Master-slave heterogeneous bilateral teleoperation control method based on improved wave variable under limited position, storage medium and robot

Abstract

The invention discloses a position-limited master-slave heterogeneous bilateral teleoperation control method based on an improved wave variable, a storage medium and a robot, and the method comprises the steps: building an improved wave variable algorithm framework, and adding a compensation wave to a slave input wave; a master robot task space reference trajectory and mapping of positions in a master-slave robot task space are constructed, and a master-slave robot joint space reference trajectory is solved based on a closed-loop inverse kinematics algorithm; based on a human operator, a far-end environment and master-slave robot characteristics, a combined robot dynamic model in a joint space is constructed; and aiming at the combined robot dynamic model, designing an adaptive neural network controller under limited joint positions based on a state transfer function. By means of the method, high-precision safety control over teleoperation of the master-slave heterogeneous robot with the limited position is effectively achieved, the stability and transparency of a teleoperation system are improved based on the designed improved wave variable algorithm, and a new safety operation method is provided for teleoperation control of the robot.

Description

Technical Field

[0001] The present invention relates to the technical field of teleoperation control of robots, and in particular to a master-slave heterogeneous bilateral teleoperation control method, a storage medium, and a robot based on improved wave variables under position constraints. Background Art

[0002] As the core carrier of the intelligent manufacturing system, the teleoperation system shows irreplaceable application value in high-risk fields such as space exploration, deep-sea operation, and nuclear industry operation. In the existing technical system, the contradictory optimization of system stability and operation transparency has always been the key problem restricting the development of technology.

[0003] Existing research mainly constructs a passivity control framework based on the wave variable algorithm, but significant defects are exposed in practical applications: First, the traditional wave variable method is only applicable to fixed time-delay scenarios and has insufficient adaptability to time-varying communication delays, resulting in difficulty in ensuring the dynamic stability of the system; Second, the heterogeneous characteristics commonly existing in master-slave robots (such as differences in joint degrees of freedom and unequal motion ranges) cause task space pose deviations in the traditional joint space mapping method, seriously affecting operation accuracy and environmental adaptability; Third, existing control models generally ignore the physical constraint characteristics of the robot body, especially the risk of mechanical interference and system instability that may be caused by joint motion range limitations.

[0004] In terms of dynamic modeling, although intelligent algorithms such as neural networks can effectively compensate for the unmodeled dynamics of nonlinear systems, in a multi-degree-of-freedom heterogeneous architecture, the problem of dimensional explosion in parameter identification significantly increases the computational complexity. At the same time, existing control strategies do not fully consider the cooperative constraint mechanism between the task space and the joint space. When facing complex environmental interactions, the system transparency index shows a non-linear decay characteristic, resulting in a significant reduction in the operational force perception presence.

[0005] It is particularly worth noting that in typical application scenarios such as nuclear fuel operation and minimally invasive surgery, millimeter-level positioning deviations of the robot end effector may cause catastrophic consequences. The position-velocity hybrid mapping method adopted by existing technologies can alleviate some motion disorder problems, but it cannot fundamentally eliminate the kinematic parameter mismatch caused by master-slave heterogeneity, and control instruction jumps are likely to occur when joint constraints are activated, leading to deterioration of the system dynamic quality.

[0006] Based on the above technical defects, how to provide a high-precision and safe control method for master-slave heterogeneous robot teleoperation under position constraints is one of the technical problems that need to be urgently solved by those skilled in the art. Summary of the Invention

[0007] The main object of the present invention is to overcome the disadvantages and deficiencies of the prior art, and to provide a master-slave heterogeneous bilateral teleoperation control method, a storage medium and a robot based on improved wave variables under position constraints. The method of the present invention aims at the problems of unknown internal dynamics, unknown human operator and environmental dynamics in the robot system, and unifies the dynamic models of the master-slave robots, human operators and the environment in the joint space; aiming at the problems of low master-slave tracking position accuracy and difficult obstacle avoidance caused by directly using joint mapping for heterogeneous master-slave robots, the present invention uses the position-position mapping method in the task space to achieve accurate mapping; aiming at the traditional wave variable algorithm in the communication link only for constant time delay and problems such as wave reflection and position drift, the present invention improves the traditional wave variable algorithm to improve the stability and transparency of the system under time-varying delay; on this basis, aiming at the limited joint position of the robot, the present invention uses the idea of state conversion to convert the limited control system into another unlimited system, and proposes an adaptive neural network controller for the unified joint space model combined with the Lyapunov stability theory, and proposes a neural network weight update law to achieve safe and stable control of the system.

[0008] To achieve the above object, the present invention adopts the following technical solutions:

[0009] In the first aspect, the present invention provides a master-slave heterogeneous bilateral teleoperation control method based on improved wave variables under position constraints, including the following steps:

[0010] S1. Establish an improved wave variable algorithm framework, and add a compensation wave to the input wave of the slave end. The compensation wave is calculated based on the position of the end effector in the task space with forward time delay transmitted by the master robot, the desired position of the end effector in the task space received by the slave robot, and the energy library.

