A rolling motion method for tensegrity robots based on closed-loop control

Abstract

The present invention discloses a rolling motion method, system and equipment for a tensegrity robot based on closed-loop control, belonging to the field of robot motion technology. The method includes selecting a rolling axis and a driving point based on the minimum direction principle according to the target point position of the tensegrity robot's rolling; planning the motion trajectory of the tensegrity robot's rolling angle and rolling radius according to the rolling axis and driving point, and calculating the expected trajectory of the center of mass motion based on the planned motion trajectory of the tensegrity robot's rolling angle and rolling radius; using the tensegrity robot dynamics model that considers contact and collision conditions, tracking the expected trajectory of the tensegrity robot's center of mass motion, and obtaining the expected instructions for the center of mass motion. The present invention solves the problems of large structural deformation of the tensegrity robot, limited motion performance and low rolling efficiency during rolling, and realizes precise control of the tensegrity robot's rolling speed.

Description

Technical Field

[0001] The present invention relates to the field of robot motion technology, and in particular to a rolling motion method, system and equipment for a tensegrity robot based on closed-loop control. Background Art

[0002] Robotics, with its unique advantages, has become an indispensable aid for humanity when exploring uncharted territory, responding to emergencies, and carrying out high-risk missions. Tensegrity robots, as an emerging robotic form, demonstrate immense potential for application in a variety of fields, including planetary exploration, post-disaster emergency rescue, and exploration of specialized environments, thanks to their unique structural characteristics and superior performance. The design concept of tensegrity robots stems from an ancient structural philosophy: the ingenious combination of tensile and compressive elements creates a stable yet flexible monolithic structure. This structure not only boasts an extremely high strength-to-weight ratio, maintaining stability and integrity in extreme environments, but also adapts to complex and changing terrain and environments through its unique deformability.

[0003] However, it is precisely this structural complexity that presents unprecedented challenges in controlling the rolling motion of tensegrity robots. Compared to traditional wheeled, tracked, or legged robots, tensegrity robots employ a more unique and complex method of locomotion. They primarily roll about their ground-contact rolling axis. This rolling method not only requires the robot to flexibly adjust its shape to accommodate varying rolling requirements, but also requires precise control of the coordinated interaction of its components to ensure smooth and efficient rolling.

[0004] Currently, the research community has proposed two main strategies for controlling the rolling motion of tensegrity robots: open-loop control methods and intelligent control methods. Open-loop control methods are usually based on a preset algorithm or model. By changing the shape of the tensegrity robot to move its center of mass to the outside of the rolling axis, single-step rolling motion is achieved. This method has the advantages of simple implementation and low computational complexity, but its rolling efficiency and stability are often limited due to the lack of real-time feedback on the external environment and in-depth consideration of structural dynamics. In practical applications, open-loop control methods have difficulty coping with complex and changing rolling environments, such as uneven terrain and varying friction coefficients, which can lead to instability or even failure of the rolling process.

[0005] In contrast, intelligent control methods attempt to improve the adaptability and robustness of rolling motion by introducing advanced artificial intelligence technologies such as machine learning and neural networks. These methods can perceive environmental changes in real time and dynamically adjust control strategies based on this perception, thereby overcoming the limitations of open-loop control methods to a certain extent. However, intelligent control methods also face many difficulties in implementation. First, the dynamic model of the tensegrity robot is complex and highly nonlinear, making it difficult to establish an accurate mathematical model. Second, the training and optimization of intelligent algorithms require a large amount of experimental data and computing resources, which poses a challenge to the rapid response and real-time control in practical applications. Therefore, while intelligent control methods can improve the adaptability and robustness of rolling motion, they ignore the unique motion analysis issues of tensegrity robots and do not fundamentally solve the deformation problem of the tensegrity structure during rolling, which still limits the improvement of its rolling efficiency and motion performance. Summary of the Invention

[0006] In view of the problem that the control methods existing in the prior art ignore the unique motion analysis problems of tensegrity robots, which make full use of the motion deformation capacity of the tensegrity structure, resulting in large structural deformation of the tensegrity robot during rolling, limited motion performance, low rolling efficiency, and inability to accurately control the rolling speed of the robot. The present invention provides a tensegrity robot rolling motion method based on closed-loop control. Based on the tensegrity robot dynamics model under contact and collision conditions, it fully considers the influence of robot dynamics and contact and collision dynamics, and realizes a tensegrity robot rolling motion with controllable motion state, strong robustness, and high efficiency and continuity, solving the problems of large structural deformation, limited motion performance, and low rolling efficiency of the tensegrity robot during rolling, and realizing accurate control of the rolling speed of the tensegrity robot.

[0007] In order to achieve the above objectives, the present invention provides the following technical solutions.

[0008] In a first aspect, the present invention provides a rolling motion method for a tensegrity robot based on closed-loop control, comprising:

[0009] According to the target point position of the tensegrity robot rolling, the rolling axis and driving point are selected based on the minimum direction principle;

[0010] planning a motion trajectory of a roll angle and a roll radius of the tensegrity robot according to the roll axis and the driving point, and calculating an expected trajectory of the center of mass motion based on the planned motion trajectory of the roll angle and the roll radius of the tensegrity robot;

[0011] Using a tensegrity robot dynamics model that considers contact and collision conditions, the desired trajectory of the tensegrity robot's center of mass motion is tracked to obtain a desired center of mass motion instruction. Based on the desired center of mass motion instruction, the tensegrity robot is controlled to achieve a desired rolling motion with a desired rolling radius and rolling angle.

[0012] When the tensegrity robot in the desired rolling motion enters the falling phase, a closed-loop control law for the rolling falling phase of the tensegrity robot is designed based on the partial feedback linearization control method to control the tensegrity robot to land.

[0013] It is determined whether the tensegrity robot reaches the set target point when landing. If it reaches the target point, it is determined that the tensegrity robot has successfully completed the continuous rolling task.

[0014] As a further improvement of the present invention, the method of selecting the rolling axis and the driving point based on the minimum direction principle according to the target point position of the tensegrity robot rolling includes:

[0015] According to the target point position of the tensegrity robot's rolling, the angle between the tensegrity robot's rolling direction and the desired direction is calculated based on the minimum direction principle. The base of the touchdown triangle around which the robot completes a single-step rolling process, corresponding to the minimum angle, is selected as the rolling axis of the tensegrity robot in the single-step rolling process.

