Online identification method for kinetic parameters of joint type series robot

Abstract

The invention discloses an online identification method for kinetic parameters of a joint type series robot, which comprises the following steps: S1, for a to-be-identified joint type series industrial robot, establishing a robot kinetic model considering friction; s2, linearization is carried out on the joint type series industrial robot dynamic model; s3, designing an optimal excitation track for an identification experiment, and obtaining an excitation track by using five finite Fourier series and solving a constrained nonlinear optimization problem; and S4, the series robot runs the excitation track, meanwhile, the position information of each joint of the robot and the driving current of each joint at each moment are collected, and to-be-identified parameters are calculated on line at each moment. According to the method, online identification of the robot kinetic parameters can be carried out, the identification efficiency of the robot kinetic parameters is improved, and the method can be used for industrial scenes such as robot moving and carrying, polishing and grinding and heavy-load hydraulic pressure.

Description

Technical Field

[0001] The invention belongs to the technical field of robots, and in particular relates to an online identification method for dynamic parameters of an articulated serial robot. Background Art

[0002] With the development of industrial automation and robotics, many application scenarios have become increasingly complex, and the requirements for robot control accuracy have become increasingly higher. In order to achieve high-precision control of robots, it is necessary to identify the robot's dynamic parameters. Traditional offline parameter identification methods usually require multiple experiments to collect enough data, which consumes a lot of time and resources. The results are highly dependent on the quality of the experimental data. If the data is noisy or not comprehensive enough, it may lead to inaccurate identification results. In addition, since the environment and operating conditions are often dynamically changing, traditional offline identification methods cannot adapt to these changes. With the advancement of computing technology, real-time data processing has become increasingly feasible. The popularity of modern sensors and computing platforms has enabled online identification algorithms to be implemented in practical applications. Summary of the invention

[0003] The main purpose of the present invention is to overcome the shortcomings and deficiencies of the prior art and to propose an online identification method for the dynamic parameters of an articulated serial robot.

[0004] In order to achieve the above object, the present invention adopts the following technical solutions:

[0005] An online identification method for dynamic parameters of an articulated serial robot comprises the following steps:

[0006] S1. For the articulated serial industrial robot to be identified, a robot dynamics model considering friction is established;

[0007] S2, linearize the dynamic model of the articulated serial industrial robot;

[0008] S3, design the optimal excitation trajectory for identification, using five finite Fourier series to obtain the excitation trajectory by solving a constrained nonlinear optimization problem;

[0009] S4: The robot runs the excitation trajectory, and at the same time collects the position information of each joint of the robot and the driving current of each joint at each moment, and calculates the parameters to be identified online at each moment.

[0010] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0011] 1. The present invention can perform online identification of robot dynamic parameters, thereby improving the efficiency of robot dynamic parameter identification, and can be used in industrial scenarios such as robot mobile handling, polishing and grinding, and heavy-load hydraulics; online identification has real-time adaptability and can update parameters in real time, so that the robot can quickly adapt to environmental changes and dynamic loads.

[0012] 2. Improve control accuracy. The online identification method of the present invention can continuously adjust parameters through real-time feedback, thereby improving the accuracy and response speed of the control system.

[0013] 3. Reduce experimental requirements. The method of the present invention does not require a large amount of offline experimental data and can perform parameter identification during normal operation, saving time and resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 It is the overall flow chart of the method of the present invention;

[0015] Figure 2 It is a flow chart of online calculation of the parameters to be identified at each moment in the method of the present invention. DETAILED DESCRIPTION

[0016] The present invention is further described in detail below in conjunction with embodiments and drawings, but the embodiments of the present invention are not limited thereto.

[0017] Example

[0018] like Figure 1 As shown, the present invention provides an online identification method of dynamic parameters of an articulated serial robot, comprising the following steps:

[0019] S1. For the articulated serial industrial robot to be identified, a robot dynamics model taking friction into consideration is established; in this embodiment, specifically:

[0020] For the articulated serial industrial robot to be identified, the robot dynamics model considering friction is established by the Newton-Euler method, which is expressed as:

[0021]

[0022] Among them, q, are the position, velocity and acceleration vector of the robot joints, M(q) is the inertia matrix of the robot arm, is the velocity term matrix related to centrifugal force and Coriolis force, g(q) is the gravity term, τ is the joint torque vector, represents friction;

[0023] For articulated serial robots, the friction at the joints needs to be considered. In the established robot dynamics model, for joint i, the Coulomb viscous friction model is used to model the joint friction, which is expressed as:

[0024]

[0025] Among them, f c is the Coulomb friction coefficient, f v is the viscous friction coefficient, b is the friction bias, and sign is the sign function.

