AGV path secondary optimization method and device based on b-spline curve and medium

Abstract

The application discloses an AGV path secondary optimization method and device based on a B-spline curve, and a medium, the method comprising: obtaining an initial fitting curve through a cubic B-spline interpolation method, and optimizing the initial fitting curve through a nonlinear programming problem to obtain a first optimized trajectory path; determining whether the angular acceleration of the AGV is suddenly changed when the AGV travels on the first optimized trajectory path; if yes, optimizing the first optimized trajectory path through the way of re-planning the fitting curve and adding a control point, and taking the result of the second optimization as the final travel path of the AGV; and if no, taking the first optimized trajectory path obtained through the first optimization as the final travel path of the AGV. The application optimizes the smoothness of the trajectory path of the AGV, reduces the error at the junction of adjacent road sections, eliminates the sudden change of the angular acceleration of the AGV and the steering wheel angle, and thus effectively improves the precision, quality and performance of the AGV path trajectory, and meets the travel requirements of the AGV.

Description

Technical Field

[0001] This invention relates to the field of AGV control technology, and in particular to a method, apparatus and medium for secondary optimization of AGV paths based on B-spline curves. Background Technology

[0002] In the field of intelligent AGV technology, path planning is the core technology of AGVs. Path planning refers to generating a static geometric trajectory line from the starting point to the ending point without collision. Trajectory planning is a type of path planning, in which the AGV plans the vehicle's travel trajectory based on the results of behavioral decisions, the vehicle's driving status, and environmental information. Typically, the evaluation indicators for path planning include whether the path is the shortest, whether the planning efficiency is the highest, and whether the path smoothness is good. In daily applications, AGVs often need to perform tasks in specific terrains and complex environments. Therefore, it is necessary to pay attention not only to path planning in complex environments but also to consider the quality of the planned path from the perspective of global path planning.

[0003] In this field, path planning methods are crucial to the performance of AGVs during operation. Two common path planning methods are polynomial curve fitting and the Bezier curve method, both of which can generate paths that meet the AGV's operational requirements. However, paths generated by fifth-order polynomial curve fitting cannot adequately constrain the directions of the start and end points, and there is a loss of accuracy during the fitting process. Furthermore, because only two of the four control points of the Bezier curve can change direction, the paths generated by the Bezier curve contain many areas of high curvature. This makes the AGV's steering wheel angular velocity and steering wheel angle extremely prone to abrupt changes during operation, leading to deviations from the path when following the trajectory.

[0004] Therefore, it is necessary to provide a method and system for reducing the impact of sudden changes in steering wheel angular velocity and angle, and improving the accuracy and quality of path planning. Summary of the Invention

[0005] The purpose of this invention is to provide a method, device, and medium for secondary optimization of AGV paths based on B-spline curves, so as to reduce the impact of sudden changes in the angular velocity and rotation angle of the AGV's steering wheel during operation, improve the accuracy and quality of path planning, and ensure the smooth operation of the AGV.

[0006] The solution to the technical problem of this invention is as follows: In a first aspect, embodiments of this application provide a method for secondary optimization of AGV paths based on B-spline curves, comprising the following steps:

[0007] S100, acquire the scene map, determine the start and end points of the AGV's travel path in the scene map, select six control points according to the length of the path to form a control point set, the control point set including the start point and the end point;

[0008] S200, the control point set is interpolated using the cubic B-spline interpolation method to obtain an initial fitting curve from the starting point to the ending point, the initial fitting curve being composed of several discrete points;

[0009] S300, determine the nonlinear programming problem based on the plurality of discrete points, and when the nonlinear programming problem has an optimal solution, construct the characteristic optimal curve based on the optimal solution;

[0010] S400, the optimal characteristic curve is used as the first optimized trajectory path of the AGV, and it is determined whether the steering wheel angular acceleration of the AGV changes abruptly during the process of the AGV traveling on the first optimized trajectory path; if not, the first optimized trajectory path is used as the final travel path of the AGV; if so, proceed to S500.

[0011] S500, when the steering wheel angular acceleration changes abruptly, the first optimized trajectory path is replanned into a new fitted curve including eight endpoints and seven path segments, and several auxiliary control points are obtained through the new fitted curve;

[0012] S600, add the aforementioned auxiliary control points as new control points to the control point set, and repeat S200-S400.

[0013] In some embodiments, in S200, the step of interpolating the control point set using a cubic B-spline interpolation method to obtain an initial fitted curve from the starting point to the ending point includes:

[0014] The cubic B-spline interpolation method is used to interpolate between two adjacent control points in the control point set to generate a fitting curve for these two adjacent control points;

[0015] Connect the fitted curves of each pair of adjacent control points to form an initial fitted curve from the starting point to the ending point.

[0016] In some embodiments, S300, determining the nonlinear programming problem based on the plurality of discrete points, and constructing an optimal characteristic curve based on the optimal solution when the nonlinear programming problem has an optimal solution, includes:

[0017] Determine the curvature of each discrete point in the initial fitted curve, and determine the total cost value based on the curvature of each discrete point;

[0018] Using the minimum total cost as the objective function, the optimization variables m and n as input variables, and the maximum value of the AGV's steering wheel rotation angular acceleration during the initial fitting curve being less than a first set threshold as the constraint function, the following nonlinear programming problem is constructed:

[0019] input m1, m2

[0020] min Cost

[0021] subjuct to C<maxA

[0022] Where m1 and m2 are both optimization variables, Cost is the total cost value, C is the maximum value of the steering wheel rotation angular acceleration of the AGV traversing the initial fitted curve, and max A is the first set threshold.

[0023] When the nonlinear programming problem has an optimal solution, construct the characteristic optimal curve based on the optimization variables m and n at this time.

[0024] In some embodiments, S400, the step of using the optimal characteristic curve as the first optimized trajectory path of the AGV and calculating the steering wheel angle and steering wheel angular acceleration of the AGV traveling on the first optimized trajectory path includes:

[0025] First, the optimal characteristic curve is used as the first optimized trajectory path of the AGV, and the coordinate position of the steering wheel relative to the starting point is defined as (x... f y f ), calculate the steering wheel angle α when the AGV is traveling on the first optimized trajectory path:

[0026]

[0027] Then, based on the steering wheel rotation angle, the angular velocity of the steering wheel is calculated:

[0028]

[0029] Among them, v s The set travel speed for the AGV, v r Angular velocity;

[0030] The time difference is calculated using the angular velocity of the steering wheel:

[0031]

[0032] Where T is the time difference and s is the length of the first optimized trajectory path;

[0033] The following steering wheel angular acceleration can then be obtained.

[0034]

[0035] Where i' is the counter symbol.

