Work space force / acceleration disturbance observer and robot including same

Abstract

Provided are a work space force / acceleration disturbance observer and a robot including same. This work space force / acceleration disturbance observer is connected to an impedance-based motion controller in a work space, and obtains a disturbance estimation value in consideration of an interaction force and acceleration applied to an end effector of a robot, wherein the disturbance estimation value is expressed by the following equation. D^ is a disturbance estimation value, Q is a Q filter, Fc' is a control input force, M^o is a mass matrix estimation value, x'' is acceleration, and Fext is an interaction force.

Description

Workspace force / acceleration disturbance observer and robot including same

[0001] The present invention relates to a workspace force / acceleration disturbance observer and a robot including the same.

[0002] Impedance control is used in a variety of applications that require simultaneous position tracking and tracking during contact. However, disturbances such as friction and model uncertainty negatively impact the performance of impedance-based motion control. To address this issue, disturbance observers, which filter out disturbances observed in the nominal model, are widely used.

[0003] The problem to be solved by the present invention is to provide a workspace force / acceleration disturbance observer that observes disturbances by utilizing both interaction force and acceleration, and a robot including the same.

[0004] The problems to be solved by the present invention are not limited to the problems mentioned above, and other problems not mentioned will be clearly understood by those skilled in the art from the description below.

[0005] According to one aspect of the present invention for solving the above-described problem, a workspace force / acceleration disturbance observer is connected to an impedance-based motion controller within a workspace, and obtains a disturbance estimation value by considering the interaction force and acceleration applied to the end effector of a robot, and the disturbance estimation value is characterized by being expressed by the following mathematical formula.

[0006] [Mathematical formula]

[0007]

[0008] The above D^ is a disturbance estimate, the above Q is a Q filter, and F c ' is the control input force, and the above M^ o is the mass matrix estimate, the x'' is the acceleration, and the F ext is the interaction force

[0009] According to another aspect of the present invention for solving the above-described problem, a robot is provided, which performs impedance-based motion control within a workspace, including a robot manipulator and a robot controller connected to the robot manipulator, wherein the robot controller includes an impedance-based motion controller and a workspace force / acceleration disturbance observer connected to the impedance-based motion controller, wherein the workspace force / acceleration disturbance observer obtains a disturbance estimate by considering an interaction force and acceleration applied to an end effector of the robot, and the disturbance estimate is characterized in that it is expressed by the following mathematical formula.

[0010] [Mathematical formula]

[0011]

[0012] The above D^ is a disturbance estimate, the above Q is a Q filter, and F c ' is the control input force, and the above M^ o is the mass matrix estimate, the x'' is the acceleration, and the F ext is the interaction force

[0013] Other specific details of the present invention are included in the detailed description and drawings.

[0014] According to the present invention, a disturbance observer loop is designed by considering both interaction force and acceleration, so precise motion tracking is possible even at low impedance gain settings.

[0015] According to the present invention, safe contact is possible while maintaining low impedance through excellent impedance rendering performance.

[0016] The effects of the present invention are not limited to the effects mentioned above, and other effects not mentioned will be clearly understood by those skilled in the art from the description below.

[0017] FIG. 1 is a schematic diagram of a robot control system including a workspace force / acceleration disturbance observer according to one embodiment of the present invention.

[0018] FIG. 2 is a schematic diagram of a robot including a workspace force / acceleration disturbance observer according to another embodiment of the present invention.

[0019] Figure 3 is a schematic diagram showing the performance of the robot of Figure 2 within its free motion space.

[0020] Figure 4 is a schematic diagram showing the performance of the robot of Figure 2 within the contact motion space.

[0021] The advantages and features of the present invention, and the methods for achieving them, will become clearer with reference to the embodiments described in detail below together with the accompanying drawings. However, the present invention is not limited to the embodiments disclosed below and may be implemented in various different forms. These embodiments are provided solely to ensure that the disclosure of the present invention is complete and to fully inform those skilled in the art of the scope of the present invention, and the present invention is defined solely by the scope of the claims.

