An adaptive hybrid impedance control method for space manipulator without force sensor

Abstract

The present invention provides a method for adaptive hybrid impedance control of a space manipulator without a force sensor, comprising: establishing a decoupled recursive adaptive dynamic controller to adaptively process the inertial parameters of the space manipulator; establishing an external force estimator without a force sensor to estimate the force at any position of the manipulator without a force sensor; establishing adaptive hybrid impedance control based on the decoupled recursive adaptive dynamic controller and the external force estimator without a force sensor to obtain an adaptive hybrid impedance control strategy without a force sensor, and realizing position control or impedance control in a task subspace divided according to requirements. The method of the present invention can significantly improve the adaptive ability of the space manipulator to the uncertainty of its own dynamic parameters and environmental parameters when performing contact tasks.

Description

Technical Field

[0001] The invention relates to an adaptive hybrid impedance control method for a space manipulator, and belongs to the technical field of space manipulator control. Background Art

[0002] High-precision force control of space manipulators generally requires six-dimensional force sensors or joint force sensors at the end to measure the magnitude of external force. However, for space manipulators, the six-dimensional force sensors or joint force sensors at the end greatly increase the weight and cost of the system, and it is difficult to meet the requirements of the space environment. Therefore, an adaptive hybrid impedance control method for space manipulators without force sensors is proposed. Based on a composite adaptive dynamics adaptive control algorithm and high-precision external force estimation, a composite adaptive hybrid impedance control method without force sensors is realized, which is of great significance to the on-orbit application of space robots.

[0003] Feedforward design based on dynamic models is an important part of the space robot control algorithm, which can greatly improve the accuracy, response speed and stability of the space robot control system. However, in actual systems, it is usually difficult to accurately obtain all inertial parameters, and there is a coupling of dynamic characteristics between various multi-body systems, which poses a great challenge to the accuracy of dynamic modeling and the real-time performance of calculations. Although the adaptive control scheme can ensure the tracking accuracy of the manipulator when the parameter uncertainty is large, the adaptive control algorithm based on the dynamic model has many parameters and is mutually coupled, making it difficult to achieve parameter adaptation. Summary of the invention

[0004] The technical problem to be solved by the present invention is: to overcome the shortcomings of the prior art, the present invention provides an adaptive hybrid impedance control method for a space robot arm that does not require a force sensor, and proposes an adaptive hybrid impedance control strategy, which significantly improves the adaptive ability of the space robot arm to the uncertainty of its own dynamic parameters and environmental parameters when performing contact tasks.

[0005] In order to achieve the above object, the main technical solution of the present invention is: an adaptive hybrid impedance control method for a space manipulator without a force sensor, comprising:

[0006] A decoupled recursive adaptive dynamics controller is established to adaptively process the inertia parameters of the space manipulator.

[0007] Establish an external force estimator without force sensor to estimate the force at any position of the robot arm without force sensor;

[0008] An adaptive hybrid impedance control is established based on a decoupled recursive adaptive dynamics controller and an external force estimator without force sensors. An adaptive hybrid impedance control strategy without force sensors is obtained to achieve position control or impedance control in task subspaces divided according to requirements.

[0009] Furthermore, the establishment of a decoupled recursive adaptive dynamics controller to adaptively process the inertia parameters of the space manipulator includes:

[0010] Establish a composite decoupling control law;

[0011] Establish a decoupled composite adaptive law to handle uncertain dynamic parameters;

[0012] The composite decoupling control law and the decoupling composite adaptive law are used to make the manipulator run along an irregular continuous trajectory in free space until the convergence of the inertia parameters meets the requirements.