[0026] S2. Linearize the dynamic model of the articulated serial industrial robot; in this embodiment, specifically:

[0027] The established dynamic model of the articulated serial industrial robot is linearized, and the full-rank minimum regression matrix expression and the corresponding minimum inertia parameter set are obtained by using methods such as QR decomposition and parameter reorganization. Finally, the following dynamic expression is obtained:

[0028]

[0029] in, is the observation matrix, β b It is the minimum inertia parameter set including the basic inertia parameters and friction parameters of each connecting rod waiting for the dynamic parameters to be identified.

[0030] S3. Design an optimal excitation trajectory for identification, using five finite Fourier series to obtain the excitation trajectory by solving a constrained nonlinear optimization problem; in this embodiment, the excitation trajectory is defined as:

[0031]

[0032]

[0033]

[0034] Among them, q i (t) and They represent the position, angular velocity and angular acceleration of joint i at time t, respectively, and q i0 is the joint angle at the initial moment; ω f is the fundamental frequency of the Fourier series excitation trajectory, a k and b k are the coefficients of the Fourier series excitation trajectory and are parameters to be determined.

[0035] Design the optimal excitation trajectory for identification, use the minimization of the observation matrix condition number as the optimization index of the excitation trajectory, and design the optimization objective function of the excitation trajectory as:

[0036]

[0037] st:

[0038] q min ≤q≤q max

[0039]

[0040]

[0041]

[0042]

[0043] Among them, cond is the observation matrix condition number, q min and q max are the lower and upper limits of the rotational positions of each joint of the robot, and are the lower and upper limits of the robot's joint speeds, and are the lower and upper limits of the acceleration of each joint of the robot respectively; are the robot joint velocity and acceleration at the initial moment, They are the velocity and acceleration of each joint of the robot at the termination moment respectively.

[0044] S4, the robot runs the excitation trajectory, and at the same time collects the position information of each joint of the robot and the driving current of each joint at each moment, and calculates the parameters to be identified online at each moment; in this embodiment, Figure 2 As shown, specifically including:

[0045] S41, robot running excitation trajectory, at the initial sampling time, set the inertial parameter β to be identified b (0) and the initial value of the covariance matrix P(0);

[0046] The second and subsequent moments include the following steps:

[0047] S42, processing the position information q and the sampling current through a first-order low-pass filter, obtaining the angular velocity and angular acceleration respectively through the second-order differentiation of the position information q, and the sampling torque of each other joint is calculated through the sampling current;

[0048] S43, based on the trajectory information q(k) of the robot joint at the current time (recorded as time k), To calculate the observation matrix at the current moment Note b (k)

[0049]

[0050] S44, based on the Φ calculated at the last moment b (k-1) and P(k-1) calculate the current moment gain matrix K(k), the formula is as follows:

[0051]

[0052] Where I is the unit diagonal matrix, is Φ b The transposed matrix of

[0053] S45, according to the inertia parameter β to be identified at the previous moment b (k-1) is used to calculate the estimation error ε, the formula is:

[0054] ε=τ(k)-Φ b (k)β b (k-1)

[0055] Among them, τ(k) is the sampling torque at the current moment;

[0056] S46, the inertial parameter β to be identified at the current moment b (k) is updated, the formula is:

[0057] β b (k) = β b (k-1)+K(k)ε

[0058] S47, calculate the covariance matrix P at the current moment, and reserve it for calculation at the next moment. The formula is:

[0059] P(k)=(IK(k)Φ b (k))P(k-1)

[0060] S48, the robot executes the excitation trajectory, cyclically recursively calculates until the trajectory ends, and finally obtains the inertial parameter identification result β b .

[0061] The method of this embodiment can perform online identification of robot dynamics parameters, thereby improving the efficiency of robot dynamics parameter identification, and can be used in industrial scenarios such as robot mobile handling, polishing and grinding, and heavy-load hydraulics.

[0062] It should also be noted that in this specification, terms such as "comprises", "includes" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of more restrictions, an element defined by the sentence "comprises a ..." does not exclude the presence of other identical elements in the process, method, article or device including the element.