[0036] In some embodiments, determining whether the steering wheel angular acceleration changes abruptly during the AGV's operation includes:

[0037] Based on the expression for the steering wheel angular velocity, the entire first optimized trajectory path is traversed to find the maximum value of the steering wheel angular acceleration. By using the maximum value of the steering wheel angular acceleration, it is determined whether the steering wheel angular acceleration changes abruptly during the AGV's operation.

[0038] Wherein, when the maximum value of the steering wheel angular acceleration is greater than the angular acceleration threshold, it is considered that the steering wheel angular acceleration has abruptly changed; when the maximum value of the steering wheel angular acceleration is not greater than the angular acceleration threshold, it is considered that the steering wheel angular acceleration has not abruptly changed.

[0039] In some embodiments, when the steering wheel angular acceleration undergoes a sudden change, the initial fitted curve is reprogrammed into a new fitted curve including eight endpoints and seven paths, and several auxiliary control points are obtained through the new fitted curve, including:

[0040] Calculate all discrete points that make up the first optimized trajectory path, select seven discrete points from all the discrete points as new path segment endpoints, and reconstruct a new fitting curve, the new fitting curve including eight endpoints and seven path segments;

[0041] Calculate the coordinates of these eight endpoints, and based on the coordinates of the eight endpoints, calculate the straight-line distances l1, l2, l3, l4, l5, l6, l7 between adjacent endpoints;

[0042] Let the endpoint vector be U n It satisfies the following formula:

[0043]

[0044] Where L = l1 + l2 + l3 + l4 + l5 + l6 + l7;

[0045] Let Δ i =U i+1 -U i Where i is the element index in the endpoint vector sequence, the following boundary conditions are obtained:

[0046]

[0047]

[0048] Where q is the coordinate of the endpoint;

[0049] The parameter 'a' for the remaining six endpoints (excluding the first and eighth endpoints) can be calculated using the following formula. i b i c i and e i i = [2, 7]:

[0050]

[0051] Through the parameter a of these eight endpoints i b i c i and e i Given i = [1, 8], we can obtain an equation with the unknown quantity being the coordinates of the current control point:

[0052]

[0053] Taking the inverse of the above formula, we get:

[0054]

[0055] Finally, solving the above formula yields the coordinates of several auxiliary control points P1, P2, ..., P... n-3 P n-2 .

[0056] Secondly, embodiments of this application provide an AGV path secondary optimization device based on B-splines, comprising:

[0057] The acquisition unit is used to acquire a scene map and determine the start and end points of the AGV's travel path in the scene map;

[0058] An interpolation generation unit is used to select multiple control points according to the length of the path to form a control point set, and to interpolate the control point set using a cubic B-spline interpolation method to obtain an initial fitting curve from the starting point to the ending point. The initial fitting curve is composed of several discrete points.

[0059] The planning unit is used to determine a nonlinear programming problem based on the plurality of discrete points, and when the nonlinear programming problem has an optimal solution, to construct an optimal characteristic curve based on the optimal solution;

[0060] The judgment unit is used to take the optimal characteristic curve as the first optimized trajectory path of the AGV, calculate the steering wheel rotation angle and steering wheel angular acceleration of the AGV traveling on the first optimized trajectory path, and determine whether the steering wheel angular acceleration changes abruptly during the travel of the AGV.

[0061] The optimization unit is used to re-plan the initial fitting curve into a new fitting curve including eight endpoints and seven paths when the steering wheel angular acceleration changes abruptly, obtain several auxiliary control points through the new fitting curve, and add the several auxiliary control points as new control points to the control point set.

[0062] The output unit is used to take the first optimized trajectory path as the final travel path of the AGV when the steering wheel angular acceleration does not change abruptly during the travel of the AGV.

[0063] Thirdly, embodiments of this application provide an AGV, including: a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the aforementioned AGV path secondary optimization method based on B-spline curves.

[0064] Fourthly, embodiments of this application provide a storage medium storing computer-executable instructions for executing the aforementioned AGV path secondary optimization method based on B-spline curves.

[0065] The beneficial effects of this invention are: it provides a method, apparatus, and medium for secondary optimization of AGV paths based on B-spline curves, which optimizes the AGV path planning method to ensure that the AGV reaches the destination point quickly; at the same time, it optimizes the smoothness of the trajectory during the AGV's travel, reduces the error at the junction of adjacent road segments, eliminates the sudden changes in the AGV's angular acceleration and steering wheel angle, ensures the stability of the AGV's travel, and avoids deviation from the path, thereby effectively improving the accuracy, quality, and performance of the AGV path trajectory and meeting the AGV's driving requirements.

[0066] Other features and advantages of this disclosure will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the disclosure. The objects and other advantages of this disclosure may be realized and obtained by means of the structures particularly pointed out in the description, claims and drawings. Attached Figure Description

[0067] The accompanying drawings are provided to further understand the technical solutions of this disclosure and constitute a part of the specification. They are used together with the embodiments of this disclosure to explain the technical solutions of this disclosure and do not constitute a limitation on the technical solutions of this disclosure.

[0068] Figure 1 A flowchart of the AGV path secondary optimization method provided in the embodiments of this application;

[0069] Figure 2 A schematic diagram of a B-spline trajectory path based on six control points provided in an embodiment of this application;

[0070] Figure 3 A schematic diagram of the first optimized trajectory path obtained after the first optimization provided in the embodiments of this application;

[0071] Figure 4A A schematic diagram of the turning radius of the first optimized path provided in the embodiments of this application;

[0072] Figure 4B A schematic diagram of the steering wheel rotation angle and angular velocity of the first optimized AGV provided in the embodiments of this application;

[0073] Figure 5 A schematic diagram of the first optimized trajectory path obtained after the second optimization provided in an embodiment of this application;

[0074] Figure 6A A schematic diagram of the turning radius of the second optimized path provided in the embodiments of this application;

[0075] Figure 6B This is a schematic diagram of the steering wheel angle and angular velocity of the second optimized AGV provided in the embodiments of this application. Detailed Implementation

[0076] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0077] The present application will be further described below with reference to the accompanying drawings and specific embodiments. The described embodiments should not be considered as limitations on the present application, and all other embodiments obtained by those skilled in the art without inventive effort are within the scope of protection of the present application.

[0078] In the following description, references are made to “some embodiments,” which describe a subset of all possible embodiments. However, it is understood that “some embodiments” may be the same subset or different subsets of all possible embodiments and may be combined with each other without conflict.

[0079] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.

[0080] Before providing a further detailed description of the embodiments of this application, the nouns and terms involved in the embodiments of this application will be explained, and the nouns and terms involved in the embodiments of this application shall be interpreted as follows.

