AGV path planning method based on two-dimensional and three-dimensional data

Abstract

The invention discloses an AGV path planning method based on two-dimensional and three-dimensional data, and aims to solve the problem that a traditional DWA algorithm cannot deal with moving obstacles and may fall into local optimum. The method specifically comprises the following steps: establishing a working space of the AGV, and adding a starting point, a target point and known obstacle information; planning a global path of the AGV by using an RRT * algorithm; training an obstacle recognition model, processing the obstacle input by the industrial camera, and determining the accurate position of the obstacle relative to the AGV in the moving process; the speed combination of the AGV is selected based on the obstacle distribution condition, then the distance and speed information of the obstacle is measured by combining the laser radar, and local path optimization is carried out through an improved DWA algorithm, so that the AGV reaches a designated position. According to the method provided by the invention, real-time local obstacle avoidance of the AGV is realized, and the automation capability of AGV path planning is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of AGV path planning, and in particular to an AGV path planning method based on two-dimensional and three-dimensional data. Background Art

[0002] As AGV technology matures, AGV is increasingly used. As a key technology, AGV path planning determines the working efficiency and safety of AGV.

[0003] AGV path planning algorithms include global planning algorithms such as Dijkstra, A*, RRT*, and D*, as well as local planning algorithms such as DWA, TEB, and MPC. The global planning algorithm only needs one planning to move the AGV from the starting point to the end point, but it can only handle situations where the size and position of the obstacle are known, and due to the design of the AGV structure, there may be an error between the theoretical position and the actual position. The local planning algorithm can detect dynamic obstacles with a relatively small error; however, the lack of global path guidance may cause the problem of lengthy paths and increase the amount of calculation. At present, there are also many algorithms that combine global planning and local planning, but these algorithms are not ideal for handling dynamic obstacles, and the amount of calculation is also relatively large. Summary of the invention

[0004] In view of the shortcomings of the prior art, the present invention proposes an AGV path planning method based on two-dimensional and three-dimensional data. By combining industrial cameras and lidars and adopting an algorithm combining RRT* and improved DWA, the obstacle avoidance function of AGV for dynamic obstacles is improved, and the rationality of speed combination selection is improved while reducing the amount of calculation.

[0005] In order to achieve the above technical objectives, the present invention provides the following technical solutions: An AGV path planning method based on two-dimensional and three-dimensional data specifically includes the following steps: S1. Establish a two-dimensional motion space for the AGV, and add the starting position, target position, and known obstacle information of the AGV in the two-dimensional motion space; S2, use the RRT* algorithm to plan a global path in the AGV's two-dimensional workspace; S3. Install an industrial camera on the AGV, control the AGV to move in a three-dimensional workspace, and collect obstacle images during the movement through the industrial camera; use the obstacle images as input to train the obstacle recognition model; use the trained model to determine the position of the obstacle relative to the AGV; S4. The AGV moves along the global path planned by the RRT* algorithm, and determines the distribution of surrounding obstacles based on the trained obstacle recognition model. The speed combination of the AGV is selected based on the obstacle distribution, and then the distance and speed information of the obstacle are measured by the lidar. The local path is optimized through the improved DWA algorithm to enable the AGV to reach the specified location.

[0006] Furthermore, step S2 specifically includes: S21, create a search tree directory Search for AGV, define the starting point of AGV as the first point of the search tree, and define the exploration step of the algorithm as Step; S22. Use the RRT* algorithm to plan the global path of the AGV, record the coordinates of each node on the path in the AGV workspace, and create a path node set J; J contains the serial number of each node on the path, the coordinates of each node, and the parent node serial number of each node.

[0007] Furthermore, step S3 specifically includes the following steps: S31. Install an industrial camera in front of the AGV to control the AGV to move in a three-dimensional workspace with obstacles and obtain a large number of images of the AGV during movement; S32, performing data processing on the image obtained by the industrial camera, marking the range of the obstacle and the position of the obstacle relative to the AGV on the image, and classifying the obstacles according to the relative position; S33, inputting the marked image data into the YOLOv5 model for training, and obtaining a model for identifying the accurate position of obstacles; S34. During the movement of the AGV, the industrial camera samples at a fixed frequency. Acquire images of the surrounding environment and input them into the trained model for recognition to obtain the accurate location and distribution of obstacles around the AGV during movement.

