Vamp roughing method and system, medium and program product

Abstract

The invention discloses a vamp roughing method and system, a medium and a program product, and relates to the field of shoemaking industrial.The method comprises the steps that a three-dimensional mesh model is generated according to collected point cloud data of a target vamp; performing feature extraction on the three-dimensional grid model, and determining a to-be-polished area of the target vamp; according to the to-be-polished area, a first motion path of the robot arm is generated; determining a roughening rate of each path segment in the first motion path, and determining a to-be-optimized path according to the roughening rate; determining a shoe tree movement path of the shoe tree according to the to-be-optimized path, and correcting the first movement path of the robot arm to obtain a second movement path; and controlling the manipulator to move along the second movement path, and controlling the shoe tree to move along the shoe tree movement path so as to thicken the vamp. By implementing the method, the motion path of the robot arm can be optimized, and the working efficiency is improved.

Description

Technical Field

[0001] The present application relates to the field of shoemaking industry, and in particular to a shoe upper roughening method, system, medium and program product. Background Art

[0002] In the shoemaking industry, roughening the upper is a key process step, which has an important impact on subsequent processes such as bonding. As the shoemaking industry's requirements for product quality and production efficiency continue to increase, traditional manual roughening methods can no longer meet the needs of the modern shoemaking industry.

[0003] In the related art, in order to improve production efficiency, an automated upper roughening method based on 3D vision technology is used. This method uses a 3D vision sensor to scan the upper with high precision to obtain a three-dimensional model of the upper; then identifies the feature points and areas on the three-dimensional model that need to be roughened, such as seams and edges; then plans the movement path of the robot arm based on this information, allows the robot arm to move according to the movement path, and controls the roughening device to polish the upper.

[0004] However, when faced with different roughening requirements, different sizes and styles of shoes, related technologies often require frequent adjustments to the movement path of the robot arm, resulting in low work efficiency. Summary of the invention

[0005] The present application provides a shoe upper roughening method, system, medium and program product for controlling the movement of a shoe last, optimizing the movement path of a robot arm and improving work efficiency.

[0006] In a first aspect, the present application provides a method for roughening a shoe upper, which is applied to a shoe upper roughening system, and the method includes: generating a three-dimensional mesh model based on the collected point cloud data of the target shoe upper; performing feature extraction on the three-dimensional mesh model to determine the area to be polished of the target shoe upper; generating a first motion path of a robot arm based on the area to be polished; determining the roughening rate of each path segment in the first motion path, and determining a path to be optimized based on the roughening rate; determining a shoe last motion path of a shoe last based on the path to be optimized, and correcting the first motion path of the robot arm to obtain a second motion path; controlling the robot arm to move along the second motion path, and at the same time controlling the shoe last to move along the shoe last motion path, so as to roughen the shoe upper.

[0007] In the above embodiment, the shoe upper roughening system generates a three-dimensional mesh model by collecting point cloud data of the target shoe upper, determines the area to be polished and generates the motion path of the robot, then determines the corresponding path to be optimized, that is, the motion path that can be optimized by running the shoe last, and corrects the robot path to obtain a second motion path, and controls the robot and the shoe last to move simultaneously to roughen the shoe upper. By introducing the movement of the shoe last to simplify the movement process of the robot, the efficiency of shoe upper roughening is effectively improved.

[0008] In combination with some embodiments of the first aspect, in some embodiments, the roughing rate of each path segment in the first motion path is determined, and the steps of determining the path to be optimized based on the roughing rate specifically include: determining the initial roughing rate of each path segment in the first motion path in which only the robot arm moves; based on the initial roughing rate, determining the rate optimization interval of each path segment in the first motion path; when the lower limit value of the rate optimization interval exceeds a preset optimization threshold, taking the corresponding path segment as the segment to be optimized; calculating and determining all the segments to be optimized to obtain the path to be optimized.

[0009] In the above embodiment, the shoe upper roughening system determines the initial roughening rate of the robot hand and determines the rate optimization interval of each segment, that is, the space that can be improved. When the lower limit of the optimization interval exceeds the preset threshold, that is, when the room for improvement reaches the specified value, the corresponding segment is determined as the segment to be optimized, and all the segments to be optimized are combined into a path to be optimized. The path segments with optimization space in the robot hand movement path can be accurately and efficiently calculated to improve the roughening work efficiency.

[0010] In combination with some embodiments of the first aspect, in some embodiments, before the step of determining the initial roughing rate of each path segment in the first motion path in which only the robot arm moves, the method also includes: calculating the curve differential of each coordinate point in the first motion path; dividing the first motion path into multiple path segments according to the curve differential; determining the segmentation type of the path segment as a smooth segmentation or a mutation segmentation; and determining the rate optimization interval of each path segment in the first motion path according to the initial roughing rate, the step specifically includes: determining the rate optimization interval according to the initial roughing rate and segmentation type of each path segment in the first motion path.

[0011] In the above embodiment, the shoe upper roughening system first performs curve differentiation on the first motion path of the robot arm, and divides the path into multiple smooth or sudden segments accordingly. When optimizing the roughening rate, the initial rate of the segment and the segment type are comprehensively considered to determine the rate optimization interval. By distinguishing between smooth segments and sudden segments, the roughening rate can be optimized more finely.

[0012] In combination with some embodiments of the first aspect, in some embodiments, the step of determining the rate optimization interval according to the initial roughening rate and segmentation type of each path segment in the first motion path specifically includes: when the segmentation type is a smooth segmentation, determining the preset upper rate interval of the target upper surface in the roughening operation; calculating the difference between the upper rate interval and the initial roughening rate to obtain the rate optimization interval; when the segmentation type is a mutation segmentation, determining the optimized segmentation of the robot arm when the target upper surface moves together with the robot arm and the shoe last; calculating the rate interval of the robot arm in the optimized segmentation to obtain the rate optimization interval.

[0013] In the above embodiment, the shoe upper roughening system determines the upper rate range of the shoe upper roughening operation for the smooth segment, and calculates the difference between the range and the initial rate as the optimization range. For the sudden segment, the robot arm's optimized segment under the joint motion of the shoe upper and the shoe last is determined, and the robot arm's rate range in the segment is calculated as the optimization range. Based on the characteristics of different path segments, the rate optimization space can be determined in a targeted manner.

[0014] In combination with some embodiments of the first aspect, in some embodiments, the movement path of the shoe last is a one-dimensional path, and the movement direction of the shoe last is the shoe last compensation direction; determining the roughening rate of each path segment in the first motion path, and determining the path to be optimized based on the roughening rate, specifically includes: extracting the path segments in the first motion path whose mapping in the shoe last compensation direction is not zero, as the segments to be optimized; calculating and determining all the segments to be optimized, and obtaining the path to be optimized.