[0081] (1) AGV (Automated Guided Vehicle) refers to a transport vehicle equipped with electromagnetic or optical automatic guidance devices, capable of traveling along a prescribed guidance path, and possessing safety protection and various transfer functions. In industrial applications, it is a driverless transport vehicle powered by rechargeable batteries. Its movement and behavior are generally controlled by a computer, or its path is established using electromagnetic tracks attached to the floor. The unmanned transport vehicle moves based on the information provided by the electromagnetic tracks. Its main characteristics include trolley programming, parking selection devices, safety protection, and various transfer functions. Under computer monitoring, it can autonomously drive according to instructions, automatically traveling along a prescribed guidance path to reach a designated location and complete a series of tasks. Its system technology and products have become important equipment and technologies for flexible production lines, flexible assembly lines, and automated warehousing and logistics systems.

[0082] (2) B-spline curve is a special form of representation in numerical analysis, a sub-discipline of mathematics. The design of B-spline curves usually requires a series of control points, a series of nodes, and a series of coefficients. Each coefficient corresponds to a control point. The curve that connects all the curve segments together and satisfies a certain continuity condition is the B-spline curve.

[0083] In the field of intelligent AGV technology, path planning is the core technology of AGVs. The global path of an AGV consists of a series of path points. These path points only need to contain spatial position information, and may also contain attitude information, but do not need to be time-dependent; these path points are called global path points. Trajectory planning is one type of path planning. Trajectory planning refers to designing certain control signals to make the system state operate stably along a set trajectory. A trajectory refers to the set of changes in the system's position, velocity, and acceleration over time, not just a time-independent path in space. Reasonable trajectory planning not only enables underactuated degrees of freedom to move according to certain rules, but also effectively improves the system's operating efficiency.

[0084] Typically, evaluation metrics for path planning include whether the path is the shortest, whether the planning efficiency is the highest, and whether the path smoothness is good. In daily applications, AGVs often need to perform tasks on specific terrains and complex maps. Therefore, it is important not only to pay attention to path planning in complex environments but also to consider the quality of the planned path from a global path planning perspective.

[0085] Since path planning methods are crucial to the performance of AGVs during operation, the following two methods are commonly used in this field to implement AGV path planning:

[0086] The main idea behind the fifth-order polynomial curve fitting method is to obtain a trajectory with continuous acceleration, where the position and velocity require suitable initial and final conditions, as well as suitable initial and final acceleration values. This results in six boundary conditions. A fifth-order polynomial is constructed using these six boundary conditions to generate a trajectory that meets the conditions, which is then fitted into a spline curve.

[0087] The Beizer curve method primarily involves iteratively calculating the optimal curve by setting a certain distance threshold and using the directional control points at the start and end points, at the cost of curvature. The cubic Beizer curve equation satisfies:

[0088] B n (t)=P0(1-t) 3 +3P1t(1-t) 2 +3P2 t 2 ( 1-t)+P3t 3 , t∈[0,1]

[0089] However, the path generated by the fifth-order polynomial curve fitting method cannot provide better constraints on the direction of the start and end points, and there is a loss of accuracy during the fitting process. Furthermore, since only two of the four control points of the Bezier curve can be variable in direction, the path generated by the Bezier curve has many areas with large curvatures. This makes the AGV's steering wheel angular acceleration and steering wheel angle extremely prone to abrupt changes during travel, leading to deviations from the path when the AGV is following the trajectory.

[0090] To address the problems existing in the prior art, this disclosure provides a method, system, and storage medium for secondary optimization of AGV paths based on B-spline curves. This method is based on the optimization method of trajectory planning and the method of inverse B-spline control points, and optimizes the AGV travel path twice to obtain a trajectory path whose performance and quality can meet the requirements of AGV driving.

[0091] Reference Figure 1 As shown, Figure 1 The diagram shows a flowchart of the AGV path secondary optimization method provided in an embodiment of this application. An embodiment of this application will be described and explained below. The AGV path secondary optimization method may include, but is not limited to, the following steps:

[0092] Step 100: Obtain the scene map, determine the start and end points of the AGV's travel path in the scene map, and select six control points based on the length of the path to form a control point set.

[0093] It should be noted that the set of control points includes both a start point and an end point.

[0094] Step 200: Use cubic B-spline interpolation to interpolate the control point set to obtain the initial fitted curve from the starting point to the ending point.

[0095] It should be noted that the initial fitted curve has several discrete points.

[0096] Step 300: Determine the nonlinear programming problem based on several discrete points. When the nonlinear programming problem has an optimal solution, construct the characteristic optimal curve based on the optimal solution.

[0097] It should be noted that nonlinear programming problems include objective functions and constraint functions. The objective function represents the length of the optimized fitted curve, while the constraint function represents the smoothness of the optimized fitted curve.

[0098] Step 400: Use the curve with the best characteristics as the first optimized trajectory path of the AGV, and determine whether the angular acceleration of the steering wheel of the AGV changes abruptly during the process of the AGV traveling on the first optimized trajectory path.

[0099] If the steering wheel angular acceleration does not change abruptly, the first optimized trajectory path will be used as the final travel path of the AGV.

[0100] If the steering wheel angular acceleration changes abruptly, proceed to step 500.

[0101] Step 500: When the steering wheel angular acceleration changes abruptly, the first optimized trajectory path is replanned into a new fitted curve, and several auxiliary control points are obtained through the new fitted curve.

[0102] It should be noted that the new fitted curve includes eight endpoints and seven path segments.

[0103] Step 600: Add several auxiliary control points as new control points to the control point set, and repeat steps 200 to 400.

[0104] This disclosure includes two optimization processes. First, a cubic B-spline interpolation method is used to generate an initial fitting curve from the starting point to the ending point, initially determining the key nodes of the AGV's travel path. Then, the first optimization process is performed: based on several discrete points in the initial fitting curve, a nonlinear programming problem is determined to facilitate subsequent optimization of the initial fitting curve, thereby optimizing the length and smoothness of the AGV's travel path to meet the AGV's travel requirements; when an optimal solution exists for the nonlinear programming problem, the optimal characteristic curve is constructed using the optimal solution, and this optimal characteristic curve is used as the first optimized trajectory path.

[0105] In the first optimized trajectory path, the differences between each path segment are significant, and unnecessary errors exist at the junctions between adjacent path segments. Furthermore, the AGV's steering wheel angle change function is not a convex function, and unnecessary errors also occur when the steering wheel angle rotates or changes clockwise or counterclockwise. This is detrimental to AGV control, causing abrupt changes in both the AGV's angular acceleration and steering wheel angle when the AGV reaches the junction between adjacent path segments. Therefore, after obtaining the first optimized trajectory path, the entire path is traversed to determine whether abrupt changes in angular acceleration and steering wheel angle occur when the AGV travels along the first optimized trajectory path. It is important to note that angular acceleration can be obtained from the steering wheel angle. Therefore, abrupt changes in steering wheel angle indicate abrupt changes in angular acceleration, and vice versa.

[0106] If the angular acceleration and steering wheel angle change abruptly during operation, and this change could occur at one or more points, a new control point is added at the point of the change. Based on the first optimization, the AGV's trajectory path is then optimized a second time. In other words, the second optimization is performed on the first optimized trajectory path, and the resulting optimized path is then re-output as the first optimized trajectory path.