[0008] More specifically, the obstacles are classified in step S32 as follows: According to the position of the obstacle relative to the AGV, the obstacles are divided into four types: left, right, middle, and full, which respectively represent the four situations where the obstacle is located on the left side, right side, and front of the AGV, and the obstacle covers the entire field of view of the industrial camera.

[0009] Furthermore, step S4 specifically includes: S41, determining the speed value range and angular velocity value range of the AGV according to the AGV's own structural limitations, motor limitations, and obstacle distance limitations; S42, set the sampling interval to group and combine the velocity values ​​and the angular velocity values, and obtain the combined velocity value ; Then take the angular velocity beyond the threshold A new set of combined velocity values ​​is formed. ; S43, based on the distribution of surrounding obstacles, select a speed combination; and calculate the next position parameter of the AGV according to the selected speed combination and the current position parameter of the AGV; S44, using a laser radar installed in front and behind the AGV to measure the distance from the surrounding obstacles to the AGV and the speed and angular velocity information of the obstacles; according to the current position, speed and angular velocity of each obstacle, calculate the passing The movement trajectory of obstacles after a certain period of time; S45, calculate each speed combination, after The moving trajectory of the AGV after a certain period of time, and calculate the score of each speed combination ; The scoring formula is: ; in, For the selected speed combination, , Represent the selected speed and angular velocity respectively; under this speed combination, is the azimuth evaluation function, is the obstacle distance evaluation function, is the speed evaluation function, is the global path distance evaluation function; is the weight coefficient of the evaluation function; Indicates normalization processing; Select the set of speeds and angular velocities with the highest score among all speed combinations as the speed combination for the AGV to move from the current position to the next position; S46. After the AGV moves into position, repeat steps S43-S45 until the AGV enters a range of radius R with the target point as the center, and selects a speed combination with the target point as the end point. The combination with the highest score is selected as the speed and angular velocity of the last step, so that the AGV reaches the target point.

[0010] More specifically, step S41 is as follows: S411: Limit the speed and angular velocity range of the AGV according to its own structural limitations The values ​​of are as follows: ; in, and It is the minimum and maximum speed that the AGV can move. and It is the minimum and maximum angular velocity that the AGV can achieve; S412: According to the AGV motor limit, limit its acceleration and angular acceleration during movement, and obtain the speed and angular velocity range under this limit The values ​​of are as follows: ; in, and is the current speed and angular velocity of the AGV, and is the maximum acceleration and maximum angular acceleration of AGV, is the time required to get from the current position to the next position; S413, consider the obstacle limitation again, that is, the moving path of the AGV from deceleration to stop is less than the shortest distance to the obstacle, and obtain the speed and angular velocity range The values ​​of are as follows: ; in, is the shortest distance between the AGV moving path and the obstacle under the current speed and angular velocity combination; S414, according to The value range of the final velocity and angular velocity for The intersection of , that is: ; ; in, and are the minimum and maximum values ​​within the speed range, and are the minimum and maximum values ​​within the angular velocity range respectively.

[0011] More specifically, step S42 is as follows: Set the speed sampling interval and angular velocity sampling interval , then the total speed value is The angular velocity values ​​are Group; the formula is: ; in, and are the minimum and maximum values ​​within the speed range, and are the minimum and maximum values ​​within the range of angular velocity respectively; This gets shared The combined speed value of the speed combination ; Then take the angular velocity greater than the threshold A new set of combined velocity values ​​is formed. .

[0012] More specifically, in step S43, based on the distribution of surrounding obstacles, the speed combination is selected as follows: Assume that at the current position, the AGV collects p images through the industrial camera, and each image has obstacles, i Indicates i images, the total number of obstacles around the current location B for: ; Calculate the proportion of obstacles located directly in front of the AGV and covering the entire field of view of the industrial camera , the calculation formula is: ; in, , They represent the number of obstacles located directly in front of the AGV and covering the entire field of view of the industrial camera; like Greater than the set threshold , then in the combined velocity value set Select the speed combination, otherwise select the value set of the combined speed Choose the speed combination.