[0011] S2. Based on the kinematic models of the master-slave robots, the joint space dynamic model of the typical non-linear teleoperation system, the dynamic model of the human operator, and the dynamic model of the environment, construct a combined joint space dynamic model of the non-linear teleoperation system.

[0012] S3. Construct a reference trajectory of the master robot in the task space and a position-position mapping in the task space of the master-slave robots; the position-position mapping in the task space of the master-slave robots is as follows:

[0013] X sd (t) = SX m (t) + T

[0014] where X m (t) is the position of the end effector of the master robot in the task space, X sd (t) is the desired position of the end effector of the slave robot in the task space, S is the scale factor, and T is the translation factor;

[0015] S4. Solve the reference trajectory of the robot joint space based on the closed-loop inverse kinematics algorithm, and for the combined joint space dynamics model, construct a neural network by dynamically arranging neurons and a state transition function to construct an adaptive neural network impedance controller under joint position constraints, so as to achieve trajectory tracking control of the robot;

[0016] The adaptive neural network impedance controller is as follows:

[0017] Z 1,i = S 1,i - S d,i

[0018] Z 1,i = S 1,i - S d,i ,Z 2.i = X 2.i - α 1,i ,

[0019]

[0020] where s dj,i is the expected value of s 1j,i , S 1,i = [s 11.i , s 12.i , …, s 1n.i T , S d1,i = [s d1,i , s d2,i , …, s dn,i T , is the derivative of the expected value matrix S d,i , q dj,i is the expected trajectory of the original state x 1j,i , τ i is the control input of the robot joint space, Φ i = diag{φ 1,i , φ 2,i ,..., φ n,i}, Φ i -1 is the inverse of the coefficient matrix Φ i , Z 1,i is the tracking error after state transformation, X 2,i = [x 21,i , x 22,i , …, x 2n,i T is the angular velocity of n joints, α 1,i ​​​is the virtual control rate, Z 2.i is the virtual error, K 1,i 、K 2,i is the controller gain matrix, is the transpose of the estimated value of the neural network weights, Υ i (β) = [γ 1,i (||β i -μ 1,i ||), …, γ N,i (||β i -μ N,i ||)] T is the Gaussian radial basis function of the neural network, μ k,i is the center point, ω k,i is the width, N i is the number of grid points of the neural network, is the input of the neural network;

[0021] Construct the weight update law of the estimated value of the neural network weights as follows:

[0022]

[0023] where, Γ i is the gain term of the weight update law, σ i is the design constant of the weight update law.

[0024] As a preferred technical solution, the improved wave variable algorithm framework is as follows:

[0025]

[0026] where, t is the time, is the velocity of the end effector of the master robot in the task space, is the desired velocity of the end effector of the slave robot in the task space, F m (t) is the force received by the end effector of the master robot, F s (t) is the force transmitted by the end effector of the slave robot, T m is the forward transmission delay from the master robot to the slave robot, T s is the backward transmission delay from the slave robot to the master robot, and b is the wave impedance.

[0027] As a preferred technical solution, adding a compensation wave to the input wave of the slave end is specifically:

[0028]

[0029]

[0030] where, us is the input wave at the slave end, is the input wave at the slave end after compensation, Δu s is the compensation wave at the slave end, v s is the output wave at the slave end, E r (t) is the energy storage, λ and μ are design constants for adjusting the compensation speed, X m (t - T m ) is the position of the end - effector task space with forward time delay transmitted by the master robot, X sd (t) is the desired position of the end - effector task space received by the slave robot.

[0031] As a preferred technical solution, step S2 is specifically as follows:

[0032] S21. The master - slave robot kinematic model is as follows:

[0033] X i = f i (q i )

[0034]

[0035] Among them, let i ∈ {m, s} represent the master and slave robots respectively, f i (q i ) is the mapping relationship between the position of the robot end - effector in the task space and the angular positions of each joint of the robot in the joint space, q i is the joint position, is the joint velocity, is the joint acceleration, X i 、 are the position, velocity, and acceleration of the robot end - effector in the task space, J i (q i ) is the Jacobian matrix, is the derivative of the Jacobian matrix;

[0036] S22. The typical non - linear teleoperation system joint - space dynamics model is as follows:

[0037]

[0038] Among them, M q,i (q i ) is the inertia matrix, is the centrifugal force matrix, G q,i (q i ) represents the gravity term, F h is the interaction force between the human operator and the master robot, F e is the interaction force between the environment and the slave robot, τ iThe control input torque for the robot;

[0039] S23. The human operator dynamics model and the environment dynamics model are as follows:

[0040]

[0041] Wherein, is the force exerted by the human operator on the master robot, is the force exerted by the environment on the slave robot, M h , B h , K h are the mass, damping, and elastic coefficient in the human operator dynamics model respectively, M e , B e , K e are the mass, damping, and elastic coefficient in the environment dynamics model respectively;

[0042] S24. The combined joint space dynamics model of the non - linear teleoperation system is as follows:

[0043]

[0044] Wherein, M x,i (q i ) is the inertia matrix under the combined model, is the centrifugal force matrix under the combined model, G x,i (q i ) is the gravity term under the combined model, and its relationship with the joint space dynamics model of the typical non - linear teleoperation system is as follows:

[0045]

[0046] As a preferred technical solution, in step S3, the master robot task - space reference trajectory is specifically:

[0047]

[0048] Wherein, is a given continuous smooth function, X′ md is the desired trajectory of the master robot task - space velocity, Y md is the desired trajectory of the master robot task - space position.