[0016] The nodes on the opposite side of the rolling axis of the tensegrity robot in single-step rolling are selected as driving points.

[0017] As a further improvement of the present invention, planning the motion trajectory of the roll angle and the roll radius of the tensegrity robot according to the roll axis and the drive point, and calculating the expected trajectory of the center of mass motion based on the planned motion trajectory of the roll angle and the roll radius of the tensegrity robot, comprises:

[0018] Plan the rolling angular motion trajectory of the tensegrity robot during the initial rolling phase based on the rolling axis and driving point and rolling radius trajectory ;

[0019] The rolling angle motion trajectory of the tensegrity robot in the initial rolling stage and rolling radius trajectory , converted into the expected motion trajectory of the robot's center of mass ;

[0020] in, is the rolling angular displacement of the robot's center of mass at time k; is the rolling radius of the robot at time k; for; for.

[0021] As a further improvement of the present invention, a tensegrity robot dynamics model considering contact and collision conditions is used to track a desired trajectory of the center of mass motion of the tensegrity robot to obtain a desired center of mass motion instruction; and the tensegrity robot is controlled to achieve a desired rolling motion with a desired rolling radius and rolling angle according to the desired center of mass motion instruction, comprising:

[0022] Using the tensegrity robot dynamics model considering contact and collision conditions, and based on the neural network adaptive sliding mode closed-loop control method, the center of mass trajectory tracking control law of the tensegrity robot in the initial rolling stage is designed to track the desired center of mass motion trajectory of the tensegrity robot and obtain the desired center of mass motion command;

[0023] The tensegrity robot is controlled to achieve a desired rolling motion with a desired rolling radius and rolling angle according to the desired center of mass motion instruction.

[0024] As a further improvement of the present invention, the method for designing the center of mass trajectory tracking control law of the tensegrity robot in the initial rolling stage based on the neural network adaptive sliding mode closed-loop control method includes:

[0025]

[0026] Where, is the control input; The prestressed cable member balances the tension and satisfies the prestressed balance condition. ; represents the expected trajectory of the center of mass motion; , is the control law parameter matrix; for estimated value.

[0027] As a further improvement of the present invention, when the tensegrity robot in the desired rolling motion rolls into the falling phase, a closed-loop control law for the rolling falling phase of the tensegrity robot is designed based on a partial feedback linearization control method to control the tensegrity robot to land, including:

[0028] When the tensegrity robot in the desired rolling motion rolls into the falling phase, a closed-loop control law for the rolling falling phase of the tensegrity robot is designed based on the partial feedback linearization control method, including the closed-loop control law for the landing shock absorption control of the tensegrity robot in the rolling falling phase and the closed-loop control law for the landing shock absorption control of the tensegrity robot in the rolling falling phase;

[0029] The closed-loop control law of the tensegrity robot's landing shock absorption control during the rolling falling stage is used to control the tensegrity robot's landing shock absorption control during the rolling falling stage.

[0030] The closed-loop control law of the landing shock absorption control of the tensegrity robot during the rolling drop phase is used to control the configuration recovery and stability control of the tensegrity robot during the rolling drop phase.

[0031] As a further improvement of the present invention, the closed-loop control law for the landing shock absorption control of the tensegrity robot during the rolling descent phase includes:

[0032]

[0033] in, represents the expected vertical displacement of the robot's landing contact point, is a positive definite matrix representing the desired damping matrix, is a positive definite matrix representing the expected stiffness matrix;

[0034] The closed-loop control law for the landing shock absorption control of the tensegrity robot during the rolling descent phase includes:

[0035]

[0036] in, represents the expected length of the cable member, is a positive definite matrix representing the desired damping matrix, is a positive definite matrix representing the desired stiffness matrix.

[0037] As a further improvement of the present invention, the step of determining whether the tensegrity robot reaches a set target point upon landing includes:

[0038] Determine whether there are three nodes of the tensegrity robot in contact with the ground at the same time when it lands;

[0039] If there are no three nodes in contact with the ground at the same time, continue execution;

[0040] If there are three nodes that are in contact with the ground at the same time, determine whether the target point is within the triangle formed by the three nodes on the bottom surface;

[0041] If the target point is within the triangle, the tensegrity robot is judged to have successfully completed a rolling motion;

[0042] If the target point position is not within the triangle, the rolling axis and driving point are reselected based on the minimum direction principle according to the target point position of the tensegrity robot rolling, and the rolling motion is continued.

[0043] In a second aspect, the present invention provides a tensegrity robot rolling motion system based on closed-loop control, comprising:

[0044] Roll axis and driving point selection module: used to select the rolling axis and driving point based on the minimum direction principle according to the target point position of the tensegrity robot;

[0045] Obtaining the expected trajectory of the center of mass motion: a module for planning the motion trajectory of the roll angle and roll radius of the tensegrity robot according to the roll axis and the driving point, and calculating the expected trajectory of the center of mass motion based on the planned motion trajectory of the roll angle and roll radius of the tensegrity robot;

[0046] The desired rolling motion module is used to track the desired trajectory of the tensegrity robot's center of mass motion using a tensegrity robot dynamics model that considers contact and collision conditions, and obtain the desired center of mass motion instructions. Based on the desired center of mass motion instructions, the tensegrity robot is controlled to achieve the desired rolling motion with the desired rolling radius and rolling angle.

[0047] Robot landing module: When the tensegrity robot in the desired rolling motion enters the falling phase, a closed-loop control law for the rolling falling phase of the tensegrity robot is designed based on the partial feedback linearization control method to control the landing of the tensegrity robot;

[0048] Rolling completion judgment module: It is used to judge whether the tensegrity robot reaches the set target point when landing. If it reaches the target point, it is determined that the tensegrity robot has successfully completed the continuous rolling task.

[0049] In a third aspect, the present invention provides an electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the steps of the rolling motion method of a tensegrity robot based on closed-loop control are implemented.

