Intelligent Fixture Path Planning Method for Industrial Robots

Abstract

The present invention discloses an intelligent fixture path planning method for industrial robots, which relates to the technical field of fixture path planning. The method includes: identifying the characteristics of the working environment space of the intelligent fixture to obtain spatial partitions; establishing spatial constraint features according to the corresponding relationship between the spatial partitions and the spatial identification features; obtaining control constraint information according to the working parameters of the intelligent fixture; performing constraint matching analysis to obtain a control parameter strategy space; decomposing the task process of the industrial robot fixture to obtain a task environment space sequence; and performing local optimization of the spatial partition strategy to obtain a path planning result. The present invention solves the technical problems existing in the prior art that the intelligent fixture lacks accuracy and efficiency in path planning in a complex industrial environment and has a collision risk, and achieves the technical effects of improving the accuracy, efficiency and safety of the intelligent fixture in path planning in a complex industrial environment.

Description

Technical Field

[0001] The present invention relates to the technical field of fixture path planning, and particularly to an intelligent fixture path planning method for industrial robots. Background Art

[0002] In the field of modern industrial automation production, industrial robots play a crucial role. As a key component for industrial robots to perform tasks such as grasping, handling, and assembling, the rationality and efficiency of the path planning of intelligent fixtures directly affect the performance of the entire production system. With the continuous development of industrial production, production scenarios are becoming increasingly complex and diverse, and the requirements for production accuracy, efficiency, and safety are also continuously rising. Traditional fixture path planning often relies on simple preset trajectories or relatively rough spatial perception and obstacle avoidance strategies. In the face of complex working environments, such as the presence of numerous static devices with various shapes, other robots or material transfer devices operating dynamically, this traditional approach exposes many drawbacks. On the one hand, it is difficult to accurately identify various obstacles in the working environment and accurately evaluate their impact on the fixture path, which easily leads to collision accidents, not only damaging equipment and workpieces but also possibly causing production interruptions and huge economic losses. On the other hand, it is unable to flexibly adjust the path planning strategy according to the working parameters of the fixture itself and task requirements, resulting in low production efficiency and unable to fully utilize the performance advantages of industrial robots and intelligent fixtures.

[0003] There are technical problems in the prior art that intelligent fixtures lack accuracy and efficiency in path planning in complex industrial environments and there is a risk of collision. Summary of the Invention

[0004] This application provides an intelligent fixture path planning method for industrial robots, which is used to solve the technical problems that intelligent fixtures lack accuracy and efficiency in path planning in complex industrial environments and there is a risk of collision in the prior art.

[0005] In view of the above problems, this application provides an intelligent fixture path planning method for industrial robots, and the method includes:

[0006] Based on the working parameters of the intelligent fixture, perform feature recognition on the working environment space of the intelligent fixture to obtain spatial partitions, where the spatial partitions are clustered and segmented according to spatial recognition features; establish spatial constraint features according to the corresponding relationship between the spatial partitions and the spatial recognition features; obtain control constraint information according to the working parameters of the intelligent fixture; perform constraint matching and parsing according to the spatial constraint features and the control constraint information to obtain a control parameter strategy space, where the control parameter strategies in the control parameter strategy space have working space position identifiers; decompose the task process of the industrial robot fixture task to obtain a task environment space sequence; perform spatial matching according to the task environment space sequence and the working space position identifiers of the control parameter strategy space, perform local optimization of the spatial partition strategy using the control parameter strategy space, and use the local optimization results to optimize the full-task path to obtain a path planning result.

[0007] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0008] Based on the working parameters of the intelligent fixture, perform feature recognition on the working environment space of the intelligent fixture to obtain spatial partitions; establish spatial constraint features according to the corresponding relationship between the spatial partitions and the spatial recognition features; obtain control constraint information according to the working parameters of the intelligent fixture; perform constraint matching and parsing to obtain a control parameter strategy space; decompose the task process of the industrial robot fixture task to obtain a task environment space sequence; perform spatial matching according to the task environment space sequence and the working space position identifiers of the control parameter strategy space, perform local optimization of the spatial partition strategy using the control parameter strategy space, and use the local optimization results to optimize the full-task path to obtain a path planning result. It achieves the technical effects of improving the accuracy, efficiency, and safety of path planning of the intelligent fixture in a complex industrial environment. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0010] Figure 1 It is a schematic flowchart of the intelligent fixture path planning method for an industrial robot provided by an embodiment of this application.

[0011] Figure 2 It is a schematic flowchart of the process of obtaining spatial partitions of the intelligent fixture path planning method for an industrial robot provided by an embodiment of this application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0012] This application provides an intelligent fixture path planning method for industrial robots, aiming to solve the technical problems in the prior art that the path planning of intelligent fixtures in complex industrial environments lacks accuracy and efficiency and has a collision risk.

[0013] Next, the technical solutions in the embodiments of this application will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of this application. Obviously, the described embodiments are only a part of the embodiments of this application, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in this application without creative efforts belong to the scope of protection of this application.

[0014] Embodiment, as Figure 1 shown, this application provides an intelligent fixture path planning method for industrial robots, and the method includes:

[0015] Step S100: Based on the working parameters of the intelligent fixture, identify the characteristics of the working environment space of the intelligent fixture to obtain a space partition, and the space partition is cluster-segmented according to the space recognition characteristics.

[0016] Specifically, based on the working parameters of the intelligent fixture, such as the clamping force magnitude, opening and closing range, movement speed, etc., comprehensively analyze its working environment space. Taking an automotive parts assembly workshop as an example, if the clamping force of the intelligent fixture is small, when identifying the working environment space, key attention should be paid to the storage area and operation space of lightweight parts; if the movement speed is slow, the space layout and obstacle distribution within a short distance need to be considered. Through these working parameters, static avoidance targets (such as fixed equipment, shelves, etc.) and dynamic avoidance targets (such as moving transport trolleys, other robots that are operating, etc.) in the working environment are identified, and their locations in the working space are determined, and then the static avoidance space characteristics and dynamic avoidance space characteristics are obtained. Then, use the minimum working space of the intelligent fixture to perform passage fitting on these characteristics to obtain passage influence characteristics, such as determining whether the fixture can pass smoothly in a narrow passage area; use the maximum working space for passage fitting to obtain operation influence characteristics, such as judging whether the movement of the fixture is restricted in an open area. Finally, perform discrete analysis of space characteristics according to the passage influence characteristics and operation influence characteristics, calculate the discrete degree of feature distribution and perform space clustering accordingly, and divide the space according to the clustering result to obtain a space partition, such as dividing the parts assembly area, material transportation area, etc. into different partitions.

[0017] Step S200: Establish space constraint characteristics according to the correspondence between the space partition and the space recognition characteristics.

[0018] Specifically, a spatial constraint feature is established based on the correspondence between the spatial partition and the spatial recognition feature. For each spatial partition, analyze how its spatial recognition feature affects the movement of the intelligent fixture. For example, in the narrow parts assembly area, the feature of narrow space corresponds to the spatial constraint feature that restricts the large-scale movement of the intelligent fixture; in the material transportation area, the dynamic feature of frequent movement of personnel and vehicles corresponds to the spatial constraint feature that the intelligent fixture needs to avoid at any time.

[0019] Step S300: Obtain control constraint information according to the working parameters of the intelligent fixture.

[0020] Specifically, obtain control constraint information according to the working parameters of the intelligent fixture. First, determine the control nodes of the intelligent fixture, such as the moving arm control node and the clamping mechanism control node, and then perform multi-node control parameter combinations according to their adjustment parameter ranges, and analyze the morphological changes of the fixture under different combinations, so as to obtain the edge space constraint, such as the space range required when the fixture opens to the maximum. At the same time, determine the passage adjustment constraint according to the adjustment parameter range of the moving arm control node, such as the limit of the turning radius at different speeds. Integrate these factors to obtain the control constraint information.

[0021] Step S400: Perform constraint matching and parsing according to the spatial constraint feature and the control constraint information to obtain a control parameter strategy space, and the control parameter strategies in the control parameter strategy space have working space position identifiers.