[0049] As a preferred technical solution, in step S4, solving the robot joint - space reference trajectory based on the closed - loop inverse kinematics algorithm is specifically:

[0050] e i (t) = X id (t) - X i (t),

[0051]

[0052] where e i (t) is the attitude error, and X id (t) is the desired position of the end effector of the robot in the task space, and ζ i , α i are design constants for adjusting the solution speed.

[0053] As a preferred technical solution, in step S4, a neural network is constructed by dynamically arranging neurons, specifically as follows:

[0054] S41. Define the parameters of the newly added neurons:

[0055] P i = <μ p,i , ω p,i , W p,i >, where μ p,i , ω p,i , W p,i are the center, width, and weight of the newly added neurons, respectively;

[0056] S42. Define the center of the newly added neurons:

[0057]

[0058] where is the average center position of the neuron set C min,i , C min,i is the set composed of δ i neurons closest to the current input, χ i is a designable parameter for determining the distance between the newly added neuron and the set C min,i ;

[0059] S43. Determine whether to add a new neuron:

[0060] Define an adjustable threshold ε i . When the input β i of the neural network is greater than the average center position min,i of the neuron set C by the threshold, a new neuron is defined according to the set parameters.

[0061] As a preferred technical solution, the joint position limitation conditions are as follows:

[0062]

[0063] where n is the number of robot joints, i represents the master-slave robot, and X 1,i = [x11,i , x 12,i , …, x 1n,i T are the n joint angle positions, θ j,i , are both constants, respectively representing the upper and lower limits of the j-th joint angle position of the master-slave robot,

[0064] The state transition function is as follows:

[0065] x 1j,i = ψ(s 1j,i ) = θ aj,i arctan(s 1j,i ) + θ bj,i

[0066] where s 1j,i represents the state corresponding to x 1j,i after system conversion, and are the conversion function coefficients obtained from the upper and lower bounds of each joint angle position.

[0067] Second, the present invention provides a computer-readable storage medium storing a program, which, when executed by a processor, implements the master-slave heterogeneous bilateral teleoperation control method based on improved wave variables under position constraints.

[0068] Third, the present invention provides a robot, which includes:

[0069] At least one processor; and,

[0070] A memory communicatively connected to the at least one processor; wherein,

[0071] The memory stores computer program instructions executable by the at least one processor, and the computer program instructions are executed by the at least one processor to enable the at least one processor to execute the master-slave heterogeneous bilateral teleoperation control method based on improved wave variables under position constraints.

[0072] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0073] 1. The present invention adopts the position-position mapping in the task space and improves the traditional wave variable algorithm, effectively solving the problems of wave reflection and position drift under time-varying time delay existing in the traditional wave variable algorithm, and improving the stability and transparency of the heterogeneous bilateral teleoperation system under time-varying time delay.

[0074] ​2. The present invention unifies the unknown internal dynamics, the unknown human operator, and the environmental dynamics existing in the master-slave robot system in the joint space, achieving precise identification of the unknown non-linear dynamics in the heterogeneous bilateral teleoperation system.

[0075] 3. The present invention uses the idea of state transformation to transform the teleoperation system with limited joint angle positions into a new system without limitations, and designs an adaptive neural network controller for the new system in combination with the Lyapunov stability theory, achieving safe and stable control of the heterogeneous bilateral teleoperation system under position limitations. BRIEF DESCRIPTION OF THE DRAWINGS

[0076] To more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0077] Figure 1 It is a system flowchart of a master-slave heterogeneous bilateral teleoperation control method based on improved wave variables under position limitations according to an embodiment of the present invention.

[0078] Figure 2 It is a schematic diagram of a link master-slave robot according to an embodiment of the present invention.

[0079] Figure 3 It is a curve graph of the change of the control input signal of the master robot according to an embodiment of the present invention.

[0080] Figure 4 It is a curve graph of the change of the control input signal of the slave robot according to an embodiment of the present invention.

[0081] Figure 5 It is an effect graph of the neural network of the master robot fitting the unknown dynamics according to an embodiment of the present invention.

[0082] Figure 6 It is an effect graph of the neural network of the slave robot fitting the unknown dynamics according to an embodiment of the present invention.

[0083] Figure 7 It is a trajectory tracking graph of the joints of the master robot according to an embodiment of the present invention.

[0084] Figure 8 It is a trajectory tracking graph of the joints of the slave robot according to an embodiment of the present invention.

[0085] Figure 9 It is a communication time delay graph of the master-slave robot according to an embodiment of the present invention.

[0086] Figure 10 It is a trajectory tracking graph of the ends of the master-slave robot according to an embodiment of the present invention.

[0087] Figure 11 This is the trajectory tracking error diagram of the master-slave robot end in the embodiment of the present invention.

