CN-117340877-B - Dead zone self-adaptive compensation control system and method for line-driven continuum robot

CN117340877BCN 117340877 BCN117340877 BCN 117340877BCN-117340877-B

Abstract

The invention discloses a dead zone self-adaptive compensation control system and method of a line-driven continuum robot, comprising a first model module, a dead zone function module, a second model module, a neural network compensator, a system feed-forward channel dead zone compensation module and a self-adaptive compensation module, wherein the first model module is used for establishing a standard dynamics model of the continuum robot, the dead zone function module is used for establishing a dead zone function for driving the line-driven continuum robot, the second model module is used for establishing an uncertain error model based on the standard dynamics model, the neural network compensator comprises two introduced RBF neural networks, one RBF neural network is used for estimating the proposed uncertain error model, the other RBF neural network is used for dead zone compensation of a system feed-forward channel to approach the established dead zone function, and the self-adaptive compensation module is used for updating the established model and the dead zone function in a self-adaptive mode to improve the control precision of the continuum robot.

Inventors

QIAN YUZHE
XIE JUNNAN
LIU WEIPENG

Assignees

河北工业大学

Dates

Publication Date: 20260512
Application Date: 20231018

Claims (7)

1. A dead zone adaptive compensation control system of a line-driven continuum robot, applied to a controller, comprising: The first model module is used for establishing a standard dynamics model of the continuum robot based on the European Bernoulli Liang Dingli and the constant curvature assumption; The dead zone function module is used for establishing a dead zone function for driving the linear driving continuum robot; a second model module for building an uncertainty error model based on the standard kinetic model; The system comprises a neural network compensator, a system feedforward channel and a system feedforward channel, wherein the neural network compensator comprises two introduced RBF neural networks, one RBF neural network is used for estimating the proposed uncertainty error model, and the other RBF neural network is used for dead zone compensation of the system feedforward channel to approximate the established dead zone function; The self-adaptive compensation module is used for updating the established model and dead zone function in a self-adaptive mode so as to improve the control precision of the continuum robot, wherein the model comprises the standard dynamics model and an uncertain error model; The uncertainty error model is designed as follows: ; Wherein, the In order to be an external disturbance, As an error signal, the signal is a signal, Representing a position vector of the continuum robot, Representing the velocity vector of the continuum robot, And (3) with The meaning is the inertial matrix of the system, And (3) with , And (3) with The meaning is the same, as the matrix related to the coriolis centripetal force of the system, And (3) with The same meaning is for a matrix related to gravity and elasticity.
2. The dead zone adaptive compensation control system of a line-driven continuum robot of claim 1, wherein said adaptively comprises: designing the self-adaption rate of the neural network; establishing an error function according to the acquired error signal; and transmitting the error function into the self-adaptive rate of the neural network to carry out compensation updating so as to update the established model and dead zone function.
3. The dead zone adaptive compensation control system of a line-driven continuum robot of claim 1, wherein the dead zone function is representable as: Wherein, the Is the control signal of the input, i.e. the control input before entering the dead zone, And Is a function of two smooth non-linearities, For the output of the control signal through the dead zone, And (3) with Is a dead zone parameter.
4. A dead zone adaptive compensation control system of a line driven continuum robot according to claim 3, characterized in that said standard dynamics model is designed to: Wherein, the Representing a position vector of the continuum robot, Representing the velocity vector of the continuum robot, Representing the acceleration vector of the continuum robot, As an inertial matrix of the system, And (3) with As a matrix related to the coriolis centripetal force of the system, Is a matrix related to gravity and elasticity.
5. The dead zone adaptive compensation control system of a line-driven continuum robot of claim 4, wherein the introduced RBF neural network algorithm is designed to: after fitting approximation, the obtained fitting function is: Wherein, the Representing the inverse of the dead zone, For the output of the controller, Is the modeling error of the neural network, And Is an ideal weight value, and the weight value is the ideal weight value, And Is the output of the radial basis function.
6. A dead zone adaptive compensation control method of a line-driven continuum robot, characterized by being applied to the dead zone adaptive compensation control system of a line-driven continuum robot according to claim 1, the method comprising: Based on Euler's Bernoulli Liang Dingli and the constant curvature assumption, a standard dynamics model of the continuum robot is established; establishing a dead zone function for a driving dead zone of the linear driving continuum robot; establishing an uncertainty error model based on the standard dynamics model; Introducing two RBF neural networks, wherein one RBF neural network is used for estimating the proposed uncertain error model, and the other RBF neural network is used for dead zone compensation of a feed-forward channel of the system so as to approximate the established dead zone function; And updating the established model and dead zone function in a self-adaptive mode to improve the control precision of the continuum robot, wherein the model comprises the standard dynamics model and an uncertain error model.
7. The dead zone adaptive compensation control method of a line-driven continuum robot of claim 6, wherein said adaptively comprises: designing the self-adaption rate of the neural network; establishing an error function according to the acquired error signal; and transmitting the error function into the self-adaptive rate of the neural network to carry out compensation updating so as to update the established model and dead zone function.