[0015] In the above embodiment, when the movement path of the shoe last is limited to a one-dimensional path, that is, when the shoe last can only move in one direction, the shoe last movement direction is defined as the compensation direction by the shoe upper roughening system. Then, the path segment with a non-zero mapping in the compensation direction is extracted from the first movement path of the robot as the segment to be optimized, that is, the movement path of the robot with a deviation in this direction, so that the calculation result is more accurate and the accuracy of roughening is improved.

[0016] In combination with some embodiments of the first aspect, in some embodiments, after the step of generating a three-dimensional mesh model based on the collected point cloud data of the target upper, the method also includes: determining the mapping relationship between the rotation of the shoe last and the movement of the target upper based on the three-dimensional mesh model; and determining the conversion formula between the upper motion curve and the shoe last motion path based on the mapping relationship.

[0017] In the above embodiment, after the upper roughening system generates the upper three-dimensional model, it determines the mapping relationship between the rotation of the shoe last and the movement of the upper according to the model. Due to the difference in the sizes of different uppers, the same rotation of the shoe last will cause different movement trajectories of the upper, which needs to be quantitatively calculated. By establishing this mapping relationship, the movement of the shoe last can be more accurately planned to optimize the roughening process.

[0018] In combination with some embodiments of the first aspect, in some embodiments, after controlling the robot arm to move in a second motion path and controlling the shoe last to move in a shoe last motion path to perform the step of roughening the upper, the method also includes: collecting point cloud data of the target upper after the roughening is completed to obtain a roughening effect model; generating a roughening quality score for the roughening of the upper based on the roughening effect model; and sending a prompt message to the user terminal when the roughening quality score is lower than a preset qualified threshold.

[0019] In the above embodiment, after the roughening is completed, the shoe upper roughening system collects point cloud data of the roughened shoe upper again to obtain a roughening effect model, and generates a roughening quality score based on it; when the score is lower than the preset threshold, the system will send a prompt message to the user terminal. Roughing quality problems can be discovered and prompted in time, improving the level of production control.

[0020] In a second aspect, an embodiment of the present application provides a shoe upper roughening system, which comprises: one or more processors and a memory; the memory is coupled to the one or more processors, the memory is used to store computer program code, the computer program code comprises computer instructions, and the one or more processors call the computer instructions to enable the shoe upper roughening system to execute the method described in the first aspect and any possible implementation method of the first aspect.

[0021] In a third aspect, an embodiment of the present application provides a computer program product comprising instructions, and when the above-mentioned computer program product runs on a shoe upper roughening system, the above-mentioned shoe upper roughening system executes the method described in the first aspect and any possible implementation method of the first aspect.

[0022] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, comprising instructions. When the above instructions are executed on a shoe upper roughening system, the above shoe upper roughening system executes the method described in the first aspect and any possible implementation method of the first aspect.

[0023] It can be understood that the shoe upper roughening system provided in the second aspect, the computer program product provided in the third aspect, and the computer storage medium provided in the fourth aspect are all used to execute the method provided in the embodiment of the present application. Therefore, the beneficial effects that can be achieved can refer to the beneficial effects in the corresponding method, which will not be repeated here.

[0024] One or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages: 1. The technical solution of introducing shoe last movement for collaborative optimization simplifies the complex process of relying solely on robot movement. Efficient roughing is achieved by synchronously controlling the movement of the robot and the shoe last, solving the problems of insufficient in-depth robot movement path planning and low operating efficiency in the prior art, thereby improving the roughing efficiency and upper product quality.

[0025] 2. For the smooth segment, the difference between the upper limit of the roughening operation on the shoe upper and the current rate is calculated. For the mutation segment, the movement complexity of the robot arm is simplified by introducing the movement of the shoe last. Then, the rate optimization space of the robot arm under the cooperation of the shoe last is calculated. This technical means accurately evaluates the rate range that can be improved, solves the problems of fuzzy rate optimization analysis and unclear optimization range in the existing technology, and maximizes efficiency while ensuring the roughening effect.

[0026] 3. Due to the technical solution of establishing the mapping relationship between the rotation of the shoe last and the movement of the upper based on the analysis of the three-dimensional model of the upper, by constructing a last-upper kinematic mapping model, the upper response under any last rotation can be accurately predicted, and then the subsequent robot path planning can be optimized. This solves the problem of ignoring the coupling of the last and upper movement and the large error of the roughening trajectory in the existing technology, realizes precise control, and improves the stability and reliability of the upper roughening process. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 It is a flow chart of a method for roughening a shoe upper in an embodiment of the present application; Figure 2 is another flow chart of the method for roughening the shoe upper in an embodiment of the present application; Figure 3 It is a schematic diagram of the structure of a physical device of the shoe upper roughening system in the embodiment of the present application. DETAILED DESCRIPTION

[0028] The terms used in the following embodiments of the present application are only for the purpose of describing specific embodiments, and are not intended to be used as limitations to the present application. As used in the specification of the present application, the singular expressions "one", "a kind of", "above", "the" and "this" are intended to also include plural expressions, unless there is a clear indication to the contrary in the context. It should also be understood that the term "and / or" used in the present application refers to any or all possible combinations comprising one or more of the listed items.

[0029] In the following, the terms "first" and "second" are used for descriptive purposes only and are not to be understood as suggesting or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features, and in the description of the embodiments of the present application, unless otherwise specified, "plurality" means two or more.

[0030] See also Figure 1 , which is a flow chart of the method for roughening the shoe upper in an embodiment of the present application.

[0031] S101: Generate a three-dimensional mesh model based on the collected point cloud data of the target shoe upper.

[0032] The target upper refers to the upper workpiece that needs to be roughened. Point cloud data refers to a collection of spatial coordinate data of a large number of points collected on the surface of a target object by a 3D scanner, which can be used to reconstruct a 3D model of the object. A 3D mesh model refers to a digital representation of a target object composed of a series of small plane units such as triangular facets, which can depict the geometric shape characteristics of the object.

[0033] Specifically, after receiving the shoe upper roughening instruction, the shoe upper roughening system first obtains the surface shape data of the target shoe upper through a 3D scanner or depth camera. Usually, this data exists in the form of a point cloud, which contains the 3D coordinate information of a large number of sampling points on the target surface. After receiving the point cloud data, the shoe upper roughening system optimizes the point cloud quality through pre-processing operations such as point cloud filtering and registration, and then uses 3D reconstruction algorithms such as triangulation and B-spline fitting to convert the discrete point cloud into a continuous and smooth 3D mesh surface model.

[0034] S102: extract features from the three-dimensional mesh model to determine the target shoe upper area to be polished.

[0035] Feature extraction refers to the process of identifying representative feature elements such as regions, lines, and points from the geometric and topological information of the 3D model. The area to be polished refers to the target area on the upper that needs to be polished, usually the seams, edges, and other parts of the upper.

