CN-121468496-B - Shape sensing method and system of rope-driven continuum robot

CN121468496BCN 121468496 BCN121468496 BCN 121468496BCN-121468496-B

Abstract

The embodiment of the specification discloses a shape sensing method and a system of a rope-driven continuum robot, wherein at least two flexible sensors fixed in a spiral manner are arranged on a flexible framework, and the method comprises the steps of establishing a general shape sensing model based on a constant curvature assumption; the method comprises the steps of driving the joint module to a test shape, solving the actual fixed position of the flexible sensor by adopting a numerical iteration method based on a general shape sensing model, driving the joint module to a working shape, solving the output value of the flexible skeleton shape parameter by adopting a numerical iteration method based on the general shape sensing model and the actual fixed position of the flexible sensor, and determining the shape of the robot based on the output value of the flexible skeleton shape parameter corresponding to each joint module. According to the invention, the fixing precision is not required to be considered when the flexible sensor is fixed, the shape change of the robot is accurately perceived through establishing and calibrating a model, and the problems of high requirement on the installation precision of the sensor and poor perception reliability of the sensor in the existing method are solved.

Inventors

SHEN WENJUN
Yao Yanyin
LIU RUIFAN

Assignees

宁波数字孪生(东方理工)研究院

Dates

Publication Date: 20260508
Application Date: 20260109

Claims (10)