[0050] In a fourth aspect, the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the optical fiber cable package defect detection and type identification method based on deep learning are implemented.

[0051] In a fifth aspect, the present invention provides a computer program product comprising computer instructions, which, when executed by a processor, implement the steps of the optical fiber cable package defect detection and type identification method based on deep learning.

[0052] Compared with the prior art, the present invention has the following beneficial effects:

[0053] The core of this invention lies in the introduction of a closed-loop control system, which addresses the unique kinematic analysis challenges of tensegrity robots and enables refined management and optimization of their rolling process. Conventional control methods often overlook the unique kinematic deformation capabilities of tensegrity structures, resulting in significant structural deformation during the robot's rolling process. This not only limits its kinematic performance, but also reduces rolling efficiency and makes precise control of rolling speed difficult. The present invention, however, cleverly leverages the characteristics of the tensegrity structure. By selecting the rolling axis and driving point based on the minimum direction principle, it ensures the optimal selection of the rolling path, effectively reducing unnecessary structural deformation and improving kinematic efficiency. Furthermore, the present invention plans the motion trajectory of the tensegrity robot's rolling angle and rolling radius and, based on this plan, calculates the desired trajectory of the center of mass motion. This step not only provides a theoretical basis for subsequent precise control but also ensures that the robot maintains a stable posture during rolling, avoiding kinematic instability caused by structural deformation. Most importantly, the present invention introduces a dynamic model for the tensegrity robot that considers contact and collision conditions. This model fully accounts for contact and collision with the ground or other obstacles during rolling, enabling more accurate simulation and calculation of dynamic parameters. Using this model, the present invention can track and adjust the robot's center of mass trajectory in real time, ensuring it consistently follows the desired trajectory during rolling. This tracking control method, based on a dynamic model, not only improves control accuracy but also significantly enhances the robot's robustness, enabling it to maintain a stable rolling state in complex and changing environments. During the descent phase of the rolling process, the present invention employs a partial feedback linearization control method to design a closed-loop control law. This method precisely controls key parameters such as landing speed and angle by monitoring and analyzing the robot's dynamic characteristics during landing in real time. This not only ensures a smooth landing but also further improves the continuity and stability of its rolling motion. Furthermore, by determining whether the robot has reached the set target point upon landing, the present invention can accurately assess the success of the rolling task, providing robust data support for subsequent rolling path planning and control strategy adjustments. Therefore, by introducing a closed-loop control system, optimizing rolling path selection, introducing a dynamic model that considers contact and collision conditions, and employing a partial feedback linearization control method, the present invention achieves comprehensive optimization and precise control of the rolling motion of a tensegrity robot. This method not only significantly reduces the structural deformation of the robot during rolling, improves its motion performance and rolling efficiency, but also achieves precise control of the rolling speed. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] The drawings described herein are for illustration purposes only and are not intended to limit the scope of the present disclosure in any way. In the drawings:

[0055] Figure 1 Schematic diagram of a flow chart of a rolling motion method of a tensegrity robot based on closed-loop control according to the present invention;

[0056] Figure 2 This is a specific flow chart of the method for controlling continuous rolling motion of a tensegrity robot according to the present invention;

[0057] Figure 3 Schematic diagram of the rolling axis selection of the tensegrity robot of the present invention;

[0058] Figure 4 Schematic diagram of the definition of the rolling angle and rolling radius of the tensegrity robot of the present invention;

[0059] Figure 5 Schematic diagram of the forward displacement tracking curve of the center of mass of the tensegrity robot in the initial rolling stage in Example 1 of the present invention;

[0060] Figure 6 Schematic diagram of the lateral displacement tracking curve of the center of mass of the tensegrity in the initial rolling stage in Example 1 of the present invention;

[0061] Figure 7 Schematic diagram of the vertical displacement tracking curve of the center of mass of the tensegrity robot in the initial rolling stage in Example 1 of the present invention;

[0062] Figure 8 3D simulation result diagram of the initial moment of rolling motion of the tensegrity robot in Example 1 of the present invention;

[0063] Figure 9 This is a three-dimensional simulation result diagram of the tensegrity robot at the end of the initial rolling phase in Example 1 of the present invention;

[0064] Figure 10 This is a three-dimensional simulation result diagram of the tensegrity robot at the highest rolling point in Example 1 of the present invention;

[0065] Figure 11 This is a three-dimensional simulation result diagram of the tensegrity robot in Example 1 of the present invention at the moment of stable landing;

[0066] Figure 12 This is a simulation diagram of the continuous rolling center of mass motion trajectory of the tensegrity robot in Example 2 of the present invention;

[0067] Figure 13 This is a schematic structural diagram of a tensegrity robot rolling motion system based on closed-loop control according to the present invention;

[0068] Figure 14 Schematic diagram of an electronic device in an embodiment of the present invention. DETAILED DESCRIPTION

[0069] In order to enable those skilled in the art to better understand the technical solutions of the present invention, the technical solutions of the present invention will be clearly and completely described below in conjunction with the accompanying drawings of the present invention. The embodiments described 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 making creative efforts should fall within the scope of protection of the present invention.

[0070] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one skilled in the art to which this invention pertains. The terms used in this specification of the present invention are for the purpose of describing specific embodiments only and are not intended to limit the present invention. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0071] The control methods in the prior art ignore the unique motion analysis problems of tensegrity robots, which make full use of the motion deformation ability of the tensegrity structure, resulting in large structural deformation of the tensegrity robot during rolling, limited motion performance, low rolling efficiency, and inability to accurately control the rolling speed of the robot. The present invention provides a rolling motion method for a tensegrity robot based on closed-loop control, such as Figure 1 As shown, the method includes:

[0072] S100: According to the target point position of the tensegrity robot rolling, the rolling axis and the driving point are selected based on the minimum direction principle;

[0073] S200: planning a motion trajectory of a roll angle and a roll radius of the tensegrity robot according to the roll axis and the driving point, and calculating an expected trajectory of the center of mass motion based on the planned motion trajectory of the roll angle and the roll radius of the tensegrity robot;

[0074] S300: using a dynamic model of the tensegrity robot that considers contact and collision conditions, tracking a desired trajectory of the center of mass motion of the tensegrity robot, and obtaining a desired center of mass motion instruction; controlling the tensegrity robot to achieve a desired rolling motion with a desired rolling radius and rolling angle according to the desired center of mass motion instruction;

[0075] S400: When the tensegrity robot in the desired rolling motion rolls into a falling phase, a closed-loop control law for the rolling falling phase of the tensegrity robot is designed based on a partial feedback linearization control method to control the tensegrity robot to land.