[0036] Specifically, after obtaining the three-dimensional mesh model of the upper, the upper roughening system performs feature recognition and extraction processing on the model. First, the mesh model is partitioned into different areas such as the upper, the sole, etc. Then, feature detection is performed on each area. By analyzing the geometric properties of the mesh such as curvature and normal vector, the protruding edges, sharp edges, and steep surface change areas on the model are found. These areas are usually where the seams of the upper are located and are the areas that need to be polished. In addition, the position of the stitching line of the upper can be directly located based on prior information such as design drawings. Finally, all the extracted feature areas are merged, and the area is appropriately expanded considering the size of the polishing head, and finally the target area for the upper roughening is determined.

[0037] S103: Generate a first motion path of the robot arm according to the area to be polished.

[0038] The robot arm refers to a robot arm used to perform shoe upper polishing operations. The first motion path represents the trajectory that the tool center point (TCP) at the end of the robot arm needs to follow, and the trajectory must cover the area to be polished and meet the kinematic constraints of the polishing head.

[0039] Specifically, after determining the target area for polishing the upper, the upper roughening system begins to plan the polishing action sequence of the robot. First, the target area is converted into a set of several polishing points, and a uniformly distributed sampling point cloud is generated in the target area through discretization and sampling methods. Then, considering the size of the polishing head, the sampling points are clustered with the radius of the polishing head as the threshold, and several cluster centers are obtained as the target polishing points of the robot. Next, based on the set of polishing points, path planning algorithms such as TSP and RRT are used to generate the optimal path covering all polishing points while satisfying the kinematic constraints of the robot, namely the range of motion, speed and acceleration of the joints. The path contains a series of target angles for the robot's joints, which can guide the robot to move smoothly along the planned trajectory and perform full coverage polishing on the target area of ​​the upper.

[0040] S104: Determine the roughening rate of each path segment in the first motion path, and determine the path to be optimized according to the roughening rate.

[0041] Among them, path segmentation means dividing the overall grinding path of the robot into several segments. The roughening rate refers to the speed of the robot moving along a certain path, which affects the grinding efficiency and effect. The path to be optimized refers to the set of key path segments extracted from the first motion path that can improve efficiency by optimizing the roughening rate.

[0042] Specifically, after obtaining the initial motion path of the robot, the shoe upper roughening system will perform speed planning and optimization on the path. First, the path is divided into several segments using information such as curvature and acceleration, with smooth segments and non-smooth segments alternating. Then, for each path segment, considering the kinematic and dynamic characteristics of the robot and the grinding tool, such as joint speed and torque limitations, the theoretical maximum speed of the segment is calculated as the initial roughening rate. Next, the speed improvement space of each segment is evaluated, focusing on the upper limit of the rate for the smooth segment and analyzing the movement restrictions of the robot for the non-smooth segment. Finally, key path segments with speed improvement potential are selected to form a set of paths to be optimized, providing input for the next step of speed optimization.

[0043] In some embodiments, the segmentation and rate initialization of the grinding path can be achieved in a variety of ways: optionally, the speed and acceleration of each trajectory point on the path are calculated, and when the speed or acceleration changes exceed the threshold, it is marked as a segmentation point to form a segmentation of the path, and then the initial roughening rate upper limit of each segment is calculated considering the robot joint speed limit; optionally, the local curvature is calculated for the sampling points on the path, and the path is segmented with the curvature extreme point as the segmentation point, and the roughening rate of the smooth segment is preset to be higher than that of the non-smooth segment, and then the initial speed scheme of each segment is planned based on the maximum speed of the grinding head; optionally, the grinding path is projected onto the XY plane under the robot base coordinate system, and the inflection point is extracted to divide the path into straight segments and arc segments. For the straight segments, only the robot speed limit is considered, and for the arc segments, the centripetal acceleration limit must also be considered, and their initial roughening rates are calculated respectively. It can be understood that other speed planning techniques, such as trapezoidal speed planning, S-type speed planning, etc., can also be applied to optimize the extraction of the roughening path segments.

[0044] S105: Determine a shoe last motion path of the shoe last according to the path to be optimized, and correct the first motion path of the robot arm to obtain a second motion path.

[0045] Among them, the shoe last refers to the bracket that fixes and supports the upper workpiece. The shoe last motion path refers to the coordinated movement of the shoe last such as rotation and translation during the robot hand grinding process, which is used to optimize the grinding angle and position. The second motion path represents the robot hand grinding trajectory after optimization and adjustment under the coordination of the shoe last movement.

[0046] Specifically, after extracting the path segments to be optimized, the upper roughening system will optimize the polishing schemes of these path segments by introducing the last motion. First, a motion coupling model between the last motion and the upper polishing is established, which can be constructed based on the last three-dimensional model. Then, for each path segment to be optimized, the theoretical polishing efficiency of the robot is evaluated under the condition that the last remains stationary. If the efficiency is low, the last motion path of the segment is generated, and the upper tangent is dynamically adjusted by rotating the last to improve the polishing posture and speed of the robot. At the same time, the displacement of the upper caused by the last motion is considered, and the original robot path is corrected to maintain accurate trajectory tracking, thereby obtaining the optimized second motion path. Finally, all optimized segmented last motion paths and the second motion path of the robot are combined to form a complete upper polishing scheme.

[0047] In some embodiments, the path optimization under the introduction of shoe last movement can be achieved in a variety of ways: Optionally, for the turning segments in the robot hand grinding path, the shoe last is rotated to keep the tangential angle between the grinding head and the upper constant, offset the posture adjustment time of the robot hand, and at the same time, the upper linear velocity caused by the rotation of the shoe last is superimposed on the robot hand velocity vector, so as to achieve smooth turning and speed improvement; Optionally, in the straight grinding section of the robot hand, the shoe last is reciprocated to form an angle with the robot hand movement direction, and the upper tangential velocity component is used to improve the relative grinding rate of the robot hand, and the straight segment of the path is optimized; Optionally, for the grinding path of the free-form surface upper, the upper posture is adjusted by rotating the shoe last, so that the grinding path is on the same surface as much as possible, reducing the position adjustment of the robot hand in the normal direction, simplifying the path and improving the grinding efficiency. It can be understood that multi-axis linkage, interpolation and other motion control technologies can also be applied to achieve dynamic optimization of the grinding trajectory under the coordinated movement of the robot hand and the shoe last.

[0048] S106, controlling the robot arm to move along the second motion path, and controlling the shoe last to move along the shoe last motion path, so as to roughen the shoe upper.

[0049] Among them, control refers to the process in which the upper roughening system sends control instructions to the robot arm and the shoe last, coordinating the two to start, run and stop synchronously according to the planned motion path. The motion in this step includes the joint motion of the robot arm and the rotation and translation of the shoe last.