[0022] The terminology used herein is for the purpose of describing embodiments only and is not intended to limit the present invention. In this specification, the singular also includes the plural unless specifically stated otherwise. As used herein, the terms "comprises" and / or "comprising" do not exclude the presence or addition of one or more other components in addition to the mentioned components. Like reference numerals refer to like components throughout the specification, and "and / or" includes each and any combination of one or more of the mentioned components. Although "first", "second", etc. are used to describe various components, these components are not limited by these terms. These terms are only used to distinguish one component from another. Therefore, it should be understood that a first component mentioned below may also be a second component within the technical spirit of the present invention.

[0023] Unless otherwise defined, all terms (including technical and scientific terms) used herein may be used in their common sense to those skilled in the art to which the present invention pertains. Furthermore, terms defined in commonly used dictionaries are not to be interpreted ideally or excessively unless explicitly and specifically defined otherwise.

[0024] In describing the present invention, if it is judged that the detailed description of related known technology is obvious to a person skilled in the art and may unnecessarily obscure the gist of the present invention, it will be omitted.

[0025] Hereinafter, embodiments of the present invention will be described in detail with reference to the attached drawings.

[0026] In this specification, "dynamics" is used in place of a dynamic model, a dynamic model formula, or a simplified expression of a dynamic equation, an equation of motion, etc. For the purpose of simplicity, the notation and / or meaning of "vector" or "matrix" is omitted in relation to some items of some mathematical formulas.

[0027] Impedance control is widely used in contact-based applications. Impedance controllers determine the force applied to an external force (external force) through rendered impedance and are used in tasks requiring follow-up motion.

[0028] However, disturbances such as friction and model uncertainty are unavoidable issues when controlling multi-link manipulators and degrade the performance of impedance-based motion control. While feedback control utilizing high impedance or feedback gain can suppress disturbances, the impedance must be designed to account for the contact between the robot and its environment to prevent damage or breakage.

[0029] Disturbance observers are widely used to remove disturbances from nominal models. They have been used to control SISO (single input, single output) systems, which require robustness. However, conventional disturbance observers are unsuitable for controlling interaction forces under contact conditions.

[0030] Human-robot interactive tasks require safety assurance against unintended contact between humans and robots. To achieve both precise motion control and safe operation in contact situations, an impedance-based, robust motion controller is required.

[0031] The workspace force / acceleration disturbance observer described below can effectively remove disturbances for precise motion control and safe contact operation. The workspace force / acceleration disturbance observer can be used in the design of low-impedance-based motion controllers.

[0032] FIG. 1 is a schematic diagram of a robot control system including a workspace force / acceleration disturbance observer according to one embodiment of the present invention.

[0033] Referring to FIG. 1, the robot control system (100) includes an impedance-based motion controller (110, Impedance Control) and a workspace force / acceleration disturbance observer (120, WFADOB).

[0034] An impedance-based motion controller (110) controls the motion of a robot based on impedance. The impedance-based motion controller (110) generates an impedance target value for an applied force.

[0035] The dynamics of a manipulator having n degrees of freedom (n is a natural number greater than or equal to 1) in joint space can be expressed by the following mathematical expression 1.

[0036] [Mathematical Formula 1]

[0037]

[0038] Here, q is the joint position, q' is the angular velocity, q'' is the angular acceleration, M(q) is the inertia matrix, C(q, q') is the Coriolis matrix, G(q) is the gravity vector, and τ f is the friction torque vector, τ c is the control torque vector, J(q) is the Jacobian matrix, and F ext Each represents an external force (i.e., interaction force) applied to the robot's end effector in Cartesian space (coordinate system).

[0039] Based on this, the dynamics of the end effector in Cartesian space can be expressed by the following mathematical equation 2.

