Dynamic model residual error compensation robot collision detection method and system considering frequency distribution difference

Abstract

The invention belongs to the technical field of robot collision detection, and discloses a kinetic model residual error compensation robot collision detection method considering frequency distribution difference, which comprises the following steps of: acquiring data through a random track, and calculating according to the acquired data to obtain input characteristic data for model training; the method comprises the following steps of: introducing time sequence information into input data in a sliding window manner to perform data enhancement processing, and inputting a residual signal into a high-frequency separation adaptive filtering framework to obtain an output target value of model training; then, an XGBoost algorithm is adopted for training, and a high-frequency residual prediction model and a low-frequency residual prediction model are trained respectively; and inputting a prediction result of the trained residual prediction model into a momentum observer in a feedforward form to obtain a compensated joint external torque signal. According to the method, the kinetic model residual contained in the joint external torque signal estimated by the momentum observer can be effectively compensated, so that the collision detection precision of the robot is improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of robot collision detection, and in particular relates to a robot collision detection method and system with dynamic model residual compensation considering frequency distribution differences. Background Art

[0002] In recent years, collaborative robots have played an important role in manufacturing, logistics, and medical fields with their flexibility and efficiency. Among them, human-robot collaboration can significantly improve the efficiency and accuracy of completing tasks. In order to ensure the safety of humans and robots when performing tasks together in the same space, robots must have collision detection capabilities. At present, collision detection methods based on robot dynamic models are widely used because they only require information from the robot's body sensors. Among them, the robot collision detection method based on momentum observer is the most classic and widely used method in this category. It estimates the external torque of the robot's joints directly through momentum, which can avoid the introduction of acceleration signals and the inversion of the inertia matrix. However, the accuracy of its external torque observation depends on the accuracy of the robot's dynamic model. Inaccurate dynamic models and dynamic parameters will seriously affect the robot's collision detection accuracy. Therefore, achieving high-precision external torque estimation under inaccurate modeling is the main research trend to improve the accuracy of robot collision detection.

[0003] Through the above analysis, the problems and defects of the prior art are as follows:

[0004] (1) It is difficult to obtain a perfect robot dynamics model and dynamics parameters only through mechanism modeling. The momentum observer constructed based on inaccurate dynamics model and parameters will inevitably contain dynamics model residuals in the estimated joint external torque, which will affect the accuracy of collision detection.

[0005] (2) The existing robot joint friction force model is difficult to fit the sudden change of friction force caused by joint switching, which will seriously affect the external torque estimation accuracy of the momentum observer.

[0006] (3) There are more and more studies on residual prediction compensation of robot dynamics models, but most of the current research on residual prediction compensation focuses on directly using machine learning or deep learning to predict and compensate the collected residuals, while ignoring the influence of measurement noise. Direct filtering will cause the commutation error of friction force to be distorted, which will seriously affect the compensation effect.

[0007] The defects and difficulties in solving the above problems are:

[0008] The residuals of the dynamic model cannot be obtained analytically, and their sources are complex and difficult to accurately describe using traditional modeling methods. The residuals not only include errors caused by unmodeled dynamic effects, but are also affected by factors such as friction nonlinearity and measurement noise. Therefore, how to construct an algorithm that can effectively predict and compensate for the residuals of the dynamic model based on the different characteristics of the residuals from various sources, thereby improving the reliability and sensitivity of collision detection, is the difficulty in solving this problem.

[0009] The significance of solving the above problems and defects is:

[0010] By constructing an algorithm that can effectively predict and compensate for the residual errors of the dynamic model, a new solution can be provided for improving the accuracy and reliability of robot collision detection. Effectively solving this problem can not only improve the safety and stability of robots in complex tasks, but also promote more advanced dynamic model residual compensation methods in robot control systems. In addition, this method can also be applied to robot friction compensation, external force detection and other fields through conversion. Summary of the invention

[0011] In view of the problems existing in the prior art, the present invention provides a robot collision detection method for dynamic model residual compensation taking into account frequency distribution differences.

[0012] The present invention is implemented as follows: a dynamic model residual compensation robot collision detection method considering frequency distribution differences comprises:

[0013] S1. Decouple the theoretical torque of the robot joints based on the Newton-Euler method and its variants to obtain the robot's momentum, Coriolis force, centrifugal force and gravity components;

[0014] S2. Solve the external torque expression output by the momentum observer through Laplace forward and inverse transform, and express the residual expression of the joint external torque accordingly;

[0015] S3. Design a feature vector based on the residual expression; extract the relevant sub-expressions of each component in the residual expression as dynamic features, and preliminarily form a feature vector with the kinematic features of the robot;

[0016] S4, build a high-frequency separation adaptive filtering training framework;

[0017] According to the differences in frequency distribution of different components of the residual signal, the residual signal is split into high-frequency and low-frequency signals through a low-pass filter for prediction;

[0018] For the training of high-frequency residual prediction model, the commutation features of robot joint friction force are constructed, and an adaptive filtering framework is built based on it;

[0019] S5, collect training data through random trajectories, introduce time series information into the training data through sliding windows for data enhancement processing, and input the residual signal into the high-frequency separation adaptive filtering framework to obtain the output target value of the model training; then use XGB oos t algorithm is used for training to obtain high-frequency and low-frequency residual prediction models respectively;

[0020] S6. The prediction result of the trained residual prediction model is input into the momentum observer in a feedforward form to obtain the compensated joint external torque signal, and then collision detection is performed based on the set threshold.

[0021] Furthermore, in step S1, based on the Newton-Euler algorithm and its variants, the expression of the theoretical joint torque of the robot can be obtained as follows:

[0022]

[0023] Among them, τ i represents the theoretical torque of the robot's i-th joint; Respectively represent the angle, velocity and acceleration of the robot's ith joint; NE α (·) indicates the standard Newton-Euler algorithm; MNE α (·) represents a variant of the Newton-Euler algorithm, which splits the joint velocity into two different inputs to obtain additional degrees of freedom, so that the output matrix can automatically satisfy the partial symmetry property; α∈{g0,0} indicates whether the gravity effect is considered in the Newton-Euler algorithm;

[0024] As a further preferred embodiment, the theoretical torque can be decoupled based on the Newton-Euler algorithm and its variants to obtain the input items of the subsequent momentum observer, which include the momentum of the robot, the Coriolis force and the centrifugal force, and the gravity component, which can be calculated by the following formula:

[0025]

[0026] Among them, β is a dynamic variable defined for simplified representation, g(q) represents the gravity vector of the robot, is the transpose of the Coriolis force and centrifugal force matrix of the robot, M(q) represents the inertia matrix of the robot, and p is the generalized momentum of the robot; the vector e n =(0 … 0 1 0 … 0) T , the nth element is 1 and the rest are 0.