[0022] Specifically, perform constraint matching and parsing based on the spatial constraint feature and the control constraint information. Determine the constraint distribution space according to the spatial constraint feature, and clarify the fixture control operation constraint space according to the control constraint information. Configure the minimum safety distance for different types of avoidance space features (static and dynamic). For example, for static equipment, determine the safety distance according to its edge shape and the response parameters of the intelligent fixture; for dynamic targets, predict the position by obtaining their operation parameters and path features, and then determine the safety distance. Then search for the feasible working parameters of the intelligent fixture in the safe operation space, construct the control parameter strategy space, and each strategy is marked with a working space position identifier for subsequent invocation.

[0023] Step S500: Decompose the task process of the industrial robot fixture to obtain a task environment space sequence.

[0024] Specifically, conduct a comprehensive and detailed breakdown of the task process for industrial robot fixtures. For example, during the assembly process of an automotive engine, the entire task can be decomposed into multiple subtasks, including grasping the engine block from the parts warehouse, transporting the block to a specific station on the assembly production line, installing internal parts such as pistons at this station, and transporting the assembled engine to the inspection area after completion. For each subtask, determine its corresponding working environment space. The subtask of grasping the engine block corresponds to the parts warehouse space, which is characterized by densely arranged parts and numerous shelves; the subtask of transporting to the assembly production line station involves the transportation passage space inside the factory, which may have limited passage width, the presence of other transportation equipment, and personnel flow; when installing internal parts at the assembly station, it corresponds to the assembly work area space, where there are various assembly tools, equipment, and strict operating space requirements; while the subtask of transporting to the inspection area corresponds to the inspection area space, which is equipped with various inspection instruments and has special regulations on the movement accuracy and operating space of the fixture. Through such a breakdown of the task process, each subtask is corresponded to its corresponding working environment space one by one, thereby obtaining a sequence of task environment spaces. This sequence accurately describes the order of different working environment spaces experienced by the industrial robot fixture during the entire task process, laying a solid foundation for subsequent spatial matching according to the sequence of task environment spaces and the control parameter strategy space, and realizing precise path planning, ensuring that the fixture can efficiently and safely complete operations in the appropriate environment space at different task stages, and improving the efficiency and quality of the entire industrial production process.

[0025] Step S600: According to the sequence of task environment spaces, perform spatial matching with the working space position identifier of the control parameter strategy space, use the control parameter strategy space for local optimization of the spatial partitioning strategy, and use the local optimization result for full-task path optimization to obtain a path planning result.

[0026] Specifically, in strict accordance with the task environment space sequence, the working environment space corresponding to each sub-task is carefully matched with the working space position identifiers in the control parameter strategy space one by one. For example, when the task proceeds to the fine assembly operation in the component assembly area, the spatial characteristics of this assembly area are identified from the task environment space sequence, and then the control parameter strategy with the position identifier of the assembly area is accurately found in the control parameter strategy space. The matching control parameter strategy space is used to conduct local optimization of the spatial partition strategy for the current spatial partition. In this specific spatial partition of the assembly area, according to the corresponding control parameter strategies, such as optimizing the motion speed curve of the fixture to meet the requirements of speed stability for precision assembly, adjusting the attitude control parameters of the fixture to ensure the accuracy of part grasping and installation, screening and optimizing the possible motion paths, calculating various performance indicators under different paths, such as path length, motion time, collision risk, etc., and finding the optimal local path strategy in this partition. Subsequently, the local optimization results of each spatial partition are used for the overall task path optimization. Considering the smooth connection between each sub-task and the goal of efficiently completing the overall task, the optimal local path strategies of different partitions are spliced and adjusted. For example, during the transportation process from the component grasping area to the assembly area, it is necessary to ensure that the transportation path is perfectly connected to the exit of the grasping area and the entrance of the assembly area, and at the same time, combined with the local optimization results of the two areas, parameters such as transportation speed and steering timing are further optimized. On the entire task path, the control parameter strategies of different spatial partitions are coordinated as a whole, and the total evaluation value of the overall task path is calculated. This value comprehensively considers various factors such as path length, motion time, collision risk, and energy consumption. Finally, after multiple iterations of optimization and evaluation, the splicing parameter strategy with the largest corrected total evaluation value is determined and used as the final path planning result. This result ensures that the industrial robot fixture can complete various operations with the optimal path, the highest efficiency, and the lowest risk throughout the task process, effectively improving the automation level and production efficiency of industrial production, reducing production costs, and improving product quality and production safety.

[0027] In a possible implementation manner, as Figure 2 shown, step S100 further includes:

[0028] Step S110: According to the working parameters of the intelligent fixture, establish the relationship between the working parameters and the volume state of the fixture body, and obtain the maximum working space and the minimum working space.

[0029] Step S120: Identify the static avoidance targets and dynamic avoidance targets in the working environment space, and determine the static avoidance space characteristics and dynamic avoidance space characteristics based on the working space where the static avoidance targets and dynamic avoidance targets are located.

[0030] Step S130: Use the minimum working space to perform a passage fitting on the static avoidance space feature and the dynamic avoidance space feature to obtain a passage influence feature.

[0031] Step S140: Use the maximum working space to perform a passage fitting on the static avoidance space feature and the dynamic avoidance space feature to obtain an operation influence feature.

[0032] Step S150: Perform a discrete analysis of the space feature according to the passage influence feature and the operation influence feature, perform a space clustering according to the feature distribution dispersion degree, and perform a space segmentation according to the space clustering result to obtain the space partition.

[0033] Specifically, based on the working parameters of the intelligent fixture, such as the maximum extension length, the maximum clamping width, the rotation angle range, etc., deeply analyze their relationship with the fixture volume state. For example, when the maximum extension length of the intelligent fixture is longer, the range it can reach in the working space is wider, and the corresponding maximum working space volume increases; when the clamping width is larger, the operation space requirements in certain directions will also increase. Through the precise calculation of these parameter relationships, determine the maximum working space and the minimum working space of the intelligent fixture in different states. The maximum working space represents the space range occupied by the fixture in the limit operation states such as fully extended and rotated to the maximum angle, and the minimum working space is the space requirement of the fixture in the minimum operation states such as contraction and homing.

[0034] Perform a comprehensive scan of the working environment space to accurately identify the static avoidance targets and dynamic avoidance targets therein. For example: in a machining workshop, the static avoidance targets may include fixed machine tools, large equipment bases, warehouse shelves, etc., which remain fixed in position during the working process; the dynamic avoidance targets are the running transport carts, operators, other collaborative robots, etc., whose positions and motion states change at any time. After determining the working space where these targets are located, further analyze to obtain the static avoidance space feature and the dynamic avoidance space feature. The static avoidance space feature mainly involves the restrictions of the shape, size, position relationship, etc. of the target on the motion path of the intelligent fixture. For example, the guardrails around the machine tool and the protruding parts of the equipment base will form specific space blocks; the dynamic avoidance space feature needs to consider factors such as the motion trajectory, speed range, motion period, etc. of the target, such as the driving route of the transport cart, the activity range of the operator and the habitual action path.

[0035] Use the previously obtained minimum working space to perform a passage fitting on the static avoidance space features and dynamic avoidance space features. Assume that the minimum working space is a cuboid with a specific shape and volume, and move it along different paths in the working environment space to simulate the situation where the intelligent fixture passes through the space around the static and dynamic avoidance targets in the minimum operating state. Through this fitting process, analyze the impacts on the fixture when passing around the avoidance target under the minimum working space, such as whether there will be collisions due to narrow space, whether it can smoothly bypass the target, etc., so as to obtain the passage impact features. These features include key information such as the feasible passage directions near the avoidance target, the minimum turning radius, and the minimum safety clearance of the minimum working space.

[0036] Adopt a similar method to perform a passage fitting on the static avoidance space features and dynamic avoidance space features using the maximum working space to obtain the operation impact features. When performing the passage fitting for the maximum working space, consider the motion of the intelligent fixture in the extreme operating state. For example, when the fixture extends to the longest length and rotates to the maximum angle, the operation trajectory and space requirements in the space around the avoidance target. The operation impact features may involve the maximum motion speed limit, the maximum operation range limit, and the maximum occupancy of the surrounding space in different avoidance target areas of the maximum working space. These features reflect the operation capabilities and space requirement characteristics of the intelligent fixture in different avoidance target environments during the normal working process.

