Method, device, equipment and medium for evaluating repair disassembly path comfort degree

Abstract

The present application belongs to the technical field of human-machine engineering and maintainability design, and provides a maintenance disassembly path comfort evaluation optimization method, device, equipment and medium, which decomposes a maintenance disassembly task to be evaluated into a series of operation points, each operation point corresponds to a human body posture node, and generates a posture vector of each human body posture node; a multi-dimensional comfort cost function is constructed by considering the instant comfort of a single point posture, the path space safety in the maintenance operation process and the duration accumulation effect of an uncomfortable posture; the multi-dimensional comfort cost function, the total maintenance time cost and the gap safety penalty are combined, the accessibility, the joint limiting and the task logic constraint are considered, a total optimization objective function is constructed and solved, a Pareto optimal solution set is output from a candidate posture sequence set, and an optimized posture sequence is output. The present application is especially suitable for human-machine engineering evaluation and optimization of a disassembly path in a virtual maintenance simulation stage of complex equipment.

Description

Technical Field

[0001] This invention belongs to the field of ergonomics and maintainability design technology, and in particular, it relates to a method, device, equipment and medium for evaluating and optimizing the comfort of maintenance and disassembly paths. Background Technology

[0002] Maintenance and disassembly operations are a critical link in the entire life cycle management of equipment. The comfort of operators directly affects maintenance efficiency, work quality, and long-term occupational health. Traditional maintenance path planning mainly focuses on task accessibility and minimizing operation time, with limited consideration for the comfort of operators' postures.

[0003] Currently, the methods for assessing human comfort during maintenance work can be mainly categorized as follows: Firstly, human posture analysis methods based on rapid assessment tools such as RULA and REBA can quickly score a single human posture. However, these methods often only output a single total score, making it difficult to accurately quantify the differentiated impact of different joint postures on comfort, and they also lack consideration for the smoothness of transitions between posture sequences.

[0004] Secondly, there are biomechanical analysis methods based on joint torques. Some studies have established human arm dynamics models, used the Kane equation to solve for joint torques, and combined them with NASA force models to obtain comfort scores. However, most of these methods only focus on the arm aspect and fail to consider the overall posture (such as torso bending and twisting) in conjunction with the arm posture. In maintenance work, the comfort impact of waist posture is often just as important as that of the arm.

[0005] Third, ergonomic optimization methods based on disassembly sequence planning. Recent studies have proposed considering ergonomic requirements in disassembly sequence planning by analyzing operator joint angles for personalized disassembly sequence planning. However, existing methods typically treat posture evaluation as a constraint on task allocation rather than an optimization objective, and are mostly geared towards robot-assisted disassembly scenarios. A systematic solution is still lacking for the continuous evaluation of human posture sequences in purely manual repair and disassembly paths.

[0006] In summary, the existing technology lacks a collaborative comfort evaluation method that can simultaneously consider the body posture (including torso and waist) and arm posture, and there is no maintenance and disassembly path optimization scheme that integrates posture comfort, path clearance safety and operation time into a multi-objective optimization framework. Summary of the Invention

[0007] This invention aims to overcome the shortcomings of the prior art and provide a method, device, equipment and medium for evaluating and optimizing the comfort of maintenance disassembly and assembly paths, especially suitable for evaluating and optimizing the ergonomics of disassembly and assembly paths for complex equipment in the virtual maintenance simulation stage.

[0008] To achieve the above objectives, the technical solution adopted by the present invention is as follows: On the one hand, the present invention provides a method for evaluating the comfort of maintenance disassembly and assembly paths, comprising the following steps: The maintenance and disassembly task to be evaluated is decomposed into a sequence of operation points, each operation point corresponds to a human posture node, and the posture vector of each human posture node is generated. At least the immediate comfort of a single-point posture, the path space safety during maintenance operations, and the cumulative effect of the duration of uncomfortable postures should be considered to construct a multidimensional comfort cost function. The overall optimal objective function is constructed by combining the multidimensional comfort cost function, the total maintenance time cost of the maintenance and disassembly task to be evaluated, and the gap safety penalty, while introducing at least the constraints including accessibility constraints, joint limit constraints, and task logic constraints. A multi-objective optimization algorithm is used to minimize the overall optimal objective function. Under the premise of satisfying the constraints, a Pareto optimal solution set is output from the candidate attitude sequence set. The output result includes at least the preferred attitude sequence.