[0107] The second optimization process is as follows: The first optimized trajectory path includes several discrete points. Eight endpoints are selected from these discrete points, and these eight endpoints are connected to form seven path segments. These eight endpoints and seven path segments together constitute a new fitted curve. Using this new fitted curve, the control points at the points of abrupt change in angular acceleration can be determined. Then, these newly obtained control points are added to the control point set, and the cubic B-spline interpolation method and nonlinear programming problem are used again to calculate the latest first optimized trajectory path. At this point, the second optimization is complete. The first optimized trajectory path obtained after the second optimization will serve as the actual driving path of the AGV.

[0108] Of course, there is also a situation in this application where the angular acceleration and steering wheel angle of the AGV do not change abruptly during the driving process. In this case, the first optimized trajectory path obtained in the first optimization will be used as the actual driving path of the AGV.

[0109] In the embodiments provided in this disclosure, improvements are made to the existing technology to address the problems existing in the background art. These improvements optimize the length of the AGV path during operation, ensuring the AGV quickly reaches its destination and eliminating unnecessary travel paths. Simultaneously, the smoothness of the trajectory is optimized during AGV operation, reducing errors at the junctions of adjacent road segments and eliminating the effects of sudden changes in AGV angular acceleration and steering wheel angle, ensuring stable AGV operation and preventing deviations from the path. Based on the method provided in this disclosure, AGV travel trajectories with higher quality, performance, and accuracy can be obtained, meeting the AGV's operational requirements.

[0110] In one embodiment of this application, the determination of the start and end points of the AGV's travel path in the scene map in step 100 will be further explained and elaborated below. In one embodiment, the AGV is equipped with a deviation measuring mechanism, which is used to measure the deviation value of the AGV from the end point. Optionally, a 3D camera is mounted on the forklift side of the forklift-type AGV, and an industrial control computer responsible for calculation is placed on the top cover of the AGV. The 3D camera and the industrial control computer are collectively referred to as the deviation measuring mechanism. The process of determining the start and end points may include, but is not limited to, the following steps:

[0111] Set the midpoint of the line connecting the centers of the rear wheel axles on both sides of the AGV as the reference point, and establish the reference coordinate system of the AGV with the reference point as the origin;

[0112] Transform the reference point from the reference coordinate system to the world coordinate system to obtain the world coordinates (x, y, y) of the reference point. start y start θ start The world coordinates of the reference point are used as the starting point of the AGV in the scene map;

[0113] Obtain the deviation value measured by the deviation measuring mechanism, convert the deviation value to the AGV's reference coordinate system, and determine the world coordinates of the measuring point as (x, y, z) based on the reference point and the deviation value. end y end θ end The world coordinates of the measurement point are used as the endpoint of the AGV in the scene map.

[0114] In one embodiment of this application, the selection of six control points based on the length of the path in step 100 to form a control point set will be further explained and elaborated below. Using the path start point as the world coordinate system of the AGV's current reference point and the path end point as the world coordinate system of the measurement point, a path from the start point to the end point can be generated. The entire path can be divided into two parts: a curved section for correction and a final fixed-length straight section. Measuring the distance L from the calculation reference point to the end of the fork, the world coordinates of the end point of the curved section are (x... end y end θend The straight section is a straight line of length L from the endpoint of the curve to the measurement endpoint. In this embodiment, the curved section is generated using cubic B-spline interpolation. Therefore, before generating the curved section, it is necessary to confirm the number of control points and their coordinates.

[0115] Specific reference Figure 2 As shown, Figure 2 This is a schematic diagram of a B-spline trajectory path based on six control points provided in an embodiment of this application. This embodiment selects six control points, and the world coordinates of the six control points are as follows:

[0116] (x start -m1*cosθ start y start -m1*sinθ start );

[0117] (x start y start );

[0118] (x start +m1*cosθ start y start +m1*sinθ staft );

[0119] (x end -m2*cosθ end y end -m2*sinθ end );

[0120] (x end y end );

[0121] (x end +m2*cosθ end y end +m2*sinθ end ).

[0122] Where m1 and m2 are modulus coefficients, representing the distance between two adjacent control points. In this embodiment, m1 and m2 are used as optimization variables. And (x start y start θ start (x) represents the starting coordinates. end y end θ end () represents the endpoint coordinates.

[0123] Based on the above embodiments, one embodiment of this application will be further described and explained below. Step 200 is path planning based on a trajectory model with six control points. Figure 1 Step 200 in the illustrated embodiment may include, but is not limited to, the following steps:

[0124] Step 210: Use cubic B-spline interpolation to interpolate between two adjacent control points in the control point set to generate a fitted curve for these two adjacent control points.

[0125] Step 220: Connect the fitted curves of each pair of adjacent control points to form an initial fitted curve from the start point to the end point.

[0126] In the above steps, firstly, a cubic B-spline curve is selected. Based on the six control points confirmed above, the expression of the fitted curve P(x, y) obtained by cubic B-spline interpolation satisfies the following formula (1):

[0127]

[0128] The four consecutive control points are represented by P0(x0, y0), P1(x1, y1), P2(x2, y2), and P3(x3, y3). The fitted curve P(x, y) is the curve segment formed by these four control points.

[0129] Then, by accumulating t from 0 to 1 with a certain step size, a fitting curve P(t) consisting of several discrete points P(i) can be generated. After completing the interpolation operation for two adjacent control points, move one control point forward and perform the interpolation operation for the next set of two adjacent control points until the last control point, thus generating the fitting curve for each set of two adjacent control points.

[0130] According to an embodiment of this application, the process of constructing the nonlinear programming problem in step 300 and obtaining the optimal solution will be further explained and elaborated below. Figure 1 Step 300 in the illustrated embodiment may include, but is not limited to, the following steps.

[0131] Step 310: Determine the curvature of each discrete point in the initial fitted curve, and determine the total cost value based on the curvature of each discrete point;

[0132] Step 320: Take the minimum total cost value as the objective function, take the optimization variables m1 and m2 as input variables, and take the maximum value of the steer wheel rotation angular acceleration of the AGV that traverses the initial fitted curve as less than the first set threshold as the constraint function to construct a nonlinear programming problem.

[0133] Step 330: When the nonlinear programming problem has an optimal solution, construct the characteristic optimal curve based on the optimization variables m1 and m2 at this time.

[0134] In this specific embodiment, a high-performance path needs to meet the following conditions: during operation, the angular acceleration of the AGV's steering wheel should not be too large, and the turning radius of the path should not be too small. Therefore, this application introduces a nonlinear programming method to find the optimal solution and obtain the optimal path for the AGV. In this embodiment, the path planning problem is transformed into a nonlinear programming problem with inputs m1 and m2, constraint functions consisting of inequality constraints on the angular acceleration of the AGV's steering wheel, and objective functions consisting of finding the minimum total cost with the sum of the turning radii of discrete points along the path.