[0088] Figure 12 This is the structural block diagram of the robot in the embodiment of the present invention. Detailed implementation manners

[0089] In order to enable those skilled in the art of this technology to better understand the solution of this application, the following will clearly and completely describe the technical solutions in the embodiments of this application with reference to the accompanying drawings in the embodiments of this application. Obviously, the described embodiments are only a part of the embodiments of this application, rather than all the embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative efforts belong to the scope of protection of this application.

[0090] In this application, referring to "embodiment" means that the specific features, structures or characteristics described in combination with the embodiment can be included in at least one embodiment of this application. The phrase appears in various positions in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment mutually exclusive with other embodiments. Those skilled in the art understand explicitly and implicitly that the embodiments described in this application can be combined with other embodiments.

[0091] As Figure 1 shown, this embodiment provides a master-slave heterogeneous bilateral teleoperation control method based on improved wave variables under position constraints. A double-link rigid robot is selected as the model, and its detailed implementation process includes:

[0092] S1. Establish an improved wave variable algorithm framework and add a compensation wave to the input wave of the slave end:

[0093] The improved wave variable algorithm framework is:

[0094]

[0095] where t is time, is the speed of the master robot end effector in the task space, is the desired speed of the slave robot end effector in the task space, F m (t) is the force received by the master robot end effector, F s (t) is the force transmitted by the slave robot end effector, T m is the forward transmission delay from the master robot to the slave robot, T s is the backward transmission delay from the slave robot to the master robot. The selected values of the delay are respectively as Figure 10 、 Figure 11 shown, and b is the wave impedance.

[0096] Further, the force transmitted from the end effector of the slave robot selected in this embodiment is:

[0097]

[0098] where X s (t) is the position of the end effector of the slave robot in the task space, is the velocity of the end effector of the slave robot in the task space, X sd (t) is the desired position of the end effector of the slave robot in the task space, K sp and K sd are the proportional gain matrix and the derivative gain matrix respectively, and is a diagonal positive definite matrix.

[0099] Even further, adding a compensation wave to the input wave of the slave side is specifically:

[0100]

[0101] where u s is the input wave of the slave side, is the compensated input wave of the slave side, Δu s is the compensation wave of the slave side, v s is the output wave of the slave side, E r (t) is the energy library, λ and μ are design constants for adjusting the compensation speed, X m (t - T m ) is the position of the end effector task space of the master robot transmitted with forward time delay, X sd (t) is the desired position of the end effector task space received by the slave robot. In a more specific embodiment, the improved wave variable algorithm and the relevant parameters of the compensation term selected are:

[0102] b = 5, K sp = 0.5, K sd = 0.3, λ = 5, μ = 0.5;

[0103] S2. Based on the characteristics of the human operator, the distal environment and the robot, construct a nonlinear teleoperation system combined joint space dynamics model, construct a reference trajectory of the master robot task space, and a position-position mapping in the master-slave robot task space, specifically including the following contents:

[0104] Based on the characteristics of the robot, the forward kinematic model of the master-slave robot is obtained as:

[0105] X i = f i (q i )

[0106]

[0107] Among them, let \(i\in\{m, s\}\) represent the master and slave robots respectively, \(f\) i (\(q\) i ) is the mapping relationship between the position of the robot end effector in the task space and the angular positions of the joints of the robot in the joint space, \(q\) i =\([q\) 1,i , q\) 2,i \) T is the joint position, is the joint velocity, joint acceleration, \(X\) i =\([x\) i , y\) i , z\) i \) T 、 are the position, velocity, and acceleration of the robot end effector in the task space, \(x\) i 、y\) i 、z\) i are the positions in three directions in the task space, \(J\) i (\(q\) i ) is the Jacobian matrix, is the derivative of the Jacobian matrix.

[0108] According to the structure of the robot, the forward kinematic model selected in this embodiment is:

[0109]

[0110] \(l\) 1,m = 0.52, \(l\) 2,m = 0.48 are the lengths of link 1 and link 2 of the master robot respectively, \(l\) 1,s = 1.04, \(l\) 2,s = 0.96 are the lengths of link 1 and link 2 of the slave robot respectively, as Figure 2 shown.

[0111] The Jacobian matrix of the robot is:

[0112]

[0113] Furthermore, the joint space dynamic model of a typical non-linear teleoperation system is:

[0114]

[0115] Among them, \(M\) q,i (\(q\) i ) is the inertia matrix, is the centrifugal force matrix, \(G\) q,i (\(q\) i ) represents the gravity term, \(F\) h is the interaction force between the human operator and the master robot, \(F\)e is the interaction force between the environment and the slave robot, τ i is the control input torque of the robot.

[0116] Furthermore, the human operator dynamics model and the environment dynamics model are as follows:

[0117]

[0118] where is the force exerted by the human operator on the master robot, is the force exerted by the environment on the slave robot, M h , B h , K h are the mass, damping, and elastic coefficient in the human operator dynamics model respectively, M e , B e , K e are the mass, damping, and elastic coefficient in the environment dynamics model respectively. The relevant parameters of the human operator dynamics model and the environment dynamics model selected in this embodiment are as follows:

[0119]

[0120] Furthermore, based on the above, the combined joint space dynamics model of the nonlinear teleoperation system is as follows:

[0121]

[0122] where M x,i (q i ) is the inertia matrix under the combined model, is the centrifugal force matrix under the combined model, G x,i (q i ) is the gravity term under the combined model, and its relationship with the joint space dynamics model of the typical nonlinear teleoperation system is as follows:

[0123]

[0124] The relevant parameters of the double-link robot model selected in this embodiment are as follows:

[0125]

[0126]

[0127] In this embodiment, it is selected that m 1,m = 0.173 kg and m 2,m = 0.175 kg are the masses of link 1 and link 2 of the master robot respectively, m 1,s = 0.316 kg and m 2,s= 0.323 kg are the masses of the robot link 1 and link 2 respectively, and g = 9.8 m / s 2 is the acceleration due to gravity.