1. The shape sensing method of the rope-driven continuum robot comprises a plurality of joint modules connected in series, and is characterized in that each joint module comprises a cylindrical flexible framework and at least two flexible sensors, each flexible sensor is fixed on the side surface of the flexible framework in a spiral mode, and the shape sensing method comprises the following steps: Based on a constant curvature assumption, a general shape perception model describing a mapping relation between a flexible skeleton shape parameter and a flexible sensor length variation of any fixed position on the flexible skeleton is established; for any joint module: Driving the joint module to a test shape corresponding to the determined flexible skeleton shape parameter, and solving the corresponding actual fixed position of each flexible sensor in the joint module by adopting a numerical iteration method based on a general shape perception model; Driving the joint module to a working shape corresponding to the flexible skeleton shape parameter to be determined, and solving an output value of the flexible skeleton shape parameter in the joint module by adopting a numerical iteration method based on a general shape perception model and an actual fixed position corresponding to each flexible sensor in the joint module; and determining the shape of the robot based on the output value of the flexible skeleton shape parameter corresponding to each joint module.
2. The method for sensing the shape of the rope-driven continuum robot according to claim 1, wherein the step of establishing a general shape sensing model describing a mapping relationship between a flexible skeleton shape parameter and a flexible sensor length variation amount at any fixed position on the flexible skeleton based on a constant curvature assumption comprises: defining a positioning parameter set describing the fixed position of the flexible sensor relative to the flexible skeleton in the joint module; Based on the constant curvature assumption, a general shape sensing model is established by the positioning parameter set of the flexible sensor and the structural parameters of the flexible skeleton.
3. The method of claim 2, wherein the defining the set of positioning parameters describing the fixed position of the flexible sensor relative to the flexible skeleton in the joint module comprises: establishing a coordinate system by taking the center of a bottom surface of the flexible framework in a vertical state as an origin, wherein the Z axis of the coordinate system coincides with the neutral axis of the flexible framework, and the X axis direction of the coordinate system is taken as a reference direction; Cutting off the side surface of the flexible framework along the direction parallel to the Z axis of the coordinate system by taking the intersection point of the X axis of the coordinate system and the side surface of the flexible framework as a starting point, and flattening the side surface of the flexible framework into a two-dimensional plane; the intercept between the length direction of the flexible sensor and the longitudinal axis of the two-dimensional plane is defined as a first positioning parameter, the included angle between the length direction of the flexible sensor and the transverse axis of the two-dimensional plane is defined as a second positioning parameter, the included angle between the projection point of any end point of the flexible sensor on the bottom surface of the flexible framework and the X axis of the coordinate system is defined as a third positioning parameter, and the central angle clamped between the projection points of the two end points of the flexible sensor, which correspond to the bottom surface of the flexible framework, is defined as a fourth positioning parameter.
4. The method for sensing the shape of the rope-driven continuum robot according to claim 2, wherein the solving the respective corresponding actual fixed position of each flexible sensor in the joint module by using a numerical iteration method based on the general shape sensing model comprises: Acquiring a first jacobian matrix of the general shape sensing model, wherein the first jacobian matrix characterizes local sensitivity of length variation of all flexible sensors in any joint module to a positioning parameter set; Based on the general shape sensing model and the first jacobian matrix, a numerical iteration method is adopted to take the length variation deviation of each flexible sensor in the joint module as a convergence target, and the calibration value of the positioning parameter set corresponding to each flexible sensor in the joint module is solved, wherein the length variation deviation is the difference value between the actual length variation obtained by the output value of the flexible sensor and the theoretical length variation obtained by the general shape sensing model.
5. The method for sensing the shape of the rope-driven continuum robot according to claim 4, wherein the calculating the calibration value of the positioning parameter set corresponding to each flexible sensor in the joint module by using the deviation of the length variation of each flexible sensor in the joint module as a convergence target based on the general shape sensing model and the first jacobian matrix by using a numerical iteration method comprises: taking the determined flexible skeleton shape parameters corresponding to the test shape and initial given values of the positioning parameter sets corresponding to each flexible sensor in the joint module as the input of a general shape sensing model to obtain theoretical length variation corresponding to each flexible sensor in the joint module; Acquiring the respective actual length variation corresponding to each flexible sensor in the joint module based on the respective output value of each flexible sensor in the joint module; judging the convergence condition of a numerical iteration method based on the difference value of the theoretical length variation and the actual length variation corresponding to each flexible sensor in the joint module; If the joint module is not converged, updating the given value of the positioning parameter set corresponding to each flexible sensor in the joint module in the next iteration based on the first jacobian matrix and the difference value of the theoretical length variation and the actual length variation corresponding to each flexible sensor in the joint module, and inputting the general shape perception model again to carry out the next iteration; If the current set value of the positioning parameter set corresponding to each flexible sensor in the joint module is converged, the current set value is used as a calibration value.
6. The method for sensing the shape of the rope-driven continuum robot according to claim 4, wherein driving the joint module to a working shape corresponding to the flexible skeleton shape parameter to be determined, solving the output value of the flexible skeleton shape parameter in the joint module by using a numerical iteration method based on the general shape sensing model and the respective corresponding actual fixed position of each flexible sensor in the joint module, comprises: Acquiring a second jacobian matrix of the general shape perception model, wherein the second jacobian matrix characterizes local sensitivity of the length variation of all flexible sensors in any joint module to flexible skeleton shape parameters; based on the general shape perception model, the second jacobian matrix and the calibration values of the positioning parameter sets corresponding to each flexible sensor in the joint module, a numerical iteration method is adopted to take the length variation deviation of each flexible sensor in the joint module as a convergence target, and the output values of the flexible skeleton shape parameters in the joint module are solved.
7. The method for sensing the shape of the rope-driven continuum robot according to claim 6, wherein the calculating the output value of the flexible skeleton shape parameter in the joint module based on the general shape sensing model, the second jacobian matrix and the calibration value of the positioning parameter set corresponding to each flexible sensor in the joint module by using the numerical iteration method and taking the deviation of the length variation of each flexible sensor in the joint module as a convergence target comprises: taking the calibration value of the positioning parameter set and the initial given value of the flexible skeleton shape parameter corresponding to each flexible sensor in the joint module as the input of a general shape sensing model to obtain the theoretical length variation corresponding to each flexible sensor in the joint module; Acquiring the respective actual length variation corresponding to each flexible sensor in the joint module based on the respective output value of each flexible sensor in the joint module; judging the convergence condition of a numerical iteration method based on the difference value of the theoretical length variation and the actual length variation corresponding to each flexible sensor in the joint module; if the flexible skeleton shape parameter is not converged, updating a given value of the flexible skeleton shape parameter in the joint module in the next iteration based on the second jacobian matrix and the difference value of the theoretical length variation and the actual length variation corresponding to each flexible sensor in the joint module, and inputting the given value into the universal shape perception model again for the next iteration; if the joint model is converged, the current given value of the flexible skeleton shape parameter in the joint model is used as an output value.
8. The method for sensing the shape of a rope-driven continuum robot of claim 1, further comprising: obtaining the shape parameter change range of the flexible skeleton and the structure parameter data of the flexible skeleton, which correspond to each joint module; for any joint module: Based on the structural parameter data of the flexible skeleton in the joint module, a particle swarm algorithm is adopted to maximize the shape parameter change range of the flexible skeleton as an optimization target, and a fixed position recommended value corresponding to each flexible sensor in the joint module is obtained; And fixing each flexible sensor in the joint module based on the recommended fixed position value corresponding to each flexible sensor in the joint module.
9. A method of shape sensing for a rope driven continuum robot according to any of claims 1-8, wherein said flexible sensor is a stretchable capacitive sensor.
10. A shape sensing system of a rope-driven continuum robot, characterized in that a shape sensing method of a rope-driven continuum robot according to any one of claims 1-9 is applied, the shape sensing system comprising a position calibration unit and a shape sensing unit; The position calibration unit is used for driving any joint module to a test shape corresponding to the determined flexible skeleton shape parameter, and solving the corresponding actual fixed position of each flexible sensor in the joint module by adopting a numerical iteration method based on the constructed general shape perception model; The shape sensing unit is used for driving any joint module to a working shape corresponding to the flexible skeleton shape parameters to be determined, solving the output value of the flexible skeleton shape parameters in the joint module by adopting a numerical iteration method based on the general shape sensing model and the actual fixed position corresponding to each flexible sensor in the joint module, and determining the shape of the robot based on the output value of the flexible skeleton shape parameters corresponding to each joint module.