[0037] Conduct a discrete analysis of the space features according to the passage impact features and operation impact features. Divide the working environment space into multiple tiny units, and calculate the distribution of the passage impact features and operation impact features in each unit, such as the distribution frequency of the minimum safety clearance in a certain area and the change trend of the maximum motion speed limit. Perform space clustering according to the discreteness of these feature distributions, and cluster the space units with similar feature distributions into one category. For example, classify the areas with a relatively small and dense distribution of the minimum safety clearance into one category, indicating that the passage of the intelligent fixture in this area is more restricted. Perform space segmentation according to the space clustering results, and divide the working environment space into different space partitions. Each space partition has relatively unified passage and operation impact features, laying a solid foundation for subsequent establishment of space constraint features, formulation of targeted control strategies, and realization of efficient path planning for the intelligent fixture.

[0038] In a possible implementation manner, step S140 further includes:

[0039] Step S141: Obtain the multi-level operation state volumes according to the relationship between the working parameters and the fixture volume state.

[0040] Step S142: Use the multi-level operation state volumes to perform passage fittings on the static avoidance space features and dynamic avoidance space features respectively to obtain multi-level operation limit features.

[0041] Step S143: Integrate the multi - level operation limit features to obtain the operation impact features.

[0042] Specifically, first, carefully study the working parameters of the intelligent fixture. These parameters cover multiple key aspects of the fixture, such as the length range of the telescopic arm, the opening and closing degree of the jaws, the angle range of the rotating joints, and the dimensional specifications of the overall structure, etc. By establishing an accurate mathematical relationship model between these working parameters and the volume state of the fixture body, the multi - level operation state volumes are determined. For example, for an intelligent fixture with a telescopic function, the extended length of the telescopic arm is an important working parameter. When the telescopic arm is in the fully retracted state, the overall volume of the fixture is small, corresponding to the volume of the first - level operation state, which is mainly determined by the space occupied by the basic structure of the fixture and the jaws in the minimum gripping state. As the telescopic arm gradually extends, every time it reaches a specific length value, the overall shape and space occupancy of the fixture change, resulting in different volume states. Suppose the telescopic arm has three adjustable extension gears, each gear corresponding to a different length value. Then when the telescopic arm extends to the length of the first gear, combined with the opening and closing degree of the jaws at this time and the angle of the rotating joint (these parameters are also within the range of working parameters and are correlated with the fixture volume), the volume of the second - level operation state is calculated. Similarly, when the telescopic arm extends to the lengths of the second and third gears, the corresponding volumes of the third - level and fourth - level operation states are calculated respectively. In addition, the opening and closing degree of the jaws also affects the volume. When the jaws open to the maximum extent to grip a workpiece with a larger size, even if the length of the telescopic arm remains unchanged, the overall volume of the fixture will increase, which also constitutes part of the volume of different levels of operation states. Through a comprehensive analysis and accurate calculation of various combinations and change situations of these working parameters, the multi - level operation state volumes that can accurately reflect the intelligent fixture in different operation modes and postures are finally obtained, laying a foundation for further in - depth analysis of the passing and operating characteristics of the fixture in the working environment.

[0043] Based on the multi-level operating state volumes obtained previously, deeply explore the operating limitations of the intelligent fixture in a complex working environment. Analyze the characteristics of the static avoidance space. In a workshop scenario filled with large fixed equipment (such as machine tools, shelves, etc.), fit the operating state volumes at each level with the spaces where these static avoidance targets are located one by one. For example, when considering the first-level operating state volume, virtually place it in the space around the machine tool, and simulate whether the fixture can smoothly bypass the protruding parts such as the corners and brackets of the machine tool and whether it can pass through the narrow passages between the machine tools in this volume state. At the same time, analyze the minimum safety distance that needs to be maintained between the fixture and the machine tool during this process. This distance is the partial operating limitation characteristic of the first-level operating state volume for this static avoidance space characteristic. For the operating state volumes of the second level and above, repeat the above-mentioned passage fitting process. As the operating state volume increases, such as when the telescopic arm extends or the gripper opens and the volume becomes larger, observe the changes in its passage situation in the space around the machine tool. It may be found that when the volume increases to a certain extent, the originally passable narrow passage becomes impassable, or a larger turning radius is required when bypassing the machine tool. These newly added limiting factors constitute the operating limitation characteristics under the static avoidance space characteristic corresponding to this level of operating state volume. Subsequently, perform similar operations for the dynamic avoidance space characteristic. In a working area with dynamic targets such as transport carts and mobile robots, fit the operating state volumes at each level with the movement spaces of these dynamic targets. Consider whether the fixture at each level of operating state volume can avoid in time when the transport cart is moving at different speeds and along different routes, and the lead and buffer spaces that need to be reserved to avoid collisions. For example, for the first-level operating state volume, when the transport cart is moving in a straight line at a low speed, the fixture may only need to maintain a small lateral avoidance distance, but when the cart makes a high-speed turn, the fixture requires a larger avoidance space. This is the partial operating limitation characteristic of the first-level operating state volume in this dynamic avoidance scenario. Similarly, for different levels of operating state volumes, as their volumes change, the operating limitation characteristics in the dynamic avoidance space also change. For example, after the volume increases, the reaction time of the fixture may become longer, and it needs to sense the movement of the dynamic target earlier and make avoidance actions, resulting in changes in its operable space and safety range in the dynamic avoidance space. Through such a comprehensive and detailed passage fitting process, for the static and dynamic avoidance space characteristics, respectively obtain the multi-level operating limitation characteristics corresponding to the operating state volumes at each level. These characteristics accurately describe the restricted conditions of the intelligent fixture in a complex working environment under different operating states, providing a key basis for subsequently integrating the complete operating influence characteristics.

[0044] Summarize and sort out the multi-level operation limit features obtained for the static avoidance space features and dynamic avoidance space features respectively before. For example, in terms of static avoidance, the limit features such as the minimum safety distance required for the fixture to bypass fixed equipment and the maximum allowable approach angle under different levels of operating state volumes are listed one by one; in terms of dynamic avoidance, the limit features such as the shortest reaction time, minimum avoidance speed, and maximum allowable collision risk coefficient when the fixture avoids moving targets corresponding to different operating state volumes are also presented centrally. Deeply analyze the internal connections and mutual influences among these multi-level operation limit features. For example, a larger safety distance requirement for the fixture during static avoidance at a certain level of operating state volume may lead to an extended reaction time during dynamic avoidance because a larger safety distance means the fixture needs to sense the movement changes of dynamic targets earlier. By establishing such a correlation model, the scattered multi-level operation limit features are organically combined. Then, remove duplicate and redundant information and optimize and integrate the mutually related limit features. For example, if the difference in the minimum avoidance speed of two adjacent operating state volumes during dynamic avoidance is extremely small, reasonable merging or taking the average value can be carried out according to the actual situation to simplify the feature description. Finally, after comprehensive integration processing, a comprehensive, complete, and accurate operation impact feature is formed. This operation impact feature covers key information such as the operation ability, limit conditions, and space requirements of the intelligent fixture in different operating states in the entire working environment space (including static and dynamic avoidance target areas), providing a solid and accurate basis for the subsequent path planning and control strategy formulation of the intelligent fixture to ensure that the fixture can operate efficiently and safely in a complex working environment.

[0045] In a possible implementation manner, step S150 further includes:

[0046] Step S151: According to the working environment space, perform position distribution annotation on the passage impact feature to obtain a passage impact distribution map.

[0047] Step S152: According to the position distribution of the operation impact feature, perform distribution annotation in the working environment space to obtain an operation impact distribution map.

[0048] Step S153: Fuse and overlay the passage impact distribution map and the operation impact distribution map, and calculate the dispersion according to the distribution annotation to obtain the feature distribution dispersion, which is used to characterize the discrete proportion of the passage impact feature and the operation impact feature in the working space.

[0049] Step S154: Perform spatial feature clustering according to the feature distribution dispersion.