[0076] S500: Determine whether the tensegrity robot reaches a set target point when landing. If so, it is determined that the tensegrity robot has successfully completed the continuous rolling task.

[0077] This method is based on the dynamic model of the tensegrity robot under contact and collision conditions, and fully considers the influence of robot dynamics and contact and collision dynamics. It realizes the rolling motion of the tensegrity robot with controllable motion state, strong robustness, and efficient continuity. It solves the problems of large structural deformation, limited motion performance, and low rolling efficiency of the tensegrity robot during the rolling process, and realizes precise control of the rolling speed of the tensegrity robot.

[0078] The present invention will be further explained below with reference to the accompanying drawings.

[0079] The present invention provides a method for rolling motion of a tensegrity robot based on closed-loop control. This method is based on the dynamic model of the tensegrity robot under contact and collision conditions, fully considering the influence of robot dynamics and contact and collision dynamics, and realizes the rolling motion of the tensegrity robot with controllable motion state, strong robustness, and high efficiency and continuity. The method flow chart is shown in the figure below. Figure 2 As shown, the specific steps are as follows:

[0080] S1: Based on the minimum direction principle and the position of the tensegrity robot's rolling target point, the robot's rolling axis and driving point in single-step rolling are selected, including:

[0081] The rolling axis is defined as the base of the ground contact triangle around which the robot completes a single-step rolling process. Figure 3 The schematic diagram of the tensegrity robot touching the ground is shown in the figure. represents the tensegrity robot's ground contact triangle, The center of mass of the robot is The projection in The scroll target point. 、 、 Perpendicular to of 、 、 The three sides are the rolling directions of the robot, with each side as the rolling axis. 、 、 The scroll direction and the desired direction By selecting the bottom side corresponding to the minimum angle as the rolling axis and the node on the opposite side of the rolling axis as the driving point, the robot can quickly approach the target point.

[0082] S2: Based on the quadratic programming method, the motion trajectory of the rolling angle and rolling radius of the tensegrity robot in the initial rolling stage is planned, and then the expected trajectory of the center of mass motion is calculated based on the expected trajectory of the rolling angle and rolling radius.

[0083] The motion trajectory of the rolling angle and rolling radius of the tensegrity robot in the initial rolling stage is planned based on the quadratic programming method, and then the expected trajectory of the center of mass motion is calculated based on the expected trajectory of the rolling angle and rolling radius.

[0084] Based on a closed-loop control approach, this application proposes a rolling motion strategy for a tensegrity robot. This strategy decomposes the rolling motion of a tensegrity robot into a rolling initialization phase (from the initial state to the moment its driving point leaves the ground) and a rolling drop phase (from the moment the driving point leaves the ground to the robot stabilizing on the ground). During the rolling initialization phase, the driving point provides the torque required to complete the robot's rolling about the rolling axis. During the rolling drop phase, the driving point leaves the ground, and the motion process can be simplified to a simple pendulum motion of the center of mass about the rolling axis.

[0085] like Figure 4 As shown in the figure The tensegrity robot touches the ground in a triangle, with the robot moving around Taking edge scrolling as an example, define The side is the rolling axis, Provide the driving point with the torque required for the robot to complete rolling, is the center of mass of the robot, for exist The projection in for A little satisfaction on the side , for and The angle between is defined as the rolling angle, for The length of the rolling radius is defined as the rolling radius. During the initial rolling phase, the robot's center of mass moves around the rolling axis, which can be described as the changes in the rolling angle and rolling radius. Trajectory planning allows the robot to enter the simple pendulum motion of the rolling drop phase at the end of the initial rolling phase (when the driving point leaves the ground) with the desired rolling angle, rolling angular velocity, and rolling radius, thus enabling the robot to complete a rolling motion around the rolling axis.

[0086] S2.1: Plan the rolling angular motion trajectory of the tensegrity robot during the initial rolling phase.

[0087] The roll angle motion of the tensegrity robot can be expressed by the following linear model:

[0088] (1)

[0089] Where, represents the roll angle motion vector, is the rolling angular displacement of the robot's center of mass at time k, , , , Indicates the simulation step size.

[0090] The expected initial stage time of robot rolling is specified as Based on formula (1), the rolling angle motion prediction model of the robot in the initial rolling stage can be obtained as follows:

[0091] (2)

[0092] Where: is the expected rolling angle trajectory of the robot in the initial rolling stage; is the solution of the quadratic programming; for; for.

[0093] The rolling angle motion trajectory is planned through the following quadratic programming problem:

[0094] (3)

[0095] Where, is the weight matrix, represents the expected rolling angular displacement at the end of the initial rolling phase, It represents the expected rolling angular velocity at the end of the initial rolling phase. Indicates the expected rolling radius at the end of the initial rolling phase. To ensure that the center of mass can cross the rolling axis to complete the rolling during the rolling drop phase, the expected angular velocity at the end The following relationship needs to be satisfied:

[0096] (4)

[0097] The solution of the quadratic programming Substituting into formula (2) we can get the expected rolling angle trajectory of the robot in the initial rolling stage: .

[0098] S2.2: Planning the rolling radius change trajectory of the tensegrity robot during the initial rolling phase

[0099] The robot rolling radius trajectory planning process is the same as the rolling angle planning method. The linear model and prediction model of the robot rolling radius can be expressed as:

[0100] (5)

[0101] (6)

[0102] Where, , is the robot rolling radius at time k, , , .

[0103] The rolling radius trajectory is planned through the following quadratic programming problem:

[0104] (7)

[0105] Where, is the weight matrix, is the forward acceleration of the center of mass, is the vertical acceleration of the center of mass, which can be expressed as:

[0106] (8)

[0107] The solution of the quadratic programming Substituting into formula (6) we can get the rolling radius trajectory of the robot in the initial rolling stage: .