[0027] Further, the step S2 specifically includes:

[0028] First, determine the expression of the momentum observer as follows:

[0029] r(t)=K0[p(t)-∫(τ m -τ f +r+β)dt]

[0030] Where r(t) represents the external joint torque output by the momentum observer at the tth moment, K0 represents the diagonal gain matrix of the momentum observer, τ m represents the motor driving torque, τ f is the calculated friction force;

[0031] Taking into account factors such as dynamic modeling error, friction error, and sensor measurement error, it is impossible to obtain a completely accurate external torque monitoring signal; incorporating the above error factors into the derivation process, the calculation formula in the actual situation can be expressed as follows:

[0032]

[0033] in, Indicates that the variable is an estimated version based on kinetic model calculations or actual readings;

[0034] By solving the above differential equations through Laplace forward and inverse transform, we can get the expression of the external torque output by the momentum observer as follows:

[0035]

[0036] In the two formulas, * represents the convolution operation; then the residual of the external torque monitoring value can be defined as follows:

[0037]

[0038] The variables in the above formula are defined as follows:

[0039]

[0040] Further, the step S3 specifically includes:

[0041] According to the defined residual expression, the expression related to the robot friction modeling error and the expression of the dynamic modeling error can be extracted as follows:

[0042]

[0043] Among them, h fe (t) represents the part of the residual that comes from the friction modeling error, Respectively The part related to friction modeling; e (t) represents the part of the residual that comes from the dynamic modeling error, and h e(t) can be further decomposed into the error parts related to momentum, Coriolis force, centrifugal force and gravity, which are expressed as follows:

[0044]

[0045] From the above formula, we can see that the residual r e Depend on and {h p ,h g ,h C ,h f}decision; among which, It can be calculated by the dynamic model and the friction model; therefore, it can be used as the feature vector input of the subsequent XGBoost, and {h p ,h g ,h C ,h f} is fitted through XGBoost training; the kinematic characteristics of the robot are added on the basis of the dynamic characteristics, and the feature vector of the dynamic residual prediction model can be initially extracted into the following form:

[0046]

[0047] Further, the step S4 specifically includes:

[0048] According to the differences in frequency distribution of different components of the residual signal, the residual signal can be split into high-frequency signal and low-frequency signal through a low-pass filter, which can be expressed as follows:

[0049]

[0050] r high =r e -r low

[0051] Wherein, LPF represents low-pass filter; r low Represents the low-frequency residual signal. The errors of the dynamic model (such as momentum and gravity errors) are concentrated in the low-frequency signal. high Represents high-frequency residual signal, measurement noise and friction commutation error are concentrated in the high-frequency signal; its essence is to further decouple the residual from the perspective of frequency;

[0052] The adaptive filtering framework specifically includes:

[0053] First, the commutation characteristics of the robot joint friction force are constructed as shown below:

[0054]

[0055] Among them, k represents the kth moment, ns represents the length of the friction mutation gap window, n a Indicates the speed reverse look-ahead window length;

[0056] According to the commutation characteristics, the high-frequency residual signal can be adaptively filtered to reduce the measurement noise contained therein while retaining the friction commutation error, so as to increase the fitting accuracy of the model; it can be expressed as follows:

[0057]

[0058] Among them, r df (k) is the high-frequency residual signal after adaptive filtering at the kth moment, r filter (k) is the high-frequency residual signal after low-pass filtering;

[0059] According to the differentiated characteristics of the frequency distribution of different components of the residual signal and the constructed friction commutation features, the extracted feature vector can be further extracted into the feature vector corresponding to the high-frequency residual and the feature vector corresponding to the low-frequency residual, which are specifically in the following form:

[0060]

[0061] Among them, feature low Represents low-frequency feature vectors, including the kinematic characteristics of the robot and related features of the dynamic model; feature high Represents high-frequency feature vectors, including the robot's kinematic characteristics, friction error terms, and friction commutation characteristics.

[0062] Further, the step S5 specifically includes:

[0063] First, let the robot run 20 randomly generated trajectories to collect the robot's angle, velocity and acceleration data, and perform low-pass filtering on the acceleration signal to filter out the measurement noise it contains; the collected data can be used according to steps S3 and S4 to obtain the input and output data sets for model training, as shown in the following formula:

[0064]

[0065] Among them, X low，i represents the low-frequency feature dataset of the robot’s i-th joint, Y low，i represents the low-frequency residual target value of the robot's i-th joint; X high，i It represents the high-frequency feature dataset of the i-th joint, Y high，i represents the high-frequency residual target value of the robot's i-th joint; n represents the data collected at n moments in total;

[0066] The main idea of ​​the data enhancement method based on sliding window is: define a sliding window with a length of L, which contains the time series feature data of the period [t-L+1, t], and use the time series feature data of the sliding window as the feature data of the current moment as the model input to achieve the effect of data enhancement; it is worth noting that if there is no data before the moment t, the missing data will be filled with 0; at the same time, the length of the window L can be selected according to different processing situations, so as to flexibly adjust the length of the retained historical time series information according to the actual situation; the feature data set at the moment t can be expressed as follows:

[0067]

[0068] X low,i (t) represents the low-frequency feature dataset of the robot’s ith joint at time t, X high,i (t) represents the high-frequency feature data set of the robot's ith joint at the tth moment; further, the overall feature data set of the ith joint can be rewritten as follows:

[0069]

[0070] Another object of the present invention is to provide a robot collision detection system with dynamic model residual compensation considering frequency distribution differences, comprising:

[0071] An acquisition module is used to decouple the theoretical torque of the robot joints based on the Newton-Euler method and its variants to obtain the momentum, Coriolis force, centrifugal force and gravity component of the robot;

[0072] A solution module is used to solve the external torque expression output by the momentum observer through Laplace forward and inverse transform, and to express the residual expression of the joint external torque accordingly;

[0073] The extraction module is used to design the feature vector according to the residual expression; the relevant sub-expressions of each component in the residual expression are extracted as dynamic features, and the feature vector is preliminarily formed with the kinematic features of the robot;

[0074] A construction module is used to construct a high-frequency separation adaptive filtering training framework; according to the differences in frequency distribution of different components of the residual signal, the residual signal is split into high-frequency and low-frequency signals through a low-pass filter for prediction; for the training of the high-frequency residual prediction model, the commutation characteristics of the robot joint friction force are constructed, and the adaptive filtering framework is constructed accordingly;

[0075] The acquisition module is used to collect training data through random trajectories, introduce time series information into the training data through a sliding window to perform data enhancement processing, and input the residual signal into the high-frequency separation adaptive filtering framework to obtain the output target value of the model training; then the XGBoost algorithm is used for training to obtain high-frequency and low-frequency residual prediction models respectively;

[0076] The collision detection module is used to input the prediction results of the trained residual prediction model into the momentum observer in a feedforward form to obtain the compensated joint external torque signal, and then perform collision detection based on the set threshold.

[0077] Another object of the present invention is to provide a computer device, comprising a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of the robot collision detection method with dynamic model residual compensation considering frequency distribution differences.

[0078] Another object of the present invention is to provide a computer-readable storage medium storing a computer program, which, when executed by a processor, enables the processor to execute the steps of the robot collision detection method with dynamic model residual compensation considering frequency distribution differences.

[0079] Another object of the present invention is to provide an information data processing terminal, which is used to implement the dynamic model residual compensation robot collision detection system that takes into account frequency distribution differences.

[0080] In combination with the above technical solutions and the technical problems solved, the advantages and positive effects of the technical solutions to be protected by the present invention are as follows:

[0081] First, the present invention proposes a robot collision detection method for dynamic model residual compensation considering frequency distribution differences. The main idea is to design feature vectors based on the constructed residual expression, and then establish a residual prediction model through XGBoost training to predict the dynamic residual contained in the momentum observer monitoring signal. In the model training process, according to the frequency distribution of the residual signal, a high-frequency separation adaptive filtering training framework is proposed to further improve the prediction ability of the model. Finally, the predicted dynamic model residual is feed-forward compensated to the momentum observer, thereby improving the collision detection accuracy of the momentum observer.

[0082] 1. This application proposes a dynamic model residual compensation method for a momentum observer. By predicting the dynamic model residual through machine learning and feedforwarding it into the momentum observer, it can effectively reduce the impact of model errors and dynamic parameter errors and improve the robot's collision detection accuracy.