[0050] Specifically, according to the actual layout and dimensions of the working environment space, detailed position distribution markings are made for the passage influence characteristics. For example, in a complex industrial workshop, the working environment space includes various equipment, passages, and operation areas. For passage influence characteristics, such as in narrow passage areas, since the passage width limits the minimum working space through which the intelligent fixture can pass, this area is marked as an area with a greater passage influence, and the marking information includes the minimum safety clearance, the feasible passage direction, etc.; while in open areas, the passage influence is relatively small, and the corresponding characteristic values are marked. By making such one-by-one markings for the entire working environment space, a passage influence distribution map is finally obtained, which visually shows the distribution of passage influence characteristics in the working environment space, enabling the staff to clearly see which areas have greater passage restrictions on the intelligent fixture and which areas are relatively more relaxed.

[0051] Similar distribution markings are made in the working environment space according to the position distribution of the operation influence characteristics. Considering the operation influence characteristics of the intelligent fixture in different operating states, for example, around large equipment, since the presence of the equipment limits the maximum operating space of the fixture, this area is marked as an area with a significant operation influence, and the marking content covers the maximum movement speed limit, the maximum operation range limit, etc.; in open areas far from the equipment, the operation influence is relatively weak, and the corresponding characteristic values are marked. After such a marking process, an operation influence distribution map is obtained, which clearly presents the distribution state of the operation influence characteristics in the working environment space, helping to comprehensively understand the operation ability and limiting conditions of the intelligent fixture at different positions.

[0052] Perform a fusion overlay operation on the obtained traffic impact distribution map and the operation impact distribution map. In this process, the two maps are integrated under the same coordinate system, so that each position point in the working environment space has the annotation information of both traffic impact characteristics and operation impact characteristics. For example, in a certain workshop area, the traffic impact distribution map indicates that the passage here is narrow and the minimum working space for passage is restricted, while the operation impact distribution map shows that this area is close to large equipment and the maximum working space for operation is restricted. Calculate the dispersion of the distribution annotation after fusion overlay. The dispersion calculation is realized through a specific mathematical algorithm, which considers the degree of difference between the traffic impact characteristics and operation impact characteristics at each position point and the distribution of these characteristics in the entire working space. During the calculation process, analyze the change frequency and amplitude of the annotation information in each area. For example, in an area, whether the transition of the traffic impact characteristics from a large restriction to a small restriction is frequent and whether it is synchronized with the change of the operation impact characteristics. Through the comprehensive evaluation of these factors, a numerical feature distribution dispersion is obtained. This feature distribution dispersion can accurately characterize the discrete proportion of the traffic impact characteristics and operation impact characteristics in the working space. If the dispersion is high, it means that these characteristics are relatively dispersed in the working space and the feature differences between different regions are obvious. For example, in a workshop, there are both open areas with relatively small traffic and operation impacts and complex equipment areas with relatively large impacts in both aspects, and the feature changes in the transition areas are drastic; on the contrary, if the dispersion is low, it indicates that the feature distribution is relatively concentrated and the traffic and operation impact characteristics in most areas are relatively similar, and perhaps the environmental restrictions on the intelligent fixture in the entire workshop are relatively unified. The feature distribution dispersion provides an important basis for subsequent spatial feature clustering, which helps to more scientifically divide the working environment space to achieve the optimal control and efficient operation of the intelligent fixture.

[0053] Perform spatial feature clustering based on the calculated feature distribution dispersion. According to the size and distribution of the dispersion, divide the working environment space into different clustering regions. In the case of a high dispersion, multiple clusters with obvious differences are formed, and the traffic impact characteristics and operation impact characteristics within each cluster are relatively similar; while in the case of a low dispersion, the number of clusters may be relatively small and the feature consistency of the spatial regions is relatively high. Through such spatial feature clustering, the complex working environment space is effectively classified according to the traffic impact and operation impact characteristics, providing an important basis for subsequent spatial segmentation and precise control of the intelligent fixture, enabling the intelligent fixture to adopt appropriate operation strategies according to the characteristics of different clustering regions, improving work efficiency and safety.

[0054] In a possible implementation manner, step S300 further includes:

[0055] Step S310: Obtain the control nodes of the intelligent fixture, where the control nodes at least include a moving arm control node and a clamping mechanism control node.

[0056] Step S320: According to the adjustment parameter ranges of the control nodes, perform multi-node control parameter combinations to obtain the fixture shape changes of each control parameter combination.

[0057] Step S330: Obtain the edge space constraints according to the fixture shape changes.

[0058] Step S340: Determine the passage adjustment constraints according to the adjustment parameter range of the moving arm control node.

[0059] Step S350: Obtain the control constraint information according to the edge space constraints and the passage adjustment constraints.

[0060] Specifically, first, it is necessary to deeply study the overall structure and functional mechanism of the intelligent fixture to accurately locate its control nodes. As a key component for an industrial robot to perform grasping and operating tasks, the control system of the intelligent fixture is relatively complex. Among them, the moving arm control node is the key to realizing the precise positioning and flexible movement of the fixture in the three-dimensional working space. It receives control instructions and precisely adjusts parameters such as the telescopic length, rotation angle, and translation speed of the moving arm, so as to ensure that the fixture can accurately reach the target position. For example, on an automotive parts assembly line, when it is necessary to grasp a part at a specific position, the moving arm control node precisely controls the moving arm to extend to an appropriate length, rotate to an accurate angle, and move at an appropriate speed according to a preset program or real-time instructions, so that the fixture can accurately approach the part. The clamping mechanism control node focuses on controlling the actions of the jaws and is the core element for realizing reliable grasping of workpieces of various shapes and sizes. It can flexibly adjust parameters such as the opening and closing degree of the jaws, the clamping force, and the opening and closing speed according to the characteristics of the workpiece, such as shape, size, and material. For example, when grasping a fragile electronic component, the clamping mechanism control node will control the jaws to gently grasp the component with a smaller clamping force and a slower opening and closing speed to prevent it from being damaged; while when grasping a heavy mechanical part, the clamping force will be increased to ensure the stable grasping of the part. Through a comprehensive analysis of the structure and function of the intelligent fixture, these key control nodes, which at least include the moving arm control node and the clamping mechanism control node, are successfully identified and obtained, laying a solid foundation for subsequent parameter analysis based on the control nodes and obtaining control constraint information, so as to realize the efficient and precise operation of the intelligent fixture in a complex industrial environment.

[0061] Define the adjustment parameter ranges for each control node, which is the basis for parameter combination. For the mobile arm control node, its adjustment parameter ranges cover key parameters such as telescopic length (e.g., adjustable between 0.5 meters and 1.5 meters), rotation angle (e.g., can rotate within the range of -180° to 180°), and moving speed (may vary between 0.1 m / s and 1 m / s); the adjustment parameter ranges for the clamping mechanism control node include the opening and closing degree of the jaws (from fully closed to the maximum opening distance, such as 0 cm to 10 cm), the clamping force magnitude (can be set between 10 N and 100 N according to different workpiece requirements), and the opening and closing speed (e.g., complete an opening and closing action in 0.05 s to 0.5 s). Based on these parameter ranges, conduct a comprehensive and systematic multi-node control parameter combination. Taking the telescopic length, rotation angle, and moving speed of the mobile arm control node, and the opening and closing degree of the jaws, clamping force magnitude, and opening and closing speed of the clamping mechanism control node as examples, if several typical values are selected from each parameter range for combination, numerous different parameter combination situations will be generated. For example, when the telescopic length of the mobile arm is 0.8 meters, the rotation angle is 45°, the moving speed is 0.3 m / s, and at the same time, the opening and closing degree of the jaws of the clamping mechanism is 5 cm, the clamping force is 30 N, and the opening and closing speed is 0.1 s, a specific control parameter combination is formed. For each control parameter combination, deeply analyze its impact on the fixture form to obtain the fixture form changes under each combination. Under the above parameter combination, the mobile arm of the fixture is in a specific extended position and angle, the jaws present corresponding opening and closing states, and the overall fixture forms a unique form in space. This form not only determines the occupied range and posture of the fixture itself in the working space but also is directly related to its interaction relationship with the surrounding environment. Through the analysis of a large number of different control parameter combinations one by one, a detailed mapping relationship between the control parameters and the fixture form changes is constructed, providing a key basis for accurately evaluating the edge space constraints of the fixture under different working conditions in the future, and further promoting the optimization process of intelligent fixture path planning.