[0009] On the other hand, the present invention provides an electronic device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the above-mentioned method for evaluating the comfort of maintenance and disassembly paths.

[0010] On the other hand, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of the above-described method for optimizing the comfort of maintenance and disassembly paths.

[0011] On the other hand, the present invention provides a computer program product stored on a computer-readable storage medium and including computer instructions that, when executed by a processor, cause a computer device to implement the steps of the above-described method for optimizing the comfort of maintenance and disassembly paths.

[0012] Compared with the prior art, the present invention has the following advantages: This invention provides a method for evaluating and optimizing the comfort of maintenance and disassembly paths. The method decomposes the maintenance and disassembly task to be evaluated into a sequence of operation points, each operation point corresponding to a human posture node. Each human posture node includes body posture parameters and arm posture parameters, generating a posture vector for each human posture node. A multi-dimensional comfort cost function is constructed, including at least joint torque deviation cost, path clearance cost, and discomfort time accumulation cost. A total optimization objective is established, comprehensively considering comfort cost, maintenance time cost, and clearance safety penalty. Under the premise of satisfying accessibility constraints, joint limit constraints, and task logic constraints, the optimal posture sequence is output. This invention, by co-modeling body posture and arm posture and introducing discomfort time accumulation cost, can more comprehensively evaluate human comfort during maintenance and disassembly processes, achieving a balance between maintenance efficiency and personnel health. It can be widely applied in the fields of virtual maintenance simulation and maintainability design.

[0013] By considering at least the immediate comfort of a single-point posture, the path space safety during maintenance operations, and the cumulative effect of uncomfortable posture duration, a multi-dimensional comfort cost function is constructed. This function not only considers the immediate comfort of a single-point posture and the path space safety during maintenance operations, but also the cumulative effect of uncomfortable posture duration. It can realistically reflect the decay process of physiological tolerance of maintenance personnel under conditions such as prolonged head tilting, bending over, and continuous arm extension, avoiding underestimation of actual operational risks due to neglecting the time factor. The further constructed overall optimization objective function combines the multi-dimensional comfort cost function, the total maintenance time cost of the maintenance and disassembly task to be evaluated, and the clearance safety penalty. This breaks the limitation of the separation between safety and efficiency, and comfort and accessibility. It can automatically weigh the relationship between them within the same optimization framework, avoiding sacrificing comfort for excessive pursuit of accessibility, or neglecting safety clearances for excessive pursuit of efficiency. This invention transforms the evaluation of comfort in maintenance and disassembly paths from a vague, subjective, and empirical problem into a multi-objective optimization problem that is calculable, constrained, optimizable, and balancing. This effectively improves the scientific rigor and accuracy of maintainability design assessment. The optimal posture sequence provided is expected to reduce the risk of strain on maintenance personnel due to prolonged uncomfortable work, while ensuring the safety and efficiency of maintenance operations. Attached Figure Description

[0014] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the structures shown in these drawings without creative effort.

[0015] Figure 1This is a flowchart of the method for evaluating the comfort of maintenance and disassembly paths provided by the present invention. Detailed Implementation

[0016] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0017] Reference Figure 1 One embodiment provides a method for optimizing the comfort of maintenance disassembly and assembly paths, including the following steps: The maintenance and disassembly task to be evaluated is decomposed into a sequence of operation points, each operation point corresponds to a human posture node, and the posture vector of each human posture node is generated. At least the immediate comfort of a single-point posture, the path space safety during maintenance operations, and the cumulative effect of the duration of uncomfortable postures should be considered to construct a multidimensional comfort cost function. The overall optimal objective function is constructed by combining the multidimensional comfort cost function, the total maintenance time cost of the maintenance and disassembly task to be evaluated, and the gap safety penalty, while introducing at least the constraints including accessibility constraints, joint limit constraints, and task logic constraints. A multi-objective optimization algorithm is used to minimize the overall optimal objective function. Under the premise of satisfying the constraints, a Pareto optimal solution set is output from the candidate attitude sequence set. The output result includes at least the preferred attitude sequence.