[0083] 2. This application decouples the theoretical torque through the Newton-Euler method and its variants, and solves the residual expression contained in the momentum observer output signal through the Laplace forward and inverse transform, and extracts the eigenvectors related to the dynamic model and the friction model. A more representative eigenvector is constructed as the model input to improve the prediction ability of the model.

[0084] 3. This application decouples the residual signal from the frequency perspective by considering the differences in the frequency distribution of different components of the residual signal. At the same time, for the influence of noise contained in the high-frequency signal, the friction commutation feature is constructed to establish an adaptive filtering framework to reduce the influence of friction on the model accuracy in the low-speed commutation stage, thereby improving the prediction accuracy of friction mutation error.

[0085] Second, as auxiliary evidence of the inventiveness of the claims of the present invention, it is also reflected in the following important aspects:

[0086] (1) The expected benefits and commercial value of the technical solution of the present invention after transformation are:

[0087] By combining machine learning methods with momentum observers, the application of the present invention in the field of robot collision detection has significant commercial value. This technology can not only significantly improve the safety performance of the robot, but also reduce the dependence on external sensors in practical applications, reducing hardware costs. At the same time, the high-precision detection capability of this solution can be applied to multiple industries, especially in the fields of automated production lines, service robots and medical robots, bringing higher economic benefits to enterprises. In addition, the present invention can reduce the false detection and missed detection rates, improve the efficiency and stability of the production line, and create more benefits for enterprises.

[0088] (2) The technical solution of the present invention fills the technical gap in the industry at home and abroad:

[0089] At present, in the field of robot collision detection at home and abroad, the conventional momentum observer method is limited by the accuracy of the dynamic model, resulting in the detection accuracy being difficult to meet the needs of practical applications. The present invention uses machine learning technology to consider the frequency distribution differences of the dynamic model residuals, successfully predicts and compensates for the dynamic model residuals, and innovatively solves this problem. In terms of the processing of friction mutation errors, traditional models cannot effectively deal with the nonlinear friction effects in the low-speed commutation stage, and the present invention fills this technical gap through a high-frequency adaptive filtering training framework. This technology has demonstrated excellent adaptability and application potential in practical robot applications.

[0090] Third, the present invention addresses the deficiencies of existing methods in robot collision detection technology and solves technical problems such as the accuracy of collision detection is limited by the accuracy of the dynamic model, the friction modeling error is difficult to predict, and the measurement noise has an adverse effect on model training. In traditional methods, due to the uncertainty of the dynamic model, a large amount of external torque residuals will be contained in the monitoring signal, which will affect the accuracy of collision detection, and the compensation for friction modeling errors has certain limitations, resulting in a high false alarm rate. The present invention constructs a method for compensating for external torque residuals that takes frequency distribution into consideration, fully utilizes the frequency distribution information of external torque residuals, separates the residual signal, and enhances the prediction accuracy of the model, thereby further improving the accuracy of collision detection and effectively improving the stability of detection. .

[0091] The technical progress of the present invention in industrial applications also lies in the use of the XGBoost algorithm combined with the designed high- and low-frequency adaptive filtering training framework for residual prediction, which overcomes the problem of insufficient extraction of input and output mapping relationships in traditional methods. The XGBoost algorithm is used to independently model and predict the separated high-frequency and low-frequency residual signals, which not only enhances the model's prediction accuracy for friction modeling errors, but also reduces the adverse effects of measurement noise on prediction accuracy. This high-frequency separated adaptive filtering training framework can more accurately predict the changes in the external torque residuals of the robot during normal motion, providing an effective solution for robot collision detection in complex industrial scenarios.

[0092] In addition, the present invention has achieved significant improvements in detection accuracy and response time. Since traditional collision detection schemes are mostly limited by the accuracy of the dynamic model, it is difficult to achieve high-precision detection under the uncertainty factors of the dynamic model, so their application in industrial automation is limited. Through frequency separation and residual compensation, the present invention not only realizes the effective processing of complex external torque residuals, but also realizes high-precision prediction of friction modeling errors in the high-frequency adaptive filtering training framework, greatly improving the sensitivity and response speed of the detection system. This technological advancement has significantly improved the safety and reliability of robots in industrial operations.

[0093] Finally, the present invention innovatively introduces a residual compensation feedforward model for the momentum observer, by predicting high and low frequency residual signals respectively and feedforwarding them into the momentum observer. This method can dynamically compensate for the residual of the robot's external torque in real time, greatly reducing the false alarm and missed alarm rates, and providing a more intelligent and accurate collision detection method for complex working condition detection in industrial robot applications. Compared with the prior art, the present invention provides a more intelligent and efficient solution, significantly improving the application value and adaptability of robots in industrial operations. BRIEF DESCRIPTION OF THE DRAWINGS

[0094] Figure 1is a flow chart of a robot collision detection method with dynamic model residual compensation considering frequency distribution differences provided by an embodiment of the present invention;

[0095] Figure 2 It is a schematic diagram of a high-frequency separation adaptive filtering training framework provided by an embodiment of the present invention;

[0096] Figure 3 is a schematic diagram of a collision detection process including a residual compensation module provided in an embodiment of the present invention;

[0097] Figure 4 is a schematic diagram of the effect of performing dynamic residual compensation on the external torque signal of the robot joint provided by an embodiment of the present invention;

[0098] Figure 5 is a schematic diagram of a robot hard collision detection experiment provided by an embodiment of the present invention;

[0099] Figure 6 is a schematic diagram of a robot soft collision detection experiment (balloon experiment) provided by an embodiment of the present invention, wherein (a) is an estimated value of the external torque of joint 2 before compensation, (b) is an estimated value of the external torque of joint 2 after compensation, (c) is the terminal Z-axis force before compensation, and (d) is the terminal Z-axis force after compensation;

[0100] Figure 7 is a schematic diagram of a balloon experiment platform provided by an embodiment of the present invention;

[0101] Figure 8 It is a structural block diagram of a robot collision detection system with residual compensation of a dynamic model taking into account frequency distribution differences provided by an embodiment of the present invention.

[0102] Fig. 9 It is a rendering of the commutation characteristics of the friction force and the adaptive filtering method provided by the example of the present invention;

[0103] Fig.10 is a schematic diagram of a data enhancement method based on a sliding window provided by an example of the present invention;

[0104] Figure 7 In: 1. Robot body; 2. Six-dimensional force sensor; 3. Balloon. DETAILED DESCRIPTION

[0105] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0106] like Figure 1As shown, a robot collision detection method for dynamic model residual compensation considering frequency distribution difference provided by an embodiment of the present invention comprises the following steps:

[0107] S1. Decouple the theoretical torque of the robot joints based on the Newton-Euler method and its variants to obtain the robot's momentum, Coriolis force, centrifugal force and gravity components;

[0108] S2. Solve the external torque expression output by the momentum observer through Laplace forward and inverse transform, and express the residual expression of the joint external torque accordingly;

[0109] S3. Design a feature vector based on the residual expression; extract the relevant sub-expressions of each component in the residual expression as dynamic features, and preliminarily form a feature vector with the kinematic features of the robot;

[0110] S4, build a high-frequency separation adaptive filtering training framework;

[0111] According to the differences in frequency distribution of different components of the residual signal, the residual signal is split into high-frequency and low-frequency signals through a low-pass filter for prediction;

[0112] For the training of high-frequency residual prediction model, the commutation features of robot joint friction force are constructed, and an adaptive filtering framework is built based on it;

[0113] S5. Collect training data through random trajectories, introduce time series information into the training data through a sliding window to perform data enhancement processing, and input the residual signal into the high-frequency separation adaptive filtering framework to obtain the output target value of the model training; then use the XGBoost algorithm for training to obtain high-frequency and low-frequency residual prediction models respectively;

[0114] S6. The prediction result of the trained residual prediction model is input into the momentum observer in a feedforward form to obtain the compensated joint external torque signal, and then collision detection is performed based on the set threshold.

