Method for constructing motion trail model of large forging manipulator

Abstract

The invention discloses a method for constructing a motion trail model of a large forging manipulator, and relates to the technical field of large forging equipment control. The method comprises the steps that a manipulator kinematics model is established based on an improved D-H parameter method, and model parameters are updated in real time through online calibration; constructing and analyzing a Jacobian matrix to identify and smoothly transit the singular configuration; performing attitude interpolation in the SE (3) space by adopting spherical linear interpolation or quaternion interpolation, and generating a smooth trajectory in combination with multi-objective optimization; a multi-source nonlinear constraint model covering machinery, driving, process and safety is established, and integrated decision making is carried out based on model predictive control; and simulating and predicting response by using finite elements and multi-body dynamics, and optimizing model parameters through Bayesian updating or a least square method based on actually measured data. According to the method, the tail end pose precision, the trajectory smoothness and the production efficiency are remarkably improved, and the problems of inaccurate modeling, sudden pose change, incomplete constraint consideration and lack of continuous optimization of a traditional method are solved.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of large-scale forging equipment control, and particularly relates to a large-scale forging manipulator motion trajectory model construction method. BACKGROUND

[0002] Large-scale forging manipulators are key equipment in modern heavy manufacturing, widely used in metallurgy, energy, aerospace and other fields. The motion trajectory quality of the manipulator directly affects the forging forming precision, surface quality and production efficiency. In recent years, with the development of automation technology, the motion control technology of forging manipulators has made significant progress.

[0003] In the aspect of forging manipulator trajectory planning, Zhongnan University disclosed an automatic forging manipulator and press linkage trajectory planning method in Chinese patent CN102091752B. The method designs a flat anvil elongation process scheme, plans the linkage trajectory of the manipulator and the press, and evaluates the rationality of the linkage trajectory by strain calculation and evaluation of deformation uniformity. This patent provides a method basis for trajectory planning of forging manipulators, but mainly focuses on process scheme design and trajectory rationality evaluation, and lacks attention to kinematic modeling accuracy and posture smooth transition.

[0004] In the aspect of manipulator motion modeling, the traditional D-H (Denavit-Hartenberg) parameter method is widely used in robot kinematic modeling. Lozano-Perez proposed a general manipulator motion planning algorithm based on configuration space in the paper "A Simple Motion-Planning Algorithm for General Robot Manipulators". Collision detection and path search are achieved through obstacle slice projection representation. However, the traditional D-H modeling method mainly considers ideal rigid link structure, and it is difficult to accurately describe complex factors such as joint flexibility, clearance and temperature drift in actual mechanical structure, which limits the modeling accuracy.

[0005] In the aspect of posture transition control, the existing technology mainly uses linear interpolation or simple angle interpolation method. Papadopoulos et al. proposed a trajectory planning method based on nonholonomic constraints for mobile manipulator systems in the paper "TRAJECTORY PLANNING AND CONTROL FOR MOBILE MANIPULATOR SYSTEMS", and designed a model controller to eliminate tracking errors. However, linear interpolation method is prone to discontinuity or mutation in posture space, especially in large-angle rotation, which may cause the manipulator motion to be not smooth, affecting the forging surface quality.

[0006] In terms of constraint processing, Zahroof et al. proposed a robot manipulator planning method under perception constraints in the paper "Perception-Constrained Robot Manipulator Planning for Satellite Servicing", which integrates sampling planning algorithms using the ROS MoveIt! framework and considers perception constraints such as field of view and line of sight. This method provides a reference for trajectory planning under constraints, but mainly targets the satellite service scenario and does not fully consider the comprehensive influence of multi-source nonlinear constraints such as mechanical structure constraints, drive system limitations, process requirements, and safety boundaries in the forging process.

[0007] In addition, the existing technology also has deficiencies in model verification. Although some studies use finite element analysis or dynamics simulation for verification, there is a lack of continuous improvement mechanism that combines simulation results with measured data and optimizes model parameters through feedback, making it difficult to continuously improve model accuracy as production practice accumulates.

