Servo motor conveying system for correcting driving path through vision AI algorithm

Abstract

The invention relates to the technical field of automobile part conveying visual path correction, in particular to a servo motor conveying system for correcting a driving path through a visual AI algorithm, and the system comprises a path visual extraction module, a path characteristic calculation module, a dynamic region recognition module, a disturbance parameter modeling module, a servo path correction module and a path priority adjustment module. According to the method, a high-precision path boundary feature matrix is generated through real-time image data acquisition and edge feature extraction, accurate data support is provided for path analysis, path feature analysis is combined with gradient change intensity and a continuity index, the integrity and accuracy of path trend expression are enhanced, and the accuracy of the path trend expression is improved. According to the dynamic region division, through comprehensive analysis of curvature changes and texture indexes, the region recognition precision is improved, the illumination gradient, the texture ratio and the obstacle offset are modeled in real time, servo signal adjustment is optimized, path correction is faster, and the method adapts to complex dynamic scenes.

Description

Technical Field

[0001] The present invention relates to the technical field of visual path correction for automobile parts transportation, and in particular to a servo motor conveying system for correcting a driving path using a visual AI algorithm. Background Art

[0002] The field of visual path correction technology for automobile parts transportation includes automation control and path planning technologies related to automobile parts transportation. The core content is to transport automobile parts efficiently, accurately and safely through automated equipment. This technical field generally involves path planning, positioning navigation, motion control and other aspects of automatic guided vehicles (AGVs). The focus is on achieving accurate control of transport vehicles through visual sensors, path detection algorithms and servo motor drives to ensure that they can adjust the path and reach the target according to the real-time environment. The systematic content of the overall technical field includes vehicle navigation control, path planning algorithms, visual information acquisition and processing, and vehicle driving parameter optimization, as well as EMS aerial conveyor lines, which are overhead conveyor systems for in-plant logistics. EMS aerial conveyor lines are an aerial suspension conveying system that runs on a preset track in the air, and multiple cargo carriers work independently. It adopts a modular design and can be designed into many operating forms according to customer needs. It has a flexible layout and is equipped with a hoisting unit to form a three-dimensional conveying system in the horizontal and vertical directions. It consists of independent vehicles that run freely on the track system. The running track is suspended from the air with a bracket. This allows the floor area under the track to be freely used for work and production purposes. The rail system allows any number of vehicles to run continuously or in cycles to achieve efficient completion of different transportation tasks.

[0003] Among them, the servo motor conveying system that uses a visual AI algorithm to correct the driving path refers to an automatic guided vehicle device that uses a visual algorithm to perceive and analyze the path offset information in the vehicle's operating environment in real time and performs path correction based on this. The technical matters targeted by this patent subject include using visual sensors to capture environmental data of the vehicle's operating path and extracting path offset information through a path detection algorithm, and correcting the servo motor's control signal based on visual information to achieve path correction. Specific methods include extracting edge points or path features in the path detection algorithm, calculating the path offset through geometric analysis, and real-time adjustment of the servo motor drive control signal to ensure that the vehicle travels along the established route.

[0004] In the path identification link, the existing technology only extracts a single feature dimension of the path data, lacks the coordinated processing of multi-dimensional parameters such as gradient, texture, and curvature, resulting in insufficient expression of the path boundary features. This lack of data granularity makes the results of edge feature extraction inconsistent in complex path environments, making it difficult to serve as an accurate basis for dynamic area division. In dynamic area division, the existing technology relies on a single edge gradient parameter for area judgment, lacks a comprehensive evaluation of the coordination of the path continuity index and multi-dimensional parameters, and easily leads to large deviations in the division results of dynamic stable areas and unstable areas. The real-time response capability to dynamic obstacles and environmental disturbances is insufficient, the compensation parameter calculation process of the existing technology has a lag, and the adjustment of the servo motor signal cannot adapt to the complex environment in real time, which may lead to delays in path correction or even cumulative deviations. The above-mentioned deficiencies make it difficult for the existing technology to cope with the path correction needs in complex environments, especially in EMS aerial conveyor lines in highly complex dynamic scenes, where the limitations of its accuracy and efficiency are more obvious. Summary of the invention

[0005] The purpose of the present invention is to solve the shortcomings of the prior art and to propose a servo motor conveying system that uses a visual AI algorithm to correct the driving path.

[0006] In order to achieve the above-mentioned purpose, the present invention adopts the following technical solution: A servo motor conveying system for correcting the driving path with a visual AI algorithm comprises: The path vision extraction module collects the path continuous image data through the visual sensor, segments the edge area of ​​the path image, identifies the gradient distribution value and boundary change characteristics of the image, calculates the range of path edge curvature change, and obtains the path boundary feature matrix; The path characteristic calculation module analyzes the boundary curvature change characteristics of the path segments based on the path boundary feature matrix, calls the gradient change intensity value and the texture continuity index, calculates the path edge change trend and continuity characteristics, and constructs a path characteristic trend data set; The dynamic region identification module formulates a curvature change trend threshold based on the path characteristic trend data set, identifies key positions in the path curvature change trend, divides the range of the texture continuity index, calibrates the parameter distribution of the dynamically stable and dynamically unstable regions, and generates a dynamic region characteristic distribution map; The disturbance parameter modeling module extracts the illumination gradient range of the dynamic unstable area based on the dynamic area characteristic distribution map, analyzes the distribution range of the texture change ratio, calculates the illumination correction value, texture compensation value and obstacle offset compensation value in combination with the dynamic obstacle offset trend data, and generates a disturbance correction parameter set; The servo path correction module calculates the power output adjustment value of the servo motor based on the disturbance correction parameter set, combines the dynamic area illumination correction value and the curvature change trend range, analyzes the steering angle range and the acceleration frequency range, optimizes the servo signal output matching value, and generates a servo path correction signal group; The path priority adjustment module calls the curvature change trend range of the dynamic area based on the servo path correction signal group, combines the interpolation calculation results of the key boundary points in the stable area, adjusts the priority distribution of the dynamically unstable area, optimizes the path morphology of the curvature change area, and generates a path correction optimization result.

[0007] As a further solution of the present invention, the path boundary feature matrix includes edge curvature distribution, gradient change intensity, and texture gradient range; the path characteristic trend data set includes edge change trend, curvature change range, and continuity characteristic distribution; the dynamic area characteristic distribution map includes stable area parameters, dynamic area parameters, and texture discontinuity area division; the disturbance correction parameter set includes illumination compensation value, texture change compensation value, and obstacle offset correction value; the servo path correction signal group includes power adjustment signal, steering angle adjustment signal, and acceleration frequency adjustment signal; the path correction optimization result includes path morphology optimization value, area priority distribution value, and curvature adjustment optimization value.