[0062] Determine the edge space constraint based on the change in the fixture form. When the fixture form changes, the actual space range it occupies in the working space and its relative position relationship with the surrounding environment also change accordingly. For example, as the gripper opens or closes, the overall outer contour size of the fixture will change correspondingly. If the gripper opens to the maximum extent to grasp a large workpiece, the extension range of the fixture in the horizontal and vertical directions increases at this time, so sufficient space needs to be reserved around it to avoid collisions with surrounding equipment, workpieces, or other obstacles. By carefully analyzing various possible changes in the fixture form, calculate the minimum safety distance that must be maintained between the fixture edge and the surrounding environment in different forms, so as to determine the edge space constraint. This constraint ensures that during the movement and operation of the fixture, its edge is always within a safe space range, preventing collision accidents caused by insufficient space.

[0063] Determine the passage adjustment constraint according to the adjustment parameter range of the mobile arm control node. The adjustment parameter range of the mobile arm, including the telescopic length, rotation angle, and moving speed, etc., directly affects the passage ability and flexibility of the fixture in the working space. For example, when the mobile arm extends to a longer length, its passage in a narrow channel may be restricted, and at this time, a larger channel width or more flexible turning space is required; if the moving speed is faster, a larger turning radius is required when turning to ensure that the fixture can pass smoothly and safely. By deeply studying the influence of each parameter of the mobile arm control node on passage under different value conditions, determine the passage adjustment constraint of the fixture in the working space under different parameter combinations, such as the minimum turning radius, channel width requirements, passage speed limits, etc. These constraint conditions provide clear guidance for the safe passage of the fixture in a complex working environment.

[0064] Integrate the edge space constraint and the passage adjustment constraint to obtain the control constraint information. The edge space constraint focuses on the space limitation brought by the change in the fixture itself form, and the passage adjustment constraint focuses on the influence of the mobile arm control node parameters on passage. The two complement each other. Integrate these constraint information together to form a comprehensive and systematic set of control constraint information. This information accurately reflects the various space limitations suffered by the intelligent fixture during the actual working process due to its own structure and motion control characteristics. In subsequent path planning and operation control, based on these control constraint information, the intelligent fixture can avoid obstacles more intelligently and efficiently, select the optimal motion path, ensure safe and stable completion of grasping and operation tasks in a complex industrial environment, and improve the automation level and work efficiency of industrial production.

[0065] In a possible implementation manner, step S400 further includes:

[0066] Step S410: Obtain the constraint distribution space according to the space constraint characteristics.

[0067] Step S420: Obtain the fixture control operation constraint space according to the control constraint information.

[0068] Step S430: Configure the minimum safety distance according to the spatial constraint features.

[0069] Step S440: Based on the minimum safety distance, use the fixture control operation constraint space to identify the operation safety space in the constraint distribution space, and obtain the safe operation space.

[0070] Step S450: Search for the working parameters of the intelligent fixture in the safe operation space, obtain all feasible working parameters, and construct the control parameter strategy space.

[0071] Specifically, obtain the constraint distribution space based on the spatial constraint features. The spatial constraint features stem from an in-depth analysis of the working environment space of the intelligent fixture, including the spatial restrictions formed by static avoidance targets (such as fixed equipment, walls, etc.) and dynamic avoidance targets (such as moving robots, transport vehicles, etc.) in the working environment. By accurately describing these constraint features, such as the position, shape, and movement trajectory of obstacles, the working space is divided into regions with different degrees of restriction, thereby determining the constraint distribution space. For example, in a machining workshop, due to the fixed position and large volume of large machine tools, restricted areas for the movement range of the intelligent fixture are formed around them, and these areas together constitute a part of the constraint distribution space.

[0072] Obtain the fixture control operation constraint space according to the control constraint information. The control constraint information is determined based on the control nodes of the intelligent fixture itself (such as the moving arm control node, the clamping mechanism control node) and their adjustment parameter ranges. Considering the morphological changes of the fixture under different control parameter combinations and the resulting edge space constraints and passage adjustment constraints, the feasible regions and limiting conditions of the fixture during movement and operation are determined. For example, when the moving arm extends to a certain length and the gripper is in a specific opening and closing state, the movable range of the fixture in the working space is restricted, and the space formed by this restricted range is the fixture control operation constraint space.

[0073] Configure the minimum safety distance according to the spatial constraint features. For static avoidance space features, such as sharp corners or protruding parts of fixed equipment, combine the running avoidance response parameters of the intelligent fixture (such as the detection accuracy of the sensor, the braking ability of the fixture, etc.) to determine a suitable minimum safety distance to prevent the fixture from colliding with these static targets during movement. For dynamic avoidance space features, such as moving robots or transport vehicles, obtain their running parameters (such as speed, movement direction, movement cycle, etc.) and analyze their running path features (such as periodic round-trip paths or random dynamic paths). For periodic paths, determine the periodic avoidance positions according to their periodic laws and combine the running avoidance response parameters of the intelligent fixture to determine the minimum safety distance; for dynamic paths, connect to the control interface of the dynamic avoidance target, obtain real-time control parameters for position prediction, and then determine the minimum safety distance according to the prediction results and the running avoidance response parameters.

[0074] Based on the configured minimum safety distance, use the fixture control running constraint space to identify the running safety space in the constraint distribution space. Combine the fixture control running constraint space with the constraint distribution space, use the minimum safety distance as a buffer, and screen out the areas that meet the fixture's own control requirements and can avoid various constraints in the working space. These areas are the safe running spaces. For example, in a workshop environment with both large equipment and moving vehicles, the safe running space is the area where the fixture can operate safely and effectively after considering the safety distance around the equipment and the predicted safety range of the vehicle's movement trajectory.

[0075] After determining the safe operating space, a comprehensive search for the working parameters of the intelligent fixture is carried out within this space. It is clear that the working parameters of the intelligent fixture cover multiple key aspects, such as the telescopic speed of the moving arm, the rotational angular velocity, the telescopic length range, as well as the gripping force of the clamping mechanism, the opening and closing speed of the jaws, and the opening and closing degree. In the safe operating space, different values of these parameters are tried one by one. For example, for the telescopic speed of the moving arm, starting from the minimum allowed value, the value is gradually increased in a certain step size, and combined with different rotational angular velocities and telescopic lengths for combined testing. Under each parameter combination, it is evaluated whether the intelligent fixture can successfully complete a series of simulated operation tasks within the safe operating space, including approaching the target workpiece, grasping the workpiece, transporting the workpiece, and releasing the workpiece at different positions and postures. Through a large number of simulated tests and precise calculations, it is judged whether each parameter combination meets the safe operating requirements, that is, whether it can avoid colliding with any obstacles within the safe operating space, and at the same time ensure that the fixture can stably and accurately grasp and operate the workpiece. The parameter combinations that meet these conditions are screened out, and these are the feasible working parameters. Finally, all the feasible working parameters are sorted and classified, and according to their applicability and effectiveness at different working space positions, a control parameter strategy space is constructed. In this space, each control parameter strategy clearly corresponds to a specific working space position identifier, so that during the actual path planning process, the intelligent fixture can quickly and accurately select the most suitable parameter strategy from the control parameter strategy space according to the current working space position it is in, thereby achieving efficient, precise, and safe operations, and improving the automation level and production efficiency of the entire industrial production process.

[0076] In a possible implementation manner, step S430 further includes:

[0077] Step S431: For the static avoidance space feature, according to the edge feature of the static avoidance target, combined with the running avoidance response parameters of the intelligent fixture, determine the minimum safe distance.

[0078] Step S432: For the dynamic avoidance space feature, take the dynamic avoidance target as a collaborative target to obtain the running parameters of the dynamic avoidance target, and obtain the running path features, where the running path features include a periodic path and a dynamic path.

[0079] Step S433: According to the periodic path, determine the periodic avoidance position, and according to the periodic avoidance position combined with the running avoidance response parameters of the intelligent fixture, determine the minimum safe distance.