[0018] This method decomposes maintenance and disassembly tasks into a sequence of operation points, each corresponding to a human posture node, generating a posture vector for each human posture node. A multidimensional comfort cost function is constructed, considering at least the immediate comfort of a single-point posture, path space safety during maintenance operations, and the cumulative effect of uncomfortable posture duration. A comprehensive optimization objective is established, incorporating comfort, maintenance time, and gap safety. The output is an optimal posture sequence that satisfies accessibility constraints, joint limit constraints, and task logic constraints, achieving a balanced optimization between maintenance efficiency and human comfort.

[0019] Furthermore, each human posture node in this invention comprises two levels: main posture parameters and arm posture parameters, which together constitute the posture vector of the human posture node. The main posture parameters include at least: torso forward tilt angle. Lateral flexion angle of the trunk Torso twist angle and hip flexion angle Knee flexion angle Arm posture parameters should include at least: shoulder flexion / extension angle. Shoulder abduction / adduction angle elbow flexion angle wrist flexion angle and wrist joint ulnar / radial deviation angle .

[0020] The operation point in the maintenance and disassembly path is decomposed into two levels: main body posture parameters and arm posture parameters. This enables collaborative modeling and comprehensive evaluation of torso / waist posture and arm posture, solving the problem in existing technologies that only focus on arm posture while ignoring the impact of main body posture on overall comfort.

[0021] For the pose vector of an individual human pose node It is expressed as follows: ; In the formula, Indicates the sequence index of human pose nodes. The first The individual's posture nodes include the trunk forward tilt angle, trunk lateral flexion angle, trunk twist angle, hip flexion angle, knee flexion angle, shoulder flexion / extension angle, shoulder abduction / adduction angle, elbow flexion angle, wrist flexion angle, and wrist ulnar / radial deviation angle.

[0022] In one embodiment, the multidimensional comfort cost function includes at least several costs, such as joint torque deviation cost, path clearance cost, and cumulative discomfort time cost for each human posture node. Specifically, those skilled in the art can also consider costs in other dimensions based on the actual situation of the maintenance and disassembly task.

[0023] The joint torque deviation cost of the human posture node is calculated as follows: For each human posture node, the deviation between the real-time joint torque of each target joint and the comfortable torque range of each target joint is calculated based on the human inverse dynamics model, and the joint torque deviation cost of the human posture node is obtained by summarizing the results.

[0024] The path gap cost is calculated based on collision detection technology in a virtual maintenance environment, which calculates the minimum gap distance between the maintenance worker's human body envelope and surrounding equipment / structures, and constructs the path gap cost. The cumulative cost of discomfort time is calculated by considering the comfort of a single-point posture and the duration of the uncomfortable posture.

[0025] The joint torque deviation cost is calculated based on a human inverse dynamics model, determining the degree of deviation between the real-time torque of each joint and the comfort torque range. A piecewise function is used to assign differentiated weighting coefficients to different deviation directions. Specifically, the calculation steps for the joint torque deviation cost are as follows: 1.1 Establish a multi-rigid-body dynamic model of the human skeletal muscle system. Divide the human body into main rigid body segments such as trunk, head, upper arm, forearm, hand, thigh, lower leg and foot. The mass, center of mass position and moment of inertia of each rigid body segment are determined according to the percentile parameters of maintenance personnel based on the anthropometry database (such as GB / T 10000-2023).

[0026] 1.2, Regarding the first Individual human pose nodes, based on their pose vectors The recursive Newton-Euler algorithm is used to solve the problem under static / quasi-static working conditions. Joint torques of each target joint in an individual human posture node For dynamic operation processes, dynamic parameters such as operating force and operating speed need to be further introduced.

[0027] 1.3. Based on human biomechanical experimental data, for the first In the individual human posture node, each target joint constructs a comfortable torque range, where the first... The first human pose node The comfort torque range for each target joint is , No. The first human pose node Joint torque of each target joint A load below this range indicates that the load is too light and uneconomical, while a load above this range indicates that the muscle load is too heavy and can easily lead to fatigue.