[0008] In summary, the existing technology has the following main problems in modeling the motion trajectory of a large forging manipulator: (1) traditional D-H modeling methods cannot accurately describe the complexity of the actual mechanical structure, including joint flexibility, clearance, and temperature drift; (2) there are discontinuities or mutations in the pose transition process, affecting the quality of the forged piece; (3) path planning does not fully consider multi-source nonlinear constraints such as mechanical structure, drive system, process requirements, and safety boundaries; (4) there is a lack of effective model verification and correction mechanism, making it difficult to achieve continuous optimization of the model. These problems result in low forging precision, poor surface quality, and low production efficiency, restricting the further development of large-scale forging technology. SUMMARY

[0009] To solve the above problems, the purpose of the present application is to provide a large forging manipulator motion trajectory model construction method, which solves the problems of traditional D-H modeling method in existing technology that cannot accurately describe the complexity of the actual mechanical structure, discontinuity or mutation in the pose transition process affecting the quality of the forged piece, path planning not fully considering multi-source nonlinear constraints, and lack of effective model verification and correction mechanism.

[0010] To achieve the above purpose, the present application provides a large forging manipulator motion trajectory model construction method, comprising the following steps: Step S1: Establishing a manipulator kinematics model based on an improved D-H parameter method, which adds joint flexibility parameters, clearance compensation terms, and temperature drift coefficients to the existing D-H parameters, and performs online calibration through laser tracking or visual measurement to update the model parameters in real time; Step S2: Constructing an analytical Jacobian matrix based on the improved D-H model, identifying singular configurations through Jacobian determinant analysis, and achieving smooth transition of singular points through joint space optimization and path re-planning; Step S3: Representing the position and attitude of the end effector in SE(3) space, performing attitude interpolation using spherical linear interpolation SLERP or quaternion interpolation method, and generating a smooth motion trajectory combining time optimization and energy optimization criteria; Step S4: Establishing a multi-source nonlinear constraint model including mechanical structure constraints, drive system constraints, process requirement constraints, and safety boundary constraints, designing a path, trajectory, and attitude integrated decision framework based on model predictive control MPC, and solving the constraint optimization problem using interior point method or sequential quadratic programming SQP; Step S5: Predicting the deformation of the forging using finite element analysis, establishing a multi-body dynamics model of the manipulator to analyze dynamic response, and optimizing and correcting the joint flexibility parameters, gap compensation terms, and temperature drift coefficients based on simulation data and measured data using Bayesian update or least squares method.

[0011] The step S1 further comprises: establishing a parameter error propagation model, analyzing the influence weight of each parameter on the end position and attitude accuracy, and determining the parameter priority of online calibration according to the influence weight.

[0012] The singular point smoothing transition in step S2 is specifically: when the absolute value of the Jacobian determinant is less than a preset threshold, it is determined as a singular configuration, and the singular region is avoided by adding redundant degree of freedom constraints or adjusting the path point attitude in the joint space.

[0013] The calculation formula of spherical linear interpolation SLERP in step S3 is: q(t)=sin[(1-t)θ] / sinθ·q0+sin(tθ) / sinθ·q1 Where q0 and q1 are the quaternion representations of the initial and final attitudes, θ is the included angle between the two quaternions, and t is the interpolation parameter and t∈[0,1].

[0014] The multi-source nonlinear constraint model in step S4 specifically includes: joint angle constraint, joint velocity constraint, joint acceleration constraint, end effector workspace constraint, driving torque constraint, collision avoidance constraint, forging process temperature field constraint, and forging deformation coordination constraint.

[0015] In step S4, the decision framework based on model predictive control MPC sets the prediction time domain range to 5 to 20 sampling periods, and the control time domain range to 1 to 5 sampling periods, and realizes real-time adjustment of path, trajectory, and attitude through rolling optimization.

[0016] The model parameter optimization correction in the step S5 includes: collecting actual operation data of the manipulator, calculating a deviation of the end pose measured value from the model predicted value, starting a parameter optimization program when the deviation exceeds a set threshold, updating the joint flexibility parameters and the gap compensation term through a Bayesian updating method, and optimizing the temperature drift coefficient through a least square method.

[0017] A large-scale forging manipulator motion trajectory model construction system for implementing the method of any one of the preceding claims, comprising: a parameter modeling and calibration module for establishing a manipulator kinematics model and performing online calibration through laser tracking or visual measurement; a kinematics analysis module for constructing an analytical Jacobian matrix based on the manipulator kinematics model, identifying singular configurations and performing smooth transition processing; a trajectory planning module for generating a smooth motion trajectory in SE(3) space using spherical linear interpolation SLERP or quaternion interpolation methods; a constraint decision module for establishing the multi-source nonlinear constraint model and performing integrated path, trajectory and attitude decision based on model predictive control MPC; a simulation verification module for evaluating trajectory performance through finite element analysis and multi-body dynamics simulation; a model correction module for optimizing model parameters through Bayesian updating or least square method based on simulation and measured data; a trajectory execution module for converting the optimized trajectory into control instructions and issuing them to the actuators of the manipulator.