[0108] S2.3: Convert the expected rolling angle trajectory and the expected rolling radius trajectory into the expected motion trajectory of the robot's center of mass.

[0109] Roll angle trajectory of the tensegrity robot and rolling radius trajectory , the center of mass motion trajectory in the initial stage of rolling can be expressed as:

[0110] (9)

[0111] in, is the forward motion trajectory in the initial stage of rolling; It is the vertical motion trajectory in the initial stage of rolling.

[0112] By taking the first-order derivative and the second-order derivative of formula (9), we can obtain the velocity trajectory and acceleration trajectory of the center of mass motion in the initial stage of the robot's rolling.

[0113] S3: Based on the neural network adaptive sliding mode closed-loop control method, a center of mass trajectory tracking control law is designed for the initial rolling stage of the tensegrity robot to achieve the desired rolling motion by tracking the center of mass motion trajectory of the robot.

[0114] A dynamic model of the tensegrity robot in the form of ordinary differential equations for control law design is established, and the center of mass trajectory tracking control law of the tensegrity robot in the initial rolling stage is designed based on the neural network adaptive sliding mode closed-loop control method. The desired center of mass motion instructions obtained in the tracking trajectory planning are used to control the robot to achieve the desired rolling motion with the desired rolling radius and rolling angle.

[0115] To design a closed-loop control law for a tensegrity robot, we first need to establish a dynamic model of the tensegrity robot that takes into account contact and collision conditions. This model is then converted into an ordinary differential equation for control law design. The closed-loop control law is then designed based on the ordinary differential equation. The specific steps are as follows:

[0116] S3.1: Establish a dynamic model of the tensegrity robot considering contact and collision conditions and transform it into an ordinary differential equation form for control law design. Specifically, the model includes:

[0117] The dynamic model of the tensegrity robot considering contact and collision conditions can be expressed as follows:

[0118] (10)

[0119] Where, represents the generalized coordinate vector of the tensegrity robot, indicating the position and orientation of the robot rod components; represents the inertia matrix; ; represents the tensegrity structure equilibrium matrix, The elements can be expressed as , Indicates the length of the cable member; represents the manipulation matrix, , , , represents the Young's modulus of the cable member, Represents the cross-sectional area of ​​the cable member. represents the control input, Indicates the original length of the cable member; represents the normal contact generalized force vector between the robot node and the environment surface, represents the tangential friction generalized force vector between the robot node and the environment surface, For the complete constraints of the generalized coordinates of the tensegrity robot rod components, is the corresponding Lagrange multiplier; , is the normal distance between the robot’s i-th node and the environment surface, is the embedding depth of the robot's i-th node along the normal direction of the environment surface; is the normal contact force of the robot's i-th node.

[0120] The contact and collision complementary relationship of the tensegrity robot described by the last two terms in formula (10) can be equivalently expressed as:

[0121] (11)

[0122] In the formula, the set Represents a set of nodes that have no contact with the environment surface. Represents the collection of nodes that are in contact with the environment surface.

[0123] Substituting formula (11) into formula (10), the dynamic model of the tensegrity robot can be expressed as:

[0124] (12)

[0125] Where, , represents the normal support force of the environment contact node, represents the tangential friction force of the environment contact node; , Represents the normal distance of the contact node relative to the environment surface, represents the tangential displacement of the contact node; , The Jacobian matrix representing the robot's constraint space.

[0126] Jacobian matrix of the constraint space The null space The projection operator can be expressed as , then its orthogonal complement space The projection operator can be expressed as .

[0127] Multiply the differential equation in formula (12) on the left by get:

[0128] (13)

[0129] Taking the first-order derivative of the equality constraint in formula (12) yields:

[0130] (14)

[0131] The generalized velocity components of the system can be expressed as:

[0132] (15)

[0133] Taking the derivative of formula (15) we can get

[0134] (16)

[0135] Where, .

[0136] By adding Equation (13) and Equation (16), we can obtain the tensegrity robot dynamic equation in the form of ordinary differential equations for control law design:

[0137] (17)

[0138] Where, , , .

[0139] S3.2: Design a center of mass trajectory tracking control law for the tensegrity robot during the initial rolling phase based on a neural network adaptive sliding mode closed-loop control method. Track the desired center of mass motion instructions obtained from trajectory planning to control the robot to achieve the desired rolling motion with the desired rolling radius and rolling angle.

[0140] The control goal of the tensegrity robot in the initial rolling phase is to control the robot's center of mass to track the desired trajectory. The robot's center of mass is selected as the task space:

[0141] (18)

[0142] Where, represents the center of mass position vector, Represents the Jacobian matrix of the robot task space.

[0143] Multiply formula (17) on the left Substituting into formula (18), the robot center of mass task space dynamics can be expressed as:

[0144] (19)

[0145] Where, , , Uncertainty It can be approximated by RBF neural network, and formula (19) can be expressed as:

[0146] (20)

[0147] Where, is the ideal weight matrix of the neural network, is the neural network approximation error vector, is the radial basis function vector, vector Each element in is selected as a Gaussian function:

[0148] (twenty one)

[0149] Where, is the input of the neural network, represents the center vector, Indicates the width of the Gaussian function.

[0150] In order to ensure the stability of the closed-loop system, the sliding surface is designed to represent the ideal closed-loop system dynamic characteristics as follows:

[0151] (twenty two)

[0152] Where, is the tracking error, is a positive definite matrix.

[0153] By deriving formula (22) and substituting it into formula (20), the sliding mode dynamics of the system can be expressed as:

[0154] (twenty three)

[0155] Design the robot neural network adaptive sliding mode closed-loop control law as:

[0156] (twenty four)

[0157] Where, Input for prestress balance control of tensegrity structures; is the robot motion torque instruction; Indicates the prestressed balanced tension of the cable member, satisfying the prestressed equilibrium condition ; represents the expected trajectory of the center of mass motion; , is the control law parameter matrix; for The estimated value of , its adaptive law is:

[0158] (25)

[0159] Where, are the control law parameters.