[0008] As a further solution of the present invention, the path vision extraction module includes an image acquisition submodule, an edge feature recognition submodule, and a curvature feature calculation submodule; The image acquisition submodule is based on a visual sensor to collect continuous image data of the path, uses a path segmented sampling mode to intercept images, extracts continuous frame images and stores them as structured files, checks and cleans up acquisition errors and duplicate images, and generates a path image data set; The edge characteristic recognition submodule performs image edge area segmentation based on the path image data set, extracts the gradient distribution value of the image edge, analyzes the boundary position characteristics according to the distribution value change rate, generates the boundary gradient distribution matrix, analyzes the pixel point change trend, and generates the path edge gradient characteristic matrix; The curvature feature calculation submodule is based on the path edge gradient characteristic matrix, segmentally fits the boundary curvature variation range, extracts the curvature variation rate, calculates the variation gradient of the edge segment, screens the nodes with sudden curvature changes, and obtains the path boundary feature matrix.

[0009] As a further solution of the present invention, the calculation formula of the change gradient value is specifically: ; in, Representative Segment Path The weighted curvature change gradient value of the point, Represents the path edge Section The gradient value of a pixel, Representative Section The weight coefficient of the point is calculated based on the gradient change range of the path area and the dynamic weight matrix. is the square root operation, is the segment number index of the path, The pixel index of the current segment.

[0010] As a further solution of the present invention, the path characteristic calculation module includes a curvature analysis submodule, a gradient and texture analysis submodule, and a trend data construction submodule; The curvature analysis submodule extracts the curvature change value of the path segment boundary based on the path boundary feature matrix, calculates the change rate of the boundary curvature, and classifies and marks it according to the rate fluctuation, calls the curvature change distribution data and performs point-by-point comparative analysis on the spatial position of the change point to generate a curvature change characteristic matrix; The gradient and texture analysis submodule extracts the gradient change intensity value based on the curvature change characteristic matrix, analyzes the fluctuation range of local changes, calculates the path segment texture continuity index, and analyzes the texture coherence feature based on the image pixel distribution as a reference, calls the distribution data to screen the texture abnormal area, and marks the characteristic area to generate the gradient and texture characteristic matrix; The trend data construction submodule calculates the path edge change trend based on the gradient and texture characteristic matrix, and performs matching analysis of characteristic parameters in combination with texture characteristics, analyzes the characteristic values ​​of trend changes, and classifies and organizes key parameters to generate a path characteristic trend data set.

[0011] As a further solution of the present invention, the dynamic region identification module includes a curvature trend threshold setting submodule, a key position identification submodule, and a dynamic region division submodule; The curvature trend threshold setting submodule calculates the curvature change range of each segment based on the path characteristic trend data set, extracts fluctuation data to set a threshold, screens the curvature segments whose fluctuation frequency exceeds the threshold, extracts the difference parameter between the change amplitude and the stable value, and generates the curvature change trend threshold; The key position identification submodule extracts the sudden change position of the curvature change of the path segment based on the curvature change trend threshold, compares the change trend value and the fluctuation characteristics, and calibrates the dynamic distribution nodes to generate dynamic key node distribution characteristics; The dynamic area division submodule classifies and marks the range changes of the texture continuity index based on the distribution characteristics of the dynamic key nodes, sets the regional stability threshold, extracts the continuity feature distribution of the stable area and the unstable area, combines the dynamic change trend, calculates the boundary range index, and generates a dynamic area characteristic distribution map.

[0012] As a further solution of the present invention, the disturbance parameter modeling module includes an illumination gradient extraction submodule, a texture change analysis submodule, and a dynamic compensation calculation submodule; The illumination gradient extraction submodule extracts the distribution data of the illumination gradient in the dynamic unstable area based on the dynamic area characteristic distribution map, identifies the grayscale value of the image pixel, calculates the illumination gradient change amplitude of each area, compares the time series characteristics of the gradient change, identifies the abnormal fluctuation area, constructs the gradient change range matrix and extracts the distribution characteristic value, and generates the illumination gradient range parameter; The texture change analysis submodule extracts the texture change characteristic value in the dynamically unstable area based on the illumination gradient range parameter, analyzes the image texture direction distribution and grayscale difference, calculates the texture change ratio, analyzes the texture fluctuation amplitude in the change area, calibrates the key fluctuation range, screens the texture distribution characteristic value associated with the illumination gradient, and generates the texture change ratio distribution; The dynamic compensation calculation submodule is based on the texture change ratio distribution and combined with the dynamic obstacle offset trend data. By comparing the interaction between regional illumination changes, texture characteristic fluctuations and offset data, it calculates the illumination correction value, texture compensation value and obstacle offset compensation value, integrates multi-dimensional disturbance parameters, and constructs a distribution matrix to generate a set of disturbance correction parameters.

[0013] As a further solution of the present invention, the texture change joint ratio calculation formula is specifically: ; in, Representative Section The joint ratio of texture changes in the region, Representative Section The texture direction gradient value of the point, Representative Region No. The gray value of the point, is the number of pixels in the area, is the index of the pixel, Represents the summation of all pixels in the area.

[0014] As a further solution of the present invention, the servo path correction module includes a power adjustment calculation submodule, a steering and acceleration range analysis submodule, and a servo signal matching optimization submodule; The power adjustment calculation submodule extracts the dynamic area illumination correction value and the dynamic obstacle offset compensation value based on the disturbance correction parameter set, calculates the adjustment amplitude of the servo motor power output value in combination with the gradient fluctuation data in the curvature change trend range, analyzes the matching relationship between the output value and the dynamic parameter, and corrects the power parameter matrix to generate the servo power adjustment value; The steering and acceleration range analysis submodule analyzes the steering angle variation range corresponding to the servo signal based on the servo power adjustment value, combined with the path curvature variation trend range and the dynamic area characteristic distribution, calculates the peak value in the acceleration frequency range, analyzes the fluctuation amplitude, compares the dynamic area signal characteristics, adjusts the steering and acceleration limit parameters, and generates the servo steering and acceleration range parameters; The servo signal matching optimization submodule calls the servo power adjustment value data matrix based on the servo steering and acceleration range parameters, performs characteristic analysis on the frequency, amplitude and output power of the servo signal, adjusts the output mode of the servo signal according to the dynamic characteristic range, constructs a signal adjustment matrix in combination with the path dynamic distribution, and generates a servo path correction signal group.