[0128] S3. Construct the reference trajectory of the master robot's task space and the position-position mapping in the master-slave robot task space;

[0129] Design the reference regression trajectory of the task space as:

[0130]

[0131] where is a given continuous smooth function, X′ md is the desired trajectory of the master robot's task space velocity, and Y md is the desired trajectory of the master robot's task space position. The selected reference trajectory of the task space in this embodiment is:

[0132] Y md = [0.8 + 0.1sin(0.5t), 0.4 + 0.1cos(0.5t)] T

[0133] Design the position-position mapping in the master-slave robot task space as:

[0134] X sd (t) = SX m (t) + T

[0135] where X m (t) is the position of the master robot's end effector in the task space, X sd (t) is the desired position of the slave robot's end effector in the task space, S is the proportional matrix, and T is the translation matrix. The parameters related to the position-position mapping in the master-slave robot task space selected in this embodiment are:

[0136]

[0137] S4. Solve the reference trajectory of the robot joint space based on the closed-loop inverse kinematics algorithm. For the combined joint space dynamics model, construct a neural network by using the method of dynamically arranging neurons, and construct an adaptive neural network impedance controller under joint position constraints based on the state transition function, specifically as follows:

[0138] Solve the reference trajectory of the robot joint space based on the closed-loop inverse kinematics algorithm as:

[0139] e i (t) = X id (t) - X i (t),

[0140]

[0141] Among them, e i (t) is the attitude error, and X id (t) is the desired position of the robot end effector in the task space. ζ i , and α i are design constants for adjusting the solution speed. The parameters related to the reference trajectory of the robot joint space solved by the closed-loop inverse kinematics algorithm selected in this embodiment are α m = α s = 2.

[0142] Further, the designed position constraint condition is:

[0143]

[0144] Among them, i represents the master-slave robot. Let X1, X2 be the state variables X 1,i = [x 11,i , x 12,i T is the joint angle position of the two-link robot, and θ j,i , are both constants, representing the upper and lower limits of the joint angle position of the j-th joint of the master and slave robots respectively. The upper and lower limits of the joint angle positions of the master and slave robots selected in this embodiment are:

[0145]

[0146] Further, the state conversion function is:

[0147] x 1j,i = ψ(s 1j,i ) = θ aj,i arctan(s 1j,i ) + θ bj,i

[0148] Among them, s 1j,i represents the state corresponding to x 1j,i after system conversion, and are the conversion function coefficients obtained from the upper and lower bounds of each joint angle position.

[0149] Design the adaptive neural network impedance controller as follows:

[0150] Z 1,i = S 1,i - S d,i

[0151] Z 1,i ​= S 1,i -S d,i , Z 2.i = X 2.i -α 1,i ,

[0152]

[0153] where s dj,i is the expected value of s 1j,i and S 1,i = [s 11.i , s 12.i T , and S d1,i = [s d1,i , s d2,i T , is the derivative of the expected value matrix S d,i , q dj,i is the expected trajectory of the original state x 1j,i , τ i is the robot joint space control input, Φ i = diag{φ 1,i , φ 2,i}, is the inverse of the coefficient matrix Φ i , Z 1,i is the tracking error after state transformation, X 2,i = [x 21,i , x 22,i T is the joint angular velocity, α 1,i is the virtual control rate, Z 2.i is the virtual error, K 1,i , K 2,i are the controller gain matrices, is the transpose of the neural network weight estimate, Υ i (β i ) = [γ 1,i (||β i - μ 1,i ||), …, γ N,i (||β i - μ N,i ||)] T is the Gaussian radial basis function of the neural network, μ k,i is the center point, ω k,i is the width, N i is the number of neural network layout points, is the input of the neural network;

[0154] ​​​Construct the weight estimation value of the neural network The weight update law is as follows:

[0155]

[0156] Among them, Γ i is the gain term of the weight update law, and σ i is the design constant of the weight update law.