[0160] The boundary layer method is used in formula (24) to reduce the system chattering caused by sliding mode control, and the saturation function is used to replace the original discontinuous term:

[0161] (26)

[0162] Where, represents the boundary layer thickness.

[0163] Substituting formula (24) into formula (23), the sliding mode dynamics of the closed-loop system can be expressed as:

[0164] (27)

[0165] Where, .

[0166] Furthermore, the stability analysis of the proposed controller is carried out and the Lyapunov function is defined as follows:

[0167] (28)

[0168] Where, , .

[0169] Formula (28) satisfies the following inequality relationship:

[0170] (29)

[0171] Where, , .

[0172] Taking the derivative of formula (28) and substituting it into formula (27) yields:

[0173] (30)

[0175] because is a constant matrix, there exists make The neural network estimation error satisfies .when When , formula (30) can be further simplified as:

[0176] (31)

[0177] Pick ,when , When , formula (31) can be further simplified as:

[0178] (32)

[0179] Where, .

[0180] According to the Lyapunov function inequality relationship shown in formula (29) and formula (32), it can be proved that Eventually uniformly bounded. Sliding mode dynamics Being able to converge to the boundary layer ensures that the closed-loop system error vector It can also be constrained to be a linear function of its boundary layer thickness.

[0181] S4: Design a closed-loop control law for the rolling descent phase of a tensegrity robot based on a partial feedback linearization control method to control the robot to maintain a balanced configuration and land smoothly.

[0182] Based on the partial feedback linearization control method, a closed-loop control law for the tensegrity robot during the rolling drop phase is designed to control the robot to achieve landing shock absorption, ensure a smooth landing, and recover and maintain a balanced configuration after landing. This includes:

[0183] S4.1: Shock absorption control during the rolling fall phase of a tensegrity robot

[0184] During the robot's rolling and falling phase, the ground contact force is regarded as an external disturbance. Based on the linear projection operator method in S3, the tensegrity robot dynamic model shown in formula (10) can be simplified as follows:

[0185] (33)

[0186] Where, , , , , , .

[0187] To ensure that the tensegrity robot can land smoothly during the rolling drop phase, the vertical displacement of the robot's landing contact point is selected as the task space:

[0188] , (34)

[0189] Where, represents the vertical displacement of the robot’s landing contact point, Represents the Jacobian matrix of the robot task space.

[0190] Multiply formula (33) on the left Substituting into formula (34), the task space dynamics of the robot during the rolling and falling phase can be expressed as:

[0191] (35)

[0192] Where, , , .

[0193] Design the robot closed-loop control law as:

[0194] (36)

[0195] Where, represents the expected vertical displacement of the robot's landing contact point, is a positive definite matrix representing the desired damping matrix, is a positive definite matrix representing the desired stiffness matrix.

[0196] Substituting formula (36) into formula (35), the task space closed-loop system can be expressed as:

[0197] (37)

[0198] By adjusting the desired damping matrix and stiffness matrix in formula (37), the robot can achieve landing shock absorption during the rolling fall phase to ensure that it can land smoothly.

[0199] S4.2: Configuration recovery control of a tensegrity robot during rolling drop

[0200] After the robot achieves rolling motion and lands smoothly, it is necessary to control the length of the robot's cable components to restore the robot to its initial configuration so that it can perform subsequent ground motion tasks.

[0201] Select the cable length of the tensegrity robot as the task space:

[0202] , (38)

[0203] Where, represents the cable member length vector, Represents the Jacobian matrix of the robot task space.

[0204] Multiply formula (33) on the left Substituting into formula (38), the task space dynamics of the robot during the rolling and falling phase can be expressed as:

[0205] (39)

[0206] Where, , , .

[0207] Design the robot closed-loop control law as:

[0208] (40)

[0209] Where, represents the expected length of the cable member, is a positive definite matrix representing the desired damping matrix, is a positive definite matrix representing the desired stiffness matrix.

[0210] Substituting formula (40) into formula (39), the task space closed-loop system can be expressed as:

[0211] (41)

[0212] By selecting The length of the cable component at the initial equilibrium configuration can realize the stable control of the robot configuration recovery during the rolling and falling stage.

[0213] S5: Determine whether there are three nodes of the tensegrity robot in contact with the ground at the same time, and determine whether the robot reaches the target point.

[0214] Determine whether there are three nodes of the tensegrity robot in contact with the ground at the same time;

[0215] If not, continue to execute the robot rolling and falling stage control task;

[0216] If so, determine whether the target point position is within the bottom triangle. If so, determine that the tensegrity robot has successfully completed a continuous rolling task; if not, return to S1 and select the rolling axis and driving point for the next rolling to continue executing the continuous rolling task.

[0217] This invention proposes a closed-loop control method for the rolling motion of a tensegrity robot. This method decomposes the robot's rolling motion into an initial rolling phase and a rolling drop phase, achieving controllable, robust, continuous, and efficient rolling motion for the tensegrity robot. Secondly, based on a quadratic programming approach, this invention provides a desired trajectory for the center of mass motion of a tensegrity robot during single-step rolling. The robot's center of mass tumbling about the rolling axis is described as the variation of the rolling angle and rolling radius. Based on this quadratic programming approach, a desired trajectory for the center of mass motion during the initial rolling phase that meets the desired constraints is designed. Furthermore, this invention proposes a neural network adaptive sliding mode control method to address the center of mass trajectory tracking problem for a tensegrity robot under contact and collision conditions. By using a radial basis function neural network adaptive law to approximate uncertainties, robust tracking of the robot's center of mass trajectory is achieved. Finally, this invention proposes a shock-absorbing control method for the rolling landing of a tensegrity robot. Based on a partial feedback linearization control method, a control law for the rolling drop phase of the tensegrity robot is designed. By incorporating a damping term into the vertical motion task space dynamics at the robot's landing contact point, this ensures a smooth landing during the rolling drop phase, thereby achieving efficient and continuous robot rolling motion.