[0028] 1.4 Calculate the first... Cost of joint torque deviation at individual body posture nodes: ; In the formula, For the first The first human pose node The real-time joint torque of the target joint and the first target joint The deviation function of the comfort torque range of each target joint For the first The first human pose node The joint weight coefficients for each target joint are assigned according to the following principles: large joints (hip, shoulder) have a higher weight than small joints (wrist, ankle); weight-bearing joints (lower back, knee) have a higher weight than non-weight-bearing joints. The weights are defined in the following segmented form: ; In the formula, For the first The first human pose node The low torque deviation of each target joint from the weighting coefficient. For the first The first human pose node The high torque deviation of each target joint is weighted by a coefficient. For the first The first human pose node Joint torque of each target joint For the first The first human pose node The minimum permissible joint torque for each target joint. For the first The first human pose node The maximum permissible joint torque for each target joint.

[0029] The path gap cost is constructed based on the minimum gap distance between the human envelope model and the obstacle, and a piecewise function is used to define the safe interval, the danger interval, and the smooth transition interval between them. In one embodiment, for the first... Individual human pose node, its path gap cost The calculation steps are as follows: 2.1 Establish a simplified human body envelope model for maintenance personnel. A sphere model, capsule model, or a more refined mesh model can be used to represent the spatial area occupied by maintenance personnel in a given posture.

[0030] 2.2, Regarding the first Each human pose node calculates the minimum gap distance between the human envelope model and all obstacles in the environment. .

[0031] 2.3 Define the safety gap threshold and dangerous gap threshold When the minimum gap distance When the distance is less than the danger threshold, the path gap cost tends to a maximum value; when the minimum gap distance is greater than the safety threshold, the path gap cost is zero; a smooth transition function is used for interpolation in the intermediate interval.

[0032] 2.4 The path gap cost function is expressed as: ; In the formula, This is the gap cost gain coefficient, which controls the rate of gain. The larger the value, the faster the path gap cost function value will increase; the smaller the value, the more gradually the path gap cost function value will increase. The maximum penalty value is preset and is a fixed constant.

[0033] Considering that prolonged operation in an uncomfortable posture can exacerbate discomfort and fatigue accumulation, this invention introduces an discomfort time accumulation cost. This cost considers not only the comfort of a single posture but also the duration of the uncomfortable posture. The discomfort time accumulation cost employs a cumulative model with a forgetting factor, weighting the accumulation of posture nodes exceeding a comfort threshold by duration to characterize the fatigue accumulation effect of continuous operation in an uncomfortable posture. Specifically, in one embodiment, for the first... The calculation steps for the cumulative cost of discomfort time at an individual human posture node are as follows: 3.1 Definition of the first The single-point comfort score of an individual's body posture node is The lower the value, the more comfortable the posture.

[0034] 3.2 Setting comfort thresholds .when At that time, the first Individual human posture nodes are considered to be in a comfortable state, not triggering time-cumulative effects; when At that time, the trigger time accumulates.

[0035] 3.3 Calculate the first... using a cumulative model with a forgetting factor. The cumulative cost of discomfort at individual body posture nodes over time: ; In the formula, For the first The duration of the task corresponding to each human posture node. Forgetting factor (TIF) ), The cumulative time window length is denoted by . The introduction of the forgetting factor makes recent discomfort contribute more to the current cost than long-term discomfort.

[0036] Finally, the total cumulative discomfort time cost of the pose sequence is the cumulative discomfort time cost of all human pose nodes. sum.

[0037] In one embodiment, the multidimensional comfort cost is combined with the maintenance time cost and the gap safety penalty to construct the following overall optimal objective function: ; in, The total comfort cost of the posture sequence is obtained by weighted summation of the costs included in the multidimensional comfort cost function; The total maintenance time cost is the sum of the standard operating times at each operation point, and may also include, if necessary, the increment of posture adjustment time introduced based on the complexity of the transition between human posture nodes: As a penalty for gap safety, it is included as an independent item in the overall objective to strengthen safety constraints. , , These are the weighting coefficients for the total comfort cost of the attitude sequence, the total maintenance time cost, and the gap safety penalty term, respectively.

[0038] In one embodiment, the total comfort cost of the posture sequence It is obtained by weighted summation of the joint torque deviation cost, path clearance cost, and cumulative discomfort time cost of each attitude node: ; The total maintenance time cost is the sum of the standard operating times at each operation point. If necessary, an attitude adjustment time increment can be introduced based on the complexity of the transition between attitude nodes. ; In the formula, For the first The baseline operation time for each operation point For from the first Individual human posture node to the first The attitude transition adjustment time of an individual human posture node.