[0013] Furthermore, the establishing of the composite decoupling control law includes:

[0014] The joint output control torque is calculated recursively according to the following control law equation:

[0015]

[0016]

[0017]

[0018]

[0019]

[0020] τ i =z6 T F ri ,

[0021] Among them, the generalized velocity of the i-th connecting rod is n is a positive integer;

[0022] v i is the linear velocity of the i-th link of the space manipulator in its body coordinate system ∑ i In the representation, ω i is the linear velocity of the i-th link of the space manipulator in its body coordinate system ∑ i In the representation;

[0023] i T i-1 is the coordinate system ∑ i-1 To coordinate system∑ i The generalized transformation matrix of :

[0024] i C i-1 is the rotation matrix, is the coordinate system ∑ i The origin is in the coordinate system ∑i-1 In the representation;

[0025] τ i is the output control torque of joint i, is the joint angular velocity The i-th component of , q is the joint angle, is the joint reference velocity The i-th component of the trajectory; z6 is a column vector, z6 = [0, 0, 0, 0, 1] T ; V ri is the reference speed of connecting rod i; V r0 V ri The initial value of F ri is the reference force of connecting rod i;

[0026] M i is the inertia matrix of link i, C i is the Coriolis and centrifugal force matrix of connecting rod i, G i is the gravity matrix of link i; Y ri is the linear regression matrix of the kinetic equation, θ i is the corresponding space manipulator inertia parameter vector; M i , C i , G i ,θ i Estimated value, K Di is a positive definite symmetric matrix;

[0027]

[0028]

[0029]

[0030] in,() × represents the cross product matrix; m i is the mass of connecting rod i; 3×3 is the identity matrix; i I i is the moment of inertia matrix of connecting rod i, is the center of mass position of connecting rod i, i C ci is the body coordinate system∑ i The rotation matrix relative to the inertial frame.

[0031] Furthermore, the establishment of a decoupled composite adaptive law to process uncertain dynamic parameters includes:

[0032] Use the mechanical energy of the robot arm to build a linear relationship with the dynamic parameters:

[0033] E j =U j (V j )θ j ,

[0034] Among them, E j is the mechanical energy of connecting rod j, U j is the regression matrix of the linear relationship between the mechanical energy and dynamic parameters of the robot;

[0035] The power of the robotic arm is:

[0036] Where τ is the output control torque τ of joint i i as a column vector of elements;

[0037] The linear relationship for the dynamics used in the composite adaptive algorithm is derived using a first-order filter:

[0038]

[0039]

[0040] Among them, λ is a positive constant, p is a differential operator, W j The linear relationship between the mechanical energy and dynamic parameters of the manipulator is about the inertia parameter θ j The regression matrix, j = 1, 2, ..., n;

[0041] The adaptive law for each joint is solved by the following formula:

[0042]

[0043] in, Sliding mode function s i =V i -V ri , P i is a positive definite matrix.

[0044] Furthermore, the composite decoupling control law and the composite decoupling adaptive law are used to make the manipulator run along an irregular continuous trajectory in free space until the convergence of the inertia parameters meets the requirements, including:

[0045] The expression of the irregular continuous trajectory is:

[0046] q d =q d (t)

[0047]

[0048] Among them, q d is the desired trajectory of the joint angle, is the expected trajectory of the joint angular velocity;

[0049] Joint reference speed Among them, Λ is a positive definite matrix;

[0050] The joint reference velocity expression Combined with the composite decoupling control law and the decoupling composite adaptive law, the inertia parameter θ of the manipulator is i Perform adaptive control.

[0051] Furthermore, the establishment of an external force estimator without a force sensor to estimate the force at any position of the manipulator without a force sensor includes:

[0052] The Lagrangian representation of the robot arm dynamics model is:

[0053]

[0054] Among them, H is the inertia matrix of the Lagrangian form equation; C is the centrifugal Coriolis force matrix of the Lagrangian form equation; G is the gravity matrix of the Lagrangian form equation; f is the joint friction force, τ e It is the value of the external force on the robot arm equivalent to the joint space;

[0055] The momentum of the robot arm is

[0056] Then the value τ equivalent to the external force on the robot arm in the joint space is estimated according to the following formula: e :

[0057]

[0058] Among them, K e is a positive constant matrix, is the estimated value of the friction force f fitted by the Coulomb friction model, p(0) is the initial value of p, and t is the time;

[0059] F e It is the combination of external force and torque on the end Cartesian space manipulator. If there is no force sensor, the estimated value of the joint external force is The estimated value is calculated by the following formula

[0060]

[0061] J is the Jacobian matrix of the robot arm.