[0040] [Equation 2]

[0041]

[0042] Here, x represents position, x' represents velocity, and x'' represents acceleration.

[0043] Here, M o(q) is the Cartesian inertia matrix (mass matrix), F f is the friction vector, F c represents the control input force vector, and N(q, q') represents other nonlinear terms, which can be expressed by the following mathematical expression 3.

[0044] [Equation 3]

[0045]

[0046] In the absence of a disturbance observer, the Cartesian coordinate system control input force F output by the impedance-based motion controller (110) c can be expressed by the following mathematical formula 4. The target values ​​of position x, velocity x', and acceleration x'' can be obtained through target error dynamics.

[0047] [Equation 4]

[0048]

[0049] Here, F p is the impedance-based motion control input, N^(q, q') is an estimate of other nonlinear terms including Coriolis force and gravity, and F c force is the external force, M^ o is the mass matrix estimate, M d is the mass target, F ext Each represents an interaction force.

[0050] In order to determine the influence of disturbances such as friction and uncertainty that interfere with impedance control, the above mathematical expression 4 can be replaced with the following mathematical expression 5.

[0051] [Equation 5]

[0052]

[0053] Here, M o is the actual mass matrix, N is other nonlinear terms, F f represent the friction force vectors respectively.

[0054] In the above mathematical expression 5, the system model Mo , N can be refined to represent model uncertainty, and thus the model uncertainty closed-loop error dynamics can be expressed as in Equation 6 below.

[0055] [Equation 6]

[0056]

[0057] Here, e is the error between the position target and the actual position (i.e., the position error), e' is the first derivative of the position error, e'' is the second derivative of the position error, and D d is the damping target, K d is the stiffness target, ΔN is the uncertainty of the nonlinear item estimate, and ΔM o represent the uncertainty of the mass matrix estimate, respectively.

[0058] In the error dynamics of the above mathematical equation 6, the friction force F f and model uncertainty (ΔN and ΔM o It includes disturbances such as x''.

[0059] Impedance-based robot motion control can ensure safe contact with low impedance due to external forces in error dynamics. However, disturbances can negatively impact impedance rendering performance. Therefore, additional control algorithms must address this issue using disturbance observers.

[0060] To this end, the robot control system (100) includes a workspace force / acceleration disturbance observer (120) that considers both interaction force and acceleration.

[0061] Unlike conventional workspace disturbance observers, the workspace force / acceleration disturbance observer (120) includes an external force loop in its nominal model. This allows the workspace force / acceleration disturbance observer (120) to simultaneously ensure robust motion control and safe contact with the environment during free motion.

[0062] For impedance-based robust motion control, the control input force can be expressed by the following mathematical expression 7.

[0063] [Equation 7]

[0064]

[0065] Here, F c ' is the control input force, F p is an impedance-based motion control input, F c force represents the external force, and D^ represents the disturbance estimate.

[0066] And, the final control input force F transmitted to the robot manipulator c can be expressed by the following mathematical formula 8.

[0067] [Equation 8]

[0068]

[0069] The above N^ represents an estimate of other nonlinear terms including the Coriolis force and gravity.

[0070] In mathematical expression 7, the robot control system (100) considers external forces as control purposes rather than disturbances. Accordingly, the estimated disturbance of the disturbance observer (120) can be expressed as mathematical expression 9 below.

[0071] [Equation 9]

[0072]

[0073] Here, Q is the Q filter, and M^ o is the mass matrix estimate, x'' is the Cartesian acceleration, F ext Each represents an interaction force.

[0074] Referring to Equation 9, an external force is input to the Q filter. Consequently, the interaction force exists within the closed-loop impedance dynamics. Substituting Equations 7 and 9 into Equation 2 yields the closed-loop dynamics, which can be expressed as Equation 10 below.

[0075] [Equation 10]

[0076]

[0077] Here, I represents the identity matrix.