[0115] The present invention proposes a robot collision detection method based on frequency distribution difference dynamic model residual compensation, which mainly improves the accuracy of collision detection by frequency division prediction and compensation of the residual signal of the robot joint torque. First, in step S1, the Newton-Euler method and its variants are used to decouple the robot joint torque to obtain the robot's momentum, Coriolis force, centrifugal force, gravity component and other dynamic information. These dynamic components provide basic data of the robot's motion state, which is convenient for subsequent residual calculation and construction of compensation model.

[0116] Next, in step S2, the external torque expression output by the momentum observer is solved by Laplace forward and inverse transform, and the residual expression of the joint external torque is constructed based on this. This step quantifies the difference between the actual external torque of the robot and the ideal dynamic model into a residual expression, thereby providing an operational target quantity for subsequent frequency separation compensation. The design of the residual expression is the core of the collision detection model, which helps to extract the difference information between the actual force and theoretical torque of the robot.

[0117] In step S3, a feature vector is designed based on the residual expression. A preliminary feature vector is formed by decomposing the components in the residual expression and extracting sub-expressions related to the robot's dynamic and kinematic characteristics. The design of the feature vector takes into account the actual situation of the robot's joint motion and covers the different components of the torque residual, so that the subsequent frequency separation can more accurately match the high-frequency and low-frequency components, thereby enhancing the model prediction effect.

[0118] In step S4, by constructing a high-frequency separation adaptive filtering training framework, based on the difference in frequency distribution, the residual signal is decomposed into high-frequency and low-frequency signals through a low-pass filter for prediction. The separation of high and low frequencies can effectively reduce the interference caused by frequency inconsistency in the residual signal. On this basis, in order to adapt to the friction characteristics of high-frequency residuals, an adaptive filtering framework is constructed to adapt to the characteristic changes of friction in commutation and ensure accurate prediction of high-frequency signals.

[0119] In step S5, training data is collected through random trajectories, and the data is introduced into the timing information through a sliding window to enhance the timeliness and robustness of the data. The residual signal is input into the high-frequency separation adaptive filtering framework, and the target training output is obtained after data preprocessing. oos The t algorithm trains high-frequency and low-frequency residual prediction models respectively, XGB oos The t model can handle the nonlinear characteristics in the residual signal and improve the prediction accuracy of high- and low-frequency residuals.

[0120] Finally, in step S6, the trained high- and low-frequency residual prediction models are fed into the momentum observer in a feedforward form to generate compensated joint external torque signals. This compensation signal can more accurately reflect the real external torque of the robot based on the ideal dynamics model, and finally perform collision judgment through the set collision detection threshold, effectively improving the accuracy and real-time performance of detection.

[0121] In step S1 provided in the embodiment of the present invention, the expression of the theoretical joint torque of the robot can be obtained based on the Newton-Euler algorithm and its variants:

[0122]

[0123] Among them, τ i represents the theoretical torque of the robot's i-th joint; Respectively represent the angle, velocity and acceleration of the robot's ith joint; NE α (·) indicates the standard Newton-Euler algorithm; MNE α (·) represents a variant of the Newton-Euler algorithm, which splits the joint velocity into two different inputs to obtain additional degrees of freedom, so that the output matrix can automatically satisfy the partial symmetry property; α∈{g0,0} indicates whether the gravity effect is considered in the Newton-Euler algorithm;

[0124] As a further preferred embodiment, the theoretical torque can be decoupled based on the Newton-Euler algorithm and its variants to obtain the input items of the subsequent momentum observer, which include the momentum of the robot, the Coriolis force and the centrifugal force, and the gravity component, which can be calculated by the following formula:

[0125]

[0126] Among them, β is a dynamic variable defined for simplified representation, g(q) represents the gravity vector of the robot, is the transpose of the Coriolis force and centrifugal force matrix of the robot, M(q) represents the inertia matrix of the robot, and p is the generalized momentum of the robot; the vector e n =(0…0 10…0) T , the nth element is 1 and the rest are 0.

[0127] Step S2 provided in the embodiment of the present invention specifically includes:

[0128] First, determine the expression of the momentum observer as follows:

[0129] r(t)=K0[p(t)-∫(τ m -τ f +r+β)dt]

[0130] Where r(t) represents the external joint torque output by the momentum observer at the tth moment, K0 represents the diagonal gain matrix of the momentum observer, τ m represents the motor driving torque, τ f is the calculated friction force;

[0131] Taking into account factors such as dynamic modeling error, friction error, and sensor measurement error, it is impossible to obtain a completely accurate external torque monitoring signal; incorporating the above error factors into the derivation process, the calculation formula in the actual situation can be expressed as follows:

[0132]

[0133] in, Indicates that the variable is an estimated version based on kinetic model calculations or actual readings;

[0134] By solving the above differential equations through Laplace forward and inverse transform, we can get the expression of the external torque output by the momentum observer as follows:

[0135]

[0136] In the two formulas, * represents the convolution operation; then the residual of the external torque monitoring value can be defined as follows:

[0137]

[0138] The variables in the above formula are defined as follows:

[0139]

[0140] Step S3 provided in the embodiment of the present invention specifically includes:

[0141] According to the defined residual expression, the expression related to the robot friction modeling error and the expression of the dynamic modeling error can be extracted as follows:

[0142]

[0143] Among them, h fe (t) represents the part of the residual that comes from the friction modeling error, Respectively The part related to friction modeling; e (t) represents the part of the residual that comes from the dynamic modeling error, and h e (t) can be further decomposed into the error parts related to momentum, Coriolis force, centrifugal force and gravity, which are expressed as follows:

[0144]

[0145] From the above formula, we can see that the residual r e Depend on and {h p ,h g ,h C ,h f}decision; among which, It can be calculated by the dynamic model and the friction model; therefore, it can be used as the feature vector input of the subsequent XGBoost, and {h p ,h g ,h C ,h f} is fitted through XGBoost training; the kinematic characteristics of the robot are added on the basis of the dynamic characteristics, and the feature vector of the dynamic residual prediction model can be initially extracted into the following form:

[0146]

[0147] Step S4 provided in the embodiment of the present invention specifically includes:

[0148] According to the differences in frequency distribution of different components of the residual signal, the residual signal can be split into high-frequency signal and low-frequency signal through a low-pass filter, which can be expressed as follows:

[0149]

[0150] r high =r e -r low

[0151] Wherein, LPF represents low-pass filter; r low Represents the low-frequency residual signal. The errors of the dynamic model (such as momentum and gravity errors) are concentrated in the low-frequency signal. high Represents high-frequency residual signal, measurement noise and friction commutation error are concentrated in the high-frequency signal; its essence is to further decouple the residual from the perspective of frequency;

[0152] The adaptive filtering framework specifically includes:

[0153] First, the commutation characteristics of the robot joint friction force are constructed as shown below:

[0154]

[0155] Among them, k represents the kth moment, n s represents the length of the friction mutation gap window, n a Indicates the speed reverse look-ahead window length;

[0156] According to the commutation characteristics, the high-frequency residual signal can be adaptively filtered to reduce the measurement noise contained therein while retaining the friction commutation error, so as to increase the fitting accuracy of the model; it can be expressed as follows:

[0157]

[0158] Among them, r df (k) is the high-frequency residual signal after adaptive filtering at the kth moment, r filter (k) is the high-frequency residual signal after low-pass filtering;

[0159] According to the differentiated characteristics of the frequency distribution of different components of the residual signal and the constructed friction commutation features, the extracted feature vector can be further extracted into the feature vector corresponding to the high-frequency residual and the feature vector corresponding to the low-frequency residual, which are specifically in the following form:

[0160]

[0161] Among them, feature low Represents low-frequency feature vectors, including the kinematic characteristics of the robot and related features of the dynamic model; feature high Represents high-frequency feature vectors, including the robot's kinematic characteristics, friction error terms, and friction commutation characteristics.