[0050] Specifically, after completing the planning of the motion paths of the robot and the shoe last, the upper roughening system starts to perform the upper roughening operation. First, the optimized robot motion path is converted into a rotation angle sequence in the joint space, and the shoe last motion path is converted into a pulse control sequence of the stepper motor. Then the two sets of control instructions are sent to the robot controller and the shoe last controller respectively, and the movement of the two mechanisms is started synchronously. During the polishing process, the system monitors the robot joint current and the shoe last motor pulse count in real time, and controls the two mechanisms through the position loop and speed loop to strictly follow the planned path until the polishing task is completed. Finally, the movement of the robot and the shoe last is stopped, and the quality of the polished upper is inspected to evaluate whether additional polishing is needed.

[0051] The following is a more detailed description of the process of the method provided by this implementation. Figure 2 , is another flow chart of the method for roughening the shoe upper in an embodiment of the present application.

[0052] S201: Generate a three-dimensional mesh model based on the collected point cloud data of the target shoe upper.

[0053] Referring to step S101, the upper roughening system generates a three-dimensional mesh model.

[0054] S202: extract features from the three-dimensional mesh model to determine the target shoe upper area to be polished.

[0055] Referring to step S102 , the shoe upper roughening system determines the target shoe upper area to be polished.

[0056] S203: Generate a first motion path of the robot arm according to the area to be polished.

[0057] Referring to step S103 , the shoe upper roughening system generates a first motion path of the robot arm.

[0058] S204: Determine an initial roughing rate of the robot arm in each path segment in the first motion path.

[0059] Among them, only the robot arm moving means that only the movement of the robot arm itself is considered in the path segment, and the coordinated movement of the shoe last is not introduced. The initial roughening rate represents the maximum grinding speed that can be achieved in the path segment under the constraints such as the limit speed of the robot arm joint.

[0060] Specifically, after obtaining the grinding motion path of the robot, the upper roughening system will first segment the path. The system calculates the curvature, velocity, and acceleration of each trajectory point on the path, and divides the path into several smooth curves according to the mutation of these parameters. Next, the system temporarily ignores the movement of the shoe last, and only evaluates the motion characteristics of each curve for the robot body. By analyzing factors such as the upper area covered by the segmented trajectory, the range of motion of the robot joints, and considering the optimal linear speed of the grinding head, the system can obtain the theoretical maximum grinding rate of this segment of the trajectory, and use it as the initial speed of this path segment. The system calculates and generates the initial roughening rate for all path segments, providing a basis for subsequent speed optimization.

[0061] In some embodiments, path segmentation and initial rate calculation can be achieved in a variety of ways: Optionally, the path is segmented using the Adpative CS method, which achieves path segmentation by adaptively setting the curvature threshold, and then applies quintic polynomial interpolation to each path segment to generate a smooth speed planning curve, and the maximum value of the curve is used as the initial rate; Optionally, a finite state machine is used to encode the path segmentation, and the state transition condition is the characteristic change of the path in the position, speed and torque space, and then the dynamic model is used to inversely solve the optimal operating speed of the robot in each state as the initial roughing rate of the corresponding segment; Optionally, the robot joint space is normalized to the unit space, the path is mapped into the space for segmentation, and then the length of each path segment is returned to the joint space, and the initial speed of the segment is calculated based on the rated speed of each joint. It can be understood that other kinematic analysis and trajectory planning techniques can also be applied to achieve segmentation and initial velocity solution of the robot grinding path, which will not be expanded here.

[0062] S205: Determine the rate optimization interval of each path segment in the first motion path according to the initial roughing rate.

[0063] The rate optimization interval indicates the numerical range of the roughening rate that can be adjusted while ensuring the grinding effect. The upper limit of the interval corresponds to the best grinding efficiency, the lower limit corresponds to the lowest optimization efficiency, and the width of the interval reflects the adjustment space of the segmented optimization rate of the path.

[0064] Specifically, after obtaining the initial roughening rate of each path segment, the upper roughening system will estimate the speed optimization potential on this basis. For each path segment, the system first analyzes the characteristics of the upper area where it is located, such as curvature, hardness, etc., and considers the movement law of the robot in this area, such as linear motion or interpolation motion, to determine the ideal grinding rate range. Then compare the ideal rate range with the initial rate of the current segment to obtain the rate optimization interval. For path segments whose initial rates are close to the upper limit of the ideal range, there is little room for rate optimization; on the contrary, for path segments whose initial rates are far below the ideal range, there is a greater potential for rate improvement, and their rate optimization interval is wider. By calculating the rate optimization interval for all path segments, the system can determine which segments are worth optimizing the speed through shoe last movement, providing a basis for the next step of screening the segments to be optimized.

[0065] In some embodiments, the determination of the rate optimization interval can be achieved in a variety of ways: optionally, using fuzzy control rules to evaluate the rate optimization space, taking the curvature, length and other characteristics of the path segment as the input of the fuzzy controller, and taking the rate optimization interval as the output, and optimizing the fuzzy rule base through training to achieve the mapping from path characteristics to the optimization space; optionally, using reinforcement learning methods such as Q-learning, modeling the rate optimization problem as a Markov decision process, and obtaining long-term benefits by exploring different rate options. After convergence, the rate adjustment strategy under each segment state is obtained, corresponding to the rate optimization interval; optionally, constructing a rate-path feature data set, using machine learning models such as support vector machines (SVM) to fit the nonlinear relationship between rate and path features, and achieving regression prediction from path segment characteristics to rate optimization intervals. It can be understood that other intelligent optimization and learning algorithms can also be applied to establish the mapping relationship between path characteristics and rate optimization space, which is not limited here.

[0066] In some embodiments, the upper roughening system determines the rate optimization interval according to the initial roughening rate and segment type of each path segment in the first motion path. The segment type is determined by: calculating the curve differential of each coordinate point in the first motion path; dividing the first motion path into multiple path segments according to the curve differential; and determining the segment type of the path segment as a smooth segment or a sudden segment.

[0067] Among them, the initial roughing rate indicates the maximum grinding speed of each path segment when only the robot motion constraints are considered. The segment type refers to the curvature change characteristics of the path segment, which is divided into two categories: smooth segment and mutation segment. Smooth segment refers to the path segment with gentle curvature change, and mutation segment refers to the path segment with large curvature change. Curve differential is used to represent the curvature change rate of each point on the path. The rate optimization range indicates the range of values ​​within which the roughing rate can be adjusted while ensuring the grinding effect.

[0068] When optimizing the speed of the robot's grinding path, the upper roughening system first segments the path. The system calculates the curve differential of each coordinate point on the path, that is, the curvature change rate. Then, according to the curvature change rate, the path is divided into smooth segments and mutation segments. Next, for each path segment, the system considers two factors, namely, the segment type and the initial roughening rate, and estimates the adjustment space of the segment rate to obtain the rate optimization range. This process can make full use of the path characteristics and adopt different rate optimization strategies for smooth segments and mutation segments. The smooth segment focuses on speed increase, while the mutation segment focuses on stability.