[0013] More specifically, each evaluation function in step S45 is: Azimuth evaluation function ; in, For selection Under the speed combination, the angle error between the AGV's movement direction from the current position to the next position and the line connecting the target point; the smaller the angle error, the higher the azimuth evaluation; Obstacle distance evaluation function ; in, j is the total number of obstacles measured by the laser radar, For the i The minimum distance between the moving trajectory of an obstacle and the moving trajectory of the AGV; Speed ​​evaluation function , the greater the speed, the higher the evaluation; The global path distance evaluation function represents the maximum distance between the end point of the AGV moving path optimized by the improved DWA algorithm and the global path planned by the RRT* algorithm; the formula is expressed as: ; in, For the i The maximum distance from the end point of the AGV moving path to the global path under the group speed combination, is the distance threshold set, k is a constant for controlling the descent speed; when the maximum distance When within the distance threshold, Set to a fixed real number N ; When the maximum distance When outside the distance threshold, the longer the distance, the lower the evaluation.

[0014] Based on the above technical solution, the method proposed in the present invention has at least the following beneficial effects: 1. The present invention aims at the problem that the DWA algorithm has no global planning guidance, which will cause a long path. It proposes a method combining the RRT* algorithm and the improved DWA algorithm to reduce the path length. It also improves the obstacle distance evaluation formula of the DWA algorithm and adds the global path distance evaluation formula to improve the rationality of the DWA algorithm.

[0015] 2. The present invention combines the information of industrial cameras and laser radars to obtain obstacle information around AGV in real time, and adjusts the speed selection range of the DWA algorithm according to the obstacle information; improves the efficiency of path planning, reduces the amount of calculation, and can better handle dynamic obstacles to ensure the safety of AGV. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings: Figure 1 This is a flow chart of the AGV path planning method proposed in the present invention; Figure 2 This is a schematic diagram of the installation of the industrial camera and laser radar involved in the method proposed in the present invention; Figure 3 Flowchart of the improved DWA algorithm designed for the present invention. DETAILED DESCRIPTION

[0017] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific implementation methods, so that the implementation process of how the present application uses technical means to solve technical problems and achieve technical effects can be fully understood and implemented accordingly.

[0018] Please refer to Figure 1-Figure 3, shows a specific implementation of this embodiment. This embodiment establishes an AGV workspace in a two-dimensional space, adds information such as the starting point, target point, and obstacles to the workspace; uses the RRT* algorithm to plan the global path of the AGV; trains the YOLOv5 model, processes the image data input by the industrial camera, and enables the model to identify the position of the obstacle relative to the AGV; combines the data output by the laser radar and the YOLOv5 model, and uses the improved DWA algorithm to plan the local path of the AGV. The specific improvements are: select different speed selection intervals according to the position of the obstacle relative to the AGV, add the evaluation of the global path distance to the speed combination evaluation, and improve the evaluation formula of the obstacle distance. Finally, the AGV can safely reach the designated location.

[0019] Please refer to Figure 1 This embodiment shows an AGV path planning method based on two-dimensional and three-dimensional data, which specifically includes the following steps: S1. Establish a two-dimensional motion space for the AGV. In this embodiment, the starting position of the AGV is defined as the coordinate origin, and the x-axis and y-axis directions of the AGV motion space are determined; and the coordinates of the target position of the AGV and the position and size information of known obstacles are added to the motion space; S2, use the RRT* algorithm to plan a global path in the AGV's workspace; As a preferred implementation, step S2 specifically includes: S21, create a search tree directory Search for AGV, define the starting point of AGV as the first point of the search tree, and define the exploration step of the algorithm as Step; S22. Use the RRT* algorithm to plan the global path of the AGV, record the coordinates of each node on the path in the AGV workspace, and create a path node set J; J contains the serial number of each node on the path, the coordinates of each node, and the parent node serial number of each node.