[0015] As a further solution of the present invention, the path priority adjustment module includes a priority distribution adjustment submodule, a dynamic area morphology optimization submodule, and a path correction result generation submodule; The priority distribution adjustment submodule calls the curvature change trend range data of the dynamic area based on the servo path correction signal group, extracts the curvature distribution in the dynamic area section by section, calculates the interpolation of the boundary points in the stable area, adjusts and calibrates the path priority parameter correction value of the dynamic unstable area, and generates the priority distribution adjustment parameter; The dynamic region morphology optimization submodule adjusts the parameters based on the priority distribution, extracts the distribution characteristics of the boundary points in the curvature change region, combines the curvature change ranges of the dynamic region and the stable region, gradually optimizes the dynamic region morphology, analyzes the offset value of the dynamic boundary point interpolation and the curvature change law, adjusts the distribution characteristics of the curvature boundary and optimizes the region morphology, and generates optimized path morphology characteristics; The path correction result generation submodule is based on the optimized path morphological characteristics, combined with the priority distribution data of the dynamic area and the curvature change boundary distribution, calculates the morphological parameters of the path after adjustment, analyzes the matching value of the servo signal to the path morphology, integrates and verifies the regional correction effect, path morphology and servo output signal, and generates the path correction optimization result.

[0016] Compared with the prior art, the advantages and positive effects of the present invention are: In the present invention, a high-precision path boundary feature matrix is ​​generated through real-time image data acquisition and edge feature extraction to provide accurate data support for path analysis. Path feature analysis is combined with gradient change intensity and continuity index to enhance the integrity and accuracy of path trend expression. Dynamic area division improves the accuracy of area recognition through comprehensive analysis of curvature change and texture index, and real-time modeling of illumination gradient, texture ratio and obstacle offset is performed to optimize servo signal adjustment. Path correction is faster and adapts to complex dynamic scenes. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0018] Figure 1 is a system flow chart of the present invention; Figure 2 It is a submodule flow chart of the present invention; Figure 3 It is a flow chart of the path vision extraction module of the present invention; Figure 4 It is a flow chart of the path characteristic calculation module of the present invention; Figure 5 It is a flow chart of the dynamic area identification module of the present invention; Figure 6 It is a flow chart of the disturbance parameter modeling module of the present invention; Figure 7 is a flow chart of a servo path correction module of the present invention; Figure 8 This is a flow chart of the path priority adjustment module of the present invention. DETAILED DESCRIPTION

[0019] The technical solution of the present invention is described below in conjunction with the accompanying drawings.

[0020] In the embodiments of the present invention, words such as "exemplarily" and "for example" are used to indicate examples, illustrations or explanations. Any embodiment or design described as "example" in the present invention should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of the word "example" is intended to present the concept in a specific way. In addition, in the embodiments of the present invention, the meaning expressed by "and / or" can be both, or it can be either of the two.

[0021] In the embodiments of the present invention, "image" and "picture" can sometimes be used interchangeably. It should be noted that when the difference between them is not emphasized, the meanings they intend to express are the same. "of", "corresponding, relevant" and "corresponding" can sometimes be used interchangeably. It should be noted that when the difference between them is not emphasized, the meanings they intend to express are the same.

[0022] In the embodiments of the present invention, sometimes a subscript such as W1 may be written as a non-subscript such as W1. When the difference is not emphasized, the meanings to be expressed are the same.

[0023] In order to make the technical problems, technical solutions and advantages to be solved by the present invention more clear, a detailed description will be given below with reference to the accompanying drawings and specific embodiments.

[0024] See also Figure 1 and Figure 2 , a servo motor conveying system with a visual AI algorithm to correct the driving path includes: The path vision extraction module collects the path continuous image data through the visual sensor, segments the edge area of ​​the path image, identifies the gradient distribution value and boundary change characteristics of the image, calculates the range of path edge curvature change, and obtains the path boundary feature matrix; The path characteristic calculation module analyzes the boundary curvature change characteristics of the path segment based on the path boundary feature matrix, calls the gradient change intensity value and texture continuity index, calculates the path edge change trend and continuity characteristics, and constructs the path characteristic trend data set; The dynamic region identification module formulates the curvature change trend threshold based on the path characteristic trend data set, identifies the key positions in the path curvature change trend, divides the range of texture continuity index, calibrates the parameter distribution of dynamically stable and dynamically unstable regions, and generates a dynamic region characteristic distribution map; The disturbance parameter modeling module extracts the illumination gradient range of the dynamic unstable area based on the dynamic area characteristic distribution map, analyzes the distribution range of the texture change ratio, and calculates the illumination correction value, texture compensation value and obstacle offset compensation value in combination with the dynamic obstacle offset trend data to generate a disturbance correction parameter set. The servo path correction module calculates the power output adjustment value of the servo motor based on the disturbance correction parameter set, combined with the dynamic area illumination correction value and the curvature change trend range, analyzes the steering angle range and acceleration frequency range, optimizes the servo signal output matching value, and generates a servo path correction signal group; The path priority adjustment module calls the curvature change trend range of the dynamic area based on the servo path correction signal group, combines the interpolation calculation results of the key boundary points in the stable area, adjusts the priority distribution of the dynamically unstable area, optimizes the path morphology of the curvature change area, and generates the path correction optimization result.

[0025] The path boundary feature matrix includes edge curvature distribution, gradient change intensity, and texture gradient range. The path characteristic trend data set includes edge change trend, curvature change range, and continuity characteristic distribution. The dynamic area characteristic distribution map includes stable area parameters, dynamic area parameters, and texture discontinuity area division. The disturbance correction parameter set includes illumination compensation value, texture change compensation value, and obstacle offset correction value. The servo path correction signal group includes power adjustment signal, steering angle adjustment signal, and acceleration frequency adjustment signal. The path correction optimization results include path morphology optimization value, area priority distribution value, and curvature adjustment optimization value.