[0135] Nonlinear programming problems satisfy the following formula:

[0136] input m1, m2

[0137] min Cost

[0138] Subjuct to C<maxA

[0139] In the above formulas, the optimization variables are represented by m1 and m2; the total cost is represented by Cost; C represents the maximum value of the steering wheel angular acceleration of the AGV traversing the initial fitted curve; max A represents the first set threshold, which is the calibrated maximum value limit of the steering wheel angular acceleration. It should be noted that the first set threshold can be set according to the actual situation, and this application does not impose specific limitations on it.

[0140] The objective function is used to characterize the length of the optimized fitted curve, and the constraint function is used to characterize the smoothness of the optimized fitted curve.

[0141] When an optimal solution exists for this nonlinear programming problem, a curve is constructed using m1 and m2 at this point. This curve is the optimal curve that meets the constraints and has the lowest cost.

[0142] In another embodiment of this application, when the nonlinear programming model does not have an optimal solution, the AGV is controlled to retreat by a first set length, and step 300 is executed with the reference point after the AGV retreats as the starting point; or, the AGV is controlled to rotate in place by a set angle in the opposite direction of the cargo offset angle and then step 300 is executed.

[0143] Furthermore, step 310, determining the total cost value, may include, but is not limited to, the following steps:

[0144] First, the coordinates of all discrete points in the initial fitted curve can be obtained through (x... N y N Let N = 1, 2, 3, ... . According to formula (1), for (x... N y NThen, increment the value of t by one step and decrement it by one step. Next, substitute the values ​​of (t-1) and (t+1) into formula (1) to obtain (x). N y N The previous point (x) N-1 y N-1 ) and the next point (x) N+1 y N+1 ).

[0145] Based on the above three points, calculate (x) N y N The turning radius at point (). Specifically:

[0146] First, calculate the previous point (x) N-1 y N-1 ) and the current point (x) N y N The distance S1 between them:

[0147]

[0148] Then, calculate the previous point (x) N-1 y N-1 ) and the next point (x) N+1 y N+1 The distance S2 between them:

[0149]

[0150] Furthermore, calculate the current point (x) N y N ) and the next point (x) N+1 y N+1 The distance S3 between them:

[0151]

[0152] The turning radius of the current point can be calculated using the three distances mentioned above.

[0153]

[0154] Then Rc is (x N y N The turning radius at point ().

[0155] In actual AGV operation, a larger turning radius reduces the likelihood of the AGV deviating from its path during turns. Therefore, in the nonlinear programming problem of this application, the derivative of the turning radius is used as the cost index for the nonlinear programming problem, i.e., the curvature k = R is used. c -1This will be used as the cost metric. Therefore, the expected total cost of the path is: This refers to the sum of the curvatures of all discrete points in the initial fitted curve. Here, N is the number of discrete points in the initial fitted curve. In solving nonlinear programming problems, finding the minimum value of Cost is equivalent to finding the maximum value of the turning radius.

[0156] The above describes the entire process of the first optimization of the AGV's travel path in the secondary optimization method disclosed in this application. (Refer to...) Figure 3 , Figures 4A to 4B , Figure 3 The diagram shown is a schematic diagram of the first optimized trajectory path obtained after the first optimization according to an embodiment of this application; Figure 4A The diagram shown is a schematic representation of the turning radius of the first optimized path provided in an embodiment of this application. Figure 4B The diagram shows the steering wheel angle and angular velocity of the first optimized AGV provided in this embodiment of the application. (By...) Figure 3 , Figure 4A and Figure 4B It can be seen that the embodiments of this application optimize the length of the AGV's travel path and the smoothness of its trajectory, which can ensure that the AGV can reach the destination quickly, improve the stability of the AGV's travel, and avoid deviation from the path.

[0157] Based on the above embodiments, for paths with a relatively long total length, the first optimized trajectory path obtained in the first optimization is a three-segment B-spline path generated through six control points. Due to the significant differences between the path segments, the steering wheel angle and angular acceleration of the AGV are prone to sudden changes when it reaches the junction of two adjacent path segments, leading to unnecessary errors. Furthermore, the steering wheel angle change function of the AGV is not a convex function; when the AGV's steering wheel turns or changes clockwise or counterclockwise, its angle and angular acceleration will also generate unnecessary errors.

[0158] In this situation, not only is it detrimental to the driving control of the AGV, but it can also cause the AGV to deviate from the path when following the trajectory.

[0159] Therefore, after obtaining the first optimized trajectory path, this application detects whether the AGV will experience a sudden change in angular acceleration while traveling on this path. If so, the first optimized trajectory path is optimized a second time to improve its accuracy. If not, the first optimized trajectory path is adopted as the final travel trajectory of the AGV.

[0160] Before performing the second optimization, it is necessary to determine whether the angular acceleration of the steering wheel changes abruptly during travel along the first optimized trajectory path. This determination process is implemented through step 400. An embodiment of this application will be further explained and elaborated below regarding step 400. Figure 1 Step 400 in the illustrated embodiment may include, but is not limited to, the following steps. Here, a single-steering-wheel AGV is used as an example:

[0161] Step 410: Use the curve with the optimal characteristics as the first optimized trajectory path of the AGV, and define the coordinate position of the steering wheel relative to the starting point as (x... f y f ), calculate the steering wheel angle when the AGV is traveling on the first optimized trajectory path.

[0162] It should be noted that the steering wheel angle satisfies formula (2):

[0163]

[0164] The steering wheel angle is represented by α. k is the curvature of the discrete point in step 310.

[0165] Step 420: Calculate the steering wheel acceleration based on the steering wheel rotation angle.

[0166] Specifically, the angular velocity v of the steering wheel is first solved using the discrete difference method. r The angular velocity satisfies formula (3):

[0167]

[0168] The AGV's set travel speed is determined by v. s To represent, v s It is a preset value.

[0169] Therefore, the time difference T can be calculated using the angular velocity as shown in the following formula (4):

[0170]

[0171] The length of the first optimized trajectory path is represented by s.

[0172] Then, the angular acceleration of the steering wheel is calculated using the following formula (5):

[0173]

[0174] Among them, the angular acceleration of the steering wheel is achieved through... Let i' represent the counter symbol.

[0175] Step 430: Based on the expression for the steering wheel angular velocity, traverse the entire first optimized trajectory path to find the maximum value of the steering wheel angular acceleration. By using the maximum value of the steering wheel angular acceleration, determine whether the steering wheel angular acceleration changes abruptly during the AGV's movement.

[0176] In this step, we traverse the entire path to find... If the maximum value is found, then the following sub-constraint function is constructed:

[0177]

[0178] Where MaxValue represents the angular velocity threshold, which is the calibrated angular acceleration. In this specific embodiment, based on the characteristics of the AGV, the value of MaxValue is taken as 40π / 180.