[0062] Furthermore, the adaptive hybrid impedance control is a dual-loop architecture, and the adaptive hybrid impedance control strategy without the need for a force sensor adopts inner-loop and outer-loop control.

[0063] Furthermore, the outer loop of the adaptive hybrid impedance control strategy obtains a reference control quantity according to the desired hybrid impedance equation, including:

[0064] Formulate the desired mixed impedance equation:

[0065]

[0066] Where S is a diagonal matrix composed of 0 and 1, which is used to divide the force control and position loop control subspaces; M d , B d , K d are the desired inertia matrix, damping matrix, and stiffness matrix of the robot arm respectively; X is the end position of the robot arm, X d is the desired terminal pose trajectory;

[0067] The reference speed of the task space is:

[0068]

[0069] Reference velocity in joint space

[0070] Furthermore, the inner loop of the adaptive hybrid impedance control strategy adopts the decoupled recursive adaptive control law of the control law equation and Substitution

[0071]

[0072] An adaptive hybrid impedance control system for a space manipulator without a force sensor, comprising:

[0073] The first module is used to establish a decoupled recursive adaptive dynamics controller to perform adaptive processing on the inertia parameters of the space manipulator;

[0074] The second module is used to establish an external force estimator without a force sensor, so as to estimate the force at any position of the robot arm without a force sensor;

[0075] The third module is used to establish adaptive hybrid impedance control based on the decoupled recursive adaptive dynamic controller and the external force estimator without force sensor, obtain an adaptive hybrid impedance control strategy without force sensor, and realize position control or impedance control in the task subspace divided according to the requirements.

[0076] The advantages of the present invention compared with the prior art are:

[0077] (1) The method proposed in the present invention can ensure the adaptation to the uncertain parameters and the convergence of the control when the dynamic parameters of the space robot are uncertain or change due to the space environment. It does not require a force sensor to realize the external force perception of any part of the space robot, and realizes the adaptive hybrid impedance control of the space robot, and realizes the coordinated control of the position and force of the robot's task space, which significantly improves the adaptive ability of the space robot to the uncertainty of its own dynamic parameters and environmental parameters when performing contact tasks.

[0078] (2) The present invention proposes a parameter adaptive controller driven by tracking error and prediction error, which decomposes a complex multi-degree-of-freedom dynamic system into multiple relatively simple dynamic subsystem modules with the same characteristics, and realizes the adaptation of control parameters for each subsystem module, thereby improving the anti-interference ability and system stability of the control system. On this basis, the external force is estimated according to the system dynamics model to realize adaptive hybrid impedance control without the need for force sensors.

[0079] (3) The method of the present invention better handles the contradiction between accuracy and stability by combining the advantages of force-position hybrid control and impedance control; designs a control algorithm based on a decoupled dynamic model, and combines the adaptive law to handle the uncertainty of modeling to obtain a high-precision dynamic model. On this basis, a dual-loop control strategy is established by estimating the external force. The strategy is based on inner and outer loop control, wherein the outer loop obtains a reference control quantity according to the desired hybrid impedance equation, and the inner loop adopts the above-mentioned recursive composite adaptive control algorithm to make the system state quantity track the reference control quantity, thereby realizing the desired hybrid impedance equation. BRIEF DESCRIPTION OF THE DRAWINGS

[0080] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creativity and labor.

[0081] Figure 1 is a coordinate system relationship diagram in the method of the present invention;

[0082] Figure 2 Schematic diagram of the inner / outer loop control strategy structure;

[0083] Figure 3 It is a schematic diagram of the initial configuration of the seven-degree-of-freedom space manipulator;

[0084] Figure 4 is the joint position error map;

[0085] Figure 5 It is the parameter convergence diagram;

[0086] Figure 6 is the trajectory (unit: m) of the end of the robot arm on the horizontal plane;

[0087] Figure 7 It is the simulation diagram of the terminal posture error;

[0088] Figure 8 This is the contact force simulation diagram in z direction;

[0089] Fig. 9 It is the joint torque simulation diagram. DETAILED DESCRIPTION

[0090] In order to better understand the technical solution of the present invention, the embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0091] It should be clear that the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0092] An adaptive hybrid impedance control method for a space manipulator without a force sensor comprises the following steps:

[0093] Step 1: Establish a decoupled recursive adaptive dynamics controller to perform adaptive processing on the inertia parameters of the space manipulator. The specific steps are:

[0094] a) Establish a composite decoupling control law.