[0026] See also Figure 3 and Figure 2 ,The path visual extraction module includes an image acquisition submodule, an edge feature recognition submodule, and a curvature feature calculation submodule; The image acquisition submodule is based on a visual sensor to collect continuous image data of the path, uses a path segmented sampling mode to intercept images, extracts continuous frame images and stores them as structured files, checks and cleans up acquisition errors and duplicate images, and generates a path image data set; By precisely controlling the stability of the sampling interval and the trigger timing, the sampling controller is used to sample and intercept the image frames in segments, and the acquired image frame data is stored in a structured file format. The file structure uses a segmented timestamp associated with a path label. The algorithm is used to detect and clean up acquisition error data and duplicate images, calculate the image frame similarity and eliminate redundant frames to ensure the validity and consistency of the continuous image sequence, and complete the generation of the path image data set.

[0027] The edge feature recognition submodule performs image edge area segmentation based on the path image data set, extracts the gradient distribution value of the image edge, analyzes the boundary position characteristics according to the distribution value change rate, generates the boundary gradient distribution matrix, analyzes the pixel point change trend, and generates the path edge gradient feature matrix; Perform image edge area segmentation and boundary position characteristic analysis, and calculate the boundary position characteristics through the change rate of edge gradient value distribution, according to the formula: ; Calculate the edge gradient distribution value, analyze the change trend of boundary pixels and generate the path edge gradient characteristic matrix.

[0028] In the formula, Represents the gradient amplitude of the pixel, Represents the gray value of the pixel. and Represent the partial derivatives of the image grayscale value in the x direction and the y direction respectively.

[0029] When collecting grayscale images, the pixels The grayscale value of is differentially calculated to calculate the gradient values ​​in the x direction and the y direction respectively. The gradient calculation formula is: ; ; Let’s take an example: Assume that the gray value of a pixel is , and their neighborhood values ​​are , , , .

[0030] Calculate the x-direction gradient: ; Calculate the y-direction gradient: ; Compute the gradient magnitude: ; The result shows that the gradient amplitude of the pixel is 22.36, indicating that it is an edge pixel. Its gradient characteristics can be used to construct the path edge gradient characteristic matrix.

[0031] The curvature feature calculation submodule is based on the path edge gradient characteristic matrix, segmentally fits the boundary curvature change range, extracts the curvature change rate, calculates the change gradient of the edge segment, screens the nodes with sudden curvature changes, and obtains the path boundary feature matrix; The specific calculation formula for the change gradient value is: ; in, Representative Segment Path The weighted curvature change gradient value of the point, Represents the path edge Section The gradient value of a pixel, Representative Section The weight coefficient of the point is calculated based on the gradient change range of the path area and the dynamic weight matrix. is the square root operation, is the segment number index of the path, The pixel index of the current segment.

[0032] Representative Segment Path The weighted curvature change gradient value of the point.

[0033] and Respectively represent The first and The gradient value of a pixel, the unit is the change of gray value.

[0034] For the Segment Path The weight coefficient of each pixel is set according to the gradient change range of the dynamic area, ranging from 0 to 1.

[0035] is the square root operator, which ensures the gradient is normalized.

[0036] Parameter calculation and data source: Gradient Value and : The grayscale value of the pixel point of the collected image is calculated by the grayscale value change of the path edge , the calculation formula is: ; The grayscale value of the image is collected through sensor data, and the unit is 0 to 255. The partial derivative of the grayscale gradient is calculated using the difference method, for example: ; This formula ensures that the gradient value truly reflects the changing characteristics of the pixel points.

[0037] Weight coefficient : Calculated by normalizing the gradient variation range of the path area. The weight value is set according to the gradient fluctuation amplitude, using the linear normalization method: ; in, and Respectively The maximum and minimum values ​​of the gradient in the segment path. The weight range is 0 to 1, the maximum gradient corresponds to weight 1, and the minimum gradient corresponds to weight 0.

[0038] Example calculation: Assume that the grayscale value of a path segment is as follows, with units ranging from 0 to 255: - , , , . - Calculate the gradient value: ; ; Calculate the weight coefficient: Assume that the maximum gradient in this path , minimum , the weight coefficient is: ; Calculate the weighted gradient value of the curvature change: ; ; Results and significance: Calculation results Indicates the current path The value of the point weighted curvature change gradient. This value is used to filter the curvature mutation nodes and provide data support for the path boundary feature matrix. By calculating point by point, the feature matrix of the complete path is obtained.

[0039] The results show that the weighted gradient value of curvature change can effectively describe the changing trend of path edge and provide data support for path optimization and dynamic adjustment.

[0040] See also Figure 4 and Figure 2 ,The path characteristic calculation module includes a curvature analysis submodule, a gradient and texture analysis submodule, and a trend data construction submodule; The curvature analysis submodule extracts the curvature change value of the path segment boundary based on the path boundary feature matrix, calculates the change rate of the boundary curvature, and classifies and marks it according to the rate fluctuation. It calls the curvature change distribution data and performs point-by-point comparative analysis on the spatial position of the change point to generate the curvature change characteristic matrix. By calculating the curvature change rate point by point, the curvature change distribution data is called by using the spatial position relationship. The change rate is calculated point by point as the ratio of the curvature change difference between the current node and its adjacent nodes. The spatial position change rate is marked as a feature label and classified. The result matrix is ​​analyzed and sorted in a point-by-point comparison manner to finally generate a curvature change characteristic matrix.

[0041] The gradient and texture analysis submodule extracts the gradient change intensity value based on the curvature change characteristic matrix, analyzes the fluctuation range of local changes, calculates the texture continuity index of the path segment, and analyzes the texture coherence characteristics based on the image pixel distribution as a reference, calls the distribution data to screen the texture abnormal area, and marks the characteristic area to generate the gradient and texture characteristic matrix; Extract the gradient change intensity value and analyze the fluctuation range of local changes, according to the formula ; Computes the path segment texture continuity index.

[0042] In the formula, represents the texture continuity index, Represents the gradient value of the i-th pixel, The number of pixels that represent the path segment.

[0043] When extracting the pixel gradient value of the path segment, the absolute value of the difference between the gradient values ​​of every two adjacent pixels is calculated and accumulated, and finally the accumulated value is divided by the number of pixels minus 1 to obtain the texture continuity index.

[0044] For example, in a path segment with a gradient value of , , , , the total number of pixels is 4.

[0045] Calculate the absolute value of the difference between adjacent gradient values: ; ; ; The total value of the gradient change is accumulated: ; Calculate the texture continuity index: ; The result shows that the texture continuity index of the path segment is 4. The lower the index value, the more continuous the texture, and the higher the index value, the greater the texture fluctuation.