[0162] Step S5 provided in the embodiment of the present invention specifically includes:

[0163] First, let the robot run 20 randomly generated trajectories to collect the robot's angle, velocity and acceleration data, and perform low-pass filtering on the acceleration signal to filter out the measurement noise it contains; the collected data can be used according to steps S3 and S4 to obtain the input and output data sets for model training, as shown in the following formula:

[0164]

[0165] Among them, X low，i represents the low-frequency feature dataset of the robot’s i-th joint, Y low，i represents the low-frequency residual target value of the robot's i-th joint; X high，i It represents the high-frequency feature dataset of the i-th joint, T high，i represents the high-frequency residual target value of the robot's i-th joint; n represents the data collected at n moments in total;

[0166] The main idea of ​​the data enhancement method based on sliding window is: define a sliding window with a length of L, which contains the time series feature data of the period [t-L+1, t], and use the time series feature data of the sliding window as the feature data of the current moment as the model input to achieve the effect of data enhancement; it is worth noting that if there is no data before the moment t, the missing data will be filled with 0; at the same time, the length of the window can be selected according to different processing situations, so as to flexibly adjust the length of the retained historical time series information according to the actual situation; the feature data set at the moment t can be expressed as follows:

[0167]

[0168] X low,i (t) represents the low-frequency feature dataset of the robot’s ith joint at time t, X high,i(t) represents the high-frequency feature data set of the robot's ith joint at the tth moment; further, the overall feature data set of the ith joint can be rewritten as follows:

[0169]

[0170] like Figure 8 As shown, a robot collision detection system for dynamic model residual compensation considering frequency distribution differences provided by an embodiment of the present invention includes:

[0171] An acquisition module is used to decouple the theoretical torque of the robot joints based on the Newton-Euler method and its variants to obtain the momentum, Coriolis force, centrifugal force and gravity component of the robot;

[0172] A solution module is used to solve the external torque expression output by the momentum observer through Laplace forward and inverse transform, and to express the residual expression of the joint external torque accordingly;

[0173] The extraction module is used to design the feature vector according to the residual expression; the relevant sub-expressions of each component in the residual expression are extracted as dynamic features, and the feature vector is preliminarily formed with the kinematic features of the robot;

[0174] A construction module is used to construct a high-frequency separation adaptive filtering training framework; according to the differences in frequency distribution of different components of the residual signal, the residual signal is split into high-frequency and low-frequency signals through a low-pass filter for prediction; for the training of the high-frequency residual prediction model, the commutation characteristics of the robot joint friction force are constructed, and the adaptive filtering framework is constructed accordingly;

[0175] The acquisition module is used to collect training data through random trajectories, introduce time series information into the training data through a sliding window to perform data enhancement processing, and input the residual signal into the high-frequency separation adaptive filtering framework to obtain the output target value of the model training; then the XGBoost algorithm is used for training to obtain high-frequency and low-frequency residual prediction models respectively;

[0176] The collision detection module is used to input the prediction results of the trained residual prediction model into the momentum observer in a feedforward form to obtain the compensated joint external torque signal, and then perform collision detection based on the set threshold.

[0177] Another object of the present invention is to provide a computer device, comprising a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of the robot collision detection method with dynamic model residual compensation considering frequency distribution differences.

[0178] Another object of the present invention is to provide a computer-readable storage medium storing a computer program, which, when executed by a processor, enables the processor to execute the steps of the robot collision detection method with dynamic model residual compensation considering frequency distribution differences.

[0179] Another object of the present invention is to provide an information data processing terminal, which is used to implement the dynamic model residual compensation robot collision detection system that takes into account frequency distribution differences.

[0180] The present invention is specifically implemented:

[0181] Step S1 specifically includes:

[0182] Based on the Newton-Euler algorithm and its variants, the expression of the theoretical joint torque of the robot can be obtained as follows:

[0183]

[0184] Among them, τ i Represents the theoretical torque of the robot's i-th joint calculated based on the dynamic model. Respectively represent the angle, velocity and acceleration of the robot’s i-th joint. α (·) indicates the standard Newton-Euler algorithm. MNE α (·) represents a variant of the Newton-Euler algorithm, which splits the joint velocity into two different inputs to obtain additional degrees of freedom, so that the output matrix can automatically satisfy the partial symmetry property. α∈{g0,0} indicates whether the gravity effect is considered in the Newton-Euler algorithm.

[0185] Based on formula (1), the theoretical torque can be decoupled to obtain the input of the subsequent momentum observer, which includes the momentum of the robot, the Coriolis force, the centrifugal force, and the gravity component, which can be calculated by the following formula:

[0186]

[0187] Among them, β is a dynamic variable defined for simplified representation, g(q) represents the gravity vector of the robot, is the transpose of the robot's Coriolis force and centrifugal force matrix, M(q) represents the robot's inertia matrix, and p is the robot's generalized momentum. The vector e n =(0 … 0 1 0 … 0) T , the nth element is 1 and the rest are 0.

[0188] Furthermore, step S2 specifically includes:

[0189] First, determine the expression of the momentum observer as follows:

[0190] r(t)=K0[p(t)-∫(τ m -τ f +r+β)dt] (3)

[0191] Where r(t) represents the external joint torque output by the momentum observer at the tth moment, K0 represents the diagonal gain matrix of the momentum observer, τ m represents the motor driving torque, τ f is the calculated friction force.

[0192] Furthermore, considering factors such as dynamic modeling error, friction error, and sensor measurement error, it is impossible to obtain a completely accurate external torque monitoring signal. Combined with formula (3), the above error factors are included in the derivation process, and the calculation formula in the actual situation can be expressed as follows:

[0193]

[0194] in, Indicates that the variable is an estimated version based on kinetic model calculations or actual readings.

[0195] Furthermore, by solving formulas (3) to (4) through Laplace forward and inverse transform, the expression of the external torque output by the momentum observer can be obtained as follows:

[0196]

[0197] In the two formulas, * represents the convolution operation. Then the residual of the external torque monitoring value can be defined as follows:

[0198]

[0199] The variables in the above formula are defined as follows:

[0200]

[0201] Furthermore, step S3 specifically includes:

[0202] According to formulas (6) to (7), the expressions related to the robot friction modeling error and the dynamic modeling error can be extracted as follows:

[0203]

[0204] Among them, h fe (t) represents the part of the residual that comes from the friction modeling error, Respectively Part related to friction modeling. e (t) represents the part of the residual that comes from the dynamic modeling error, and he (t) can be further decomposed into the error parts related to momentum, Coriolis force, centrifugal force and gravity, which are expressed as follows:

[0205]

[0206] From formulas (8) to (9), we can see that the residual r e Depend on and {h p ,h g ,h C ,h f} decision. Among them, It can be calculated by the dynamic model and the friction model. Therefore, it can be used as the feature vector input of the subsequent XGBoost, and {h p ,h g ,h C ,h f} can be fitted through XGBoost training. By adding the robot's kinematic characteristics to the dynamic characteristics, the feature vector of the dynamic residual prediction model can be initially extracted into the following form:

[0207]

[0208] Furthermore, step S4 constructs a high-frequency separation adaptive filtering training framework, the framework structure is as follows Figure 2 As shown, specifically including:

[0209] According to the differences in frequency distribution of different components of the residual signal, the residual signal can be split into high-frequency signal and low-frequency signal through a low-pass filter, which can be expressed as follows:

[0210]

[0211] Wherein, LPF stands for low-pass filter. low Represents the low-frequency residual signal. The dynamic model errors (such as momentum and gravity errors) are concentrated in the low-frequency signal. high Represents the high-frequency residual signal, and the measurement noise and friction commutation error are concentrated in the high-frequency signal. Formula (11) essentially further decouples the residual from the perspective of frequency.