[0157] Furthermore, the initial values of each state and parameter settings are as follows: The initial value of the main robot joint angle position is [x 11,m (0), x 12,m (0)] T = [-0.5, 1.0] T , the initial value of the slave robot joint angle position is [x 11,s (0), x 12,s (0)] T = [0.5, 0.5] T , the initial value of the main robot joint angular velocity is [x 21,m (0), x 22,m (0)] T = [0, 0] T , the initial value of the slave robot joint angular velocity is [x 21,s (0), x 22,s (0)] T = [0, 0] T , the initial values of the neural networks of the master and slave robots are The initial node center of the main robot neural network is μ 1,m = [-0.5, 0, -0.5, 0; 0, 0.5, 0.5, 0; 0.5, 0.5, 0.5, 0.5] T , the number of neurons δ min,m in the neuron set C m = 3, the designable parameter χ min,m that determines the distance between the newly added neuron and the set C m = 0.2, the adjustable threshold ε m = 0.5, the neuron width is η m = [0.65, 0.65, 0.65, 0.65] T , the initial node center of the slave robot neural network is μ 1,s = [0, 0, 0, 0; 0, 0.5, 0.5, 0; 0.5, 0.5, 0.5, 0.5] T , the number of neurons δ min,s in the neuron set C s = 3, the designable parameter χ min,s that determines the distance between the newly added neuron and the set C s= 0.3, adjustable threshold ε s = 0.5, neuron width is η s = [0.8, 0.8, 0.8, 0.8] T , the neural network update rate parameter Γ of the master robot m = diag{2, 4}, design constant σ of the neural network weight update law m = 0.001, controller gain is K 1,m = diag{6, 8}, K 2,m = diag{5, 4}, the neural network update rate parameter Γ of the slave robot s = diag{4, 5}, design constant σ of the neural network weight update law s = 0.001, controller gain is K 1,s = diag{8, 10}, K 2,s = diag{5, 5}.

[0158] In this embodiment, the simulation sampling step is set to 0.01 s, and the simulation duration is 50 s. Figure 2 This is a schematic diagram of the link-closed robot according to the embodiment of the present invention, where the first joint is located at the origin of the task space coordinate system.

[0159] Figure 3 This is a graph of the change of the control input signal of the master robot according to the embodiment of the present invention. Figure 4 This is a graph of the change of the control input signal of the joints of the master robot according to the embodiment of the present invention. It can be seen from the figure that the control input signal oscillates when there is an external disturbance, but the time is extremely short, and it is smooth and continuous throughout, ensuring the stability of the system. Figure 5 This is an effect diagram of the neural network of the master robot fitting the unknown dynamics of the system model according to the embodiment of the present invention. Figure 6 This is an effect diagram of the neural network of the slave robot fitting the unknown dynamics of the system model according to the embodiment of the present invention. It can be seen that the neural network output achieves a good approximation of the unknown dynamics. Figure 7 This is a trajectory tracking graph of the joints of the master robot according to the embodiment of the present invention. Figure 8 This is a trajectory tracking graph of the joints of the slave robot according to the embodiment of the present invention. It can be seen that the positions of the robot joints all move under the established constraint conditions, ensuring the safe operation of the robot. Figure 9 This is a communication time-delay graph of the master-slave robots according to the embodiment of the present invention. It can be seen from the figure that there is a time-varying time delay between the master and slave robots and it changes irregularly. Figure 10 This is a trajectory tracking graph of the ends of the master-slave robots according to the embodiment of the present invention. Figure 11 This is a trajectory tracking error graph of the ends of the master-slave robots according to the embodiment of the present invention. It can be seen that the slave robot can accurately track the motion trajectory of the master robot under the condition of time-varying time delay, realizing good transparency of teleoperation.

[0160] Please refer to Figure 12 In one embodiment, a robot implementing a master-slave heterogeneous bilateral teleoperation control method based on improved wave variables under position constraints is provided. The robot 100 may include a first processor 101, a first memory 102, and a bus. It may also include a computer program stored in the first memory 102 and executable on the first processor 101, such as a master-slave heterogeneous bilateral teleoperation control program 103 based on improved wave variables under position constraints.

[0161] Among them, the first memory 102 includes at least one type of readable storage medium, and the readable storage medium includes flash memory, mobile hard disk, multimedia card, card-type memory (such as SD or DX memory, etc.), magnetic memory, magnetic disk, optical disk, etc. In some embodiments, the first memory 102 may be an internal storage unit of the robot 100, such as the mobile hard disk of the robot 100. In some other embodiments, the first memory 102 may also be an external storage device of the robot 100, such as a plug-in mobile hard disk, a Smart Media Card (SMC), a Secure Digital (SD) card, a Flash Card, etc. equipped on the robot 100. Further, the first memory 102 may also include both the internal storage unit of the robot 100 and the external storage device. The first memory 102 can be used not only to store application software installed on the robot 100 and various types of data, such as the code of the master-slave heterogeneous bilateral teleoperation control program 103 based on improved wave variables under position constraints, but also to temporarily store data that has been output or will be output.

[0162] In some embodiments, the first processor 101 may be composed of integrated circuits. For example, it may be composed of a single packaged integrated circuit, or may be composed of multiple integrated circuits with the same or different functions, including a combination of one or more central processing units (CPUs), microprocessors, digital processing chips, graphics processors, and various control chips. The first processor 101 is the control core (Control Unit) of the robot, connecting various components of the entire robot through various interfaces and circuits, and executing various functions of the robot 100 and processing data by running or executing programs or modules stored in the first memory 102 and calling data stored in the first memory 102.

[0163] Figure 12 Only the robot with components is shown. Those skilled in the art can understand that Figure 12The structures shown do not constitute a limitation on the robot 100, and it may include fewer or more components than shown, or combine certain components, or have different component arrangements.

