reduced, improving path smoothness and reducing the computational cost of subsequent path smoothing. The target node data is obtained by reducing the initial node data in the above manner. Here, l is the sampling step size. This is the maximum steering angle, ranging from 30° to 45°.

[0037] Step 1043: Obtain the initial optimized path based on the target node data and the dynamic map data.

[0038] Specifically, all nodes in the target node data are sequentially concatenated to form a complete path from the starting coordinates to the target coordinates, and then mapped onto the dynamic map data to obtain the initial optimized path.

[0039] In an optional embodiment of the present invention, step 105, smoothing the initial optimized path to obtain the target optimized path, may include: Step 1051, through The fitted coordinates are obtained; where, To fit the coordinates, n is the number of nodes in the initial optimized path minus 1. The node coordinates are for the initial optimized path. The basis functions are 3rd order B-spline functions; Step 1052: Obtain the target optimized path based on the fitted coordinates.

[0040] Specifically, in order to smooth out the few polyline nodes in the initial optimization path so that the target optimization path is free of polylines and has continuous turning, the node coordinates in the initial optimization path are fitted using the formula in step 1051, and the fitted coordinates are sequentially stitched together in the map of the dynamic map data to obtain the target optimization path.

[0041] A specific embodiment of the laser-guided vehicle path optimization method according to this invention includes: Step 111: Acquire real-time reference point scanning data and real-time environmental scanning data using lidar; This step provides a data foundation for subsequent path optimization by collecting real-time reference point scanning data and real-time environmental scanning data (such as reflector distance and scanning angle) from the lidar.

[0042] Step 112: By performing compensation processing on the real-time scanning data, the positioning coordinate data of the automated guided vehicle (AGV) is obtained; The positioning accuracy of laser-guided AGVs directly affects the path optimization effect. The positioning error mainly originates from the pseudorange error of the laser radar (the distance measurement error between the laser radar and the reflector / natural feature). This step compensates for the pseudorange error in real time, outputting high-precision AGV self-positioning coordinates, providing a reliable positioning foundation for subsequent path optimization.

[0043] Step 113: Obtain dynamic map data based on the real-time environmental scanning data, the AGV positioning coordinate data, and the preset map data; Based on the high-precision positioning coordinates of AGVs and combined with real-time environmental data scanned by LiDAR (static obstacles, dynamic obstacles, narrow passages, etc.), a grid map of the AGV operating environment is constructed. At the same time, constraints for path optimization (smoothness constraints, obstacle avoidance constraints, path length constraints, AGV motion constraints) are set to clarify the boundaries of the path optimization objectives and avoid planning paths that do not conform to the actual movement capabilities of AGVs (such as excessively large turning angles or passages that are too narrow to pass through).

[0044] Step 114: Obtain the initial optimized path based on the AGV positioning coordinate data, the dynamic map data, and the preset target coordinate data; This algorithm searches for an initial optimized path that satisfies all constraints in dynamic map data, thus solving the problems of non-smooth paths and insufficient real-time performance in traditional algorithms.

[0045] Step 115: Smooth the initial optimized path to obtain the target optimized path.

[0046] Although the initial optimized path meets the basic constraints, it may still contain a small number of polygonal nodes, requiring further smoothing. This step performs smooth fitting on the nodes of the initial path to ensure that the path is free of polygons and that turning is continuous. At the same time, it ensures that the fitted path does not deviate from the free grid (satisfying obstacle avoidance constraints), adapts to the actual motion characteristics of the AGV, and reduces the frequency of start-stop and turning.

[0047] The laser-guided vehicle path optimization method of this invention improves path smoothness by more than 30%, reduces AGV turning frequency, and lowers energy consumption and equipment wear; path replanning time is ≤200ms, and it has rapid dynamic obstacle avoidance capability; it integrates laser positioning error compensation, with path deviation ≤5mm, and is suitable for complex scenarios such as narrow passages and high-density warehouses; it supports multi-AGV collaborative path optimization, avoids path conflicts, and improves overall operating efficiency.

[0048] like Figure 2 As shown, an embodiment of the present invention proposes a laser-guided vehicle path optimization device 200, comprising: Acquisition module 201 is used to acquire real-time reference point scanning data and real-time environmental scanning data through lidar; The processing module 202 is used to obtain AGV positioning coordinate data by compensating the real-time scanning data; obtain dynamic map data based on the real-time environmental scanning data, the AGV positioning coordinate data, and preset map data; obtain an initial optimized path based on the AGV positioning coordinate data, the dynamic map data, and preset target coordinate data; and obtain a target optimized path by smoothing the initial optimized path.