[0046] The trend data construction submodule calculates the path edge change trend based on the gradient and texture characteristic matrix, and performs matching analysis of characteristic parameters in combination with texture characteristics, analyzes the characteristic values ​​of trend changes, and classifies and organizes key parameters to generate a path characteristic trend data set; The gradient values ​​and texture characteristic data in the characteristic matrix are called to calculate the intensity and direction of the path edge change trend. The trend characteristic matching analysis is performed by comparing the changes between adjacent characteristic parameters and combining the spatial position difference. The changing characteristic values ​​are classified and sorted according to the fluctuation intensity and direction, and a path characteristic trend data set containing trend characteristics and parameter classification is generated.

[0047] See also Figure 5 and Figure 2 ,The dynamic region recognition module includes a curvature trend threshold setting submodule, a key location recognition submodule, and a dynamic region division submodule; The curvature trend threshold setting submodule calculates the curvature change range of each segment based on the path characteristic trend data set, extracts fluctuation data to set the threshold, filters the curvature segments whose fluctuation frequency exceeds the threshold, extracts the difference parameters between the change amplitude and the stable value, and generates the curvature change trend threshold; The curvature variation range of each path segment is extracted, and the frequency and amplitude of the curvature variation are analyzed based on the extracted fluctuation data. The curvature segments exceeding the threshold are screened out by setting the curvature fluctuation threshold. The difference parameters between the curvature variation amplitude and the stable value are calculated, and each curvature variation segment is marked and sorted. Finally, the curvature change trend threshold is generated for subsequent analysis.

[0048] The key position identification submodule extracts the sudden change position of the curvature change of the path segment based on the curvature change trend threshold, compares the change trend value and fluctuation characteristics, and calibrates the dynamic distribution nodes to generate dynamic key node distribution characteristics; Extract the sudden change position of the curvature change of the path segment and analyze the dynamic distribution nodes. According to the formula ; Calculate the mutation rate of curvature change.

[0049] In the formula, represents the rate of change of curvature, represents the curvature value of the i-th node, Represents the distance between adjacent nodes.

[0050] The curvature value is extracted through the characteristic trend data in the path data, and the curvature change rate is obtained by dividing the absolute difference of the curvature change values ​​of adjacent nodes by the distance between the nodes.

[0051] For example, the curvature values ​​of two adjacent nodes in the path are and , the distance between nodes .

[0052] Calculate the rate of change of curvature: ; The result shows that the curvature change rate is 0.15, indicating that the curvature changes slowly between the two nodes. The curvature change rate result is used to calibrate the dynamic distribution nodes and further generate the dynamic key node distribution characteristics.

[0053] The dynamic region division submodule classifies and marks the range change of texture continuity index based on the distribution characteristics of dynamic key nodes, sets the regional stability threshold, extracts the continuity feature distribution of stable and unstable regions, combines the dynamic change trend, calculates the boundary range index, and generates a dynamic region characteristic distribution map; The range change data of texture continuity index is extracted, and each texture feature is classified and marked in combination with the regional stability threshold. The continuity feature distribution of stable and unstable regions is statistically analyzed, and the regional boundary range index is calculated using the dynamic change trend. By gradually calibrating and adjusting the boundary points of regional divisions, a characteristic distribution map of the dynamic region is formed to clarify the characteristic information of regional changes.

[0054] See also Figure 6 and Figure 2 ,The disturbance parameter modeling module includes the illumination gradient extraction submodule, the texture change analysis submodule, and the dynamic compensation calculation submodule; The illumination gradient extraction submodule extracts the distribution data of illumination gradient in the dynamically unstable area based on the dynamic area characteristic distribution map, identifies the grayscale value of image pixels, calculates the amplitude of illumination gradient change in each area, compares the time series characteristics of gradient change, identifies abnormal fluctuation areas, constructs the gradient change range matrix and extracts the distribution characteristic values, and generates illumination gradient range parameters; The distribution of pixel grayscale values ​​is analyzed in blocks, the difference between the maximum and minimum values ​​of the pixel grayscale gradient change in each block is calculated, and the fluctuation of the illumination gradient change is calculated for each block in the time series. The area range of abnormal fluctuations is screened out, and the distribution eigenvalues ​​are extracted using the gradient distribution matrix of each block. The calculation results of the illumination gradient range are combined with the distribution eigenvalues ​​to generate the illumination gradient range parameters.

[0055] The texture change analysis submodule extracts the texture change feature values ​​in the dynamically unstable area based on the illumination gradient range parameters, analyzes the image texture direction distribution and grayscale difference, calculates the texture change ratio, analyzes the texture fluctuation amplitude in the change area, calibrates the key fluctuation range, screens the texture distribution characteristic values ​​associated with the illumination gradient, and generates the texture change ratio distribution; The formula for calculating the texture change joint ratio is as follows: ; in, Representative Section The joint ratio of texture changes in the region, Representative Section The texture direction gradient value of the point, Representative Region No. The gray value of the point, is the number of pixels in the area, is the index of the pixel, Represents the summation of all pixels in the area.

[0056] Representative Section The ratio of texture variation in a region.

[0057] Representative Section The texture directional gradient value of the point is obtained by calculating the directional change of the texture.

[0058] Representative Region No. The gray value of a point is obtained through the collected image pixel data, and the unit is gray value.

[0059] is the total number of pixels in the current area.

[0060] is the index of the pixel, Indicates the summation of all pixels.

[0061] Parameter calculation and acquisition method: Texture direction gradient value : It is calculated through the directional distribution of the image gradient, and the gradient direction is calculated using the grayscale changes of adjacent pixels: ; The grayscale value data of the image is collected by the image sensor, and the gradient direction is calculated using the pixel neighbor difference method. For example: ; Gray value : Directly collected through image data, the unit is grayscale value from 0 to 255. The collection of grayscale value is based on the photosensitivity of the sensor and takes the regional average value as a reference.

[0062] Example calculation: Assume Section The grayscale value and gradient data of the pixel points in the area are as follows: -Grayscale value: - Texture direction gradient value: -Total number of pixels:

[0063] Calculate the absolute value of the gradient change of adjacent pixels: ; ; ; Calculate the absolute value of the grayscale change of adjacent pixels: ; ; ; Calculate the numerator part: ; ; Calculate the denominator: ; ; ; Calculate the joint ratio: ; Result description: Calculation results Indicates the ratio of texture change to grayscale change in the current dynamic area. This result is used to calibrate the texture fluctuation amplitude in the dynamically unstable area and screen the texture characteristics related to the illumination gradient, providing basic data for subsequent analysis.