[0212] However, high The measurement noise contained in cannot be directly fitted due to its randomness, so it will affect the fitting accuracy of the model. Directly using filtering will cause friction mutation error distortion and thus affect the compensation effect. Therefore, the noise signal can be removed by constructing an adaptive filtering method, and the specific steps are as follows:

[0213] First, the commutation characteristics of the robot joint friction force are constructed as shown below:

[0214]

[0215] Among them, k represents the kth moment, n s represents the length of the friction mutation gap window, n a Indicates the speed reverse look-ahead window length.

[0216] Furthermore, according to the commutation characteristics, the high-frequency residual signal can be adaptively filtered to reduce the measurement noise contained therein while retaining the friction commutation error, so as to increase the fitting accuracy of the model. It can be expressed as follows:

[0217]

[0218] Among them, r df (k) is the high-frequency residual signal after adaptive filtering at the kth moment, r filter (k) is the high-frequency residual signal after low-pass filtering. The effect of the adaptive filtering method is as follows: Fig. 9 As shown in the figure, after the high-frequency residual signal is adaptively filtered, the commutation error of the friction force is well retained, while the measurement noise is correspondingly removed by low-pass filtering.

[0219] According to the differentiated characteristics of the frequency distribution of different components of the residual signal and the constructed friction commutation features, the extracted feature vector can be further extracted into the feature vector corresponding to the high-frequency residual and the feature vector corresponding to the low-frequency residual, which are specifically in the following form:

[0220]

[0221] Among them, feature low Represents low-frequency feature vectors, including the robot's kinematic characteristics and related features of the dynamics model. high Represents high-frequency feature vectors, including the robot's kinematic characteristics, friction error terms, and friction commutation characteristics.

[0222] Further, step S5 includes:

[0223] First, the diagonal elements of the momentum observer's diagonal gain matrix K0 are set to [10, 10, 15, 30, 30, 30]. The robot is allowed to run 20 randomly generated trajectories to collect joint angle, velocity, acceleration data and the output of the momentum observer. The acceleration signal is low-pass filtered to filter out the measurement noise it contains. Substituting the collected data into formulas (8) to (14), the following formula is obtained:

[0224]

[0225] Among them, X low，i represents the low-frequency feature dataset of the robot’s i-th joint, Y low，i represents the low-frequency residual target value of the robot's i-th joint. high，i It represents the high-frequency feature dataset of the i-th joint, Y high，i represents the high-frequency residual target value of the robot's ith joint. n represents the data collected at n moments in total.

[0226] In order to establish an effective residual prediction model, a large amount of high-quality data is required to ensure the accuracy and robustness of machine learning training. To this end, a data enhancement method based on a sliding window is used to reduce the difficulty of obtaining a large number of data samples and improve the quality of sample data. The schematic diagram of this method is shown in Fig.10 As shown in the figure, the main idea is to define a sliding window of length L, which contains the time series feature data of the period [t-L+1, t], and use the time series feature data of the sliding window as the feature data of the current moment as the model input to achieve the effect of data enhancement. It is worth noting that if there is no data before the moment t, the missing data will be filled with 0. At the same time, the length of the window L can be selected according to different processing situations, so as to flexibly adjust the length of the retained historical time series information according to the actual situation. The feature data set at moment t can be expressed as follows:

[0227]

[0228] X low,i (t) represents the low-frequency feature dataset of the robot’s ith joint at time t, X high,i (t) represents the high-frequency feature data set of the robot's ith joint at time t. Furthermore, the overall feature data set of the ith joint can be rewritten as follows:

[0229]

[0230] Further, step S6 includes:

[0231] After obtaining the trained residual prediction model, it can be added to the momentum observer as an independent residual compensation module. The overall structure of collision detection is as follows: Figure 3 As shown, the residual prediction model receives the joint angle, velocity and acceleration information, and calculates the low-frequency features and high-frequency features through formulas (8)(9)(12)(14) and inputs them into the corresponding residual prediction model. Then the high-frequency residual prediction results and the low-frequency residual prediction results are jointly compensated to the output signal of the momentum observer. Finally, the collision detection function is input to determine whether a collision occurs. The collision detection function can be expressed as follows:

[0232]

[0233] Among them, r(t) is the external torque signal at the tth moment, ε DTh , ε UTh The upper and lower limits of the collision detection threshold are set. Once the external torque signal exceeds the set threshold, it means that the robot has collided.

[0234] like Figure 4 The figure shows the comparison of the external torque estimation of the momentum observer before and after adding the residual compensation module. The robot is monitored in a collision-free state, so the closer the estimated external torque is to 0, the higher the accuracy. The proposed dynamic model residual compensation method considering the frequency distribution difference shows a superior compensation effect and effectively improves the external torque estimation accuracy of the momentum observer.

[0235] like Figure 5 The figure shows the results of the robot hard collision experiment. The characteristics of hard collision are: the external force changes rapidly and is an impact force. The experimental method is to use the human body to quickly apply impact force to the robot to generate a hard collision signal. Figure 5 It can be seen that after adding the residual compensation module, the threshold accuracy of the robot changes from [-11NM, 9NM] before compensation to [-4.5NM, 5.5NM], which is greatly improved, and the accuracy and response speed of collision detection are also further improved. Therefore, the dynamic model residual compensation robot collision detection method considering frequency distribution differences proposed in the present invention shows excellent performance in the detection of hard collisions.

[0236] like Figure 6 The following is the result of the robot soft collision experiment. The characteristics of soft collision are: the external force changes slowly and is a squeezing force. The experiment is carried out through a balloon experiment. The experimental platform is as follows Figure 7 As shown, the robot end is equipped with a six-dimensional force sensor, and the linear speed of the end in the Z axis is set to 0.1m / s, so that the end is vertically pressed on the balloon to produce a soft collision effect on the joint. When the monitored external torque of the joint exceeds the set threshold, the robot responds and returns to the initial point. For this reason, the smaller the contact force detected by the end sensor in the Z direction, the higher the collision detection accuracy and the faster the response speed. Figure 6It can be seen that before adding the residual compensation module, the external torque output by the robot contains the residual of the dynamic model, so the threshold is set to [-3NM, 8NM]. When a collision is detected, the end force sensor reading is: 9.61N. After adding the residual compensation module, the dynamic model residual is compensated, and the accuracy is further improved. The threshold is set to [0NM, 7NM]. When a collision is detected, the end force sensor reading is: 6.90N. Therefore, the dynamic model residual compensation robot collision detection method proposed in the present invention considering the frequency distribution difference shows excellent performance for soft collision detection.

[0237] The present invention focuses on improving the collision detection accuracy of robots and is applicable to specific fields such as industrial robots and service robots. In industrial applications, such as assembly robots on automated production lines, the accuracy of collision detection technology is directly related to production efficiency and equipment safety. Through the technical solution of the present invention, the external torque residual output by the momentum observer can be effectively compensated, thereby achieving high-precision collision detection. This high-precision collision detection method can significantly reduce equipment damage and production delays caused by collisions, thereby reducing the economic losses of enterprises.