[0049] Optionally, real-time reference point scanning data and real-time environmental scanning data can be acquired via LiDAR, including: Real-time reference point scanning data is acquired using a lidar; the real-time reference point scanning data includes the distance values ​​between the lidar and multiple preset reference points. Real-time environmental scanning data is acquired using lidar; the real-time environmental scanning data includes the coordinates of static obstacles, the coordinates of dynamic obstacles, and the movement speed of dynamic obstacles.

[0050] Optionally, by performing compensation processing on the real-time scanning data, the positioning coordinate data of the automated guided vehicle (AGV) is obtained, including: pass Obtain compensation data; among which, To compensate for the i-th distance value in the data, Let k be the i-th distance value in the real-time scan data, and k be the environmental correction coefficient. The distance error to the i-th reference point; The compensation data is fitted to obtain the positioning coordinate data of the automated guided vehicle (AGV).

[0051] Optionally, dynamic map data is obtained based on the real-time environmental scanning data, the AGV positioning coordinate data, and the preset map data, including: Acquire preset map data; the preset map data is a global raster map of a preset scene; The preset map data is initialized based on the AGV positioning coordinate data to obtain an initialized grid map; Obstacle marking is performed on the initial grid map based on the real-time environmental scan data to obtain dynamic map data.

[0052] Optionally, based on the AGV positioning coordinate data, the dynamic map data, and the preset target coordinate data, an initial optimized path is obtained, including: Initial node data is obtained based on the AGV positioning coordinate data, the dynamic map data, and the preset target coordinate data; The initial node data is pruned to obtain the target node data; An initial optimized path is obtained based on the target node data and the dynamic map data.

[0053] Optionally, initial node data is obtained based on the AGV positioning coordinate data and preset target coordinate data, including: Candidate nodes are obtained by randomly sampling based on the dynamic map data; Based on the candidate nodes and the preset search radius, the surrounding nodes are obtained; pass Obtain the total path cost; where Cost is the total path cost. , , All are dynamic weighting factors, where L is the total path length. The average steering angle of the path, For obstacle avoidance safety redundancy; The initial node data is obtained based on the total path cost and the surrounding nodes.

[0054] Optionally, the initial optimized path is smoothed to obtain the target optimized path, including: pass The fitted coordinates are obtained; where, To fit the coordinates, n is the number of nodes in the initial optimized path minus 1. The node coordinates are for the initial optimized path. The basis functions are 3rd order B-spline functions; The target optimized path is obtained based on the fitted coordinates.

[0055] The laser-guided vehicle path optimization device of this invention acquires real-time reference point scanning data and real-time environmental scanning data through a laser radar. By compensating the real-time scanning data, it obtains AGV positioning coordinate data. Based on the real-time environmental scanning data, the AGV positioning coordinate data, and preset map data, it obtains dynamic map data. Based on the AGV positioning coordinate data, the dynamic map data, and preset target coordinate data, it obtains an initial optimized path. Finally, it smooths the initial optimized path to obtain the target optimized path, thereby improving the efficiency and accuracy of AGV path optimization.

[0056] It should be noted that this device corresponds to the method described above, and all implementations in the method embodiments described above are applicable to the embodiments of this device and can achieve the same technical effect. Further details are omitted in this embodiment.

[0057] This invention also provides a computing device, including: a processor and a memory storing a computer program, wherein the computer program, when executed by the processor, performs the method as described in any of the above embodiments. All implementations in the above method embodiments are applicable to the embodiments of this device and can achieve the same technical effects. Further details are omitted in this embodiment.

[0058] This invention also provides a computer-readable storage medium storing instructions that, when executed on a computer, cause the computer to perform the method as described in any of the above embodiments. All implementations in the above method embodiments are applicable to the embodiments of this device and can achieve the same technical effects. Further details are omitted in this embodiment.

[0059] It should be noted that in the apparatus and method of the present invention, the components or steps can obviously be decomposed and / or recombined. These decompositions and / or recombinations should be considered equivalent solutions of the present invention. Furthermore, the steps for performing the above series of processes can naturally be performed in the order described and in chronological order, but are not necessarily required to be performed in chronological order. Some steps can be performed in parallel, overlapping, or independently of each other.