[0039] Among them, gap safety penalty item To strengthen security constraints, it is included as an independent item in the overall objective, as follows: ; in For the first The gap safety penalty for individual human posture nodes. Use an exponential penalty function: ; In the formula, To penalize the steepness coefficient, For the first The minimum clearance distance between the maintenance worker's human body envelope model corresponding to an individual human pose node and all obstacles in the environment. The threshold value is the danger gap threshold. The gap safety penalty term uses an exponential penalty function, applying a penalty when the minimum gap distance between the human body envelope and the obstacle is less than the danger threshold.

[0040] Weighting coefficient , , and sub-cost weights , , Adjustments can be made based on specific application scenarios: in scenarios where work efficiency is prioritized, additional steps can be taken. The weighting can be increased in scenarios with limited operating space and high collision risk. The weighting; in scenarios where operator health protection is emphasized, it can be increased. The weight.

[0041] Based on the above overall objective function, three types of constraints are introduced—reachability constraints, joint limit constraints, and task logic constraints—to filter and optimize the attitude sequence.

[0042] The accessibility constraint means that for each human posture node, the maintenance worker's operating point (such as the hand gripping point) must be able to reach the target maintenance position. The accessibility constraint verifies through forward kinematics whether the distance between the hand end position and the target maintenance position is less than a tolerance threshold.

[0043] Specifically, the accessibility constraint is verified using forward and inverse kinematics methods, including: first, solving for the position coordinates of the hand end-effector in the global coordinate system based on the posture vector of the human posture node using forward kinematics; second, calculating the Euclidean distance between the hand end-effector position and the target repair position; and finally, requiring that this distance be less than the accessibility tolerance. .

[0044] The joint constraint refers to the requirement that the angle of each joint in each human posture node must be within the physiologically permissible range of the joint. That is, for the first joint... All target joints in an individual human pose node. In the formula, and The first The first human pose node The minimum and maximum joint angle limits for each target joint can be determined based on standards such as human anatomical data.

[0045] Task logic constraints refer to the requirement that the arrangement of attitude sequences must conform to the inherent logical order of maintenance and disassembly tasks. These constraints include: disassembly sequence constraints, tool usage constraints, and continuous operation constraints. Continuous operation constraints include angular velocity constraints on the range of joint angle changes between adjacent attitude nodes.

[0046] The disassembly sequence constraint includes the requirement that obstructing components must be removed before the target component can be accessed; The tool usage constraints include the requirement to complete the tool picking posture before using the tool and to complete the tool returning posture after use. The continuous operation constraint includes that the change in joint angle between adjacent human posture nodes must not exceed the product of the human's maximum angular velocity and the time between nodes, that is: In the formula, For the first The maximum safe angular velocity of each target joint For from the first Individual human posture node to the first Individual human posture nodes The time for posture transition adjustment.

[0047] A multi-objective optimization algorithm (such as NSGA-II, MOPSO, or multi-objective simulated annealing algorithm) is used to optimize the overall objective. The process involves minimizing the input while satisfying the three types of constraints described above, outputting the Pareto optimal solution set from the candidate attitude sequence set. The final output is the preferred attitude sequence. This includes the corresponding multi-dimensional comfort score details, total maintenance time estimate, and path clearance safety assessment report.

[0048] The above method constructs a three-dimensional comfort cost function that includes joint torque deviation cost, path clearance cost, and discomfort time accumulation cost. The discomfort time accumulation cost innovatively introduces a time accumulation model with a forgetting factor, which can effectively characterize the fatigue accumulation effect caused by continuous operation in uncomfortable postures. This is a technical feature that has not been disclosed in existing maintenance path comfort evaluation methods.

[0049] The overall optimization objective incorporates three factors simultaneously: comfort cost, maintenance time cost, and gap safety penalty. Furthermore, the constraints integrate accessibility constraints, joint limitation constraints, and task logic constraints, forming a unified optimization framework with multiple objectives and constraints. This framework can dynamically adjust the weights of each objective according to different application scenarios, achieving a balanced optimization between maintenance efficiency and human comfort.