[0018] The parameter modeling and calibration module includes an error analysis unit for establishing a parameter error propagation model and analyzing the influence weight of each parameter on the end pose accuracy.

[0019] The constraint decision module uses an interior point method or a sequential quadratic programming SQP algorithm to solve the constraint optimization problem, and the optimization objective function comprehensively considers trajectory smoothness, motion time and energy consumption.

[0020] The beneficial effects of the present application are: 1. The present application can more accurately describe the complex characteristics of the actual mechanical structure by adding joint flexibility parameters, gap compensation terms and temperature drift coefficients based on the existing D-H parameters, significantly improving the end pose accuracy, thereby improving the forging forming precision. 2. The present application uses spherical linear interpolation SLERP or quaternion interpolation method in SE(3) space for attitude interpolation, avoiding the discontinuity or mutation problem of attitude transition in the existing method, ensuring the smoothness of the motion trajectory, and effectively improving the forging surface quality. 3. The present application comprehensively considers multi-source nonlinear constraints: a multi-source nonlinear constraint model including mechanical structure, drive system, process requirements and safety boundary is established, and integrated path, trajectory and attitude decision is made based on model predictive control MPC, which fully considers various constraint conditions in the actual forging process, improves the feasibility and safety of trajectory planning. 4. The present application analyzes the parameter influence weight by establishing a parameter error transmission model, and realizes adaptive optimization and long-term precision maintenance of model parameters by combining online calibration and model correction mechanism based on simulation and measured data, ensuring the stability and reliability of the system in the long-term operation process. 5. The present application shortens the forging cycle and improves the production efficiency while ensuring the trajectory quality through the comprehensive optimization of singular point smooth transition processing and time optimal and energy optimal criteria. BRIEF DESCRIPTION OF DRAWINGS

[0021] Figure 1 The flowchart of the present application is a large-scale forging manipulator motion trajectory model construction method.

[0022] Figure 2 The detailed flowchart of the present application is a large-scale forging manipulator motion trajectory model construction method. DETAILED DESCRIPTION

[0023] In order to make the content, technical scheme and advantages of the present application clearer, the present application will be further described in detail below according to specific embodiments of the present application, wherein: Example 1 As shown in Figure 1 , Figure 2 , the present application provides a large-scale forging manipulator motion trajectory model construction method, comprising the following steps: Step S1: establishing a manipulator kinematics model based on the improved D-H parameter method, the improved D-H parameter method is to add joint flexibility parameters, gap compensation terms and temperature drift coefficients based on the existing D-H parameters, and to perform online calibration through laser tracking or visual measurement to update the model parameters in real time; Step S2: constructing an analytical Jacobian matrix based on the improved D-H model, identifying singular configurations through Jacobian determinant analysis, and realizing smooth transition of singular points through joint space optimization and path re-planning; Step S3: In the SE(3) space, the position and attitude of the end effector are uniformly represented. The attitude is interpolated by spherical linear interpolation SLERP or quaternion interpolation method. The smooth motion trajectory is generated by combining the time-optimal and energy-optimal criteria. Step S4: Establish a multi-source nonlinear constraint model including mechanical structure constraints, drive system constraints, process requirement constraints, and safety boundary constraints. Based on the model predictive control (MPC) design path, trajectory, and attitude integrated decision framework, use the interior point method or sequential quadratic programming (SQP) to solve the constraint optimization problem. Step S5: Predict the deformation of the forging using finite element analysis, establish a multibody dynamics model of the manipulator to analyze the dynamic response, and optimize and correct the joint flexibility parameters, clearance compensation term and temperature drift coefficient based on simulation data and measured data using Bayesian update or least squares method.

[0024] Step S1 further includes: establishing a parameter error propagation model, analyzing the influence weight of each parameter on the end pose accuracy, and determining the parameter priority for online calibration based on the influence weight.