[0069] Specifically, the upper roughening system first samples the polishing path and calculates the curve differential of each sampling point. By setting the curvature change threshold, the path segments whose curvature change rate exceeds the threshold are marked as mutation segments, and the rest are marked as smooth segments. The system then loads the initial roughening rate of each path segment. For the smooth segment, it focuses on the speed-up potential. By analyzing the limiting speed of the robot and the polishing tool, it estimates the rate improvement space of the segment and obtains the upper limit of rate optimization. For the mutation segment, it focuses on analyzing the movement restrictions of the robot, estimates the stable rate range of the segment through kinematic calculations, and obtains the lower limit of rate optimization. Finally, the system integrates the upper and lower limits of the rate optimization of each segment into a rate optimization interval, which can be used to guide the coordinated optimization of the robot movement and the shoe last movement.

[0070] In some embodiments, the upper roughening system will determine the preset upper rate range of the target upper surface in the roughening operation when the segmentation type is smooth segmentation; calculate the difference between the upper rate range and the initial roughening rate to obtain the rate optimization range; when the segmentation type is mutation segmentation, determine the optimized segmentation of the robot arm when the robot arm and the shoe last move together; calculate the rate range of the robot arm in the optimized segmentation, and calculate the rate optimization range.

[0071] Among them, the upper rate range indicates the theoretically highest grinding rate range that can be achieved in the smoothing segment while ensuring the grinding effect. The optimized segment refers to the robot movement segment in the mutation segment where the rate can be improved after the shoe last movement is introduced. The rate range is used to indicate the speed range in which the robot can stably operate in the optimized segment under the cooperation of the shoe last movement.

[0072] After obtaining the segmentation type of the robot's grinding path, the shoe upper roughening system will adopt targeted rate optimization strategies according to the segmentation characteristics. For smooth segments, due to the smooth curvature change and simple grinding trajectory, the system will focus on evaluating the robot's speed-up potential, which can also be provided by the opposite movement of the shoe last. The upper rate range of the smooth segment is summarized through experiments or experience, and then the range is compared with the initial rate of the current segment. The difference reflects the rate optimization space of the segment. For mutation segments, due to the large change in curvature and the complex movement of the robot, simple speed-up is prone to cause vibration and overshoot. At this time, the system will focus on introducing the shoe last movement for coordination, and simplify the movement path of the robot by rotating the shoe last. The system first identifies the key movement segments of the robot in the mutation segment as the optimization segment, and then analyzes the stability of the robot's rate in the optimization segment with the cooperation of the shoe last movement, and obtains the rate fluctuation range, which corresponds to the rate optimization space of the optimization segment.

[0073] Specifically, for smooth segments, the upper roughening system first obtains the ideal upper and lower limits of the grinding rate by table lookup or calculation based on factors such as the grinding head type, workpiece material, and path curvature. Then, the upper rate range is compared with the initial rate of the current segment, and the size of the rate optimization space is estimated using the interval analysis method. For mutation segments, the system first extracts the critical path segments in which the robot moves more, and the segments in which the movement speed and acceleration changes exceed a certain threshold are marked as optimized segments. Then the system switches to the simulation mode of the coordinated movement of the robot and the shoe last, and discretely samples the shoe last angle with a certain step size in the optimized segment, and evaluates the rate-acceleration stability domain of the robot movement under each sampling condition. The intersection of the stability domains of multiple conditions is the rate optimization interval of the optimized segment.

[0074] In some embodiments, the determination of the rate optimization interval of the smooth segment and the mutation segment can be achieved in a variety of ways: Optionally, the maximum principle is adopted for the smooth segment, that is, the upper limit of the rate of the segment is determined by the speed limit of the robot joint and the speed limit of the shoe last (when moving in opposite directions, the final rate of relative motion is equal to the sum of the two), and the maximum value of the path rate is solved under the condition that the terminal speed is continuous, so as to construct the rate optimization interval; the minimum energy principle is adopted for the mutation segment, and after the shoe last movement is introduced, the large change of the joint space speed is preferentially reduced, and the speed optimal target and the acceleration optimal target are weightedly combined to obtain the rate optimization interval through convex optimization solution; Optionally, the speed- The torque constraint diagram determines the rate optimization space. In the process of robot speed planning, the grinding force is used as the optimization target, and the speed-torque constraint is converted into an inequality constraint for speed optimization. Under this constraint, the speed is gradient-enhanced. The reduction ratio optimization method is used for the mutation segment. By searching for the optimal reduction ratio of each axis of the robot, the joint speed change is minimized, and the stable rate interval of the optimized segment is calculated under the optimal reduction ratio. Optionally, the same hierarchical optimization strategy is used for the smooth segment and the mutation segment, with the entire optimization space as the global and the key point rate as the local. The interval analysis method is alternately applied to search for the global optimal interval and the local optimal rate. The convergence is judged by the KKT condition to obtain the global optimal solution of the segment rate. It can be understood that other trajectory optimization algorithms such as genetic algorithms, particle swarm algorithms, etc. can also be applied to solve the rate optimization space of the smooth segment and the mutation segment under the constraints of the coordinated motion of the robot and the shoe last. It will not be repeated here.

[0075] S206: When the lower limit value of the rate optimization interval exceeds a preset optimization threshold, the corresponding path segment is used as the segment to be optimized.

[0076] The preset optimization threshold refers to the minimum value of the rate improvement range that is preset according to the requirements of the upper roughening process. When the lower limit of the rate optimization space of a certain path segment exceeds the threshold, it can be considered that the roughening rate of this segment can be significantly improved through the last movement coordination, so it is marked as a segment to be optimized.

[0077] Specifically, after obtaining the rate optimization interval of each path segment, the upper roughening system will screen out segments with significant optimization value according to preset conditions. The system first sets the optimization threshold for rate improvement based on previous polishing experience and test data, usually 10% to 20%. Then the system judges the rate optimization interval of each path segment to see whether its lower limit exceeds the optimization threshold. If it exceeds, it means that the segment has great potential for rate improvement, and it is worth the cost to introduce shoe last movement for optimization; if it does not exceed, it means that the segment has limited room for rate improvement, and continued optimization is not cost-effective. By threshold judgment and screening of all segments, the system finally obtains a batch of segments with the greatest rate optimization potential to be optimized, laying the foundation for the subsequent introduction of shoe last movement to formulate a path optimization plan.

[0078] S207: Calculate and determine all the segments to be optimized to obtain the paths to be optimized.

[0079] The path to be optimized refers to a set of robot trajectory segments that can be introduced into the shoe last movement for rate improvement after rate optimization screening. Compared with the original complete polishing path, the path to be optimized focuses more on the key sections that need and are worth improving.