[0080] Step S434: According to the dynamic path, connect the control interface of the dynamic avoidance target, and obtain real-time control parameters for position prediction.

[0081] Step S435: Determine the minimum safety distance based on the position prediction result and in combination with the running avoidance response parameters of the intelligent fixture.

[0082] Specifically, for static avoidance targets such as large machine tools and fixed shelves in the workshop, precise measurement is first carried out. Professional measuring tools are used to obtain their three-dimensional dimensions, including length, width, height, and the coordinate positions of each corner and other detailed geometric information. Then, computer-aided design (CAD) or 3D modeling software is used to construct an accurate three-dimensional model of the static avoidance target. At the same time, a detailed model of the intelligent fixture itself is also built, including information such as the shape, size, and extension range of the jaws, as well as the length and rotation angle range of the moving arm, to ensure that the model can accurately reflect the spatial form of the fixture in different working states. According to the working tasks and operation processes of the intelligent fixture, analyze its possible movement trajectories when approaching, bypassing, or operating near the static avoidance target. For example, when the fixture needs to grasp a workpiece near a static obstacle from one side, determine the extension direction of its moving arm, the change in rotation angle, and the opening and closing path of the jaws, etc. Based on the above analysis, in combination with the model of the static avoidance target and the movement trajectory of the fixture, the safety distance is determined through geometric calculations. Considering the uncertainties during the movement of the fixture, such as possible slight offsets or vibrations, the safety distance is usually set as the sum of the maximum dimension of the static avoidance target (such as the longest side length, maximum diameter, etc.) and the maximum extension length of the fixture in the corresponding direction, and an additional safety margin is added. For example, if the longest side length of the static obstacle is 2 meters and the maximum extension length of the fixture in this direction is 1 meter, according to experience and safety standards, an additional safety margin of 0.5 meters may be added, thus determining the minimum safety distance to be 3.5 meters.

[0083] Using advanced collision detection algorithms, the models of static avoidance targets and intelligent fixtures are imported into a dedicated simulation software environment. These algorithms can precisely handle complex geometries, considering not only the basic shapes of objects but also accurately judging collisions for details such as corners and depressions. In the simulation software, various possible motion scenarios of the intelligent fixture around the static avoidance target are set, including parameter combinations such as different speeds, accelerations, starting positions, and motion directions. The simulated fixture starts moving from the initial position and gradually approaches the static avoidance target. At each time step during the motion, the collision detection algorithm calculates the distance between the fixture model and the static avoidance target model and determines whether a collision occurs. Through a large number of simulation tests, the distances between the fixture and the static avoidance target when a collision is about to occur in different scenarios are recorded. The results of multiple simulation tests are statistically analyzed to find the minimum distance value. Considering various factors in the actual working environment, such as sensor errors and environmental interference, this minimum distance value is appropriately corrected and finally determined as the minimum safety distance for the static avoidance space feature. For example, after multiple simulations, it is found that the closest distance between the fixture and the static avoidance target in a specific motion scenario is 0.3 meters. However, considering the actual situation, the minimum safety distance is set to 0.5 meters to ensure a sufficient safety margin. Deeply study the motion speed characteristics of the intelligent fixture, including its maximum speed, average speed, and speed change rules in different working modes (such as rapid grasping and precise placement). At the same time, accurately measure and analyze the running avoidance response parameters of the intelligent fixture to determine the time (response time) required for it to start taking braking or avoidance measures from detecting an obstacle, as well as parameters such as deceleration during the braking process. According to the kinematic formula, calculate the distance that the fixture may move during the response time at the current motion speed of the fixture. For example, if the maximum motion speed of the fixture is 1 m / s and the response time is 0.5 s, the fixture may move 0.5 meters during this time. At the same time, considering that the fixture will continue to move a certain distance during the braking process, calculate the braking distance according to the braking deceleration. Assuming the braking deceleration is 2 m / s², calculate the braking distance as 0.125 meters according to the formula. Add the distance moved during the response time and the braking distance to obtain the size of the buffer area as 0.625 meters. Set this size of buffer area around the static avoidance target to ensure that the fixture has enough space and time to stop or adjust the motion direction to avoid collisions.

[0084] For dynamic avoidance targets, such as moving transport trolleys, other robots in operation, etc., their operating parameters are collected in real time through multiple sensors (such as vision sensors, lidar, position sensors, etc.). The vision sensor can obtain information such as the external contour and position change of the dynamic avoidance target. The lidar can accurately measure the distance and relative speed between it and the intelligent fixture, and the position sensor provides the coordinate position information in the working space. A data acquisition system is established to collect, process, and store the data obtained by these sensors in real time to ensure the accuracy and timeliness of the data. According to the collected operating parameters, the operating path characteristics of the dynamic avoidance target are analyzed. For dynamic avoidance targets with periodic motion laws, such as transport trolleys that move back and forth periodically on a fixed track, by analyzing the data of their position changing with time, their periodic path characteristics such as the cycle length, motion direction change points, speed change laws, etc. are determined. For example, it is found that the transport trolley completes a round trip every 20 seconds, with a speed of 0 at both ends of the track and a constant speed of 0.5 m / s in the middle section. For dynamic avoidance targets with irregular motion trajectories, such as manually operated handling robots, by tracking their position changes in real time and using data processing algorithms (such as Kalman filtering, etc.) to predict and analyze their motion trajectories, their dynamic path characteristics are determined, including the general trend of the motion direction, the speed change range, etc.

[0085] According to the periodic path characteristics of the dynamic avoidance target, the key periodic avoidance positions are determined. For example, for the above-mentioned transport trolley, both ends of its track (the positions where the speed is 0 and the motion direction changes) and the positions in the middle of the track where the motion path of the intelligent fixture may cross are regarded as periodic avoidance positions. These positions are areas with a relatively high risk of collision and need special attention. Considering the motion of the intelligent fixture near these periodic avoidance positions and combining its operating avoidance response parameters, the minimum safety distance is calculated. Similar to the static avoidance space characteristics, the distance that the fixture may move during the response time at the current motion speed and the braking distance are calculated. Assume that the motion speed of the fixture when approaching the periodic avoidance position of the transport trolley is 0.8 m / s, the response time is 0.4 s, and the braking deceleration is 1.5 m / s². The calculated moving distance during the response time is 0.32 m, and the braking distance is 0.21 m. The sum of the two is 0.53 m. Then, according to factors such as the residence time of the transport trolley at the periodic avoidance position and the starting acceleration, an additional safety margin, such as 0.2 m, is added, and the minimum safety distance at this periodic avoidance position is finally determined to be 0.73 m.

[0086] For an avoidance target with a dynamic path, its precise control parameters, such as motor drive signals, speed set values, steering commands, etc., are obtained in real time by connecting its control interface (if possible) or using advanced communication technologies. These parameters can more accurately reflect the motion intention and future motion trend of the dynamic avoidance target. Based on the obtained real-time control parameters, a prediction algorithm is used to predict the future position of the dynamic avoidance target. For example, according to the current speed set value and steering command, combined with its kinematic model, the position change in the next few seconds is predicted. At the same time, considering the interference factors in the actual environment, such as changes in ground friction and load changes, the prediction results are corrected and optimized.

[0087] According to the position prediction results, analyze the possible collision situations between the intelligent fixture and the dynamic avoidance target in the future. Considering the operation avoidance response parameters of the intelligent fixture, calculate the minimum distance required for the fixture to safely stop or avoid before the predicted collision moment. For example, it is predicted that the dynamic avoidance target will intersect with the current motion path of the fixture after 3 seconds. At this time, the motion speed of the fixture is 0.6 m / s, the response time is 0.35 s, and the braking deceleration is 1.2 m / s². The calculated moving distance within the response time is 0.21 m, the braking distance is 0.15 m, and a total safe distance of 0.36 m is required. At the same time, according to the motion uncertainty of the dynamic avoidance target (such as possible sudden acceleration or steering), an additional safety factor, such as 0.15 m, is added, and the final determined minimum safe distance is 0.51 m. Moreover, during the motion of the dynamic avoidance target, its position prediction and minimum safe distance calculation are updated in real time to ensure that the intelligent fixture always maintains a safe distance from the dynamic avoidance target and guarantees the operation safety. Through the above detailed methods for static and dynamic avoidance space characteristics, the minimum safe distance of the intelligent fixture in a complex working environment can be accurately determined, providing a solid guarantee for the safe and efficient path planning of the intelligent fixture, effectively avoiding collision accidents, and improving the automation level and reliability of industrial production.