[0095] The coordinate system of the spatial manipulator with n degrees of freedom is shown in Figure 1 , record the linear velocity and angular velocity of the i-th link of the space manipulator in its body coordinate system ∑ i In the figure, they are represented as v i ∈ 3×1 and ω i ∈ 3×1 , define the generalized velocity V of the i-th connecting rod i is the combination of linear velocity and angular velocity:

[0096]

[0097] i=1, 2, ..., n; n is a positive integer;

[0098] definition i T i-1 is the coordinate system ∑ i-1 To coordinate system∑ i The generalized transformation matrix of :

[0099]

[0100] in, i C i-1 is the rotation matrix, is the coordinate system ∑ i The origin is in the coordinate system ∑ i-1 In the representation.

[0101] The joint output control torque is calculated recursively according to the following control law equation:

[0102]

[0103]

[0104]

[0105]

[0106] τ i =z6 T F ri (7)

[0107] Among them, τ i is the output control torque of joint i, is the joint angular velocity The i-th component of (joint angle is denoted as q), is the joint reference velocity The i-th component of the trajectory will be defined in c) and step 3; z6 is a column vector, z6 = [0, 0, 0, 0, 1] T , V ri is the reference speed of connecting rod i; V r0 V ri The initial value of F ri is the reference force of connecting rod i; M i , C i , G i They are the inertia matrix, Coriolis and centrifugal force matrix, and gravity matrix of connecting rod i, Y ri is the linear regression matrix of the kinetic equation, θ i is the corresponding inertia parameter vector; It is M i , C i , G i ,θ i Estimated value, K Di ∈ 6×6 is a positive definite symmetric matrix.

[0108] M i , C i , G i The specific form of is as follows, wherei I i is the moment of inertia matrix of connecting rod i, is the center of mass position of connecting rod i, i C ci is the body coordinate system∑ i The rotation matrix relative to the inertial frame.

[0109]

[0110]

[0111]

[0112] in,() × represents the cross product matrix; m i is the mass of connecting rod i; 3×3 is the unit matrix; g is the gravitational acceleration;

[0113] b) Establish a decoupled composite adaptive law to deal with uncertain dynamic parameters.

[0114] Since the dynamic parameters are unknown, the above control law uses the estimated values ​​of the inertia parameters. In order to establish a recursive composite adaptive law, the mechanical energy of the manipulator is used to build a linear relationship with the dynamic parameters:

[0115]

[0116] E j is the mechanical energy of connecting rod j, E j (0) is a constant, U j It is the regression matrix of the linear relationship between the mechanical energy and dynamic parameters of the robot.

[0117] The derivative of mechanical energy is the power of the robot arm, that is, the power of the joint work:

[0118]

[0119] τ is the output control torque τ of joint i i as a column vector of elements;

[0120] It can be seen that the above power is a computable quantity. In order to avoid using the joint angular acceleration, a first-order filter is used to obtain a linear relationship of dynamics that can be applied to the composite adaptive algorithm:

[0121]

[0122] λ is a positive constant, p is a differential operator, W j The linear relationship between the mechanical energy and dynamic parameters of the manipulator is about the inertia parameter θ jThe regression matrix, j = 1, 2, ..., n, the adaptive law of each joint can be solved by the following formula:

[0123]

[0124] in, Sliding mode function s i =V i -V ri , P i is a positive definite matrix.

[0125] c) Using the above composite decoupling control law and decoupling composite adaptive law, the robot arm is allowed to run along an irregular continuous trajectory in free space for a sufficient time until the convergence of the inertial parameters meets the requirements.