[0080] Specifically, after all the segments to be optimized are screened out through the threshold conditions, the upper roughening system needs to organize these scattered path fragments into a structured data representation. The system traverses each screened segment to be optimized, extracts its starting and ending point information in the original path, and the location of the upper area covered by the segment. At the same time, the system also calculates the mean of the rate improvement space for each segment as the optimization potential score of the segment. Finally, the system summarizes the key information of each segment to be optimized, and splices it to form an optimized path that is isomorphic to the original path. This path retains the topological structure of the complete path, but the segments in it are all path fragments with the greatest potential for rate optimization.

[0081] In some embodiments, the generation of the path to be optimized can be achieved in a variety of ways: Optionally, the path to be optimized is constructed in a stacking manner, and the starting point, end point, rate improvement space and other information of each segment to be optimized are encapsulated into a structure, sorted according to the optimization potential score, and then these structures are stacked in order from high to low to form a weighted path to be optimized; Optionally, the path to be optimized is constructed in a sparse storage manner, and a sparse matrix of the same length as the original path is designed, with non-segments to be optimized corresponding to zero values, and segments to be optimized corresponding to rate improvement spaces, so as to achieve in-situ marking of key path information; Optionally, the path to be optimized is constructed in a graph structure manner, and a path graph is constructed with the original polished path as the skeleton, and the segments to be optimized are mapped to key nodes and edges on the path graph, and the optimization potential is used as the edge weight, and the key path is searched through a graph algorithm. It can be understood that other data structures and algorithms can also be applied to characterize and search for the optimal path segment set, and are not limited to the above implementation methods.

[0082] In some embodiments, the movement path of the shoe last is a one-dimensional path, and the movement direction of the shoe last is the shoe last compensation direction; at this time, the upper roughening system will extract the path segments in the first movement path whose mapping in the shoe last compensation direction is not zero as the segments to be optimized; calculate and determine all the segments to be optimized to obtain the path to be optimized.

[0083] Among them, the one-dimensional path means that the shoe last reciprocates in only one direction without multi-axis linkage. The shoe last compensation direction refers to the synthetic direction of the shoe last movement, in which the displacement of the upper caused by the shoe last movement is most significant. Mapping refers to the projection length obtained by mapping the robot path to the direction of shoe last movement. The segments to be optimized refer to the path segments whose projection in the shoe last compensation direction is not zero, and the rates of these segments are greatly affected by the shoe last movement. The path to be optimized is the rate optimization key path formed by connecting all the segments to be optimized in sequence.

[0084] When introducing the movement of the shoe last for rate optimization, the upper roughening system needs to evaluate the degree to which each segment of the robot path is affected by the movement of the shoe last. Considering the limited freedom of the shoe last, the system usually limits the movement path of the shoe last to a simple one-dimensional linear motion, that is, the up and down adjustment movement of the height, or the forward and backward movement. At this time, the dominant direction of the shoe last movement can be determined, which is called the shoe last compensation direction. The system maps each segment in the robot path to the shoe last compensation direction and calculates its projection length. For segments with a projection length that is not zero, their rate will be significantly affected by the movement of the shoe last, and these segments are marked as segments to be optimized. After evaluating and screening each segment, the system finally obtains the path to be optimized formed by connecting all the segments to be optimized in sequence. This path is the focus of subsequent optimization of the coordinated movement of the robot and the shoe last.

[0085] Specifically, the upper roughening system first determines the degrees of freedom of movement of the shoe last according to the shoe last structure, such as determining the shoe last movement path by combining the rotation axis and the direction of the slide rail. The system then calculates the angle between the shoe last movement path and the robot arm path in the base coordinate system, and defines the component of the shoe last movement in the direction of the angle as the shoe last compensation direction. Next, the system uses the compensation direction as the normal vector, performs vector projection on each segment of the robot arm path in turn, and calculates the projection length of each segment in the compensation direction. For path segments whose projection length is greater than the threshold, the system marks them as segments to be optimized. Finally, the system splices all segments to be optimized into a complete path to be optimized in the original order of the path. This is equivalent to extracting the key path that best reflects the movement effect of the shoe last from the robot arm path.

[0086] S208: Determine a mapping relationship between the rotation of the shoe last and the movement of the target shoe upper according to the three-dimensional grid model.

[0087] Among them, the rotation of the shoe last refers to the movement mode of changing the direction of the upper by rotating the shoe last during the robot hand grinding process. The upper movement refers to the movement trajectory of each point on the upper under the action of the shoe last rotation. The mapping relationship refers to the unique determination of the displacement of each point on the upper given the rotation amount of the shoe last in a certain direction. This relationship is usually closely related to the shape of the upper.

[0088] Specifically, after obtaining the path to be optimized, the upper roughening system needs to find a specific solution for rate optimization using the last motion. The system first loads the three-dimensional mesh model corresponding to the upper polishing and extracts the last model. Then the system selects a suitable rotation axis on the last model, usually the direction from the heel to the toe. With this axis as the center, the system applies an equally spaced rotation transformation in the range of -60° to 60° on the last model. For each rotation angle, the system calculates the coordinate transformation of each vertex on the upper mesh model to obtain the motion displacement of each point on the upper at the current angle. Through multiple simulation rotation and displacement calculations, the system calculates the pairing data of the last rotation angle and the upper displacement. Finally, these data are curve fitted to obtain the mapping equation between the last rotation angle and the upper vertex displacement. This mapping equation intuitively describes the influence of the last motion on the upper motion.

[0089] In some embodiments, the modeling of the shoe last-upper mapping relationship can be achieved in a variety of ways: optionally, a kinematic method is used, based on the three-dimensional model of the shoe last, the shoe last and the upper are regarded as rigid bodies, and a kinematic transformation matrix from the shoe last coordinate system to the upper coordinate system is established by setting a rotation pair and a translation pair, and the matrix parameters and the shoe last angle constitute a mapping equation; optionally, a finite element simulation method is used to divide the upper mesh into a number of units, and a rotation boundary condition is imposed on the shoe last model. The displacement field of the unit node is calculated by displacement transfer, and then the functional mapping of the displacement field and the angle is fitted; optionally, a machine learning method is used to collect a certain number of shoe last angle-upper displacement data pairs through virtual simulation or physical experiments, and then a neural network model is used to fit the nonlinear mapping relationship from angle to displacement. It can be understood that other physical modeling, simulation analysis and data-driven methods can also be applied to characterize the constraint relationship between shoe last movement and upper response.

[0090] S209: Determine a conversion formula between the upper motion curve and the last motion path according to the mapping relationship.

[0091] The upper motion curve refers to the motion trajectory of the key points of the upper polishing during the rotation of the shoe last. These key points are usually located on the path segment to be optimized. The last motion path represents a series of rotation and positioning actions that the shoe last needs to perform during the polishing process in order to achieve the desired upper motion curve. The conversion formula defines the mathematical relationship between the upper motion description and the last motion description.