[0064] The dynamic compensation calculation submodule is based on the texture change ratio distribution and combines the dynamic obstacle offset trend data. By comparing the interaction between regional illumination changes, texture characteristic fluctuations and offset data, it calculates the illumination correction value, texture compensation value and obstacle offset compensation value, integrates multi-dimensional disturbance parameters, and constructs a distribution matrix to generate a disturbance correction parameter set. Extract the dynamic obstacle offset trend data, make a joint comparison of the illumination gradient change, texture fluctuation characteristics and offset data of each section, calculate the illumination correction value through time series analysis, calculate the texture compensation value by using the fluctuation amplitude difference between illumination change and texture characteristics, and calculate the offset compensation value by combining the obstacle offset trend difference parameters. Combined with the characteristics of the elevated suspension conveying system of the EMS aerial conveyor line, in a modularly designed three-dimensional conveying environment, the independent operation characteristics of the carrier vehicle are used to achieve accurate compensation of multi-dimensional disturbance parameters in the dynamic area. By utilizing the stability of the track system suspension design, the illumination gradient change, texture fluctuation characteristics and offset trend can be efficiently adapted and adjusted in the conveying system in different directions, ensuring the continuity and dynamic adaptability of the compensation calculation. Finally, all multi-dimensional disturbance parameters are integrated into the distribution matrix, combined with the flexible layout and continuous cycle operation characteristics of the EMS system, the disturbance correction performance of the dynamic area is further optimized, and a disturbance correction parameter set is formed for dynamic environmental characteristic compensation calculation.

[0065] See also Figure 7 and Figure 2 ,The servo path correction module includes a power adjustment calculation submodule, a steering and acceleration range analysis submodule, and a servo signal matching optimization submodule; The power adjustment calculation submodule extracts the dynamic area illumination correction value and the dynamic obstacle offset compensation value based on the disturbance correction parameter set, calculates the adjustment amplitude of the servo motor power output value by combining the gradient fluctuation data in the curvature change trend range, analyzes the matching relationship between the output value and the dynamic parameters, and corrects the power parameter matrix to generate the servo power adjustment value; Calculate the adjustment range of the servo motor power output value according to the formula ; Calculate the servo motor power adjustment value.

[0066] In the formula, Represents the servo motor power adjustment value, Represents the dynamic area lighting correction value, Represents the dynamic obstacle offset compensation value, Represents the gradient fluctuation amplitude within the curvature change trend range.

[0067] The servo motor power adjustment value is calculated by integrating the illumination correction value and the obstacle offset compensation value, and combining the curvature gradient fluctuation amplitude. The illumination correction value is obtained by analyzing the dynamic area illumination parameters, the obstacle offset compensation value comes from the dynamic distribution characteristic data, and the curvature gradient fluctuation amplitude is the calculation result of the gradient change of the path curvature range.

[0068] For example, the dynamic area light correction value is , the dynamic obstacle offset compensation value is , the curvature gradient fluctuation amplitude is .

[0069] Calculate the servo motor power adjustment value: ; The result shows that the servo motor power adjustment value is 57.5, which is used to correct the dynamic power output of the servo system and ensure that the power adjustment adapts to the parameter changes in the dynamic environment.

[0070] The steering and acceleration range analysis submodule analyzes the steering angle variation range corresponding to the servo signal based on the servo power adjustment value, combined with the path curvature variation trend range and dynamic area characteristic distribution, calculates the peak value in the acceleration frequency range, analyzes the fluctuation amplitude, compares the dynamic area signal characteristics, adjusts the steering and acceleration limit parameters, and generates the servo steering and acceleration range parameters; By parsing the input parameters of the servo signal, the steering angle variation range under the action of the servo signal is extracted, the peak value in the acceleration frequency range is analyzed and the acceleration variation amplitude is calculated in combination with the time series characteristics. By comparing the dynamic area distribution characteristics with the action range of the servo signal, the input limit value of the servo steering signal is adjusted and the acceleration range parameters are updated. Finally, the servo steering and acceleration range parameters are generated for dynamic path characteristic correction.

[0071] The servo signal matching optimization submodule calls the servo power adjustment value data matrix based on the servo steering and acceleration range parameters, performs characteristic analysis on the frequency, amplitude and output power of the servo signal, adjusts the output mode of the servo signal according to the dynamic characteristic range, builds a signal adjustment matrix based on the path dynamic distribution, and generates a servo path correction signal group; The servo power adjustment value data matrix is ​​parsed, and the frequency and amplitude of the servo signal are optimized and adjusted through the data matching process of the dynamic area characteristic distribution. The output mode of the servo signal is calculated according to the dynamic path range, and the signal adjustment matrix is ​​constructed by combining the distribution of signal frequency and the disturbance characteristics in the dynamic area. The output value of the servo signal is associated with the dynamic path characteristic distribution to generate a servo path correction signal group.

[0072] See also Figure 8 and Figure 2 ,The path priority adjustment module includes a priority distribution adjustment submodule, a dynamic area morphology optimization submodule, and a path correction result generation submodule; The priority distribution adjustment submodule, based on the servo path correction signal group, calls the curvature change trend range data of the dynamic area, extracts the curvature distribution in the dynamic area section by section, calculates the interpolation of the boundary points in the stable area, adjusts and calibrates the path priority parameter correction value in the dynamic unstable area, and generates the priority distribution adjustment parameter; The curvature change trend range of the dynamic area is extracted, the curvature distribution data in the dynamic area is analyzed section by section, the boundary points in the stable area are interpolated using the piecewise interpolation method, and the path priority parameters of the unstable part in the dynamic area are corrected in combination with the curvature distribution law. The priority of each path segment is adjusted and calibrated through the correction value, and the priority distribution adjustment parameters are generated for further analysis of the path dynamic area optimization.

[0073] The dynamic region morphology optimization submodule adjusts parameters based on priority distribution, extracts the distribution characteristics of the boundary points in the curvature change region, combines the curvature change range of the dynamic region and the stable region, gradually optimizes the dynamic region morphology, analyzes the offset value of the dynamic boundary point interpolation and the curvature change law, adjusts the distribution characteristics of the curvature boundary and optimizes the regional morphology, and generates the optimized path morphology characteristics; The dynamic area morphology is gradually optimized according to the formula ; Calculate the rate of change of the curvature boundary distribution characteristics of the dynamic region.