[0020] In this embodiment, after planning the global path, a mathematical expression of the path is calculated to calculate the distance between the end point of the speed combination in the DWA algorithm and the global path, that is, after step S22: Divide the global path into segments according to nodes and find the expression of the path between every two adjacent nodes .

[0021] S3, such as Figure 2 As shown, an industrial camera is installed on the AGV to control the AGV to move in a three-dimensional workspace, and the obstacle images during the movement are collected by the industrial camera; the obstacle recognition model is trained with the obstacle images as input; and the position of the obstacle relative to the AGV is determined using the trained model; As a preferred implementation, step S3 specifically includes: S31. Install an industrial camera in front of the AGV to control the AGV to move in a three-dimensional workspace with obstacles and obtain a large number of images of the AGV during movement; S32, performing data processing on the image obtained by the industrial camera, marking the range of the obstacle and the position of the obstacle relative to the AGV on the image, and classifying the obstacles according to the relative position; S33, inputting the marked image data into the YOLOv5 model for training, and obtaining a model for identifying the accurate position of obstacles; S34. During the movement of the AGV, the industrial camera samples at a fixed frequency. Acquire images of the surrounding environment and input them into the trained model for recognition to obtain the accurate location and distribution of obstacles around the AGV during movement.

[0022] S4, AGV moves according to the global path planned by the RRT* algorithm, and determines the distribution of surrounding obstacles based on the trained obstacle recognition model; selects the speed combination of AGV based on the obstacle distribution, and then combines the distance and speed information of obstacles measured by laser radar, through Figure 3 The improved DWA algorithm shown performs local path optimization to enable the AGV to reach the specified location; As a preferred implementation, step S4 specifically includes: S41, determining the speed value range and angular velocity value range of the AGV according to the AGV's own structural limitations, motor limitations, and obstacle distance limitations; More specifically, step S41 specifically includes: S411: Limit the speed and angular velocity range of the AGV according to its own structural limitations The values ​​of are as follows: ; in, and It is the minimum and maximum speed that the AGV can move. and It is the minimum and maximum angular velocity that the AGV can achieve; S412: According to the AGV motor limit, limit its acceleration and angular acceleration during movement, and obtain the speed and angular velocity range under this limit The values ​​of are as follows: ; in, and is the current speed and angular velocity of the AGV, and is the maximum acceleration and maximum angular acceleration of AGV, is the time required to get from the current position to the next position; S413, consider the obstacle limitation again, that is, the moving path of the AGV from deceleration to stop is less than the shortest distance to the obstacle, and obtain the speed and angular velocity range The values ​​of are as follows: ; in, is the shortest distance between the AGV moving path and the obstacle under the current speed and angular velocity combination; S414, according to The value range of the final velocity and angular velocity for The intersection of , that is: ; ; in, and are the minimum and maximum values ​​within the speed range, and are the minimum and maximum values ​​within the angular velocity range respectively.

[0023] S42, set the sampling interval to group and combine the velocity values ​​and the angular velocity values, and obtain the combined velocity value ; Then take the angular velocity beyond the threshold group, forming a new set of combined velocity values ; In this embodiment, step S42 is specifically as follows: Set the speed sampling interval and angular velocity sampling interval , then the total speed value is The angular velocity values ​​are Group; the formula is: ; in, and are the minimum and maximum values ​​within the speed range, and are the minimum and maximum values ​​within the range of angular velocity respectively; This gets shared The combined speed value of the speed combination ; Then take the angular velocity greater than the threshold group, forming a new set of combined velocity values ; This way the grouping can be done directly in the collection when a large angle rotation is required The speed combination is selected from the above, thereby reducing the meaningless score calculation for the combination with smaller angular velocity, that is, reducing the overall calculation amount.