[0179] Then, the sub-constraint function is used to determine whether the steering wheel angular acceleration has abruptly changed. Specifically, if the maximum value of the steering wheel angular acceleration is greater than the angular acceleration threshold, it is considered that the steering wheel angular acceleration has abruptly changed; if the maximum value of the steering wheel angular acceleration is less than or equal to the angular acceleration threshold, it is considered that the steering wheel angular acceleration has not abruptly changed.

[0180] When it is determined that the AGV's steering wheel angular acceleration undergoes a sudden change during travel along the first optimized trajectory path, step 500 is executed. Steps 500-600 constitute the second optimization process for the AGV's trajectory path, which is implemented by adding new control points. An embodiment of this application will be further explained and elaborated below regarding steps 500-600. Figure 1 Step 500 in the illustrated embodiment may include, but is not limited to, the following steps.

[0181] First, all discrete points constituting the trajectory path are calculated according to formula (1), that is, all discrete points constituting the first optimized trajectory path. The first optimized trajectory path includes three path segments, which are calculated using six control points. Seven discrete points are selected from all discrete points as new path segment endpoints to reconstruct a new fitting curve. The new fitting curve includes eight endpoints and seven path segments.

[0182] In this specific embodiment, each path segment consists of 100 discrete points, so the first optimized trajectory path has a total of 300 discrete points. Given that 7 path segments are suitable for most AGV application scenarios in practice, this embodiment uses the 1st, 43rd, 86th, 129th, 172nd, 215th, 258th, and 300th discrete points as the endpoints of new path segments, reconstructing a trajectory path composed of 7 path segments. The starting point is the 1st endpoint, and the ending point is the last endpoint, i.e., the 300th endpoint.

[0183] Then, the coordinates of these eight endpoints, namely the coordinates of the 1st, 43rd, 86th, 129th, 172nd, 215th, 258th, and 300th discrete points, are calculated using formula (1). Furthermore, based on the coordinates of these eight endpoints, the straight-line distances l1, l2, l3, l4, l5, l6, and l7 between adjacent endpoints are calculated. In this specific embodiment:

[0184] l1 is the straight-line distance between the 1st discrete point and the 43rd discrete point;

[0185] l2 is the straight-line distance between the 43rd discrete point and the 86th discrete point;

[0186] l3 is the straight-line distance between the 86th discrete point and the 129th discrete point;

[0187] l4 is the straight-line distance between the 129th discrete point and the 172nd discrete point;

[0188] l5 is the straight-line distance between the 172nd discrete point and the 215th discrete point;

[0189] l6 is the straight-line distance between the 215th discrete point and the 258th discrete point;

[0190] l7 is the straight-line distance between the 258th and 300th discrete points.

[0191] The sum of these seven distances is L = l1 + l2 + l3 + l4 + l5 + l6 + l7.

[0192] Let the endpoint vector be:

[0193]

[0194] Wherein, the endpoint vector passes through U n Let Δ be used to represent this. i =U i+1 -U i Where i represents the element index in the endpoint vector sequence, the boundary conditions are obtained as shown in formulas (6)-(7):

[0195]

[0196]

[0197] Where q represents the coordinates of the endpoints. This allows us to obtain the parameters a1, b1, c1, and e1 of the first endpoint and the parameters a8, b8, c8, and e8 of the last endpoint.

[0198] Apart from the boundary endpoints, namely the start and end points, the parameters of the other endpoints can be calculated using the following formula (8):

[0199]

[0200] Where i = [2, 7]. By repeating the above formula, all parameters a from the second endpoint to the seventh endpoint can be obtained. i b i c i and e i i = [2, 7]. The parameter a of these eight endpoints... i b i c i and e i Substituting i = [1, 8] into the following formula, we obtain an equation with the coordinates of the control point as the unknown quantity, as shown in formula (9):

[0201]

[0202] Inverting formula (9) yields formula (10):

[0203]

[0204] Finally, solving formula (10) above yields the coordinates of several auxiliary control points P1, P2, ..., P... n-3 P n-2 .

[0205] After obtaining the coordinates of the auxiliary control points, these coordinates are added to the control point set. Repeating steps 200 to 400 yields the second optimization result, which is the optimized first trajectory path. The above describes the entire process of the second optimization of the AGV's travel path in the secondary optimization method disclosed in this application.

[0206] Reference Figure 5 , Figure 6A and Figure 6B As shown, Figure 5 The diagram shown is a schematic diagram of the first optimized trajectory path obtained after the second optimization provided in the embodiment of this application; Figure 6A The diagram shown is a schematic representation of the turning radius of the second optimized path provided in an embodiment of this application. Figure 6B The diagram shows the steering wheel angle and angular velocity of the second optimized AGV provided in this embodiment. Figure 5 , Figure 6A and Figure 6BAs can be seen, after the second optimization, the abrupt changes in the first optimized trajectory path obtained after the first optimization disappeared, and the smoothness of the trajectory path was further improved. In other words, this application reduces the error at the junction of adjacent road segments, eliminates the influence of abrupt changes in the AGV's angular acceleration and steering wheel angle, ensures the stability of the AGV's travel, and avoids deviation from the path.

[0207] Based on the above embodiments, the principle and execution process of the technical solution of this application will be illustrated below with an example. In this example, the task of the AGV is to deliver goods from a set starting point to a destination, that is, the endpoint.

[0208] First, a map of the scene where the AGV is located is obtained, and the delivery start and end points of the AGV are marked. Six control points are selected based on the length of this path, forming a control point set. Then, a cubic B-spline interpolation method based on the six control points is used to interpolate the control point set to obtain an initial fitted curve. In existing technologies, after obtaining the fitted curve through cubic B-spline interpolation, this curve is usually used as the AGV's travel path. However, there are many areas with large curvatures on the fitted curve, making it extremely easy for the AGV's steering wheel angular acceleration and steering wheel angle to change abruptly during travel, which can lead to the AGV deviating from the path when following the trajectory.

[0209] Therefore, the initial fitted curve needs to be optimized first, i.e., the first optimization. The following nonlinear programming problem is constructed:

[0210] input m1, m2

[0211] min Cost

[0212] subjuct to C<maxA

[0213] Wherein, the curvature k = R of discrete points is used. c -1 As an indicator of total cost, Cost satisfies When the nonlinear programming problem has an optimal solution, an optimal characteristic curve is constructed based on m1 and m2. This optimal characteristic curve may represent the AGV's travel path.

[0214] Due to the significant differences between path segments, the steering wheel angle and angular acceleration of the AGV can easily change abruptly when it reaches the junction of two adjacent path segments, leading to unnecessary errors. Furthermore, the steering wheel angle change function of the AGV is not a convex function; when the AGV's steering wheel turns or changes clockwise or counterclockwise, its angle and angular acceleration will also generate unnecessary errors. Therefore, before using the characteristic optimization curve as the AGV's final travel path, it is necessary to check whether the AGV's angular acceleration undergoes abrupt changes when traveling on this characteristic optimization curve.