[0060] It should be noted that in the above embodiments, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element. Furthermore, it should be noted that the scope of the methods and apparatuses in the embodiments described above is not limited to performing functions in the order shown or discussed, but may also include performing functions substantially simultaneously or in the reverse order, depending on the functions involved. For example, the described methods may be performed in a different order than described, and various steps may be added, omitted, or combined. Additionally, features described with reference to certain examples may be combined in other examples.

[0061] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for optimizing the path of an automated guided vehicle (AGV) based on laser navigation, characterized in that, include: Real-time reference point scanning data and real-time environmental scanning data are acquired using lidar. By performing compensation processing on the real-time scanning data, the positioning coordinate data of the automated guided vehicle (AGV) is obtained. Dynamic map data is obtained based on the real-time environmental scanning data, the AGV positioning coordinate data, and the preset map data; Based on the AGV positioning coordinate data, the dynamic map data, and the preset target coordinate data, an initial optimized path is obtained; The initial optimized path is smoothed to obtain the target optimized path.

2. The laser-guided vehicle path optimization method according to claim 1, characterized in that, Real-time reference point scanning data and real-time environmental scanning data are acquired through LiDAR, including: Real-time reference point scanning data is acquired using a lidar; the real-time reference point scanning data includes the distance values ​​between the lidar and multiple preset reference points. Real-time environmental scanning data is acquired using lidar; the real-time environmental scanning data includes the coordinates of static obstacles, the coordinates of dynamic obstacles, and the movement speed of dynamic obstacles.

3. The laser-guided vehicle path optimization method according to claim 1, characterized in that, By compensating the real-time scanning data, the positioning coordinate data of the automated guided vehicle (AGV) is obtained, including: pass Obtain compensation data; among which, To compensate for the i-th distance value in the data, Let k be the i-th distance value in the real-time scan data, and k be the environmental correction coefficient. The distance error to the i-th reference point; The compensation data is fitted to obtain the positioning coordinate data of the automated guided vehicle (AGV).

4. The laser-guided vehicle path optimization method according to claim 1, characterized in that, Based on the real-time environmental scanning data, the AGV positioning coordinate data, and the preset map data, dynamic map data is obtained, including: Acquire preset map data; the preset map data is a global raster map of a preset scene; The preset map data is initialized based on the AGV positioning coordinate data to obtain an initialized grid map; Obstacle marking is performed on the initial grid map based on the real-time environmental scan data to obtain dynamic map data.

5. The laser-guided vehicle path optimization method according to claim 1, characterized in that, Based on the AGV positioning coordinate data, the dynamic map data, and the preset target coordinate data, an initial optimized path is obtained, including: Initial node data is obtained based on the AGV positioning coordinate data, the dynamic map data, and the preset target coordinate data; The initial node data is pruned to obtain the target node data; An initial optimized path is obtained based on the target node data and the dynamic map data.

6. The laser-guided vehicle path optimization method according to claim 1, characterized in that, Based on the AGV positioning coordinate data and the preset target coordinate data, initial node data is obtained, including: Candidate nodes are obtained by randomly sampling based on the dynamic map data; Based on the candidate nodes and the preset search radius, the surrounding nodes are obtained; pass Obtain the total path cost; where Cost is the total path cost. , , All are dynamic weighting factors, where L is the total path length. The average steering angle of the path, For obstacle avoidance safety redundancy; The initial node data is obtained based on the total path cost and the surrounding nodes.

7. The laser-guided vehicle path optimization method according to claim 1, characterized in that, The initial optimized path is smoothed to obtain the target optimized path, including: pass The fitted coordinates are obtained; where, To fit the coordinates, n is the number of nodes in the initial optimized path minus 1. The node coordinates are for the initial optimized path. The basis functions are 3rd order B-spline functions; The target optimized path is obtained based on the fitted coordinates.

8. A laser-guided vehicle path optimization device, characterized in that, include: The acquisition module is used to acquire real-time reference point scanning data and real-time environmental scanning data through LiDAR; The processing module is used to obtain AGV positioning coordinate data by compensating the real-time scanning data; to obtain dynamic map data based on the real-time environmental scanning data, the AGV positioning coordinate data, and preset map data; to obtain an initial optimized path based on the AGV positioning coordinate data, the dynamic map data, and preset target coordinate data; and to obtain a target optimized path by smoothing the initial optimized path.

9. A computing device, characterized in that, include: A processor, a memory storing a computer program, wherein the computer program, when executed by the processor, performs the method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The system stores instructions that, when executed on a computer, cause the computer to perform the method as described in any one of claims 1 to 7.

Images