[0126] The expression of irregular continuous trajectory is:

[0127]

[0128] q d is the desired trajectory of the joint angle, is the expected trajectory of the joint angular velocity;

[0129] Then let the Calculate according to the following formula

[0130]

[0131] Among them, Λ is a positive definite matrix. Combining the above control law and adaptive law, we can get the complete manipulator inertia parameter θ i Identification process control algorithm. The above adaptive algorithm uses tracking error and prediction error as two types of driving signals for adaptive control. The parameter convergence speed and accuracy are higher than the general model reference adaptive algorithm, and the calculation amount is O(n), which is one order lower than the general composite adaptive algorithm.

[0132] Step 2: Establish an external force estimator without the need for a force sensor to estimate the force at any position of the robotic arm without the need for a force sensor.

[0133] For the joints of the robot without joint force sensors, estimate the joint output torque and the six-dimensional force on the robot in space. According to the above steps, an accurate dynamic model of the robot is obtained, and its Lagrangian expression is:

[0134]

[0135] H, C, G are the inertia matrix, centrifugal Coriolis force matrix, and gravity matrix of the Lagrangian equation respectively. f is the joint friction force, τ e It is the value of the external force on the robot arm equivalent to the joint space.

[0136] The momentum of the robot arm is

[0137]

[0138] Then the value τ equivalent to the external force on the robot arm in the joint space is estimated according to the following formula: e :

[0139]

[0140] Among them, K e is a positive constant matrix; is the estimated value of the friction force f fitted by the Coulomb friction model, which can be calculated using the joint angular velocity; the output torque τ is the result calculated by the control algorithm and is a known quantity; p(0) is the set initial value of p, and t is the time.

[0141] F required for adaptive hybrid impedance control e It is the combination of the external force and torque on the end Cartesian space manipulator. If there is no force sensor, the joint external force can be estimated Calculated by the following formula:

[0142]

[0143] J is the Jacobian matrix of the robot;

[0144] Step 3, based on the decoupled recursive adaptive dynamic controller in step 1 and the external force estimator without force sensor in step 2, an adaptive hybrid impedance control is established to obtain an adaptive hybrid impedance control strategy without force sensor, and position control or impedance control is realized in the task subspace divided according to the requirements.

[0145] The adaptive hybrid impedance control architecture is established as a dual-loop architecture, such as Figure 2 As shown, the strategy is based on inner and outer loop control.

[0146] (1) The outer loop obtains the reference control quantity according to the desired mixed impedance equation.

[0147] First, establish the desired mixed impedance equation:

[0148]

[0149] Where S is a diagonal matrix composed of 0 and 1, which is used to divide the force control and position loop control subspaces; M d , B d , K d are the desired inertia matrix, damping matrix, and stiffness matrix of the robot arm, respectively. X is the end position of the robot arm, which can be calculated by the joint angle according to the kinematic equation; X dis the desired end pose trajectory.

[0150] The reference speed of the task space is defined as:

[0151]

[0152] The reference velocity in joint space is:

[0153]

[0154] (2) The inner loop adopts the decoupled recursive adaptive control law of formulas (1) to (7), and substitutes formula (23) into formula (5).

[0155] The above together constitute an adaptive hybrid impedance method for a space manipulator without the need for a force sensor.

[0156] An adaptive hybrid impedance control system for a space manipulator without a force sensor, comprising:

[0157] The first module is used to establish a decoupled recursive adaptive dynamics controller to perform adaptive processing on the inertia parameters of the space manipulator;

[0158] The second module is used to establish an external force estimator without a force sensor, so as to estimate the force at any position of the robot arm without a force sensor;

[0159] The third module is used to establish adaptive hybrid impedance control based on the decoupled recursive adaptive dynamic controller and the external force estimator without force sensor, obtain an adaptive hybrid impedance control strategy without force sensor, and realize position control or impedance control in the task subspace divided according to the requirements.

[0160] Example:

[0161] The large robotic arm of the Chinese space station is used as a simulation model. Figure 3 , the decoupled recursive adaptive control algorithm is used to perform trajectory tracking control on the robot arm, and the joint position error is obtained as Figure 4 , the parameter convergence is as follows Figure 5 .from Figure 5 It can be seen that the algorithm has advantages in the convergence accuracy and convergence time of trajectory tracking error, and the parameters can also converge to the true value.