[0024] S43, based on the distribution of surrounding obstacles, select a speed combination; in this embodiment, assuming that at the current position, the AGV collects p images through the industrial camera, each image has obstacles, i Indicates i images, the total number of obstacles around the current location B for: ; Calculate the proportion of obstacles located directly in front of the AGV and covering the entire field of view of the industrial camera , the calculation formula is: ; in, , They represent the number of obstacles located directly in front of the AGV and covering the entire field of view of the industrial camera; like Greater than the set threshold , then in the combined velocity value set Select the speed combination, otherwise select the value set of the combined speed Select a speed combination; according to the above selection rules, the selected speed combination is combined with the current position parameters of the AGV to calculate the next position parameters of the AGV; specifically: Assume that the current position parameters of AGV are , then according to the selected speed and angular velocity, the next position parameters of the AGV can be calculated : ; And calculate the movement route of the AGV under this speed and angular velocity.

[0025] S44, using a laser radar installed in front and behind the AGV to measure the distance from the surrounding obstacles to the AGV and the speed and angular velocity information of the obstacles; according to the current position, speed and angular velocity of each obstacle, calculate the passing The movement trajectory of obstacles after a certain period of time; S45, calculate each speed combination, after The moving trajectory of the AGV after a certain period of time, and calculate the score of each speed combination ; The scoring formula is: ; in, For the selected speed combination, , Represent the selected speed and angular velocity respectively; under this speed combination, is the azimuth evaluation function, is the obstacle distance evaluation function, is the speed evaluation function, is the global path distance evaluation function; in this embodiment, each evaluation function is specifically: Azimuth evaluation function ; in, For selection Under the speed combination, the angle error between the AGV's movement direction from the current position to the next position and the line connecting the target point; the smaller the angle error, the higher the azimuth evaluation; Obstacle distance evaluation function ; in, j is the total number of obstacles measured by the laser radar, For the i The minimum distance between the moving trajectory of an obstacle and the moving trajectory of the AGV; During the movement of the AGV, multiple obstacles may be encountered. When the distance between these obstacles and the AGV is relatively large, the score will be higher, otherwise it will be lower. However, in order to prevent the situation where other obstacles are far away but one obstacle is very close, which may also cause danger, the present application designs the above obstacle distance evaluation function. When one obstacle is very close to the AGV, it will have a great impact on the score, which will greatly reduce the score and better ensure the safety of the moving path. Speed ​​evaluation function , the greater the speed, the higher the evaluation; The global path distance evaluation function represents the maximum distance between the end point of the AGV moving path optimized by the improved DWA algorithm and the global path planned by the RRT* algorithm; the formula is expressed as: ; in, For the i The maximum distance from the end point of the AGV moving path to the global path under the group speed combination, is the distance threshold set, k is a constant for controlling the descent speed; when the maximum distance When within the distance threshold, Set to a fixed real number N ; When the maximum distance When outside the distance threshold, the farther the distance, the lower the evaluation; Setting the global path evaluation function can ensure that when the DWA algorithm is used to optimize the local path, the score is fixed when the deviation from the global path is not large. After exceeding this threshold, the farther the score is, the lower the score is. This ensures that as long as it is within a certain range of global path planning, the evaluation function can be played, and the paths with too long distances can be removed. Several other evaluation functions can be used as the evaluation criteria within a reasonable range.

[0026] The present invention improves the rationality of the DWA algorithm in optimizing the local path by improving the obstacle distance evaluation formula of the DWA algorithm and adding the global path distance evaluation formula; is the weight coefficient of the evaluation function; Indicates normalization processing; The normalization formula is as follows: ; ; ; ; in, m Represents the total number of speed values ​​in the speed value set, e Indicates e Group value; Through normalization processing, it is ensured that each item in the evaluation function can play a normal role, and the influence of other items will not be eliminated because one of them is too large.

[0027] Select the set of speeds and angular velocities with the highest score among all speed combinations as the speed combination for the AGV to move from the current position to the next position; S46. After the AGV moves into position, repeat steps S43-S45 until the AGV enters a range of radius R with the target point as the center, and selects a speed combination with the target point as the end point. The combination with the highest score is selected as the speed and angular velocity of the last step, so that the AGV reaches the target point.

[0028] The AGV path planning method proposed in this embodiment is based on the data of industrial cameras and laser radars. In view of the problem that the DWA algorithm may cause a lengthy path, a method combining the RRT* algorithm and the improved DWA algorithm is proposed. The industrial camera and laser radar data are combined to avoid dynamic obstacles, and the scoring calculation formula of the DWA algorithm is changed, so that the AGV can choose a more reasonable speed combination during movement.