[0074] In the formula, represents the rate of change of the curvature distribution characteristic, Representative The curvature value of the boundary point, Represents the total number of boundary points.

[0075] The curvature boundary distribution characteristic change rate is calculated by calculating the ratio of the absolute value of the difference in curvature values ​​of adjacent boundary points to the total curvature. The curvature data of the boundary points in the dynamic area is extracted, and the characteristic changes of the boundary points are analyzed by the curvature change law.

[0076] For example, the curvature value of a region boundary point is , , , , the total number of boundary points is 4.

[0077] Calculate the absolute value of the curvature difference between adjacent points: ; ; ; The curvature difference is accumulated: ; Compute the sum of the curvatures: ; Calculate the rate of change of the curvature distribution characteristic: ; The results show that the change rate of the curvature distribution characteristics of the dynamic area is 0.259, which can be used to adjust the curvature boundary characteristics of the dynamic area morphology and optimize the regional distribution characteristics.

[0078] The path correction result generation submodule is based on the optimized path morphological characteristics, combined with the priority distribution data of the dynamic area and the curvature change boundary distribution, calculates the morphological parameters of the path after adjustment, analyzes the matching value of the servo signal to the path morphology, integrates and verifies the regional correction effect, path morphology and servo output signal, and generates the path correction optimization result; The priority distribution data of the dynamic area is called to analyze the distribution characteristics of the curvature change boundary. By integrating the action mode of the servo signal and combining the path morphology parameter adjustment data, the dynamic matching value of the path morphology and the servo signal is analyzed. Combined with the system characteristics of the EMS aerial conveyor line, through its elevated suspension conveying design, the servo signal and path correction can be more efficiently dynamically adapted in the aerial track operation environment. The path morphology parameters are flexibly adjusted by using independently operated cargo carriers, and the curvature change area is accurately matched and corrected and optimized in the complex three-dimensional conveying system in the horizontal and vertical directions. The corrected path morphology parameters are verified with the servo signal characteristic distribution, and combined with the stability of the aerial track suspension and the characteristics of continuous cycle operation, the optimization of the servo path signal and the efficiency of the dynamic area path correction are further ensured. Finally, the path correction optimization results are generated for the application and analysis of the servo path signal in the dynamic area.

[0079] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A servo motor conveying system with a visual AI algorithm to correct the driving path, characterized by: The system comprises: The path vision extraction module collects the path continuous image data through the visual sensor, segments the edge area of ​​the path image, identifies the gradient distribution value and boundary change characteristics of the image, calculates the range of path edge curvature change, and obtains the path boundary feature matrix; The path characteristic calculation module analyzes the boundary curvature change characteristics of the path segments based on the path boundary feature matrix, calls the gradient change intensity value and the texture continuity index, calculates the path edge change trend and continuity characteristics, and constructs a path characteristic trend data set; The dynamic region identification module formulates a curvature change trend threshold based on the path characteristic trend data set, identifies key positions in the path curvature change trend, divides the range of the texture continuity index, calibrates the parameter distribution of the dynamically stable and dynamically unstable regions, and generates a dynamic region characteristic distribution map; The disturbance parameter modeling module extracts the illumination gradient range of the dynamic unstable area based on the dynamic area characteristic distribution map, analyzes the distribution range of the texture change ratio, calculates the illumination correction value, texture compensation value and obstacle offset compensation value in combination with the dynamic obstacle offset trend data, and generates a disturbance correction parameter set; The servo path correction module calculates the power output adjustment value of the servo motor based on the disturbance correction parameter set, combines the dynamic area illumination correction value and the curvature change trend range, analyzes the steering angle range and the acceleration frequency range, optimizes the servo signal output matching value, and generates a servo path correction signal group; The path priority adjustment module calls the curvature change trend range of the dynamic area based on the servo path correction signal group, combines the interpolation calculation results of the key boundary points in the stable area, adjusts the priority distribution of the dynamically unstable area, optimizes the path morphology of the curvature change area, and generates a path correction optimization result.

2. The servo motor conveying system with visual AI algorithm to correct the driving path according to claim 1 is characterized by: The path boundary feature matrix includes edge curvature distribution, gradient change intensity, and texture gradient range; the path characteristic trend data set includes edge change trend, curvature change range, and continuity characteristic distribution; the dynamic area characteristic distribution map includes stable area parameters, dynamic area parameters, and texture discontinuity area division; the disturbance correction parameter set includes illumination compensation value, texture change compensation value, and obstacle offset correction value; the servo path correction signal group includes power adjustment signal, steering angle adjustment signal, and acceleration frequency adjustment signal; the path correction optimization result includes path morphology optimization value, area priority distribution value, and curvature adjustment optimization value.

3. The servo motor conveying system with visual AI algorithm to correct the driving path according to claim 1 is characterized by: The path vision extraction module includes an image acquisition submodule, an edge feature recognition submodule, and a curvature feature calculation submodule; The image acquisition submodule is based on a visual sensor to collect continuous image data of the path, uses a path segmented sampling mode to intercept images, extracts continuous frame images and stores them as structured files, checks and cleans up acquisition errors and duplicate images, and generates a path image data set; The edge characteristic recognition submodule performs image edge area segmentation based on the path image data set, extracts the gradient distribution value of the image edge, analyzes the boundary position characteristics according to the distribution value change rate, generates the boundary gradient distribution matrix, analyzes the pixel point change trend, and generates the path edge gradient characteristic matrix; The curvature feature calculation submodule fits the boundary curvature variation range in segments based on the path edge gradient characteristic matrix, extracts the curvature variation rate, calculates the variation gradient of the edge segment, screens the nodes with sudden curvature changes, and obtains the path boundary feature matrix.

4. The servo motor conveying system with visual AI algorithm to correct the driving path according to claim 3 is characterized by: The calculation formula of the change gradient value is specifically: ； in, Representative Segment Path The weighted curvature change gradient value of the point, Represents the path edge Section The gradient value of a pixel, Representative Section The weight coefficient of the point is calculated based on the gradient change range of the path area and the dynamic weight matrix. is the square root operation, is the segment number index of the path, The pixel index of the current segment.