[0162] On this basis, adaptive hybrid impedance control simulation is carried out. The robot arm is force controlled in the vertical direction and position controlled in other directions. The motion constraint is to keep the end attitude angle unchanged and track a heart-shaped line in the horizontal direction. The trajectory equation is:

[0163]

[0164] t is time in seconds. The vertical direction is given by F downward cz =10N command.

[0165] Figure 6 The trajectory of the end of the robot arm in the XY plane is given, and it can be seen that the trajectory tracking task is well completed in the position control subspace. Figure 8 It is the variation curve of the contact force in the z direction. At the beginning of the simulation, the end of the robot arm is not in contact with the ground. Under the action of the bias force command, the end moves downward with the set impedance and contacts the ground after about 12 seconds, with a slight impact. Figure 7 The tracking error in the position control direction is given. On the basis of ensuring the convergence of the contact force in the vertical direction, the horizontal direction still has a high control accuracy. The position control error is less than 0.01mm, and the angle control error is better than 3×10 -5 rad, the maximum error occurs at the moment of contact. The z direction belongs to the force control subspace, Figure 8 The deviation is the deviation from the initial position. Fig. 9 It is the output torque of the joint. The torque size meets the requirements of the motor output and is relatively smooth without high-frequency oscillation.

[0166] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

[0167] The contents not described in detail in the specification of the present invention belong to the common knowledge of those skilled in the art.

Claims

1. An adaptive hybrid impedance control method for a space manipulator without a force sensor, characterized in that: include: A decoupled recursive adaptive dynamics controller is established to adaptively process the inertia parameters of the space manipulator. Establish an external force estimator without force sensor to estimate the force at any position of the robot arm without force sensor; Based on the decoupled recursive adaptive dynamic controller and the external force estimator without force sensor, an adaptive hybrid impedance control is established, and an adaptive hybrid impedance control strategy without force sensor is obtained to realize position control or impedance control in the task subspace divided according to the requirements; The method of establishing a decoupled recursive adaptive dynamics controller to adaptively process the inertia parameters of the space manipulator includes: Establish a composite decoupling control law; Establish a decoupled composite adaptive law to handle uncertain dynamic parameters; The composite decoupling control law and the decoupling composite adaptive law are used to make the manipulator run along an irregular continuous trajectory in free space until the convergence of the inertia parameters meets the requirements; The establishing of the composite decoupling control law comprises: The joint output control torque is calculated recursively according to the following control law equation: Among them, the generalized velocity of the i-th connecting rod is n is a positive integer; v i is the linear velocity of the i-th link of the space manipulator in its body coordinate system Σ i In the representation, ω i is the linear velocity of the i-th link of the space manipulator in its body coordinate system Σ i In the representation; i T i-1 is the coordinate system Σ i-1 To coordinate system∑ i The generalized transformation matrix of : i C i-1 is the rotation matrix, is the coordinate system ∑ i The origin is in the coordinate system ∑ i-1 In the representation; τ i is the output control torque of joint i, is the joint angular velocity The i-th component of , q is the joint angle, is the joint reference velocity The i-th component of the trajectory; z6 is a column vector, z6 = [0,0,0,0,0,1] T ; V ri is the reference speed of connecting rod i; V r0 V ri The initial value of F ri is the reference force of connecting rod i; M i is the inertia matrix of link i, C i is the Coriolis and centrifugal force matrix of connecting rod i, G i is the gravity matrix of link i; Y ri is the linear regression matrix of the kinetic equation, θ i is the corresponding space manipulator inertia parameter vector; M i ,C i ,G i ,θ i Estimated value, K Di is a positive definite symmetric matrix; Among them, m i is the mass of connecting rod i; 3×3 is the identity matrix; i I i is the moment of inertia matrix of connecting rod i, is the center of mass position of connecting rod i, i C ci is the body coordinate system∑ i The rotation matrix relative to the inertial frame.