[0088] In a possible implementation manner, step S600 further includes:

[0089] Step S610: Sort the control parameter policy space according to the spatial position relationship of the task environment space sequence, and establish a mapping association between the task space sequence relationship and the control parameter policy space.

[0090] Step S620: Decompose the spatial task with the task execution target to obtain the target tasks of each space.

[0091] Step S630: Optimize the parameters of the control parameter policy space according to the target task, screen the policies according to the optimization results, obtain the control parameter policies that meet the evaluation threshold, and sort the policies according to the evaluation values, where the control parameter policies have optimization evaluation annotations.

[0092] Step S640: Piece together the full-task parameter policies of the task environment space sequence according to the optimization evaluation annotations to obtain the total evaluation value of the pieced-together parameter policies.

[0093] Step S650: Use the space connection adjustment coefficient of the task environment space sequence as the parameter piecing evaluation coefficient to correct the total evaluation value of the pieced-together parameter policies, and obtain the path planning result, where the path planning result is the pieced-together parameter policy with the largest corrected total evaluation value.

[0094] Specifically, conduct an in-depth analysis of the control parameter policy space, which contains numerous parameter combination policies for the intelligent fixture in different working states and spatial positions, and each policy has unique performance characteristics and applicable scenarios. At the same time, study the task environment space sequence, which details the various working space areas and their sequence that the intelligent fixture goes through during the entire task process. Then, according to the task environment space sequence, carefully sort the various policies in the control parameter policy space according to their most suitable spatial positions. For example, in an industrial scenario involving parts processing and assembly, if the task starts in the raw material storage area, the control parameter policies that match the operating characteristics of this area (such as the parameter combinations of the initial grasping speed and clamping force of the fixture in this area) will be ranked at the front of the sequence; when the task progresses to the processing area, the policies corresponding to the precise positioning, high-speed movement, etc. characteristics required for the processing operation are arranged in sequence. Through such a sorting process, a clear mapping relationship is established one by one between each space in the task environment space sequence and the corresponding policies in the control parameter policy space. This mapping relationship ensures that when the intelligent fixture executes the task, it can quickly and accurately obtain and apply the most suitable control parameter policy according to the specific task environment space it is in, laying a solid foundation for subsequent efficient path planning and precise operation.

[0095] Clarify the task execution goal, which is the core orientation of the entire task and covers the expected results from the starting operation to the final completion. For example, in the task of assembling automotive parts, the task execution goal may be to assemble specific types of parts into a complete automotive component according to precise positions and sequences, ensuring that the assembly quality meets strict process standards and completing the assembly task within the specified time to meet the production rhythm requirements. Then, based on the task execution goal, conduct a comprehensive and detailed spatial task decomposition of the entire task. Considering that automotive part assembly involves multiple different work areas, such as part storage areas, pre-treatment areas, assembly areas, and inspection areas, etc., determine the corresponding target tasks according to the functional characteristics and operation requirements of each area. In the part storage area, the target task may be to accurately identify and grasp the required parts, which requires the intelligent fixture to quickly locate the target parts in the complex spatial environment of the storage area and grasp them with appropriate postures and forces, ensuring a stable grasping process without damaging the parts; in the assembly area, the target task changes to assembling the grasped parts with other components according to precise positions and angles, and the intelligent fixture needs to have high-precision positioning capabilities and flexible posture adjustment capabilities to achieve a perfect fit between the parts. Through such a spatial task decomposition process, the complex overall task is refined into specific target tasks for each space, providing a clear direction and basis for the subsequent targeted optimization of the control parameter strategy space according to the target tasks, helping the intelligent fixture to efficiently and accurately complete various operations in different work spaces, thus ensuring the smooth achievement of the entire task.

[0096] For each space target task obtained from the decomposition of space missions, deeply analyze its unique operation requirements, precision standards, and environmental constraints. For example, in the task of assembling precision electronic components, in the component picking space, the target task requires the fixture to gently and precisely grasp the tiny components, which requires a control parameter strategy to ensure an appropriate clamping force, a smooth opening and closing speed of the gripper jaws, and an extremely high positioning accuracy; while in the component assembly space, the target task is to accurately install the components at the specified positions on the circuit board. At this time, the control parameter strategy needs to focus on the movement trajectory accuracy of the moving arm and the accuracy of attitude adjustment. Based on the specific requirements of these target tasks, parameter optimization is fully carried out in the control parameter strategy space, and numerous control parameter combinations in the strategy space (covering parameters such as the speed, acceleration, and rotation angle of the moving arm, and the clamping force and opening and closing speed of the gripper jaws) are carefully evaluated. Measure the performance of each parameter combination in achieving the target task through simulation experiments, such as the accuracy, efficiency, stability of the operation, and the impact on the workpiece and the fixture itself. Then, strict strategy screening is carried out according to the optimization results, and a series of evaluation thresholds are set, which are determined based on the key performance indicators of the task. For example, in the component picking space, the component grasping position error shall not exceed ±0.05 mm, and the clamping force fluctuation range needs to be controlled within a minimum value; in the component assembly space, the assembly position accuracy requirement reaches ±0.02 mm, etc. Only the control parameter strategies that meet these evaluation thresholds are retained. For the screened strategies, further ranking is carried out according to the comprehensive evaluation value. The comprehensive evaluation value comprehensively considers various factors, such as the weight of operation accuracy, the impact of operation efficiency, the energy consumption situation, and the equipment wear degree. Each control parameter strategy is given an optimization evaluation label, clearly recording its detailed performance on various evaluation indicators, so as to intuitively compare the advantages and disadvantages of different strategies. Through such ranking, the control parameter strategy most suitable for each target task can be quickly determined, providing high-quality strategy selection for the subsequent full-task path planning.

[0097] According to the optimized evaluation labels obtained in the previous steps, the optimal control parameter strategies corresponding to each target task in the task environment space sequence are spliced in an orderly manner. For example, in a complex electronic product assembly task, if the task environment space sequence includes a component picking area, a circuit board installation area, a housing assembly area, etc., the optimal parameter strategies found in the component picking area (such as the combination of parameters like the clamping force and opening / closing speed suitable for grasping tiny components), the precise positioning and installation parameter strategies in the circuit board installation area, and the assembly parameter strategies in the housing assembly area will be spliced together in the order of task execution to form a complete full-task parameter strategy chain. During the splicing process, the transition connection between each space should be fully considered. For example, from the component picking area to the circuit board installation area, it is necessary to ensure that parameters such as the speed and attitude adjustment of the fixture during movement can transition smoothly, avoiding instability or collision risks caused by parameter mutations. Then, calculate the overall evaluation value of this spliced parameter strategy. The calculation of the overall evaluation value comprehensively considers the various performance manifestations of each target task under its corresponding parameter strategy. For example, for the component picking area, factors such as picking accuracy, efficiency, and the degree of protection of components will be considered; for the circuit board installation area, the focus is on the accuracy of the installation position and the stability of the installation process; for the housing assembly area, the emphasis is on the tightness of the assembly and the appearance integrity. These factors are assigned different weights according to their importance, and the overall evaluation value of the spliced parameter strategy is obtained through calculation methods such as weighted summation. This overall evaluation value comprehensively reflects the overall performance level of the entire task path planning under the current parameter strategy combination, provides an important reference basis for further optimizing the path planning in the future, helps to judge whether the current spliced parameter strategy meets the task requirements, and whether adjustments or re-optimization are needed.