[0092] Specifically, after establishing the mapping relationship between the rotation of the shoe last and the movement of the upper, the upper roughening system needs to further explore the practical application value of the mapping relationship. The system first extracts several key points for upper polishing based on the path to be optimized. These points are usually the starting point, end point or intermediate bending point of the segment to be optimized. Then the system studies the change law of the motion trajectory of the key points under different shoe last rotation states. For a given target upper motion curve, the system calculates a set of shoe last rotation sequences that can generate the curve by solving the inverse problem of the mapping equation. Furthermore, the system fine-tunes the angular velocity and time of the above-mentioned shoe last rotation sequence in combination with the speed planning of the robot on the segmented path to be optimized, and finally obtains a shoe last motion instruction set synchronized with the robot movement.

[0093] S210: Determine a shoe last motion path of the shoe last according to the path to be optimized, and correct the first motion path of the robot arm to obtain a second motion path.

[0094] Referring to step S105, the upper roughening system determines the shoe last movement path and the second movement path.

[0095] S211, controlling the robot arm to move along the second motion path, and controlling the shoe last to move along the shoe last motion path, so as to roughen the shoe upper.

[0096] Referring to step S106, the upper roughening system will control the roughening.

[0097] In some embodiments, the upper roughening system will collect point cloud data again on the target upper after the roughening is completed to obtain a roughening effect model; based on the roughening effect model, a roughening quality score for the upper roughening is generated; when the roughening quality score is lower than a preset qualified threshold, a prompt message is sent to the user terminal.

[0098] Among them, the roughening effect model refers to a digital model obtained by reconstructing the shape of the upper after grinding in three dimensions. This model can accurately depict the processing effect of the upper. The roughening quality score is a score obtained by quantitatively evaluating the roughening quality of the upper by comparing the ideal model with the actual roughening effect model. The preset qualified threshold represents the lower limit of the upper quality score set manually in advance. If it is lower than this threshold, the upper roughness is considered to be substandard. The user terminal contains a human-computer interaction interface, through which the system can report roughening quality problems to the user.

[0099] After a complete upper roughening task is completed, the upper roughening system will perform a quality inspection on the roughening effect. The system re-scans the polished upper in three dimensions to obtain its point cloud data. The data is used to reconstruct the upper solid model to obtain the roughening effect model. The system then compares and analyzes the effect model with the ideal design model, calculates the shape error between the two models, and generates a quantitative score for the upper roughening quality based on indicators such as the error size and distribution. The system compares the score with the pre-set qualified threshold. If the score is lower than the threshold, it means that the roughening quality does not meet the standard. The system will send a warning message to the human-computer interaction interface to remind the operator that the upper roughness exceeds the standard and needs to be reworked. Automated quality inspection and feedback can help to promptly discover and solve the problem of unqualified upper roughening and improve production yield.

[0100] Specifically, after the upper roughening system completes all the grinding tasks of an upper workpiece, it automatically transfers the workpiece to the 3D scanning station. The same point cloud acquisition equipment and algorithm as the initial scan are used to measure the surface shape of the upper after grinding with high precision to obtain the effect point cloud data. Using point cloud reconstruction technology such as Poisson surface reconstruction, the point cloud data is converted into a triangular patch mesh model with attributes such as normal vectors and curvature, which is the roughening effect model. Next, the system loads the design CAD model of the upper and achieves the optimal registration of the CAD model and the roughening effect model through the ICP algorithm. Then the system uniformly samples the corresponding point pairs on the registered model, calculates the distance deviation between each point pair, and performs statistical analysis on the deviation value to obtain the mean, variance, distribution histogram, etc. of the deviation. Finally, the system compares each deviation index with the empirical threshold, and designs a weighted summation formula to calculate the comprehensive score of the upper roughening quality. If the comprehensive score is lower than the preset qualified threshold, the system will immediately pop up a warning reminder and upload the problem signal to the MES system and alarm terminal through the industrial bus gateway.

[0101] In some embodiments, the detection and evaluation of the quality of the roughening of the shoe upper can be achieved in a variety of ways: optionally, a structured light three-dimensional imaging solution is used to obtain high-precision point cloud data, a spherical harmonic function method is used to reconstruct the shoe upper effect model, the M-estimation method is used to eliminate large errors when evaluating the model deviation, a scoring function that comprehensively considers various indicators of shape deviation is designed, and the qualified threshold of the logistic function is determined by the validation set data; optionally, a lidar scanning is used to obtain depth map data, the shoe upper space is discretized by the voxel grid method, the effect model is represented by the SDF implicit function, the model deviation calculation is efficiently implemented by the ray casting algorithm, the anomaly detection of the deviation indicator is based on the isolation forest algorithm, the deviation pattern is divided by the clustering algorithm, and the clustering result is mapped to a quality score; optionally, a joint inverse design method is used, the design parameters are used as the estimated quantities, the shoe upper polishing process is regarded as a parameter transfer and disturbance process, the process is simulated by Monte Carlo sampling, the posterior distribution differences between the roughening effect model and the design parameters are compared, a Bayesian estimation framework is constructed to recursively predict the quality status, and whether to trigger an alarm is determined based on the status prediction result. It is understandable that deep learning methods can also be introduced to perform end-to-end training on a quality prediction network that takes a rough effect model as input and a quality score as output, thereby achieving automatic mapping from the effect model to the quality assessment.

[0102] In the embodiment of the present application, due to the use of three-dimensional vision-based upper reconstruction and polishing process monitoring technology, combined with trajectory model constraint optimization, adaptive trajectory generation and closed-loop error compensation algorithms, the robot upper polishing trajectory planning is automated and refined, effectively solving the problems of poor versatility of traditional offline programming solutions and difficulty in quickly adapting to changes in upper styles. The present application also proposes strategies such as collaborative optimization of shoe last motion and robot motion and optimization of process parameters based on simulation, further expanding the flexibility of the trajectory and the optimization space of the polishing process, greatly improving the work efficiency and workpiece quality of robot upper polishing.

[0103] The following describes the shoe upper roughening system in the embodiment of the present invention from the perspective of hardware processing. Figure 3 , which is a schematic diagram of the structure of a physical device of the shoe upper roughening system in an embodiment of the present application.

[0104] It should be noted that Figure 3 The structure of the shoe upper roughening system shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.

[0105] like Figure 3 As shown, the upper roughening system includes a central processing unit (CPU) 301, which can perform various appropriate actions and processes according to the program stored in the read-only memory (ROM) 302 or the program loaded from the storage part 308 to the random access memory (RAM) 303, such as executing the method described in the above embodiment. In the RAM 303, various programs and data required for system operation are also stored. The CPU 301, the ROM 302 and the RAM 303 are connected to each other through the bus 304. The input / output (I / O) interface 305 is also connected to the bus 304.