[0025] The singularity smooth transition in step S2 specifically involves: when the absolute value of the Jacobian determinant is less than a preset threshold, it is determined to be a singular configuration, and singular region avoidance is achieved by adding redundant degree-of-freedom constraints in the joint space or adjusting the orientation of the path points.

[0026] The formula for calculating the spherical linear interpolation SLERP in step S3 is as follows: q(t)=sin[(1-t)θ] / sinθ·q0+sin(tθ) / sinθ·q1 Where q0 and q1 are quaternion representations of the initial and final attitudes, θ is the angle between the two quaternions, and t is the interpolation parameter and t∈[0,1].

[0027] The multi-source nonlinear constraint model in step S4 specifically includes: joint angle constraint, joint velocity constraint, joint acceleration constraint, end effector workspace constraint, driving torque constraint, collision avoidance constraint, forging process temperature field constraint, and forging deformation coordination constraint.

[0028] In step S4, the decision framework based on model predictive control (MPC) sets the prediction time domain range to 5 to 20 sampling periods and the control time domain range to 1 to 5 sampling periods, and achieves real-time adjustment of path, trajectory and attitude through rolling optimization.

[0029] The model parameter optimization and correction in step S5 includes: collecting actual operating data of the manipulator, calculating the deviation between the measured end pose value and the model prediction value, starting the parameter optimization program when the deviation exceeds a set threshold, updating the joint flexibility parameters and gap compensation terms through the Bayesian update method, and optimizing the temperature drift coefficient through the least squares method.

[0030] A system for constructing motion trajectory models of large forging manipulators, used to implement the method described in any of the above-mentioned embodiments, comprising: The parameter modeling and calibration module is used to establish the kinematic model of the manipulator and perform online calibration through laser tracking or visual measurement. The kinematics analysis module is used to construct an analytical Jacobian matrix based on the kinematic model of the manipulator, identify singular configurations, and perform smooth transition processing. The trajectory planning module is used to generate smooth motion trajectories in SE(3) space using spherical linear interpolation SLERP or quaternion interpolation methods; The constraint decision module is used to establish the multi-source nonlinear constraint model and make integrated decisions on path, trajectory and attitude based on model predictive control (MPC). The simulation verification module is used to evaluate trajectory performance through finite element analysis and multibody dynamics simulation. The model correction module is used to optimize model parameters based on simulation and measured data using Bayesian updates or least squares methods. The trajectory execution module is used to convert the optimized trajectory into control commands and send them to the actuators of the manipulator.

[0031] The parameter modeling and calibration module includes an error analysis unit, which is used to establish a parameter error propagation model and analyze the influence weight of each parameter on the end pose accuracy.

[0032] The constraint decision module uses the interior point method or the sequential quadratic programming (SQP) algorithm to solve the constraint optimization problem. Its optimization objective function comprehensively considers trajectory smoothness, motion time, and energy consumption.

[0033] Example 2 Using the motion trajectory model construction method of a large forging manipulator as described in Example 1, this example details the implementation process of the improved DH parameter method based on trajectory planning using the improved DH parameter method and SLERP interpolation.

[0034] First, based on the existing DH parameter method, a joint flexibility parameter (stiffness coefficient k = 1.2 × 10⁻⁶) is added. 6 N·m / rad), gap compensation term (δ=0.05 mm), and temperature drift coefficient (α=1.5×10). -5 / °C); Online calibration was performed using a laser tracker (Leica AT960) with a sampling frequency of 100 Hz; A parameter error propagation model was established to analyze the influence weight of each parameter on the end pose accuracy, and the calibration priority was determined as: joint flexibility parameters > clearance compensation term > temperature drift coefficient; An analytical Jacobian matrix is ​​constructed based on the improved DH model. When the absolute value of the Jacobian determinant is less than 0.001, it is determined to be a singular configuration. This can be avoided by adding redundant degrees of freedom constraints in the joint space. Quaternion interpolation is performed in the SE(3) space using the SLERP formula: q(t)=sin[(1-t)θ] / sinθ·q0+sin(tθ) / sinθ·q1, where θ=30°, t∈[0,1]; A multi-source nonlinear constraint model was established, including joint angle constraints (±90°), joint velocity constraints (0.5 rad / s), and end-effector workspace constraints (2m×2m×1m). Based on the MPC framework, the prediction time domain was set to 10 sampling periods, and the control time domain was set to 3 sampling periods. The deformation of the forging was predicted by finite element analysis (maximum deformation 0.8 mm), and the dynamic response was analyzed by establishing a multibody dynamics model (maximum vibration amplitude 0.02 mm). Based on the measured data, parameter optimization was initiated when the end pose deviation exceeded 0.1 mm, and the joint flexibility parameters and clearance compensation terms were updated by Bayesian update method.