[0050] By treating path gap safety as both an independent objective and a constraint, a safety assurance mechanism combining soft and hard constraints is formed. This mechanism constrains collision risk with an exponential penalty function in the objective function and imposes hard constraints through gap distance thresholds in the reachability constraints.

[0051] This invention, by co-modeling the body posture and arm posture and introducing the cumulative cost of discomfort time, can more comprehensively evaluate human comfort during the maintenance and disassembly process, and achieve a balance between maintenance efficiency and personnel health. It can be widely applied in the fields of virtual maintenance simulation and maintainability design.

[0052] To demonstrate the effectiveness of the method proposed in this invention, this embodiment uses the disassembly and maintenance of an accessory housing for a certain type of aero-engine as an example for verification.

[0053] Step 1: Decompose the disassembly and maintenance tasks of the aircraft engine accessory housing.

[0054] The disassembly and maintenance task of the aircraft engine accessory housing is broken down into 7 operation points, including: ① approaching the housing and unlocking the fixing bolts; ② disconnecting the upper connecting pipes; ③ disconnecting the side wiring harness plugs; ④ removing the left fixing bolts; ⑤ removing the right fixing bolts; ⑥ removing the housing body; ⑦ transporting it to the maintenance station.

[0055] Each operation point corresponds to a human posture node. The actual posture data of the maintenance personnel is obtained through a motion capture system or virtual simulation software, resulting in a complete parameter vector of 7 human posture nodes.

[0056] Step 2: Human pose node modeling.

[0057] Taking operation point ③ (removing the side wiring harness connector) as an example, the attitude parameters of the human posture node corresponding to operation point ③ are as follows: Main body posture parameters: torso forward tilt angle Lateral flexion angle of the trunk Torso twist angle Hip flexion angle Knee flexion angle .

[0058] Arm posture parameters: shoulder joint flexion angle shoulder abduction angle elbow flexion angle wrist flexion angle ulnar deviation of the wrist joint .

[0059] Step 3: Calculate the cost of multidimensional comfort.

[0060] A multi-rigid-body dynamic model of the maintenance worker's human body was established, and the mass and geometric parameters of each rigid body segment were set based on the 50th percentile anthropometric parameters of adult Chinese males. The recursive Newton-Euler algorithm was used to calculate the torques at each joint.

[0061] Taking the shoulder joint as an example, the shoulder joint flexion torque in this posture is The comfortable torque range for this joint is set to [15, 28] N·m. Because... This triggers a deviation penalty, and the deviation component of the shoulder joint is calculated to be 0.38 when substituted into the deviation function.

[0062] Similarly, the torque deviation components of other joints are calculated, and after weighted summation, the joint torque deviation cost of the human posture node corresponding to operation point ③ is obtained. (After dimensionless conversion).

[0063] Regarding path clearance cost, the minimum clearance distance between the human body envelope and the engine casing and surrounding piping is calculated in a virtual maintenance environment. Set a safety gap threshold. Dangerous gap threshold Since 87 mm lies between the two, substituting it into the gap cost function yields the following result. .

[0064] Regarding the cumulative cost of discomfort time, the single-point comfort score of the human posture node corresponding to operation point ③ is... Exceeding the comfort threshold Duration of the operation Forgetting factor Taking 0.95, and setting the cumulative time window length L to 3 nodes, we substitute these values ​​into the cumulative model to obtain the result. .

[0065] Step 4: Calculate the overall optimal target.

[0066] Set weighting coefficients , , Sub-cost weights , , The total comfort cost of this posture sequence .

[0067] Total repair time Standardized processing of repair time costs .

[0068] Gap safety penalty items (All nodes) (All are above the danger threshold, so the penalty is relatively small).

[0069] Substituting into the overall optimization objective, we get .

[0070] Step 5: Constraint verification.

[0071] Accessibility constraint verification: Using the attitude parameters of the human posture node corresponding to operation point ③, the coordinates of the hand end are calculated through forward kinematics. The distance between the hand end and the center coordinates of the harness plug is 23.5 mm, which is less than the tolerance. The accessibility requirement is met.

[0072] Joint constraint verification: All joint angles are within the physiologically permissible range. For example, the trunk torsion angle of 15.8° is less than the maximum limit of 45°, which meets the requirements.