2. The method for adaptive hybrid impedance control of a space manipulator without a force sensor according to claim 1, characterized in that: The method of establishing a decoupled composite adaptive law to process uncertain dynamic parameters includes: Use the mechanical energy of the robot arm to build a linear relationship with the dynamic parameters: Yes j =You j (V j )θ j , Among them, E j is the mechanical energy of connecting rod j, U j is the regression matrix of the linear relationship between the mechanical energy and dynamic parameters of the robot; The power of the robotic arm is: Where τ is the output control torque τ of joint i i as a column vector of elements; The linear relationship for the dynamics used in the composite adaptive algorithm is derived using a first-order filter: Among them, λ is a positive constant, p is a differential operator, W j The linear relationship between the mechanical energy and dynamic parameters of the manipulator is about the inertia parameter θ j The regression matrix, j = 1, 2, ..., n; The adaptive law for each joint is solved by the following formula: in, Sliding mode function s i =V i -V ri , P i is a positive definite matrix.

3. The method for adaptive hybrid impedance control of a space manipulator without a force sensor according to claim 2, characterized in that: The composite decoupling control law and the composite decoupling adaptive law are used to make the manipulator run along an irregular continuous trajectory in free space until the convergence of the inertial parameters meets the requirements, including: The expression of the irregular continuous trajectory is: q d =q d (t) Among them, q d is the desired trajectory of the joint angle, is the expected trajectory of the joint angular velocity; Joint reference speed Among them, Λ is a positive definite matrix; The joint reference velocity expression Combined with the composite decoupling control law and the decoupling composite adaptive law, the inertia parameter θ of the manipulator is i Perform adaptive control.

4. The method for adaptive hybrid impedance control of a space manipulator without a force sensor according to claim 3, characterized in that: The method of establishing an external force estimator without a force sensor to estimate the force at any position of the robot arm without a force sensor includes: The Lagrangian representation of the robot arm dynamics model is: Among them, H is the inertia matrix of the Lagrangian form equation; C is the centrifugal Coriolis force matrix of the Lagrangian form equation; G is the gravity matrix of the Lagrangian form equation; f is the joint friction force, τ e It is the value of the external force on the robot arm equivalent to the joint space; The momentum of the robot arm is Then the value τ equivalent to the external force on the robot arm in the joint space is estimated according to the following formula: e : Among them, K e is a positive constant matrix, is the estimated value of the friction force f fitted by the Coulomb friction model, p(0) is the initial value of p, and t is the time; F e It is the combination of external force and torque on the end Cartesian space manipulator. If there is no force sensor, the estimated value of the joint external force is The estimated value is calculated by the following formula J is the Jacobian matrix of the robot arm.

5. The method for adaptive hybrid impedance control of a space manipulator without a force sensor according to claim 4, characterized in that: The adaptive hybrid impedance control is a dual-loop architecture, and the adaptive hybrid impedance control strategy without the need for a force sensor adopts inner-loop and outer-loop control.

6. The method for adaptive hybrid impedance control of a space manipulator without a force sensor according to claim 5, characterized in that: The outer loop of the adaptive hybrid impedance control strategy obtains a reference control quantity according to the desired hybrid impedance equation, including: Formulate the desired mixed impedance equation: Where S is a diagonal matrix composed of 0 and 1, which is used to divide the force control and position loop control subspaces; M d , B d , K d are the desired inertia matrix, damping matrix, and stiffness matrix of the robot arm respectively; X is the end position of the robot arm, X d is the desired terminal pose trajectory; The reference speed of the task space is: Reference velocity in joint space 7. The method for adaptive hybrid impedance control of a space manipulator without a force sensor according to claim 6, characterized in that: The inner loop of the adaptive hybrid impedance control strategy adopts the decoupled recursive adaptive control law of the control law equation and Substitution 8. A control system for implementing the adaptive hybrid impedance control method for a space manipulator without a force sensor as claimed in any one of claims 1 to 7, characterized in that: include: The first module is used to establish a decoupled recursive adaptive dynamics controller to perform adaptive processing on the inertia parameters of the space manipulator; The second module is used to establish an external force estimator without a force sensor, so as to estimate the force at any position of the robot arm without a force sensor; The third module is used to establish adaptive hybrid impedance control based on the decoupled recursive adaptive dynamic controller and the external force estimator without force sensor, obtain an adaptive hybrid impedance control strategy without force sensor, and realize position control or impedance control in the task subspace divided according to the requirements.

Images