Description

Shape sensing method and system of rope-driven continuum robot Technical Field Various embodiments of the present description relate to the field of rope-driven continuum robots, and in particular to optimization of shape-aware methods of rope-driven continuum robots. Background Rope-driven continuum robots are a class of flexible robots employing lightweight rope drives that achieve continuous deformability through a modular design, typically consisting of multiple joint modules in series. Each joint module comprises a base platform, a movable platform, a flexible framework 1 and a plurality of driving ropes, wherein the flexible framework 1 is used as a passive supporting structure, has the characteristics of small bending rigidity, large stretching rigidity and large torsional rigidity, and enables the robot to adapt to complex environments and realize high-flexibility movement. Currently, the shape sensing method of the rope-driven continuum robot is mainly implemented depending on a plurality of flexible sensors 2 disposed on the surface of the flexible skeleton 1. As shown in fig. 1 and 2, for a joint module having two degrees of freedom bending motion, a flexible sensor 2 is generally disposed in a bendable direction of a surface of a flexible skeleton 1, and a longitudinal direction of the sensor is kept parallel to an axis of the flexible skeleton 1. The arrangement mode simplifies the flexible sensor 2 into a straight line along the axial direction of the framework or an arc when the flexible sensor is bent, and the shape parameter is calculated by measuring the deformation of the flexible sensor 2, so that the indirect sensing of the shape of the robot is realized. However, the scheme of arranging the flexible sensor 2 in the parallel direction has extremely high precision requirement on the fixed position of the sensor, the actual fixed position of the sensor is required to be completely consistent with the ideal fixed position, once the actual installation position of the sensor is deviated, particularly when the fixed angle is deviated, the sensor cannot be simplified into a straight line, the precision of the shape sensing model can be greatly reduced, and the reliability of the rope-driven continuum robot in a fine operation scene is limited. Disclosure of Invention The embodiment of the specification provides a shape sensing method and a system of a rope-driven continuum robot, the scheme does not need to consider the fixing precision when fixing a flexible sensor, and the problem that the existing shape sensing method has high requirement on the sensor installation precision and poor sensing reliability is solved by arranging at least two flexible sensors which are arranged in a spiral way in each joint module and establishing and calibrating a general shape sensing model of the flexible sensor which is arranged in the spiral way so as to accurately sense the shape change of the robot. The technical scheme is as follows: In a first aspect, embodiments of the present disclosure provide a shape sensing method of a rope-driven continuum robot, the robot including a plurality of joint modules connected in series, each joint module including a cylindrical flexible skeleton and at least two flexible sensors, each flexible sensor being fixed to a side of the flexible skeleton in a spiral manner; The shape sensing method comprises the following steps: Based on a constant curvature assumption, a general shape perception model describing a mapping relation between a flexible skeleton shape parameter and a flexible sensor length variation of any fixed position on the flexible skeleton is established; for any joint module: Driving the joint module to a test shape corresponding to the determined flexible skeleton shape parameter, and solving the corresponding actual fixed position of each flexible sensor in the joint module by adopting a numerical iteration method based on a general shape perception model; Driving the joint module to a working shape corresponding to the flexible skeleton shape parameter to be determined, and solving an output value of the flexible skeleton shape parameter in the joint module by adopting a numerical iteration method based on a general shape perception model and an actual fixed position corresponding to each flexible sensor in the joint module; and determining the shape of the robot based on the output value of the flexible skeleton shape parameter corresponding to each joint module. As a preferred solution, the establishing a general shape sensing model describing a mapping relationship between a flexible skeleton shape parameter and a flexible sensor length variable of any fixed position on the flexible skeleton based on a constant curvature assumption includes: defining a positioning parameter set describing the fixed position of the flexible sensor relative to the flexible skeleton in the joint module; Based on the constant curvature assumption, a general shape sensi