[0106] The following components are connected to the I / O interface 305: an input section 306 including an audio input device, a button switch, etc.; an output section 307 including a liquid crystal display (LCD) and an audio output device, an indicator light, etc.; a storage section 308 including a hard disk, etc.; and a communication section 309 including a network interface card such as a LAN (Local Area Network) card, a modem, etc. The communication section 309 performs communication processing via a network such as the Internet. A drive 310 is also connected to the I / O interface 305 as needed. A removable medium 311, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 310 as needed so that a computer program read therefrom is installed into the storage section 308 as needed.

[0107] In particular, according to an embodiment of the present invention, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present invention includes a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a computer program for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through the communication part 309, and / or installed from a removable medium 311. When the computer program is executed by the central processing unit (CPU) 301, various functions defined in the present invention are performed.

[0108] It should be noted that specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM), a flash memory, an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In the present invention, a computer-readable storage medium may be any tangible medium containing or storing a program that may be used by or in combination with an instruction execution system, apparatus, or device.

[0109] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present invention. Each box in the flowchart or block diagram may represent a module, a program segment, or a part of a code, and the above-mentioned module, program segment, or a part of a code contains one or more executable instructions for implementing the specified logical functions. It should also be noted that in some alternative implementations, the functions marked in the box may also occur in an order different from that marked in the accompanying drawings.

[0110] Specifically, the shoe upper roughening system of this embodiment includes a processor and a memory. The memory stores a computer program. When the computer program is executed by the processor, the shoe upper roughening method provided in the above embodiment is implemented.

[0111] As another aspect, the present invention further provides a computer-readable storage medium, which may be included in the shoe upper roughening system described in the above embodiment; or may exist independently without being assembled into the shoe upper roughening system. The above storage medium carries one or more computer programs, and when the above one or more computer programs are executed by a processor of the shoe upper roughening system, the shoe upper roughening system implements the shoe upper roughening method provided in the above embodiment.

[0112] As described above, the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present application.

[0113] As used in the above embodiments, the term "when..." may be interpreted to mean "if..." or "after..." or "in response to determining..." or "in response to detecting...", depending on the context. Similarly, the phrases "upon determining..." or "if (the stated condition or event) is detected" may be interpreted to mean "if determining..." or "in response to determining..." or "upon detecting (the stated condition or event)" or "in response to detecting (the stated condition or event)", depending on the context.

[0114] Those skilled in the art can understand that to implement all or part of the processes in the above-mentioned embodiments, the processes can be completed by computer programs to instruct related hardware, and the programs can be stored in computer-readable storage media. When the programs are executed, they can include the processes of the above-mentioned method embodiments. The aforementioned storage media include: ROM or random access memory RAM, magnetic disk or optical disk and other media that can store program codes.

Claims

1. A method for roughening a shoe upper, characterized in that: Applied to the shoe upper roughening system, the method comprises: Generate a three-dimensional mesh model based on the collected point cloud data of the target shoe upper; Performing feature extraction on the three-dimensional mesh model to determine the area to be polished of the target shoe upper; Generating a first motion path of the robot according to the area to be polished; Determine the roughing rate of each path segment in the first motion path, and determine the path to be optimized according to the roughing rate; Determining a shoe last motion path of the shoe last according to the path to be optimized, and correcting the first motion path of the robot arm to obtain a second motion path; The robot arm is controlled to move along the second motion path, and the shoe last is controlled to move along the shoe last motion path, so as to roughen the shoe upper.

2. The method according to claim 1, characterized in that The step of determining the roughing rate of each path segment in the first motion path and determining the path to be optimized according to the roughing rate specifically includes: Determine an initial roughing rate for each path segment in the first motion path in which only the robot arm moves; Determining a rate optimization interval for each path segment in the first motion path according to the initial punching rate; When the lower limit value of the rate optimization interval exceeds the preset optimization threshold, the corresponding path segment is used as the segment to be optimized; All the segments to be optimized are determined by calculation to obtain the paths to be optimized.

3. The method according to claim 2, characterized in that Before the step of determining the initial roughing rate of each path segment in the first motion path when only the robot hand moves, the method further includes: Calculating the curve differential of each coordinate point in the first motion path; dividing the first motion path into a plurality of path segments according to the curve differentiation; Determining whether the segmentation type of the path segmentation is a smooth segmentation or a sudden segmentation; The step of determining the rate optimization interval of each path segment in the first motion path according to the initial roughing rate specifically includes: A rate optimization interval is determined according to the initial punching rate and the segment type of each path segment in the first motion path.

4. The method according to claim 3, characterized in that The step of determining the rate optimization interval according to the initial roughing rate of each path segment in the first motion path and the segment type specifically includes: When the segmentation type is smooth segmentation, determining a preset upper rate range of the target shoe upper in the roughening operation; Calculate the difference between the upper rate range and the initial roughing rate to obtain a rate optimization range; When the segmentation type is a sudden segmentation, determining an optimized segmentation of the robot arm when the target shoe upper moves together with the robot arm and the shoe last; The speed interval of the robot arm in the optimization segment is calculated to obtain the speed optimization interval.

5. The method according to claim 1, characterized in that The movement path of the shoe last is a one-dimensional path, and the movement direction of the shoe last is a shoe last compensation direction; The step of determining the roughing rate of each path segment in the first motion path and determining the path to be optimized according to the roughing rate specifically includes: Extracting, in the first motion path, path segments whose mapping in the shoe last compensation direction is not zero as segments to be optimized; All the segments to be optimized are determined by calculation to obtain the paths to be optimized.

6. The method according to claim 1, characterized in that After the step of generating a three-dimensional mesh model according to the collected point cloud data of the target shoe upper, the method further includes: Determining a mapping relationship between a rotation of the shoe last and a movement of the target shoe upper according to the three-dimensional grid model; According to the mapping relationship, a conversion formula between the upper motion curve and the last motion path is determined.

7. The method according to claim 1, characterized in that After the step of controlling the robot arm to move along the second motion path and controlling the shoe last to move along the shoe last motion path to roughen the shoe upper, the method further includes: The point cloud data of the target shoe upper after roughening is collected again to obtain a roughening effect model; Generating a roughening quality score of the upper roughening according to the roughening effect model; When the roughing quality score is lower than a preset qualified threshold, a prompt message is sent to the user terminal.

8. A shoe upper roughening system, characterized in that: The shoe upper roughening system includes: one or more processors and a memory; the memory is coupled to the one or more processors, the memory is used to store computer program code, the computer program code includes computer instructions, and the one or more processors call the computer instructions to enable the shoe upper roughening system to execute the method described in any one of claims 1-7.

9. A computer-readable storage medium comprising instructions, characterized in that: When the instructions are executed on a shoe upper roughening system, the shoe upper roughening system is caused to execute the method according to any one of claims 1 to 7.

10. A computer program product, characterized in that When the computer program product runs on a shoe upper roughening system, the shoe upper roughening system is enabled to perform the method according to any one of claims 1-7.

Images