Description

Dead zone self-adaptive compensation control system and method for line-driven continuum robot Technical Field The invention relates to the technical field of continuum robots, in particular to a dead zone self-adaptive compensation control system and method of a line-driven continuum robot. Background The continuum robot is a novel robot inspired by a biological bionic structure and is made of elastic materials, can generate continuous deformation and deform greatly, and has the advantages of strong structural flexibility, high flexibility and the like. The continuum robot can make up for the defect of insufficient capability of the traditional rigid robot to cope with complex environments, so the continuum robot is widely applied to the fields of industry, disaster search and rescue, equipment overhaul, aerospace, medical treatment and the like, and becomes a hot spot for the development of a new generation of robots. Compared with the traditional rigid robot, the continuous robot has high flexibility and softness, but the structural complexity of the continuous robot is greatly increased. The continuum robot model involves complex factors, and the dynamics model has the characteristics of strong nonlinearity, high coupling and the like, and inevitably causes modeling errors. In addition, the continuum robot can also receive a large amount of external interference in actual work, and these interference and error seriously influence the control accuracy of continuum robot under dynamic model, can even influence global system's stability. Therefore, the conventional robot control method has difficulty in handling the control problem of the continuum robot. The continuum robot using the line driving has a driving dead zone phenomenon because of its driving manner. Compensation for the "dead zone" becomes very difficult due to the non-minimally nonlinear nature of the "dead zone" and thus increases the complexity of the driveline control. However, if the influence of the dead zone cannot be eliminated, besides the output error, the high-precision control system of the continuum robot can generate limit cycle oscillation, and the performance is reduced or even becomes unstable. Disclosure of Invention Aiming at the technical defects in the prior art, the embodiment of the invention aims to provide a dead zone self-adaptive compensation control system and method for a line-driven continuum robot, so as to overcome the defects that the control system is affected by driving dead zone and modeling inaccuracy in the prior art. To achieve the above object, in a first aspect, an embodiment of the present invention provides a dead zone adaptive compensation control system for a line-driven continuum robot, applied to a controller, including: The first model module is used for establishing a standard dynamics model of the continuum robot based on the European Bernoulli Liang Dingli and the constant curvature assumption; The dead zone function module is used for establishing a dead zone function for driving the linear driving continuum robot; a second model module for building an uncertainty error model based on the standard kinetic model; The system comprises a neural network compensator, a system feedforward channel and a system feedforward channel, wherein the neural network compensator comprises two introduced RBF neural networks, one RBF neural network is used for estimating the proposed uncertainty error model, and the other RBF neural network is used for dead zone compensation of the system feedforward channel to approximate the established dead zone function; The self-adaptive compensation module is used for updating the established model and dead zone function in a self-adaptive mode so as to improve the control precision of the continuum robot, wherein the model comprises the standard dynamics model and an uncertain error model. As a preferred embodiment of the present application, the adaptive manner specifically includes: designing the self-adaption rate of the neural network; establishing an error function according to the acquired error signal; and transmitting the error function into the self-adaptive rate of the neural network to carry out compensation updating so as to update the established model and dead zone function. In a second aspect, an embodiment of the present invention further provides a dead zone adaptive compensation control method of a line-driven continuum robot, which is applied to the dead zone adaptive compensation control system of the line-driven continuum robot in the first aspect, where the method includes: Based on Euler's Bernoulli Liang Dingli and the constant curvature assumption, a standard dynamics model of the continuum robot is established; establishing a dead zone function for a driving dead zone of the linear driving continuum robot; establishing an uncertainty error model based on the standard dynamics model; Introducing two RBF neural networks, wherein one RBF neural net