[0238] In the field of service robots, the technology of the present invention also has important applications. For example, in a medical environment, when a service robot is responsible for drug delivery or patient care, a collision may occur due to the complex environment. Through the collision detection scheme of the present invention, the robot can quickly detect the occurrence of a collision after the collision occurs, so as to effectively take measures to reduce potential damage to surrounding equipment and patients. This not only improves the operational safety of the robot in a complex environment, but also enhances its response capability in emergency situations.

[0239] In addition, the present invention can be applied to robot scheduling and management in intelligent manufacturing systems, helping enterprises achieve more efficient resource allocation and risk control by improving collision detection accuracy. Combined with the technology of the present invention, robots can accurately feedback collision detection information when performing tasks, allowing the management system to quickly adjust the workflow and ensure the smooth progress of production activities.

[0240] These specific application embodiments demonstrate the innovation and practical value of the present invention in improving the accuracy of robot collision detection, and provide a solid foundation for the further development of robot technology.

[0241] like Figure 4 The figure shows the comparison of the external torque estimation of the momentum observer before and after adding the residual compensation module. The robot is monitored in a collision-free state, so the closer the estimated external torque is to 0, the higher the accuracy. The proposed dynamic model residual compensation method considering the frequency distribution difference shows a superior compensation effect and effectively improves the external torque estimation accuracy of the momentum observer.

[0242] like Figure 5 The figure shows the results of the robot hard collision experiment. The characteristics of hard collision are: the external force changes rapidly and is an impact force. The experimental method is to use the human body to quickly apply impact force to the robot to generate a hard collision signal. Figure 5 It can be seen that after adding the residual compensation module, the threshold accuracy of the robot changes from [-11NM, 9NM] before compensation to [-4.5NM, 5.5NM], which is greatly improved, and the accuracy and response speed of collision detection are also further improved. Therefore, the dynamic model residual compensation robot collision detection method considering frequency distribution differences proposed in the present invention shows excellent performance in the detection of hard collisions.

[0243] like Figure 6 The following is the result of the robot soft collision experiment. The characteristics of soft collision are: the external force changes slowly and is a squeezing force. The experiment is carried out through a balloon experiment. The experimental platform is as follows Figure 7 As shown, the robot end is equipped with a six-dimensional force sensor, and the linear speed of the end in the Z axis is set to 0.1m / s, so that the end is vertically pressed on the balloon to produce a soft collision effect on the joint. When the monitored external torque of the joint exceeds the set threshold, the robot responds and returns to the initial point. For this reason, the smaller the contact force detected by the end sensor in the Z direction, the higher the collision detection accuracy and the faster the response speed. Figure 6 It can be seen that before adding the residual compensation module, the external torque output by the robot contains the residual of the dynamic model, so the threshold is set to [-3NM, 8NM]. When a collision is detected, the end force sensor reading is: 9.61N. After adding the residual compensation module, the residual of the dynamic model is compensated, and the accuracy is further improved. The threshold is set to [0NM, 7NM]. When a collision is detected, the end force sensor reading is: 6.90N. Therefore, the dynamic model residual compensation robot collision detection method proposed in the present invention considering the frequency distribution difference shows excellent performance for soft collision detection.

[0244] like Fig. 9 The figure shows the friction commutation characteristics and the effect of the adaptive filtering method. Fig. 9 It can be seen that the friction commutation feature can well locate the friction commutation error; at the same time, after the high-frequency residual signal is adaptively filtered, the commutation error of the friction force is well retained, and the measurement noise is correspondingly removed by low-pass filtering.

[0245] It should be noted that the embodiments of the present invention can be implemented by hardware, software, or a combination of software and hardware. The hardware part can be implemented using dedicated logic; the software part can be stored in a memory and executed by an appropriate instruction execution system, such as a microprocessor or dedicated design hardware. It can be understood by a person of ordinary skill in the art that the above-mentioned devices and methods can be implemented using computer executable instructions and / or contained in a processor control code, such as a carrier medium such as a disk, CD or DVD-ROM, a programmable memory such as a read-only memory (firmware), or a data carrier such as an optical or electronic signal carrier. Such code is provided on the carrier medium. The device and its modules of the present invention can be implemented by hardware circuits such as very large-scale integrated circuits or gate arrays, semiconductors such as logic chips, transistors, etc., or programmable hardware devices such as field programmable gate arrays, programmable logic devices, etc., can also be implemented by software executed by various types of processors, and can also be implemented by a combination of the above-mentioned hardware circuits and software, such as firmware.

[0246] The above description is only a specific implementation mode of the present invention, but the protection scope of the present invention is not limited thereto. Any modifications, equivalent substitutions and improvements made by any technician familiar with the technical field within the technical scope disclosed by the present invention and within the spirit and principle of the present invention should be covered by the protection scope of the present invention.

Claims

1. A robot collision detection method based on dynamic model residual compensation considering frequency distribution difference, characterized in that: The following steps are involved: S1. Decouple the theoretical torque of the robot joints based on the Newton-Euler method and its variants to obtain the robot's momentum, Coriolis force, centrifugal force and gravity components; S2. Solve the external torque expression output by the momentum observer through Laplace forward and inverse transform, and express the residual expression of the joint external torque accordingly; S3. Design a feature vector based on the residual expression; extract the relevant sub-expressions of each component in the residual expression as dynamic features, and preliminarily form a feature vector with the kinematic features of the robot; S4, build a high-frequency separation adaptive filtering training framework; According to the differences in frequency distribution of different components of the residual signal, the residual signal is split into high-frequency and low-frequency signals through a low-pass filter for prediction; For the training of high-frequency residual prediction model, the commutation features of robot joint friction force are constructed, and an adaptive filtering framework is built based on it; S5. Collect training data through random trajectories, introduce time series information into the training data through a sliding window to perform data enhancement processing, and input the residual signal into the high-frequency separation adaptive filtering framework to obtain the output target value of the model training; then use the XGBoost algorithm for training to obtain high-frequency and low-frequency residual prediction models respectively; S6. The prediction result of the trained residual prediction model is input into the momentum observer in a feedforward form to obtain the compensated joint external torque signal, and then collision detection is performed based on the set threshold.

2. The robot collision detection method with dynamic model residual compensation considering frequency distribution difference as claimed in claim 1, characterized in that: In step S1, the expression of the theoretical joint torque of the robot can be obtained based on the Newton-Euler algorithm and its variants: Among them, τ i represents the theoretical torque of the robot's i-th joint; q i , Respectively represent the angle, velocity and acceleration of the robot's ith joint; NE α (·) indicates the standard Newton-Euler algorithm; MNE α (·) represents a variant of the Newton-Euler algorithm, which splits the joint velocity into two different inputs to obtain additional degrees of freedom, so that the output matrix can automatically satisfy the partial symmetry property; α∈{g0,0} indicates whether the gravity effect is considered in the Newton-Euler algorithm; As a further preferred embodiment, the theoretical torque can be decoupled based on the Newton-Euler algorithm and its variants to obtain the input items of the subsequent momentum observer, which include the momentum of the robot, the Coriolis force and the centrifugal force, and the gravity component, which can be calculated by the following formula: Among them, β is a dynamic variable defined for simplified representation, g(q) represents the gravity vector of the robot, is the transpose of the Coriolis force and centrifugal force matrix of the robot, M(q) represents the inertia matrix of the robot, and p is the generalized momentum of the robot; the vector e n =(0...0 1 0...0) T , the nth element is 1 and the rest are 0.