[0073] Task logic constraint verification: The fixing bolts were unlocked before the wiring harness plug was removed, which is in line with the task logic sequence; the joint angular velocities between adjacent attitude nodes did not exceed the safety limits.

[0074] Step 6: Optimize the solution.

[0075] Using the sequence of 7 human pose nodes as the initial solution, a multi-objective optimization algorithm (NSGA-II algorithm, population size 100, evolutionary generations 200) is employed for optimization. After optimization, a Pareto optimal solution set is obtained, from which the overall optimal pose sequence is selected.

[0076] The results before and after optimization are compared as follows: Total comfort cost before optimization After optimization The decrease was 28.6%; the total maintenance time before optimization was... After optimization The reduction was 8.0%; the minimum clearance distance before optimization. After optimization The safety margin has increased by 20.7%.

[0077] On the other hand, in one embodiment, a device for evaluating and optimizing the comfort of maintenance and disassembly paths is provided, comprising: The posture vector generation module is used to decompose the maintenance and disassembly task to be evaluated into a sequence of operation points, each operation point corresponding to a human posture node, and generate the posture vector of each human posture node. A multidimensional comfort cost function construction module is used to construct a multidimensional comfort cost function, taking into account at least the instantaneous comfort of a single-point posture, the path space safety during maintenance operations, and the cumulative effect of the duration of uncomfortable postures. The objective function construction module is used to combine the multidimensional comfort cost function, the total maintenance time cost of the maintenance and disassembly task to be evaluated, and the gap safety penalty, while introducing at least the constraints including accessibility constraints, joint limit constraints, and task logic constraints to construct the overall optimal objective function; The objective function solution and result output module is used to minimize the overall preferred objective function using a multi-objective optimization algorithm. Under the premise of satisfying the constraints, it outputs the Pareto optimal solution set from the candidate attitude sequence set, and the output result includes at least the preferred attitude sequence.

[0078] The specific implementation methods of the above modules can be referred to the corresponding implementation schemes in the foregoing method embodiments. The operations performed by each module correspond one-to-one with the steps of the foregoing method, and will not be repeated here. Those skilled in the art should understand that all or part of the steps in the foregoing method embodiments can be implemented by hardware modules, software modules, or a combination of hardware and software. The division of each module is not limited to the above functional division, and can be merged or split according to the actual system design requirements.

[0079] On the other hand, the present invention provides a computer device including a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the steps of the maintenance and disassembly path comfort evaluation optimization method provided in any of the above embodiments. The computer device may be a server. The computer device includes a processor, a memory, a network interface, and a database connected via a system bus. The processor of the computer device provides computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and computer program in the non-volatile storage medium. The database of the computer device stores sample data. The network interface of the computer device is used for communication with external terminals via a network connection.

[0080] On the other hand, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, it implements the steps of the preferred method for evaluating the comfort of maintenance disassembly and assembly paths provided in any of the above embodiments.

[0081] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0082] Matters not covered in this invention are common knowledge.

[0083] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0084] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these modifications and improvements all fall within the protection scope of this application.