[0098] Clarify the meaning of the spatial connection adjustment coefficient, which is a quantitative index comprehensively considering the transition characteristics between the task environment spaces. For example, in an industrial scenario involving multiple processing stations and material transfer areas, from one processing station to the adjacent material transfer area, the fixture needs to change the movement speed, adjust the posture, and adapt to different spatial layouts, and these factors will all affect the performance of the overall path planning. Then, introduce this coefficient as a parameter to splice the evaluation coefficient. The total evaluation value of the splicing parameter strategy has been obtained before, and this total evaluation value reflects the overall performance of the parameter strategy combination corresponding to each target task under ideal conditions. However, various factors in the actual spatial connection process will affect the total evaluation value. At this time, the spatial connection adjustment coefficient is used to correct the total evaluation value. For example, if the transition between two adjacent spaces is relatively complex and the fixture needs to make large-scale posture adjustments and speed changes, then the corresponding spatial connection adjustment coefficient may reduce the total evaluation value; conversely, if the transition is relatively smooth, the impact on the total evaluation value is relatively small. By calculating and correcting the spatial connection adjustment coefficient and the total evaluation value, the corrected total evaluation value is obtained. Then, among many splicing parameter strategies, select the strategy with the largest corrected total evaluation value, which is the final path planning result. This path planning result fully considers the characteristics of each area in the task environment space sequence and the connection relationship between spaces, ensuring that the intelligent fixture can operate smoothly in the complex working space with optimal performance during task execution, effectively avoiding collision risks, improving the accuracy and efficiency of operations, and meeting the high requirements for the path planning of intelligent fixtures in industrial production.

[0099] It should be noted that the above sequence of embodiments of the present application is only for description and does not represent the superiority or inferiority of the embodiments. And the above specific embodiments of this specification have been described. In addition, the processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0100] The above are only the preferred embodiments of the present application and are not intended to limit the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

[0101] This specification and the drawings are only exemplary descriptions of the present application and are considered to have covered any and all modifications, variations, combinations, or equivalents within the scope of the present application. Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalent technologies, the present application is intended to include these changes and modifications.

Claims

1. An intelligent fixture path planning method for an industrial robot, characterized in that: The method comprises: Based on the working parameters of the intelligent fixture, feature recognition is performed on the working environment space of the intelligent fixture to obtain space partitions, wherein the space partitions are clustered and segmented according to the space recognition features; Establishing spatial constraint features according to the correspondence between the spatial partitions and the spatial identification features; Obtaining control constraint information according to the working parameters of the intelligent fixture; According to the spatial constraint feature and the control constraint information, constraint matching analysis is performed to obtain a control parameter strategy space, wherein the control parameter strategy in the control parameter strategy space has a workspace location identifier; Decompose the task process of the industrial robot fixture task to obtain the task environment space sequence; According to the task environment space sequence, spatial matching is performed with the workspace position identifier of the control parameter strategy space, the control parameter strategy space is used to perform local optimization of the spatial partition strategy, and the local optimization result is used to perform full task path optimization to obtain a path planning result; The method of performing feature recognition on the working environment space of the intelligent fixture based on the working parameters of the intelligent fixture to obtain space partitions includes: According to the working parameters of the intelligent fixture, a relationship between the working parameters and the volume state of the fixture is established to obtain a maximum working space and a minimum working space; Identify static avoidance targets and dynamic avoidance targets in the work environment space, and determine static avoidance space characteristics and dynamic avoidance space characteristics based on the work space where the static avoidance targets and dynamic avoidance targets are located; Using the minimum working space, the static avoidance space characteristics and the dynamic avoidance space characteristics are subjected to traffic fitting to obtain traffic impact characteristics; Using the maximum working space, the static avoidance space characteristics and the dynamic avoidance space characteristics are fitted to obtain operation impact characteristics; A spatial feature discrete analysis is performed according to the traffic impact characteristics and the operation impact characteristics, spatial clustering is performed according to the feature distribution discreteness, and spatial segmentation is performed according to the spatial clustering results to obtain the spatial partition.

2. The intelligent fixture path planning method for an industrial robot according to claim 1, characterized in that: The obtaining of the operation impact characteristics further includes: According to the relationship between the working parameters and the clamp volume state, a multi-level operating state volume is obtained; Using the multi-level operation state volume, respectively performing pass fitting on the static avoidance space feature and the dynamic avoidance space feature to obtain a multi-level operation restriction feature; The multi-level operation restriction characteristics are integrated to obtain the operation impact characteristics.

3. The intelligent fixture path planning method for an industrial robot according to claim 1, characterized in that: According to the traffic impact characteristics and the operation impact characteristics, spatial feature discrete analysis is performed, and spatial clustering is performed according to the feature distribution dispersion, including: According to the working environment space, the location distribution of the traffic impact features is marked to obtain a traffic impact distribution map; According to the position distribution of the operation impact features, distribution annotation is performed in the work environment space to obtain an operation impact distribution map; The traffic impact distribution map and the operation impact distribution map are merged and superimposed, and the dispersion is calculated according to the distribution annotation to obtain the feature distribution dispersion, which is used to characterize the discrete proportion of the traffic impact feature and the operation impact feature in the workspace; Spatial feature clustering is performed according to the feature distribution discreteness.

4. The intelligent fixture path planning method for an industrial robot according to claim 1, characterized in that: According to the working parameters of the intelligent fixture, control constraint information is obtained, including: Obtaining control nodes of the intelligent clamp, wherein the control nodes at least include a moving arm control node and a clamping mechanism control node; According to the adjustment parameter range of the control node, a multi-node control parameter combination is performed to obtain the fixture shape change of each control parameter combination; According to the change of the fixture morphology, edge space constraints are obtained; Determining a traffic adjustment constraint according to an adjustment parameter range of the mobile arm control node; The control constraint information is obtained according to the edge space constraint and the traffic adjustment constraint.

5. The intelligent fixture path planning method for an industrial robot according to claim 1, characterized in that: According to the spatial constraint feature and the control constraint information, constraint matching analysis is performed to obtain a control parameter strategy space, including: Obtaining a constraint distribution space according to the spatial constraint characteristics; According to the control constraint information, obtaining the fixture control operation constraint space; According to the spatial constraint characteristics, a minimum safety distance is configured; Based on the minimum safety distance, the fixture is used to control the operation constraint space to identify the operation safety space in the constraint distribution space to obtain a safe operation space; The working parameters of the intelligent fixture are searched in the safe operation space, all feasible working parameters are obtained, and the control parameter strategy space is constructed.

6. The intelligent fixture path planning method for an industrial robot according to claim 5, characterized in that: According to the spatial constraint characteristics, the minimum safety distance is configured, including: For the static avoidance space characteristics, the minimum safety distance is determined according to the edge characteristics of the static avoidance target and the operation avoidance response parameters of the intelligent fixture; For the dynamic avoidance space feature, the dynamic avoidance target is taken as the collaborative target to obtain the operation parameters of the dynamic avoidance target, and the operation path feature is obtained, wherein the operation path feature includes a periodic path and a dynamic path; Determine a periodic avoidance position according to the periodic path, and determine the minimum safety distance according to the periodic avoidance position combined with the operation avoidance response parameter of the intelligent fixture; According to the dynamic path, connecting to the control interface of the dynamic avoidance target, obtaining real-time control parameters for position prediction; The minimum safety distance is determined according to the position prediction result and in combination with the operation avoidance response parameters of the intelligent fixture.

7. The intelligent fixture path planning method for an industrial robot according to claim 1, characterized in that: The obtaining of the path planning result includes: The control parameter strategy space is sorted according to the task environment space sequence to establish a task space sequence relationship and a mapping association with the control parameter strategy space; Based on the task execution goal, the spatial tasks are decomposed to obtain the target tasks of each space; Optimizing the control parameter strategy space according to the target task, screening strategies according to the optimization results, obtaining control parameter strategies that meet the evaluation threshold, and sorting strategies according to the evaluation values, wherein the control parameter strategies have optimization evaluation labels; According to the optimization evaluation annotation, all task parameter strategies of the task environment space sequence are spliced ​​to obtain the total evaluation value of the spliced ​​parameter strategy; The spatial connection adjustment coefficient of the task environment spatial sequence is used as the parameter splicing evaluation coefficient, and the total evaluation value of the splicing parameter strategy is corrected to obtain the path planning result. The path planning result is the splicing parameter strategy with the largest total evaluation value after correction.

Images