[0035] Technical results: End-effector pose accuracy reaches ±0.08 mm (compared to ±0.15 mm using traditional methods), attitude transition smoothness is improved by 40%, singularity avoidance success rate is 100%, and computational efficiency is improved by 25%. Example 3 Using the motion trajectory model construction method for a large forging manipulator as described in Example 1, this example employs trajectory planning based on quaternion interpolation and SQP optimization. Quaternion interpolation is used instead of SLERP in this example, with other conditions remaining the same as in Example 2. The quaternion interpolation parameters are set to θ = 45°, t ∈ [0, 1]. Constraint optimization uses the Sequential Quadratic Programming (SQP) algorithm, with the objective function comprehensively considering trajectory smoothness (weight 0.4), motion time (weight 0.3), and energy consumption (weight 0.3). Forging process temperature field constraints (800-1200°C) and deformation compatibility constraints (maximum allowable deformation 1 mm) are added to the multi-source constraints. The MPC framework is set to a prediction time domain of 15 sampling periods and a control time domain of 5 sampling periods.

[0036] Technical Results: Under complex process constraints, the trajectory planning success rate is increased to 98% (95% in Example 1), energy consumption is reduced by 15%, and temperature field stability is improved by 20%. Example 4 Using the motion trajectory model construction method for a large forging manipulator as described in Example 1, trajectory optimization under high load conditions is performed. This example is specifically designed for high-load conditions of large forgings weighing over 50 tons. The joint flexibility parameter is adjusted to 2.0 × 10⁻⁶. 6 The N·m / rad interval compensation term is increased to 0.1 mm. The singularity threshold is relaxed to 0.005 to accommodate larger deformations. The MPC prediction time domain is shortened to 5 sampling periods to improve real-time performance. A hybrid optimization strategy is adopted: interior point method is used in the coarse planning stage, and SQP is used in the fine planning stage.

[0037] Technical results: Under a 50-ton load, the system stability remains above 95%, the trajectory tracking error is controlled within ±0.15mm, and the real-time calculation delay is <50 ms.

[0038] Actual test of the method of the present invention: Experimental conditions: using the parameter settings of Example 2, the test platform is a 20MN forging manipulator, the forging material is Ti-6Al-4V, the test temperature is 1000°C, and the experimental results are: end pose accuracy ±0.08 mm, attitude transition time 0.8 s, energy consumption 85 kJ, and calculation time 12 s.

[0039] Test using the existing DH parameter method: Experimental conditions: Same test platform and operating conditions, using the existing standard DH parameter method, Euler angle attitude interpolation, and simple constraint considerations. Experimental results: End-effector pose accuracy ±0.15 mm, attitude transition time 1.2 s, energy consumption 100 kJ, computation time 16 s.

[0040] This invention, by adding joint flexibility parameters, clearance compensation terms, and temperature drift coefficients to the existing DH parameters, can more accurately describe the complex characteristics of actual mechanical structures, significantly improve end-effector pose accuracy, and thus enhance forging forming accuracy.

[0041] Obviously, the specific embodiments described above are only some of the embodiments of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the protection scope of the present invention.

Claims

1. A method for constructing a motion trajectory model of a large forging manipulator, characterized in that: Includes the following steps: Step S1: Establish a kinematic model of the manipulator based on the improved DH parameter method. The improved DH parameter method adds joint flexibility parameters, clearance compensation terms and temperature drift coefficients to the existing DH parameters, and performs online calibration through laser tracking or visual measurement to update the model parameters in real time. Step S2: Construct an analytical Jacobian matrix based on the kinematic model of the manipulator, identify singular configurations through Jacobian determinant analysis, and achieve smooth transition of singular points through joint space optimization and path replanning; Step S3: In the SE(3) space, the position and attitude of the end effector are uniformly represented. The attitude is interpolated by spherical linear interpolation SLERP or quaternion interpolation method. The smooth motion trajectory is generated by combining the time-optimal and energy-optimal criteria. Step S4: Establish a multi-source nonlinear constraint model including mechanical structure constraints, drive system constraints, process requirement constraints, and safety boundary constraints. Based on the model predictive control (MPC) design path, trajectory, and attitude integrated decision framework, use the interior point method or sequential quadratic programming (SQP) to solve the constraint optimization problem. Step S5: Predict the deformation of the forging using finite element analysis, establish a multibody dynamics model of the manipulator to analyze the dynamic response, and optimize and correct the joint flexibility parameters, clearance compensation term and temperature drift coefficient based on simulation data and measured data using Bayesian update or least squares method.