[0029] Those skilled in the art can understand that all or part of the steps in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a program, so the present application can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0030] Each embodiment in this specification is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referred to each other. For the above embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiments.

[0031] The above implementation methods have been described in detail. Specific examples are used herein to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea. At the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as a limitation on the present invention.

Claims

1. An AGV path planning method based on two-dimensional and three-dimensional data, characterized in that: The specific steps include: S1. Establish a two-dimensional motion space for the AGV, and add the starting position, target position, and known obstacle information of the AGV in the two-dimensional motion space; S2, use the RRT* algorithm to plan a global path in the AGV's two-dimensional workspace; S3. Install an industrial camera on the AGV, control the AGV to move in a three-dimensional workspace, and collect obstacle images during the movement through the industrial camera; use the obstacle images as input to train the obstacle recognition model; use the trained model to determine the position of the obstacle relative to the AGV; S4. The AGV moves along the global path planned by the RRT* algorithm, and determines the distribution of surrounding obstacles based on the trained obstacle recognition model. The speed combination of the AGV is selected based on the obstacle distribution, and then the distance and speed information of the obstacle are measured by the lidar. The local path is optimized through the improved DWA algorithm to enable the AGV to reach the specified location.

2. The AGV path planning method based on two-dimensional and three-dimensional data according to claim 1, characterized in that: Step S2 specifically includes the following steps: S21, create a search tree directory Search for AGV, define the starting point of AGV as the first point of the search tree, and define the exploration step of the algorithm as Step; S22. Use the RRT* algorithm to plan the global path of the AGV, record the coordinates of each node on the path in the AGV workspace, and create a path node set J; J contains the serial number of each node on the path, the coordinates of each node, and the parent node serial number of each node.

3. The AGV path planning method based on two-dimensional and three-dimensional data according to claim 1, characterized in that: Step S3 specifically includes the following steps: S31. Install an industrial camera in front of the AGV to control the AGV to move in a three-dimensional workspace with obstacles and obtain a large number of images of the AGV during movement; S32, performing data processing on the image obtained by the industrial camera, marking the range of the obstacle and the position of the obstacle relative to the AGV on the image, and classifying the obstacles according to the relative position; S33, inputting the marked image data into the YOLOv5 model for training, and obtaining a model for identifying the accurate position of obstacles; S34. During the movement of the AGV, the industrial camera samples at a fixed frequency. Acquire images of the surrounding environment and input them into the trained model for recognition to obtain the accurate location and distribution of obstacles around the AGV during movement.

4. The AGV path planning method based on two-dimensional and three-dimensional data according to claim 3, characterized in that: The classification of obstacles in step S32 is specifically as follows: According to the position of the obstacle relative to the AGV, the obstacles are divided into four types: left, right, middle, and full, which respectively represent the four situations where the obstacle is located on the left side, right side, and front of the AGV, and the obstacle covers the entire field of view of the industrial camera.

5. The AGV path planning method based on two-dimensional and three-dimensional data according to claim 1, characterized in that: Step S4 specifically includes: S41, determining the speed value range and angular velocity value range of the AGV according to the AGV's own structural limitations, motor limitations, and obstacle distance limitations; S42, set the sampling interval to group and combine the velocity values ​​and the angular velocity values, and obtain the combined velocity value ; Then take the angular velocity beyond the threshold group, forming a new set of combined velocity values ; S43, based on the distribution of surrounding obstacles, select a speed combination; and calculate the next position parameter of the AGV according to the selected speed combination and the current position parameter of the AGV; S44, using a laser radar installed in front and behind the AGV to measure the distance from the surrounding obstacles to the AGV and the speed and angular velocity information of the obstacles; according to the current position, speed and angular velocity of each obstacle, calculate the passing The movement trajectory of obstacles after a certain period of time; S45, calculate each speed combination, after The moving trajectory of the AGV after a certain period of time, and calculate the score of each speed combination ; The scoring formula is: ； in, For the selected speed combination, , Represent the selected speed and angular velocity respectively; under this speed combination, is the azimuth evaluation function, is the obstacle distance evaluation function, is the speed evaluation function, is the global path distance evaluation function; is the weight coefficient of the evaluation function; Indicates normalization processing; Select the set of speeds and angular velocities with the highest score among all speed combinations as the speed combination for the AGV to move from the current position to the next position; S46. After the AGV moves into position, repeat steps S43-S45 until the AGV enters a range of radius R with the target point as the center, and selects a speed combination with the target point as the end point. The combination with the highest score is selected as the speed and angular velocity of the last step, so that the AGV reaches the target point.