[0215] Specifically, by solving the sub-constraint functions This will reveal whether the AGV's angular acceleration changes abruptly. When the AGV travels along this characteristic optimization curve, its angular acceleration will not change abruptly. Therefore, this characteristic optimization curve obtained from the first optimization can be used as the AGV's final travel path.

[0216] However, when the AGV travels on this characteristic optimization curve, its angular acceleration changes abruptly. Therefore, it is necessary to optimize the characteristic optimization curve obtained from the first optimization again, i.e., the second optimization.

[0217] The principle of the second optimization is as follows: based on the initial optimization curve, a new fitted curve is reconstructed, which is a trajectory path composed of 7 path segments and 8 endpoints. Then, the endpoint vector U is set. n And relation Δ i =U i+1 -U i Boundary conditions are constructed, and the parameters a1, b1, c1, and e1 of the first endpoint and the parameters a8, b8, c8, and e8 of the last endpoint are calculated. Then, the parameters a1, b1, c1, and e8 of the other endpoints are calculated using a specific formula. i b i c i and e i Let i = [2, 7]. Based on all the parameters of these eight endpoints, obtain an equation with the unknown quantity being the coordinates of the control points. Invert this equation to obtain the inverse equation. Finally, solve the inverse equation to obtain multiple auxiliary control points, which are added to the control point set. After adding the auxiliary control points to the control point set, repeat steps 200 to 400, that is, re-apply the cubic B-spline interpolation method to obtain the initial fitted curve, and then solve the optimal solution of the nonlinear programming problem to obtain the characteristic optimal curve. Then, the characteristic optimal curve at this time is both the result obtained from the second optimization and the final driving path of the AGV.

[0218] Based on the above embodiments, this disclosure also provides an AGV path secondary optimization device based on B-splines. The device includes:

[0219] The acquisition unit is responsible for acquiring the scene map and determining the start and end points of the AGV's travel path within the scene map.

[0220] The interpolation generation unit selects multiple control points based on the length of the path, forming a control point set. It then uses cubic B-spline interpolation to interpolate this control point set, obtaining an initial fitted curve from the starting point to the ending point. The initial fitted curve consists of several discrete points.

[0221] The planning unit is used to determine the nonlinear programming problem based on several discrete points. When the nonlinear programming problem has an optimal solution, it constructs the characteristic optimal curve based on the optimal solution.

[0222] The judgment unit is responsible for taking the optimal characteristic curve as the first optimized trajectory path of the AGV, calculating the steering wheel angle and steering wheel angular acceleration of the AGV traveling on the first optimized trajectory path, and judging whether the steering wheel angular acceleration changes abruptly during the AGV's travel.

[0223] The optimization unit's function is to re-plan the initial fitted curve into a new fitted curve including eight endpoints and seven paths when the steering wheel angular acceleration changes abruptly. Several auxiliary control points are obtained through the new fitted curve, and these auxiliary control points are added to the control point set as new control points.

[0224] The output unit's function is to use the first optimized trajectory path as the AGV's final travel path when the steering wheel angular acceleration does not change abruptly during the AGV's travel.

[0225] In addition, this application also provides an AGV, including: a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements an AGV path quadratic optimization method based on B-spline curves.

[0226] In addition, this application embodiment also provides a storage medium storing computer-executable instructions for executing a secondary optimization method for AGV paths based on B-spline curves.

[0227] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented, for example, in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “having,” and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatuses.

[0228] It should be understood that in this application, "at least one (item)" means one or more, and "more than" means two or more. "And / or" is used to describe the relationship between related objects, indicating that three relationships can exist. For example, "A and / or B" can represent three cases: only A exists, only B exists, and both A and B exist simultaneously, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one (item) of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one (item) of a, b, or c can represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.

[0229] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be indirect coupling or communication connection through some interfaces, apparatuses, or units, and may be electrical, mechanical, or other forms.

[0230] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0231] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0232] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0233] The step numbers in the above method embodiments are set only for ease of explanation and do not limit the order of the steps. The execution order of each step in the embodiments can be adaptively adjusted according to the understanding of those skilled in the art.

Claims

1. A method for quadratic optimization of AGV paths based on B-spline curves, characterized in that, Includes the following steps: S100, acquire the scene map, determine the start and end points of the AGV's travel path in the scene map, select six control points according to the length of the path to form a control point set, the control point set including the start point and the end point; S200, the control point set is interpolated using the cubic B-spline interpolation method to obtain an initial fitting curve from the starting point to the ending point, the initial fitting curve being composed of several discrete points; S300, determine the nonlinear programming problem based on the plurality of discrete points, and when the nonlinear programming problem has an optimal solution, construct the characteristic optimal curve based on the optimal solution; S400, the optimal characteristic curve is used as the first optimized trajectory path of the AGV, and it is determined whether the steering wheel angular acceleration of the AGV changes abruptly during the process of the AGV traveling on the first optimized trajectory path; if not, the first optimized trajectory path is used as the final travel path of the AGV. If so, switch to S500; S500, when the steering wheel angular acceleration changes abruptly, the first optimized trajectory path is replanned into a new fitted curve including eight endpoints and seven path segments, and several auxiliary control points are obtained through the new fitted curve; S600, add the aforementioned auxiliary control points as new control points to the control point set, and repeat S200-S400 in a loop; Specifically, when the steering wheel angular acceleration undergoes a sudden change, the initial fitted curve is reprogrammed into a new fitted curve comprising eight endpoints and seven paths. Several auxiliary control points are obtained through the new fitted curve, including: Calculate all discrete points that make up the first optimized trajectory path, select seven discrete points from all the discrete points as new path segment endpoints, and reconstruct a new fitting curve, the new fitting curve including eight endpoints and seven path segments; Calculate the coordinates of the eight endpoints, and based on these coordinates, calculate the straight-line distance between any two adjacent endpoints. ; Let the endpoint vector be It satisfies the following formula: ； in, ; set up Where i is the element index in the endpoint vector sequence, the following boundary conditions are obtained: ， ； Where q is the coordinate of the endpoint; This represents the coordinates of the first of the eight endpoints. This represents the coordinates of the second endpoint out of the eight endpoints. This represents the coordinates of the seventh endpoint out of the eight endpoints. This represents the coordinates of the eighth endpoint out of the eight endpoints; The parameters of the remaining six endpoints (excluding the first and eighth endpoints) out of the eight endpoints can be calculated using the following formula. and i = [2, 7]: ； Through the parameters of these eight endpoints and Given i = [1, 8], we can obtain an equation with the unknown quantity being the coordinates of the current control point: ； Taking the inverse of the above formula, we get: ； Finally, solving the above formula yields the coordinates of several auxiliary control points. .