[0218] In summary, the present invention designs a continuous rolling motion strategy for a tensegrity robot based on a closed-loop control method, improving the tensegrity robot's rolling efficiency and enabling continuous and efficient ground rolling motion for the tensegrity robot. Furthermore, a desired trajectory for the tensegrity robot's center of mass motion in single-step rolling is provided based on a quadratic programming method. The motion process of the robot's center of mass rolling around the rolling axis is described as the changes in the rolling angle and rolling radius. During the rolling motion of the center of mass along this trajectory, a rolling torque is provided by the driving point, enabling the robot to roll with minimal structural deformation, thereby improving rolling efficiency. Furthermore, the present invention designs a closed-loop center of mass motion trajectory tracking control law for the tensegrity robot's initial rolling phase based on a neural network adaptive sliding mode control method. By compensating for uncertainties using the neural network adaptive law, the robot's center of mass motion robustly tracks the desired trajectory, thereby achieving controllable rolling of the tensegrity robot. Finally, the present invention designs a closed-loop control law for the tensegrity robot's rolling descent phase based on a partial feedback linearization control method. This enables the robot to recover and maintain a stable configuration and achieves shock absorption during the tensegrity robot's rolling landing, ensuring a smooth landing for continuous rolling.

[0219] The present invention is further explained below with reference to specific embodiments.

[0220] Example 1: Simulation of Single-Step Rolling Motion Control of a Six-Rod Spherical Tensegrity Robot

[0221] The following uses the simulation results of the rolling motion control of a six-rod spherical tensegrity robot as an example to illustrate the effectiveness of the proposed solution. The six-rod spherical tensegrity robot consists of 6 rod components and 24 cable components. Its specific structure is as follows: Figure 8 shown.

[0222] The simulation process parameters are set as follows: the total simulation time is 2 s, the simulation step is 0.005 s, and the initial state of the tensegrity robot is stationary on the horizontal ground.

[0223] The physical parameters of the robot and the environment are as follows: The length of the rod is , the quality is ; The Young's modulus of the cable member is , the cross-sectional area is ; The coefficient of friction of the ground is , the acceleration due to gravity .

[0224] The parameters related to the robot's center of mass motion trajectory planning are defined as follows: , , , , , , , , , .

[0225] The relevant parameters of the tensegrity robot control law are: , , , , The initial value of is 0. , , , .

[0226] Under the above settings, the simulation results are as follows Figures 5 to 11 shown.

[0227] Figures 5 to 7 Figure 3 shows the center of mass trajectory tracking curve for the tensegrity robot during the initial rolling phase (0-0.2s). The dashed line represents the expected center of mass trajectory obtained through trajectory planning, while the solid line represents the actual center of mass trajectory. Simulation results show that during the initial rolling phase, the robot's forward, vertical, and lateral displacements are able to quickly and stably track the desired instructions, with tracking errors remaining within a small range. This allows the robot to complete the initial rolling phase with the desired rolling angular velocity and rolling radius, ultimately achieving the desired rolling motion.

[0228] Figures 8 to 11 This is a 3D motion state simulation diagram of the tensegrity robot's rolling motion process. Figure 8 for Simulation results at time t (initial time), the tensegrity robot is stationary on the horizontal ground; Figure 9 for Simulation results at time t (end of the initial rolling phase). The tensegrity robot completes the center of mass trajectory tracking task in the initial rolling phase. The driving point C leaves the ground, and the robot enters the rolling descent phase. Figure 10 for The simulation results at the moment (the highest point of rolling) show that the robot's center of mass moves to just above its rolling axis. After this moment, the robot completes the rolling and falling phase under the action of gravity. Figure 11 for The simulation results at time t (stable landing moment) show that the robot completes the rolling task and finally stabilizes on the horizontal ground. Under the action of the closed-loop control law during the rolling and falling phase, the robot achieves landing shock absorption, quickly stabilizes on the ground, and is able to restore and maintain its initial equilibrium configuration.

[0229] In summary, the closed-loop control method for the rolling motion of the tensegrity robot designed in the present invention can realize the tracking of the robot's center of mass motion trajectory in the initial stage of rolling, thereby controlling the robot to roll with a desired rolling angle and rolling radius, and can achieve rapid and smooth landing in the rolling drop stage. During the entire rolling motion process, the deformation of the tensegrity robot is kept within a small range, realizing the rolling motion of the tensegrity robot with controllable motion state and high rolling efficiency, which has guiding significance for the design of the tensegrity robot motion control system.

[0230] Example 2: Simulation of Continuous Rolling Motion Control of a Six-Rod Spherical Tensegrity Robot

[0231] The simulation results of continuous rolling motion control of a six-rod spherical tensegrity robot illustrate the effectiveness of the control strategy proposed in this paper.

[0232] The physical parameters related to the robot and the environment, the parameters related to the robot center of mass motion trajectory planning, and the parameters related to the tensegrity robot control law are the same as those in Example 1.

[0233] The simulation process parameters are set as follows: the total simulation time is 7.5s, the simulation step is 0.005s, the initial state of the tensegrity robot is stationary on the horizontal ground, and the initial center of mass position is , the scroll target point position is .

[0234] Under the above settings, the center of mass position of the tensegrity robot changes during continuous rolling as follows: Figure 12 As shown in the figure, the gray triangle represents the trajectory of the tensegrity robot's contact triangle during continuous rolling, the black dot represents the horizontal position of the robot's center of mass during continuous rolling, the * represents the horizontal position of the rolling target, and the dashed line represents the horizontal displacement of the robot's center of mass during continuous rolling. To quickly approach the target, the robot performed five consecutive rolls, each time selecting an axis that shortened the distance between the robot and the target. During the rolling process, the change in the contact triangle is as follows: ( Indicates that the ground contact triangle is an equilateral triangle. (denoting an isosceles contact triangle). By planning and tracking the center of mass trajectory during the initial rolling phase, the robot can complete rolling motion along the edges of various contact triangles. After the fifth roll, the target point enters the robot's contact triangle, completing a continuous rolling task, with its center of mass near the target point.

[0235] In summary, the closed-loop control method for rolling motion of a tensegrity robot designed in the present invention can realize continuous rolling motion of a tensegrity robot with controllable motion state and strong robustness.