2. The method for constructing a motion trajectory model of a large forging manipulator according to claim 1, characterized in that: Step S1 further includes: establishing a parameter error propagation model, analyzing the influence weight of each parameter on the end pose accuracy, and determining the parameter priority for online calibration based on the influence weight.

3. The method for constructing a motion trajectory model of a large forging manipulator according to claim 1, characterized in that: The singularity smooth transition in step S2 specifically involves: when the absolute value of the Jacobian determinant is less than a preset threshold, it is determined to be a singular configuration, and singular region avoidance is achieved by adding redundant degree-of-freedom constraints in the joint space or adjusting the orientation of the path points.

4. The method for constructing a motion trajectory model of a large forging manipulator according to claim 1, characterized in that: The formula for calculating the spherical linear interpolation SLERP in step S3 is as follows: q(t)=sin[(1-t)θ] / sinθ·q0+sin(tθ) / sinθ·q1 Where q0 and q1 are quaternion representations of the initial and final attitudes, θ is the angle between the two quaternions, and t is the interpolation parameter and t∈[0,1].

5. The method for constructing a motion trajectory model of a large forging manipulator according to claim 1, characterized in that: The multi-source nonlinear constraint model in step S4 specifically includes: joint angle constraint, joint velocity constraint, joint acceleration constraint, end effector workspace constraint, driving torque constraint, collision avoidance constraint, forging process temperature field constraint, and forging deformation coordination constraint.

6. The method for constructing a motion trajectory model of a large forging manipulator according to claim 1, characterized in that: In step S4, the decision framework based on model predictive control (MPC) sets the prediction time domain range to 5 to 20 sampling periods and the control time domain range to 1 to 5 sampling periods, and achieves real-time adjustment of path, trajectory and attitude through rolling optimization.

7. The method for constructing a motion trajectory model of a large forging manipulator according to claim 1, characterized in that: The model parameter optimization and correction in step S5 includes: collecting actual operating data of the manipulator, calculating the deviation between the measured end pose value and the model prediction value, starting the parameter optimization program when the deviation exceeds a set threshold, updating the joint flexibility parameters and gap compensation terms through the Bayesian update method, and optimizing the temperature drift coefficient through the least squares method.

8. A motion trajectory model construction system for a large forging manipulator, used to implement the method according to any one of claims 1-7, characterized in that, include: The parameter modeling and calibration module is used to establish the kinematic model of the manipulator and perform online calibration through laser tracking or visual measurement. The kinematics analysis module is used to construct an analytical Jacobian matrix based on the kinematic model of the manipulator, identify singular configurations, and perform smooth transition processing. The trajectory planning module is used to generate smooth motion trajectories in SE(3) space using spherical linear interpolation SLERP or quaternion interpolation methods; The constraint decision module is used to establish the multi-source nonlinear constraint model and make integrated decisions on path, trajectory and attitude based on model predictive control (MPC). The simulation verification module is used to evaluate trajectory performance through finite element analysis and multibody dynamics simulation. The model correction module is used to optimize model parameters based on simulation and measured data using Bayesian updates or least squares methods. The trajectory execution module is used to convert the optimized trajectory into control commands and send them to the actuators of the manipulator.

9. The motion trajectory model construction system for a large forging manipulator according to claim 8, characterized in that: The parameter modeling and calibration module includes an error analysis unit, which is used to establish a parameter error propagation model and analyze the influence weight of each parameter on the end pose accuracy.

10. The motion trajectory model construction system for a large forging manipulator according to claim 8, characterized in that: The constraint decision module uses the interior point method or the sequential quadratic programming (SQP) algorithm to solve the constraint optimization problem. Its optimization objective function comprehensively considers trajectory smoothness, motion time, and energy consumption.

Images