[0078] The error dynamics of Equation 10 within the frequency range below the cutoff frequency of the Q filter (low-pass filter) can be expressed by Equation 11 below.

[0079] [Equation 11]

[0080]

[0081] Here, by substituting Equation 4 and Equation 12 below regarding impedance-based motion control input, the closed-loop impedance dynamics of the robot can be calculated using Equation 13 below.

[0082] [Equation 12]

[0083]

[0084] [Equation 13]

[0085]

[0086] According to this, since the disturbance observer (120) considers external force for the purpose of control, the interaction force is maintained according to the target impedance setting, thereby ensuring safe contact.

[0087] In order to explain the gist of the present invention, there are some unexplained contents related to the system (100) illustrated in FIG. 1, but the implementation process and operational effects thereof will be sufficiently understandable to those skilled in the art.

[0088] FIG. 2 is a schematic diagram of a robot including a workspace force / acceleration disturbance observer according to another embodiment of the present invention.

[0089] Referring to FIG. 2, the robot (300) includes a robot controller (100) and a robot manipulator (200).

[0090] A robot (300) according to an embodiment of the present invention may be a robot that performs impedance-based motion control within a work space.

[0091] The robot controller (100) is connected to the robot manipulator (200) and controls the motion of the robot manipulator (200).

[0092] The robot controller (100) can be configured substantially identically to the robot control system described with reference to FIG. 1.

[0093] Although a simplified robot manipulator (200) is illustrated in FIG. 2, the robot manipulator (200) is not limited thereto, and embodiments of the present invention can be implemented using various robot manipulators such as a horizontal or vertical multi-joint robot manipulator having any degree of freedom, an orthogonal robot manipulator, a SCARA robot manipulator, a delta robot manipulator, etc.

[0094] The robot (300) may include a force / torque sensor (not shown, F / T sensor) for measuring interaction force.

[0095] Figure 3 is a schematic diagram showing the performance of the robot of Figure 2 within its free motion space.

[0096] Referring to FIG. 3, the results of the motion of the robot by impedance-based control (a) and the results of the motion of the robot including the workspace force / acceleration disturbance observer according to an embodiment of the present invention (b) are shown.

[0097] Free motion space refers to a condition where there is no contact between the robot and the external environment.

[0098] Using only impedance-based control results in the error dynamics of Equation 6, which incorporates disturbances such as friction and model uncertainty. Since this approach lacks error integration, friction degrades tracking performance, and model uncertainty has a similar effect. Consequently, disturbances in the error dynamics negatively impact impedance rendering performance, potentially leading to positional errors within free-motion space, as shown in (a).

[0099] These obstacles can be overcome by using additional control algorithms, such as a workspace force / acceleration disturbance observer according to an embodiment of the present invention. According to an embodiment of the present invention, the workspace force / acceleration disturbance observer eliminates disturbances, thereby ensuring highly precise tracking performance, as illustrated in (b).

[0100] Figure 4 is a schematic diagram showing the performance of the robot of Figure 2 within the contact motion space.

[0101] Referring to FIG. 4, the results of the motion of a robot by a conventional workspace disturbance observer (a) and the results of the motion of a robot including a workspace force / acceleration disturbance observer according to an embodiment of the present invention (b) are shown.

[0102] Contact motion space refers to the conditions under which there is contact between the robot and the external environment.

[0103] Conventional workspace disturbance observers, because they remove the influence of external forces from error dynamics, can potentially cause damage to the robot or the environment due to interaction forces in contact situations. That is, conventional workspace disturbance observers generate control inputs solely to reduce errors without considering safety or tracking, which can cause damage to the environment, as shown in (a).

[0104] This problem can be solved by using a workspace force / acceleration disturbance observer according to an embodiment of the present invention. According to an embodiment of the present invention, since the workspace force / acceleration disturbance observer considers external forces for control purposes, the interaction force is maintained according to the impedance target value setting, ensuring safe contact as illustrated in (b).