6. The AGV path planning method based on two-dimensional and three-dimensional data according to claim 5, characterized in that: Step S41 is specifically as follows: S411: Limit the speed and angular velocity range of the AGV according to its own structural limitations The values ​​of are as follows: ； in, and It is the minimum and maximum speed that the AGV can move. and It is the minimum and maximum angular velocity that the AGV can achieve; S412: According to the AGV motor limit, limit its acceleration and angular acceleration during movement, and obtain the speed and angular velocity range under this limit The values ​​of are as follows: ； in, and is the current speed and angular velocity of the AGV, and is the maximum acceleration and maximum angular acceleration of AGV, is the time required to get from the current position to the next position; S413, consider the obstacle limitation again, that is, the moving path of the AGV from deceleration to stop is less than the shortest distance to the obstacle, and obtain the speed and angular velocity range The values ​​of are as follows: ； in, is the shortest distance between the AGV moving path and the obstacle under the current speed and angular velocity combination; S414, according to The value range of the final velocity and angular velocity for The intersection of , that is: ； ； in, and are the minimum and maximum values ​​within the speed range, and are the minimum and maximum values ​​within the angular velocity range respectively.

7. The AGV path planning method based on two-dimensional and three-dimensional data according to claim 5, characterized in that: Step S42 is specifically as follows: Set the speed sampling interval and angular velocity sampling interval , then the total speed value is The angular velocity values ​​are Group; the formula is: ； in, and are the minimum and maximum values ​​within the speed range, and are the minimum and maximum values ​​within the range of angular velocity respectively; This gets shared The combined speed value of the speed combination ; Then take the angular velocity greater than the threshold group, forming a new set of combined velocity values .

8. The AGV path planning method based on two-dimensional and three-dimensional data according to claim 5, characterized in that: In step S43, based on the distribution of surrounding obstacles, the speed combination is selected as follows: Assume that at the current position, the AGV collects p images through the industrial camera, and each image has obstacles, i Indicates i images, the total number of obstacles around the current location B for: ； Calculate the proportion of obstacles located directly in front of the AGV and covering the entire field of view of the industrial camera , the calculation formula is: ； in, , They represent the number of obstacles located directly in front of the AGV and covering the entire field of view of the industrial camera; like Greater than the set threshold , then in the combined velocity value set Select the speed combination, otherwise select the value set of the combined speed Choose the speed combination.

9. The AGV path planning method based on two-dimensional and three-dimensional data according to claim 5, characterized in that: The evaluation functions in step S45 are specifically: Azimuth evaluation function ; in, For selection Under the speed combination, the angle error between the AGV's movement direction from the current position to the next position and the line connecting the target point; the smaller the angle error, the higher the azimuth evaluation; Obstacle distance evaluation function ; in, j is the total number of obstacles measured by the laser radar, For the i The minimum distance between the moving trajectory of an obstacle and the moving trajectory of the AGV; Speed ​​evaluation function , the greater the speed, the higher the evaluation; The global path distance evaluation function represents the maximum distance between the end point of the AGV moving path optimized by the improved DWA algorithm and the global path planned by the RRT* algorithm; the formula is expressed as: ； in, For the i The maximum distance from the end point of the AGV moving path to the global path under the group speed combination, is the distance threshold set, k is a constant for controlling the descent speed; when the maximum distance When within the distance threshold, Set to a fixed real number N ; When the maximum distance When outside the distance threshold, the longer the distance, the lower the evaluation.

Images