2. The AGV path quadratic optimization method based on B-spline curves according to claim 1, characterized in that, In S200, the step of interpolating the control point set using the cubic B-spline interpolation method to obtain an initial fitted curve from the starting point to the ending point includes: The cubic B-spline interpolation method is used to interpolate between two adjacent control points in the control point set to generate a fitting curve for these two adjacent control points; Connect the fitted curves of each pair of adjacent control points to form an initial fitted curve from the starting point to the ending point.

3. The AGV path quadratic optimization method based on B-spline curves according to claim 1, characterized in that, In S300, the step of determining the nonlinear programming problem based on the plurality of discrete points, and when the nonlinear programming problem has an optimal solution, constructing an optimal characteristic curve based on the optimal solution, includes: Determine the curvature of each discrete point in the initial fitted curve, and determine the total cost value based on the curvature of each discrete point; The minimum value of the total cost is taken as the objective function, and the optimization variables are... m1 , m2 As input variables, the maximum value of the steer wheel rotation angular acceleration of the AGV traversing the initial fitted curve is less than a first set threshold, which is used as a constraint function to construct the following nonlinear programming problem: Where m1 and m2 are both optimization variables, Cost This is the total cost value. C To determine the maximum value of the steering wheel angular acceleration of the AGV that iterates through the initially fitted curve, max A Set a threshold for the first one; When the nonlinear programming problem has an optimal solution, the optimization variables at this time are... m1 and m2 Construct the curve with optimal characteristics.

4. The AGV path quadratic optimization method based on B-spline curves according to claim 3, characterized in that, The step of determining the curvature of each discrete point in the initial fitted curve and determining the total cost value based on the curvature of each discrete point includes: definition Let the coordinates of all discrete points in the initial fitted curve be given. By adding and subtracting the step size from the t value, we obtain... and ; N =1,2,3,…; according to , and ,get Turning radius Rc at the point: ； ； ； ； Curvature of discrete points As an indicator of the total cost value, the total cost value is ,in N The number of discrete points in the initial fitted curve.

5. The AGV path quadratic optimization method based on B-spline curves according to claim 4, characterized in that, In S400, the step of using the optimal characteristic curve as the first optimized trajectory path of the AGV and calculating the steering wheel angle and steering wheel angular acceleration of the AGV traveling on the first optimized trajectory path includes: First, the optimal characteristic curve is used as the first optimized trajectory path of the AGV, and the coordinate position of the steering wheel relative to the starting point is defined as follows: Calculate the steering wheel angle when the AGV is traveling on the first optimized trajectory path. : ； Then, based on the steering wheel rotation angle, the angular velocity of the steering wheel is calculated: ； in, Set the travel speed for the AGV. Angular velocity; The time difference is calculated using the angular velocity of the steering wheel: ； Where T is the time difference and s is the length of the first optimized trajectory path; The following steering wheel angular acceleration can then be obtained. : ； Where i' is the counter symbol.

6. The AGV path quadratic optimization method based on B-spline curves according to claim 5, characterized in that, The determination of whether the steering wheel angular acceleration changes abruptly during the AGV's operation includes: Based on the expression for the steering wheel angular velocity, the entire first optimized trajectory path is traversed to find the maximum value of the steering wheel angular acceleration. By using the maximum value of the steering wheel angular acceleration, it is determined whether the steering wheel angular acceleration changes abruptly during the AGV's operation. Wherein, when the maximum value of the steering wheel angular acceleration is greater than the angular acceleration threshold, it is considered that the steering wheel angular acceleration has abruptly changed; when the maximum value of the steering wheel angular acceleration is not greater than the angular acceleration threshold, it is considered that the steering wheel angular acceleration has not abruptly changed.

7. A secondary optimization device for AGV paths based on B-splines, characterized in that, include: The acquisition unit is used to acquire a scene map and determine the start and end points of the AGV's travel path in the scene map; An interpolation generation unit is used to select multiple control points according to the length of the path to form a control point set, and to interpolate the control point set using a cubic B-spline interpolation method to obtain an initial fitting curve from the starting point to the ending point. The initial fitting curve is composed of several discrete points. The planning unit is used to determine a nonlinear programming problem based on the plurality of discrete points, and when the nonlinear programming problem has an optimal solution, to construct an optimal characteristic curve based on the optimal solution; The judgment unit is used to take the optimal characteristic curve as the first optimized trajectory path of the AGV, calculate the steering wheel rotation angle and steering wheel angular acceleration of the AGV traveling on the first optimized trajectory path, and determine whether the steering wheel angular acceleration changes abruptly during the travel of the AGV. The optimization unit is used to re-plan the initial fitting curve into a new fitting curve including eight endpoints and seven paths when the steering wheel angular acceleration changes abruptly, obtain several auxiliary control points through the new fitting curve, and add the several auxiliary control points as new control points to the control point set. The output unit is used to take the first optimized trajectory path as the final travel path of the AGV when the steering wheel angular acceleration does not change abruptly during the travel of the AGV. Specifically, when the steering wheel angular acceleration undergoes a sudden change, the initial fitted curve is reprogrammed into a new fitted curve comprising eight endpoints and seven paths. Several auxiliary control points are obtained through the new fitted curve, including: Calculate all discrete points that make up the first optimized trajectory path, select seven discrete points from all the discrete points as new path segment endpoints, and reconstruct a new fitting curve, the new fitting curve including eight endpoints and seven path segments; Calculate the coordinates of the eight endpoints, and based on these coordinates, calculate the straight-line distance between any two adjacent endpoints. ; Let the endpoint vector be It satisfies the following formula: ； in, ; set up Where i is the element index in the endpoint vector sequence, the following boundary conditions are obtained: ， ； Where q is the coordinate of the endpoint; This represents the coordinates of the first of the eight endpoints. This represents the coordinates of the second endpoint out of the eight endpoints. This represents the coordinates of the seventh endpoint out of the eight endpoints. This represents the coordinates of the eighth endpoint out of the eight endpoints; The parameters of the remaining six endpoints (excluding the first and eighth endpoints) out of the eight endpoints can be calculated using the following formula. and i = [2, 7]: ； Through the parameters of these eight endpoints and Given i = [1, 8], we can obtain an equation with the unknown quantity being the coordinates of the current control point: ； Taking the inverse of the above formula, we get: ； Finally, solving the above formula yields the coordinates of several auxiliary control points. .

8. An AGV, comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, when the processor executes the computer program, it implements the AGV path quadratic optimization method based on B-spline curves as described in any one of claims 1 to 6.

9. A storage medium storing computer-executable instructions, characterized in that, The computer-executable instructions are used to execute the AGV path quadratic optimization method based on B-spline curves as described in any one of claims 1 to 6.

Images