[0085] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for evaluating and optimizing the comfort of maintenance and disassembly paths, characterized in that, Includes the following steps: The maintenance and disassembly task to be evaluated is decomposed into a sequence of operation points, each operation point corresponding to a human posture node. A posture vector is generated for each human posture node. Each human posture node contains two levels: main posture parameters and arm posture parameters. The main posture parameters include at least the following: trunk forward tilt angle, trunk lateral flexion angle, trunk twist angle, hip flexion angle, and knee flexion angle. The arm posture parameters include at least the following: shoulder flexion / extension angle, shoulder abduction / adduction angle, elbow flexion angle, wrist flexion angle, and wrist ulnar / radial deviation angle. Considering at least the immediate comfort of a single-point posture, the path space safety during maintenance operations, and the cumulative effect of the duration of uncomfortable postures, a multidimensional comfort cost function is constructed. This multidimensional comfort cost function includes at least several costs, such as the joint torque deviation cost, path clearance cost, and cumulative discomfort time cost for each human posture node. For each human posture node, the deviation between the real-time joint torque of each target joint and the comfortable torque range of each target joint is calculated based on the human inverse dynamics model, and the joint torque deviation cost of the human posture node is obtained by summarizing these values. Based on collision detection technology in a virtual maintenance environment, the minimum clearance distance between the maintenance personnel's human body envelope and surrounding equipment / structures is calculated, and the path clearance cost is constructed. Considering the comfort of a single-point posture and the duration of uncomfortable postures, the cumulative discomfort time cost is calculated. By combining the multidimensional comfort cost function, the total maintenance time cost of the maintenance and disassembly task to be evaluated, and the gap safety penalty, and by introducing at least the following constraints, including accessibility constraints, joint limit constraints, and task logic constraints, a total optimal objective function is constructed as follows: in, The total comfort cost of the posture sequence is obtained by weighted summation of the costs included in the multidimensional comfort cost function; The total maintenance time cost is the sum of the standard operating times at each operation point, and also includes the increment of posture adjustment time introduced based on the complexity of the transition between human posture nodes. As a penalty for gap safety, it is included as an independent item in the overall objective to strengthen safety constraints. , , These are the weighting coefficients for the total comfort cost, total maintenance time cost, and gap safety penalty term of the attitude sequence, respectively. A multi-objective optimization algorithm is used to minimize the overall optimal objective function. Under the premise of satisfying the constraints, a Pareto optimal solution set is output from the candidate attitude sequence set. The output result includes at least the preferred attitude sequence.

2. The method for evaluating and optimizing the comfort of maintenance and disassembly paths according to claim 1, characterized in that, The reachability constraint means that for each human posture node, the operator's operating point must be able to reach the target maintenance position; the joint limit constraint means that the angle of each joint in each human posture node must be within the physiologically permissible range of the joint; the task logic constraint means that the arrangement of the posture sequence must conform to the inherent logical order of the maintenance disassembly and assembly task, including disassembly sequence constraints, tool use constraints, and continuous operation constraints. The disassembly sequence constraint includes that obstructing parts must be removed before contacting the target part; the tool use constraint includes that the tool picking posture must be completed before using the tool and the tool returning posture must be completed after using the tool; the continuous operation constraint includes that the change in joint angle between adjacent human posture nodes must not exceed the product of the human's maximum angular velocity and the time between nodes.

3. The method for evaluating and optimizing the comfort of maintenance and disassembly paths according to claim 1, characterized in that, Gap safety penalty items as follows: in For the first The gap safety penalty for individual human posture nodes. Use an exponential penalty function: In the formula, To penalize the steepness coefficient, For the first The minimum clearance distance between the maintenance worker's human body envelope model corresponding to an individual human pose node and all obstacles in the environment. This is the threshold for dangerous gaps.

4. The method for evaluating and optimizing the comfort of maintenance and disassembly paths according to claim 1, characterized in that, The overall optimal objective function is solved using NSGA-II, MOPSO, or multi-objective simulated annealing algorithms.

5. A device for evaluating and optimizing the comfort of maintenance and disassembly routes, used to implement the method for evaluating and optimizing the comfort of maintenance and disassembly routes as described in any one of claims 1 to 4, characterized in that, include: The posture vector generation module is used to decompose the maintenance and disassembly task to be evaluated into a sequence of operation points, each operation point corresponding to a human posture node, and generate the posture vector of each human posture node. A multidimensional comfort cost function construction module is used to construct a multidimensional comfort cost function, taking into account at least the instantaneous comfort of a single-point posture, the path space safety during maintenance operations, and the cumulative effect of the duration of uncomfortable postures. The objective function construction module is used to combine the multidimensional comfort cost function, the total maintenance time cost of the maintenance and disassembly task to be evaluated, and the gap safety penalty, while introducing at least the constraints including accessibility constraints, joint limit constraints, and task logic constraints to construct the overall optimal objective function; The objective function solution and result output module is used to minimize the overall preferred objective function using a multi-objective optimization algorithm. Under the premise of satisfying the constraints, it outputs the Pareto optimal solution set from the candidate attitude sequence set, and the output result includes at least the preferred attitude sequence.

6. An electronic device comprising a memory and a processor, the memory storing a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the preferred method for evaluating the comfort of maintenance disassembly and assembly paths as described in any one of claims 1 to 4.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the preferred method for evaluating the comfort of maintenance disassembly and assembly paths as described in any one of claims 1 to 4.

Images