5. The servo motor conveying system with visual AI algorithm to correct the driving path according to claim 1 is characterized by: The path characteristic calculation module includes a curvature analysis submodule, a gradient and texture analysis submodule, and a trend data construction submodule; The curvature analysis submodule extracts the curvature change value of the path segment boundary based on the path boundary feature matrix, calculates the change rate of the boundary curvature, and classifies and marks it according to the rate fluctuation, calls the curvature change distribution data and performs point-by-point comparative analysis on the spatial position of the change point to generate a curvature change characteristic matrix; The gradient and texture analysis submodule extracts the gradient change intensity value based on the curvature change characteristic matrix, analyzes the fluctuation range of local changes, calculates the path segment texture continuity index, and analyzes the texture coherence feature based on the image pixel distribution as a reference, calls the distribution data to screen the texture abnormal area, and marks the characteristic area to generate the gradient and texture characteristic matrix; The trend data construction submodule calculates the path edge change trend based on the gradient and texture characteristic matrix, and performs matching analysis of characteristic parameters in combination with texture characteristics, analyzes the characteristic values ​​of trend changes, and classifies and organizes key parameters to generate a path characteristic trend data set.

6. The servo motor conveying system with visual AI algorithm to correct the driving path according to claim 1 is characterized by: The dynamic area identification module includes a curvature trend threshold setting submodule, a key position identification submodule, and a dynamic area division submodule; The curvature trend threshold setting submodule calculates the curvature change range of each segment based on the path characteristic trend data set, extracts fluctuation data to set a threshold, screens the curvature segments whose fluctuation frequency exceeds the threshold, extracts the difference parameter between the change amplitude and the stable value, and generates the curvature change trend threshold; The key position identification submodule extracts the sudden change position of the curvature change of the path segment based on the curvature change trend threshold, compares the change trend value and the fluctuation characteristics, and calibrates the dynamic distribution nodes to generate dynamic key node distribution characteristics; The dynamic area division submodule classifies and marks the range changes of the texture continuity index based on the distribution characteristics of the dynamic key nodes, sets the regional stability threshold, extracts the continuity feature distribution of the stable area and the unstable area, combines the dynamic change trend, calculates the boundary range index, and generates a dynamic area characteristic distribution map.

7. The servo motor conveying system with visual AI algorithm to correct the driving path according to claim 1 is characterized by: The disturbance parameter modeling module includes an illumination gradient extraction submodule, a texture change analysis submodule, and a dynamic compensation calculation submodule; The illumination gradient extraction submodule extracts the distribution data of the illumination gradient in the dynamic unstable area based on the dynamic area characteristic distribution map, identifies the grayscale value of the image pixel, calculates the illumination gradient change amplitude of each area, compares the time series characteristics of the gradient change, identifies the abnormal fluctuation area, constructs the gradient change range matrix and extracts the distribution characteristic value, and generates the illumination gradient range parameter; The texture change analysis submodule extracts the texture change characteristic value in the dynamically unstable area based on the illumination gradient range parameter, analyzes the image texture direction distribution and grayscale difference, calculates the texture change ratio, analyzes the texture fluctuation amplitude in the change area, calibrates the key fluctuation range, screens the texture distribution characteristic value associated with the illumination gradient, and generates the texture change ratio distribution; The dynamic compensation calculation submodule is based on the texture change ratio distribution and combined with the dynamic obstacle offset trend data. By comparing the interaction between regional illumination changes, texture characteristic fluctuations and offset data, it calculates the illumination correction value, texture compensation value and obstacle offset compensation value, integrates multi-dimensional disturbance parameters, and constructs a distribution matrix to generate a set of disturbance correction parameters.

8. The servo motor conveying system with visual AI algorithm to correct the driving path according to claim 7 is characterized by: The texture change joint ratio calculation formula is specifically: ； in, Representative Section The joint ratio of texture changes in the region, Representative Section The texture direction gradient value of the point, Representative Region No. The gray value of the point, is the number of pixels in the area, is the index of the pixel, Represents the summation of all pixels in the area.

9. The servo motor conveying system with visual AI algorithm to correct the driving path according to claim 1 is characterized by: The servo path correction module includes a power adjustment calculation submodule, a steering and acceleration range analysis submodule, and a servo signal matching optimization submodule; The power adjustment calculation submodule extracts the dynamic area illumination correction value and the dynamic obstacle offset compensation value based on the disturbance correction parameter set, calculates the adjustment amplitude of the servo motor power output value in combination with the gradient fluctuation data in the curvature change trend range, analyzes the matching relationship between the output value and the dynamic parameter, and corrects the power parameter matrix to generate the servo power adjustment value; The steering and acceleration range analysis submodule analyzes the steering angle variation range corresponding to the servo signal based on the servo power adjustment value, combined with the path curvature variation trend range and the dynamic area characteristic distribution, calculates the peak value in the acceleration frequency range, analyzes the fluctuation amplitude, compares the dynamic area signal characteristics, adjusts the steering and acceleration limit parameters, and generates the servo steering and acceleration range parameters; The servo signal matching optimization submodule calls the servo power adjustment value data matrix based on the servo steering and acceleration range parameters, performs characteristic analysis on the frequency, amplitude and output power of the servo signal, adjusts the output mode of the servo signal according to the dynamic characteristic range, constructs a signal adjustment matrix in combination with the path dynamic distribution, and generates a servo path correction signal group.

10. The servo motor conveying system with visual AI algorithm to correct the driving path according to claim 1, characterized in that: The path priority adjustment module includes a priority distribution adjustment submodule, a dynamic area morphology optimization submodule, and a path correction result generation submodule; The priority distribution adjustment submodule calls the curvature change trend range data of the dynamic area based on the servo path correction signal group, extracts the curvature distribution in the dynamic area section by section, calculates the interpolation of the boundary points in the stable area, adjusts and calibrates the path priority parameter correction value of the dynamic unstable area, and generates the priority distribution adjustment parameter; The dynamic region morphology optimization submodule adjusts the parameters based on the priority distribution, extracts the distribution characteristics of the boundary points in the curvature change region, combines the curvature change ranges of the dynamic region and the stable region, gradually optimizes the dynamic region morphology, analyzes the offset value of the dynamic boundary point interpolation and the curvature change law, adjusts the distribution characteristics of the curvature boundary and optimizes the region morphology, and generates optimized path morphology characteristics; The path correction result generation submodule is based on the optimized path morphological characteristics, combined with the priority distribution data of the dynamic area and the curvature change boundary distribution, calculates the morphological parameters of the path after adjustment, analyzes the matching value of the servo signal to the path morphology, integrates and verifies the regional correction effect, path morphology and servo output signal, and generates the path correction optimization result.

Images