3. The robot collision detection method with dynamic model residual compensation considering frequency distribution difference as claimed in claim 1, characterized in that: The step S2 specifically includes: First, determine the expression of the momentum observer as follows: r(t)=K0[p(t)-∫(τ m -τ f +r+β)dt] Where r(t) represents the external joint torque output by the momentum observer at the tth moment, K0 represents the diagonal gain matrix of the momentum observer, τ m represents the motor driving torque, τ f is the calculated friction force; Taking into account factors such as dynamic modeling error, friction error, and sensor measurement error, it is impossible to obtain a completely accurate external torque monitoring signal; incorporating the above error factors into the derivation process, the calculation formula in the actual situation can be expressed as follows: in, Indicates that the variable is an estimated version based on kinetic model calculations or actual readings; By solving the above differential equations through Laplace forward and inverse transform, we can get the expression of the external torque output by the momentum observer as follows: In the two formulas, * represents the convolution operation; then the residual of the external torque monitoring value can be defined as follows: The variables in the above formula are defined as follows:

4. The robot collision detection method with dynamic model residual compensation considering frequency distribution difference as claimed in claim 1, characterized in that: The step S3 specifically includes: According to the defined residual expression, the expression related to the robot friction modeling error and the expression of the dynamic modeling error can be extracted as follows: Among them, h fe (t) represents the part of the residual that comes from the friction modeling error, h f (t) respectively represent The part of r related to friction modeling; h e (t) represents the part of the residual that comes from the dynamic modeling error, and h e (t) can be further decomposed into the error parts related to momentum, Coriolis force, centrifugal force and gravity, which are expressed as follows: From the above formula, we can see that the residual r e Depend on and {h p ,h g ,h C ,h f }decision; among which, It can be calculated by the dynamic model and the friction model; therefore, it can be used as the feature vector input of the subsequent XGBoost, and {h p ,h g ,h C ,h f } is fitted through XGBoost training; the kinematic characteristics of the robot are added on the basis of the dynamic characteristics, and the feature vector of the dynamic residual prediction model can be initially extracted into the following form:

5. The robot collision detection method with dynamic model residual compensation considering frequency distribution difference as claimed in claim 1, characterized in that: The step S4 specifically includes: According to the differences in frequency distribution of different components of the residual signal, the residual signal can be split into high-frequency signal and low-frequency signal through a low-pass filter, which can be expressed as follows: r high ＝r e -r low Wherein, LPF represents low-pass filter; r low Represents the low-frequency residual signal. The errors of the dynamic model (such as momentum and gravity errors) are concentrated in the low-frequency signal. high Represents high-frequency residual signal, measurement noise and friction commutation error are concentrated in the high-frequency signal; its essence is to further decouple the residual from the perspective of frequency; The adaptive filtering framework specifically includes: First, the commutation characteristics of the robot joint friction force are constructed as shown below: Among them, k represents the kth moment, n s represents the length of the friction mutation gap window, n a Indicates the speed reverse look-ahead window length; According to the commutation characteristics, the high-frequency residual signal can be adaptively filtered to reduce the measurement noise contained therein while retaining the friction commutation error, so as to increase the fitting accuracy of the model; it can be expressed as follows: Among them, r df (k) is the high-frequency residual signal after adaptive filtering at the kth moment, r filter (k) is the high-frequency residual signal after low-pass filtering; According to the differentiated characteristics of the frequency distribution of different components of the residual signal and the constructed friction commutation features, the extracted feature vector can be further extracted into the feature vector corresponding to the high-frequency residual and the feature vector corresponding to the low-frequency residual, which are specifically in the following form: Among them, feature low Represents low-frequency feature vectors, including the kinematic characteristics of the robot and related features of the dynamic model; feature high Represents high-frequency feature vectors, including the robot's kinematic characteristics, friction error terms, and friction commutation characteristics.

6. The robot collision detection method with dynamic model residual compensation considering frequency distribution difference as claimed in claim 1, characterized in that: The step S5 specifically includes: First, let the robot run 20 randomly generated trajectories to collect the robot's angle, velocity and acceleration data, and perform low-pass filtering on the acceleration signal to filter out the measurement noise it contains; the collected data can be used according to steps S3 and S4 to obtain the input and output data sets for model training, as shown in the following formula: Among them, X low，i represents the low-frequency feature dataset of the robot’s i-th joint, Y low，i represents the low-frequency residual target value of the robot's i-th joint; X high，i It represents the high-frequency feature dataset of the i-th joint, Y high，i represents the high-frequency residual target value of the robot's i-th joint; n represents the data collected at n moments in total; The main idea of ​​the data enhancement method based on sliding window is: define a sliding window with a length of L, which contains the time series feature data of the period [t-L+1, t], and use the time series feature data of the sliding window as the feature data of the current moment as the model input to achieve the effect of data enhancement; it is worth noting that if there is no data before the moment t, the missing data will be filled with 0; at the same time, the length of the window L can be selected according to different processing situations, so as to flexibly adjust the length of the retained historical time series information according to the actual situation; the feature data set at the moment t can be expressed as follows: X low,i (t) represents the low-frequency feature dataset of the robot’s ith joint at time t, X high，i (t) represents the high-frequency feature data set of the robot's ith joint at the tth moment; further, the overall feature data set of the ith joint can be rewritten as follows:

7. A robot collision detection system with dynamic model residual compensation considering frequency distribution difference, which implements the robot collision detection method with dynamic model residual compensation considering frequency distribution difference as claimed in any one of claims 1 to 6, characterized in that: The dynamic model residual compensation robot collision detection system considering frequency distribution differences includes: An acquisition module is used to decouple the theoretical torque of the robot joints based on the Newton-Euler method and its variants to obtain the momentum, Coriolis force, centrifugal force and gravity component of the robot; A solution module is used to solve the external torque expression output by the momentum observer through Laplace forward and inverse transform, and to express the residual expression of the joint external torque accordingly; The extraction module is used to design the feature vector according to the residual expression; the relevant sub-expressions of each component in the residual expression are extracted as dynamic features, and the feature vector is preliminarily formed with the kinematic features of the robot; A construction module is used to construct a high-frequency separation adaptive filtering training framework; according to the differences in frequency distribution of different components of the residual signal, the residual signal is split into high-frequency and low-frequency signals through a low-pass filter for prediction; for the training of the high-frequency residual prediction model, the commutation characteristics of the robot joint friction force are constructed, and the adaptive filtering framework is constructed accordingly; The acquisition module is used to collect training data through random trajectories, introduce time series information into the training data through a sliding window to perform data enhancement processing, and input the residual signal into the high-frequency separation adaptive filtering framework to obtain the output target value of the model training; then the XGBoost algorithm is used for training to obtain high-frequency and low-frequency residual prediction models respectively; The collision detection module is used to input the prediction results of the trained residual prediction model into the momentum observer in a feedforward form to obtain the compensated joint external torque signal, and then perform collision detection based on the set threshold.

8. A computer device, characterized in that: The computer device includes a memory and a processor, the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of the robot collision detection method with dynamic model residual compensation considering frequency distribution differences as described in any one of claims 1-6.

9. A computer-readable storage medium storing a computer program, wherein when the computer program is executed by a processor, the processor executes the steps of the robot collision detection method with residual compensation of a dynamic model considering frequency distribution differences as described in any one of claims 1 to 6.

10. An information data processing terminal, characterized in that: The information data processing terminal is used to implement the robot collision detection system with residual compensation of the dynamic model considering the frequency distribution difference as described in claim 7.

Images