[0105] According to the present invention, a disturbance observer loop is designed by considering both interaction force and acceleration, so precise motion tracking is possible even at low impedance gain settings.

[0106] According to the present invention, safe contact is possible while maintaining low impedance through excellent impedance rendering performance.

[0107] While the embodiments of the present invention have been described above with reference to the attached drawings, those skilled in the art will appreciate that the present invention can be implemented in other specific forms without altering the technical concept or essential features thereof. Therefore, the embodiments described above should be understood to be illustrative in all respects and not restrictive.

Claims

1. Connected to an impedance-based motion controller within the workspace, Obtain disturbance estimates by considering the interaction force and acceleration applied to the robot's end effector. A workspace force / acceleration disturbance observer, characterized in that the above disturbance estimate is expressed by the following mathematical expression 1. [Mathematical Formula 1] The above D^ is a disturbance estimate, the above Q is a Q filter, and F c ' is the control input force, and the above M^ o is the mass matrix estimate, the x'' is the acceleration, and the F ext is the interaction force 2. In paragraph 1, The above control input force is expressed by the following mathematical expression 2: Workspace force / acceleration disturbance observer. [Mathematical formula 2] F above p is an impedance-based motion control input by the impedance-based motion controller, and the F c force is an external force 3. In paragraph 2, Final control input force F transmitted to the robot manipulator c is expressed by the following mathematical formula 3, Workspace force / acceleration disturbance observer. [Mathematical Formula 3] The above N^ is an estimate of other nonlinear terms including the Coriolis force and gravity.

4. In paragraph 2, The above external force is expressed by the following mathematical formula 4: Workspace force / acceleration disturbance observer. [Mathematical formula 4] Above M^ o is the mass matrix estimate, and the above M d is the mass target 5. In paragraph 2, The above impedance-based motion control input is expressed by the following mathematical expression 5: Workspace force / acceleration disturbance observer. [Mathematical Formula 5] Above M^ o is the mass matrix estimate, and the above x'' d is the acceleration target value, and the above M d is the mass target, e is the position error, e' is the first derivative of the position error, and D d is the damping target, and K d is the stiffness target 6. In paragraph 5, The closed loop dynamics is expressed by the following mathematical equation 6: Workspace force / acceleration disturbance observer. [Mathematical Formula 6] Above M^ o is the mass matrix estimate, the x'' is the acceleration, the I is the identity matrix, and the ΔM o is the uncertainty of the mass matrix estimate, and F f is the frictional force, and ΔN is the uncertainty in the estimate of the nonlinear terms including the Coriolis force and gravity.

7. In paragraph 6, In the frequency range below the cutoff frequency of the above Q filter, the closed loop dynamics are expressed by the following mathematical expression 7: Workspace force / acceleration disturbance observer. [Mathematical formula 7] 8. In paragraph 7, The above closed loop dynamics is expressed by the following mathematical equation 8: Workspace force / acceleration disturbance observer. [Mathematical formula 8] Above M d is the mass target, e is the position error, e' is the first derivative of the position error, e'' is the second derivative of the position error, and D d is the damping target value, and the above K d is the stiffness target 9. A robot that performs impedance-based motion control within a workspace, Robot manipulator; and A robot controller connected to the above robot manipulator is included, The above robot controller comprises an impedance-based motion controller and a workspace force / acceleration disturbance observer connected to the impedance-based motion controller, The above workspace force / acceleration disturbance observer is, Obtain disturbance estimates by considering the interaction force and acceleration applied to the robot's end effector. A robot characterized in that the above disturbance estimate is expressed by the following mathematical expression 9. [Mathematical formula 9] The above D^ is a disturbance estimate, the above Q is a Q filter, and F c ' is the control input force, and the above M^ o is the mass matrix estimate, the x'' is the acceleration, and the F ext is the interaction force

Images