[0063] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. An online identification method for dynamic parameters of an articulated serial robot, characterized in that: The following steps are involved: S1. For the articulated serial industrial robot to be identified, a robot dynamics model considering friction is established; S2, linearize the dynamic model of the articulated serial industrial robot; S3, design the optimal excitation trajectory for identification, using five finite Fourier series to obtain the excitation trajectory by solving a constrained nonlinear optimization problem; S4: The robot runs the excitation trajectory, and at the same time collects the position information of each joint of the robot and the driving current of each joint at each moment, and calculates the parameters to be identified online at each moment.

2. The method for online identification of dynamic parameters of an articulated serial robot according to claim 1, characterized in that: Step S1 is specifically as follows: For the articulated serial industrial robot to be identified, the robot dynamics model considering friction is established by the Newton-Euler method, which is expressed as: Among them, q, are the position, velocity and acceleration vector of the robot joints, M(q) is the inertia matrix of the robot arm, is the velocity term matrix related to centrifugal force and Coriolis force, g(q) is the gravity term, τ is the joint torque vector, represents friction; For articulated serial robots, the friction at the joints needs to be considered. In the established robot dynamics model, for joint i, the Coulomb viscous friction model is used to model the joint friction, which is expressed as: Among them, f c is the Coulomb friction coefficient, f v is the viscous friction coefficient, b is the friction bias, and sign is the sign function.

3. The method for online identification of dynamic parameters of an articulated serial robot according to claim 2, characterized in that: Step S2 is specifically as follows: The established dynamic model of the articulated serial industrial robot is linearized, and the full-rank minimum regression matrix expression and the corresponding minimum inertia parameter set are obtained by using the QR decomposition parameter reorganization method. Finally, the following dynamic expression is obtained: in, is the observation matrix, β b It is the minimum inertia parameter set of the dynamic parameters to be identified. The dynamic parameters to be identified include but are not limited to the basic inertia parameters and friction parameters of each connecting rod.

4. The method for online identification of dynamic parameters of an articulated serial robot according to claim 3, characterized in that: In step S3, the excitation trajectory is defined as: Among them, q i (t) and They represent the position, angular velocity and angular acceleration of joint i at time t, respectively, and q i0 is the joint angle at the initial moment; ω f is the fundamental frequency of the Fourier series excitation trajectory, a k and b k are the coefficients of the Fourier series excitation trajectory and are parameters to be determined.

5. The method for online identification of dynamic parameters of an articulated serial robot according to claim 4, characterized in that: In step S3, the optimal excitation trajectory for identification is designed, and the minimization of the observation matrix condition number is used as the optimization index of the excitation trajectory. The optimization objective function of the excitation trajectory is designed as: Among them, cond is the observation matrix condition number, q min and q max are the lower and upper limits of the rotational positions of each joint of the robot, and are the lower and upper limits of the robot's joint speeds, and are the lower and upper limits of the acceleration of each joint of the robot respectively; are the robot joint velocity and acceleration at the initial moment, They are the velocity and acceleration of each joint of the robot at the termination moment respectively.

6. The method for online identification of dynamic parameters of an articulated serial robot according to claim 5, characterized in that: Step S4 specifically includes: S41, robot running excitation trajectory, at the initial sampling time, set the inertial parameter β to be identified b (0) and the initial value of the covariance matrix P(0); The second and subsequent moments include the following steps: S42, processing the position information q and the sampling current through a first-order low-pass filter, obtaining the angular velocity and angular acceleration respectively through the second-order differentiation of the position information q, and the sampling torque of each other joint is calculated through the sampling current; S43, according to the trajectory information q(k) of the robot joint at the current time, i.e., time k, To calculate the observation matrix at the current moment Note Φ b (k) S44, based on the Φ calculated at the last moment b (k-1) and P(k-1) calculate the current moment gain matrix K(k), the formula is as follows: Where I is the unit diagonal matrix, Φ b The transposed matrix of S45, according to the inertia parameter β to be identified at the previous moment b (k-1) is used to calculate the estimation error ε, the formula is: ε=τ(k)-Φ b (k)b b (k-1) Among them, τ(k) is the sampling torque at the current moment; S46, the inertial parameter β to be identified at the current moment b (k) is updated, the formula is: b b (k)=β b (k-1)+K(k)e S47, calculate the covariance matrix P at the current moment, and reserve it for calculation at the next moment. The formula is: P(k)=(I-K(k)Φ b (k))P(k-1) S48, the robot executes the excitation trajectory, cyclically recursively calculates until the trajectory ends, and finally obtains the inertial parameter identification result β b .

Images