[0164] The master-slave heterogeneous bilateral teleoperation control program 103 based on the improved wave variable stored in the first memory 102 in the robot 100 is a combination of multiple instructions. When running in the first processor 101, it can achieve:

[0165] S1. Establish an improved wave variable algorithm framework, and add a compensation wave to the input wave at the slave end. The compensation wave is calculated based on the end-effector task space position with forward time delay transmitted by the master robot, the desired end-effector task space position received by the slave robot, and the energy library.

[0166] S2. Based on the master-slave robot kinematic model, the typical non-linear teleoperation system joint space dynamics model, the human operator dynamics model, and the environment dynamics model, construct a non-linear teleoperation system combined joint space dynamics model.

[0167] S3. Construct a master robot task space reference trajectory and a position-position mapping in the master-slave robot task space.

[0168] S4. Solve the robot joint space reference trajectory based on the closed-loop inverse kinematics algorithm. For the combined joint space dynamics model, construct a neural network by means of dynamically arranging neurons, and construct an adaptive neural network impedance controller under joint position constraints based on the state transition function to achieve bilateral control of the robot.

[0169] Furthermore, if the modules / units integrated in the robot 100 are implemented in the form of software functional units and sold or used as independent products, they can be stored in a non-volatile computer-readable storage medium. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disc, computer memory, read-only memory (ROM, Read-Only Memory).

[0170] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The program can be stored in a non-volatile computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, storage, database, or other medium used in the various embodiments provided in the present application can include non-volatile and / or volatile memories. Non-volatile memories can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memories can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and Rambus dynamic RAM (RDRAM), etc.

[0171] The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity of description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope described in this specification.

[0172] The above embodiments are preferred embodiments of the present invention, but the embodiments of the present invention are not limited to the above embodiments. Any other changes, modifications, substitutions, combinations, and simplifications made without departing from the spirit and principle of the present invention should be equivalent replacement methods and are all included in the protection scope of the present invention.

Claims

1. A master-slave heterogeneous bilateral teleoperation control method based on improved wave variables under limited position, characterized in that It includes the following steps: S1. Establish an improved wave variable algorithm framework, and add a compensation wave to the input wave at the slave end. The compensation wave is calculated based on the end effector task space position with forward time delay transmitted by the master robot, the expected position of the end effector task space received by the slave robot, and the energy library; S2. Based on the master-slave robot kinematic model, the typical non-linear teleoperation system joint space dynamics model, the human operator dynamics model, and the environment dynamics model, construct a non-linear teleoperation system combined joint space dynamics model; S3. Construct a reference trajectory in the master robot task space and a position-position mapping in the master-slave robot task space. The position-position mapping in the master-slave robot task space is as follows: X sd S_X(t) = m S(t) + T where X m (t) is the position of the end effector of the master robot in the task space, and X sd (t) is the desired position of the end effector of the slave robot in the task space, S is the scale factor, and T is the translation factor; S4. Solve the reference trajectory of the robot joint space based on the closed-loop inverse kinematics algorithm, and for the combined joint space dynamics model, construct a neural network by means of dynamically arranging neurons and construct an adaptive neural network impedance controller under joint position constraints to achieve trajectory tracking control of the robot; The adaptive neural network impedance controller is as follows: Z 1,i = S 1,i - S d,i Z 1,i = S 1,i - S d,i ,Z 2.i = X 2.i - α 1,i , Among them, s dj,i is the expected value of s 1j,i , S 1,i = [s 11.i , s 12.i , …, s 1n.i T , S d1,i = [s d1,i , s d2,i , …, s dn,i T , is the derivative of the expected value matrix S d,i , q dj,i is the expected trajectory of the original state x 1j,i , τ i is the robot joint space control input, Φ i = diag{φ 1,i , φ 2,i ,..., φ n,i}, Φ i -1 is the inverse of the coefficient matrix Φ i , Z 1,i is the tracking error after state transformation, X 2,i = [x 21,i , x 22,i , …, x 2n,i T is the angular velocity of n joints, α 1,i is the virtual control rate, Z 2.i is the virtual error, K 1,i , K 2,i are the controller gain matrices, is the transpose of the neural network weight estimate value, Υ i (β) = [γ 1,i (||β i - μ 1,i ||), …, γ N,i (||β i - μ N,i ||)] T is the Gaussian radial basis function of the neural network, μ k,i is the center point, ω k,i is the width, N i is the number of neural network layout points, is the input of the neural network;​​​ Construct the estimated value of the neural network weights The weight update law is as follows: Among them, Γ i is the gain term of the weight update law, and σ i is the design constant of the weight update law.

2. The master-slave heterogeneous bilateral teleoperation control method based on improved wave variables under position limitation according to claim 1, wherein, The improved wave variable algorithm framework is as follows: where t is time, is the velocity of the end - effector of the master robot in the task space, is the desired velocity of the end - effector of the slave robot in the task space, F m (t) is the force received by the end - effector of the master robot, F s (t) is the force transmitted by the end - effector of the slave robot, T m is the forward transmission delay from the master robot to the slave robot, T s is the backward transmission delay from the slave robot to the master robot, and b is the wave impedance.

3. The master-slave heterogeneous bilateral teleoperation control method based on improved wave variables under limited position according to claim 1, characterized in that The addition of the compensation wave to the input wave at the slave end is specifically as follows: Among them, u s is the input wave at the slave end, is the compensated input wave at the slave end, Δu s is the compensation wave at the slave end, v s is the output wave at the slave end, E r (t) is the energy storage, λ and μ are design constants for adjusting the compensation speed, X m (t - T m ) is the position of the end effector task space with forward time delay transmitted by the master robot, X sd (t) is the expected position of the end effector task space received by the slave robot.

4. The master-slave heterogeneous bilateral teleoperation control method based on improved wave variables under limited position according to claim 1, characterized in that Step S2 is specifically as follows: S21. The master-slave robot kinematic model is as follows: X i = f i (q i ) Among them, let \(i\in\{m, s\}\) represent the master and slave robots respectively, \(f\) i \((q\) i ) represents the mapping relationship between the position of the robot end effector in the task space and the angular positions of the joints of the robot in the joint space, \(q\) i is the joint position, is the joint velocity, joint acceleration, \(X\) i 、 are the position, velocity, and acceleration of the robot end effector in the task space, \(J\) i (q\) i ) is the Jacobian matrix, is the derivative of the Jacobian matrix; S22. The typical non-linear teleoperation system joint space dynamics model is as follows: where, M q,i (q i ) is the inertia matrix, is the centrifugal force matrix, G q,i (q i ) represents the gravity term, F h is the interaction force between the human operator and the master robot, F e is the interaction force between the environment and the slave robot, τ i is the control input torque of the robot; S23. The human operator dynamics model and the environment dynamics model are as follows: Among them, is the force exerted by the human operator on the master robot, F e * is the force exerted by the environment on the slave robot, M h 、B h 、K h are the mass, damping, and elastic coefficient in the human operator's dynamics model, M e 、B e 、K e are the mass, damping, and elastic coefficient in the environment dynamics model respectively; S24. The non-linear teleoperation system combined joint space dynamics model is as follows: Among them, M x,i (q i ) is the inertia matrix under the combined model, is the centrifugal force matrix under the combined model, G x,i (q i ) is the gravity term under the combined model, and its relationship with the typical non-linear teleoperation system joint space dynamics model is as follows:

5. The position-constrained master-slave heterogeneous bilateral teleoperation control method based on improved wave variables according to claim 1, wherein In step S3, the reference trajectory in the master robot task space is specifically: Among them, is a given continuous and smooth function, X′ md is the expected trajectory of the master robot's task space velocity, Y md is the expected trajectory of the master robot's task space position.

6. The method for master-slave heterogeneous bilateral teleoperation control based on improved wave variables under limited position according to claim 1, wherein In step S4, solving the reference trajectory of the robot joint space based on the closed-loop inverse kinematics algorithm is specifically: e i (t) = X id (t) - X i (t), where, e i (t) is the attitude error, X id (t) is the desired position of the robot end effector in the task space, ζ i , α i are design constants for adjusting the solution speed.

7. The master-slave heterogeneous bilateral teleoperation control method based on improved wave variables under position constraints according to claim 1, characterized in that In step S4, constructing a neural network by means of dynamically arranging neurons is specifically: S41. Define the parameters of the newly added neurons: P i = < μ p,i , ω p,i , W p,i > where μ p,i , ω p,i , W p,i are the center, width, and weight of the newly added neuron, respectively; S42. Define the centers of the newly added neurons: Among them, is the average center position of the neuron set C min,i and C min,i is a set composed of δ i neurons closest to the current input, and χ i is a designable parameter for determining the distance between the newly added neuron and the set C min,i ; S43. Determine whether to add new neurons: Define the adjustable threshold ε i , when the input β of the neural network i and the average center position of the neuron ensemble C min,i is greater than the threshold, new neurons are defined according to the set parameters. ​ 8. The master-slave heterogeneous bilateral teleoperation control method based on improved wave variables under position limitation according to claim 1, characterized in that The joint position constraint conditions are as follows: where n is the number of robot joints, i represents the master-slave robots, X 1,i = [x 11,i , x 12,i , …, x 1n,i T are the angular positions of n joints, θ j,i , are both constants, representing the upper and lower limits of the angular position of the j-th joint of the master-slave robots respectively,​ The state transition function is as follows: x 1j,i = ψ(s 1j,i ) = θ aj,i arctan(s 1j,i ) + θ bj,i Among them, s 1j,i represents the state corresponding to x 1j,i after system conversion, and are the conversion function coefficients obtained from the upper and lower bounds of the joint angle positions.

9. A computer-readable storage medium storing a program, characterized in that, When the program is executed by a processor, it implements the master-slave heterogeneous bilateral teleoperation control method based on the improved wave variable under position constraints described in any one of claims 1-8.

10. A robot, characterized in that, The robot includes: At least one processor; and, A memory communicatively connected to the at least one processor; wherein, The memory stores computer program instructions executable by the at least one processor. The computer program instructions are executed by the at least one processor so that the at least one processor can execute the master-slave heterogeneous bilateral teleoperation control method based on the improved wave variable under position constraints described in any one of claims 1-8.

Images