[0236] The second object of the present invention is to propose a tensegrity robot rolling motion system based on closed-loop control, such as Figure 13 As shown, including:

[0237] The rolling axis and driving point selection module 101 is used to select the rolling axis and driving point based on the minimum direction principle according to the target point position of the tensegrity robot;

[0238] Obtaining the expected trajectory of the center of mass movement module 201: used to plan the trajectory of the roll angle and roll radius of the tensegrity robot according to the roll axis and the driving point, and calculate the expected trajectory of the center of mass movement based on the planned trajectory of the roll angle and roll radius of the tensegrity robot;

[0239] Desired rolling motion module 301: used to track the desired trajectory of the tensegrity robot's center of mass motion using a tensegrity robot dynamics model that considers contact and collision conditions, and obtain a desired center of mass motion instruction; and control the tensegrity robot to achieve the desired rolling motion with a desired rolling radius and rolling angle according to the desired center of mass motion instruction;

[0240] Robot landing module 401: when the tensegrity robot in the desired rolling motion enters the falling phase, a closed-loop control law for the rolling falling phase of the tensegrity robot is designed based on the partial feedback linearization control method to control the tensegrity robot to land;

[0241] The module 501 for determining the completion of rolling is used to determine whether the tensegrity robot has reached the set target point when landing. If so, it is determined that the tensegrity robot has successfully completed the continuous rolling task.

[0242] like Figure 14 As shown, a third object of the present invention is to provide an electronic device, comprising: a processor 601, a memory 602, and a display screen 603. The memory 602 and the display screen 603 are both connected to the processor 601, for example, via a bus 604. Optionally, the electronic device may further include a transceiver 605. It should be noted that in actual applications, the number of transceivers 605 is not limited to one, and the structure of the electronic device does not constitute a limitation on the embodiments of the present application.

[0243] Processor 601 can be a CPU (Central Processing Unit), a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It can implement or execute the various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of this application. Processor 601 can also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, etc.

[0244] Bus 604 may include a path for transmitting information between the aforementioned components. Bus 604 may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus. Bus 604 may be divided into an address bus, a data bus, a control bus, and the like.

[0245] The memory 602 may be a ROM (Read Only Memory) or other type of static storage device that can store static information and instructions, a RAM (Random Access Memory) or other type of dynamic storage device that can store information and instructions, an EEPROM (Electrically Erasable Programmable Read Only Memory), a CD-ROM (Compact Disc Read Only Memory) or other optical disk storage, an optical disc storage (including a compact disc, laser disc, optical disc, digital versatile disc, Blu-ray disc, etc.), a magnetic disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited thereto.

[0246] The memory 602 is used to store application code for executing the solution of the present application, and the execution is controlled by the processor 601. The processor 601 is used to execute the application code stored in the memory 602 to implement the content shown in the above method embodiment.

[0247] Figure 14 The electronic device shown is merely an example and should not limit the functions and scope of use of the embodiments of the present application.

[0248] The fourth object of the present invention is to provide a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, wherein the computer program is stored thereon, and when the program is executed by a processor, the computer program is realized as described above. Figures 1 to 2 The various processes of the illustrated method embodiment include, for example, a memory including instructions, and the instructions can be executed by a processor of an electronic device to perform the above method.

[0249] A computer-readable storage medium may be a tangible device that holds and stores instructions for use by an instruction execution device. A computer-readable storage medium may be, but is not limited to, an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any combination thereof. Specifically, a computer-readable storage medium may be a portable computer disk, a hard drive, a USB flash drive, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a rostrum random access memory (SRAM), a portable compact disc read-only memory (CD-ROM), a digital versatile disc (DVD), a memory stick, a floppy disk, an optical disc, a magnetic disk, a mechanical encoding device, or any combination thereof.

[0250] The fifth object of the present invention is to provide a computer program product comprising computer instructions, which, when executed by a processor, implement the above Figures 1 to 2 The various processes of the method embodiment shown can achieve the same technical effect, and to avoid repetition, they will not be described here.

[0251] Many embodiments and applications beyond the examples provided will be apparent to those skilled in the art upon reading the foregoing description. Therefore, the scope of the present teachings should be determined not with reference to the foregoing description, but rather with reference to the preceding claims, along with the full scope of equivalents to which such claims are entitled. For the purpose of completeness, all articles and references, including the disclosures of patent applications and publications, are incorporated herein by reference. The omission of any aspect of the subject matter disclosed herein from the preceding claims is not a disclaimer of such subject matter, nor should it be interpreted that the applicants did not consider such subject matter to be part of the disclosed inventive subject matter.

[0252] The above content is a further detailed description of the present invention, and it cannot be considered that the specific implementation methods of the present invention are limited to these. For ordinary technicians in the technical field to which the present invention belongs, they can make several simple deductions or substitutions without departing from the concept of the present invention, which should be regarded as belonging to the scope of protection of the present invention determined by the submitted claims.

Claims

1. A rolling motion method for a tensegrity robot based on closed-loop control, characterized in that: include: According to the target point position of the tensegrity robot rolling, the rolling axis and driving point are selected based on the minimum direction principle; planning a motion trajectory of a roll angle and a roll radius of the tensegrity robot according to the roll axis and the driving point, and calculating an expected trajectory of the center of mass motion based on the planned motion trajectory of the roll angle and the roll radius of the tensegrity robot; Using a tensegrity robot dynamics model that considers contact and collision conditions, the desired trajectory of the tensegrity robot's center of mass motion is tracked to obtain a desired center of mass motion instruction. Based on the desired center of mass motion instruction, the tensegrity robot is controlled to achieve a desired rolling motion with a desired rolling radius and rolling angle. When the tensegrity robot in the desired rolling motion enters the falling phase, a closed-loop control law for the rolling falling phase of the tensegrity robot is designed based on the partial feedback linearization control method to control the tensegrity robot to land. Determine whether the tensegrity robot reaches the set target point when landing. If so, it is determined that the tensegrity robot has successfully completed the continuous rolling task; The method of selecting the rolling axis and the driving point based on the minimum direction principle according to the target point position of the tensegrity robot rolling includes: According to the target point position of the tensegrity robot's rolling, the angle between the tensegrity robot's rolling direction and the desired direction is calculated based on the minimum direction principle. The base of the touchdown triangle around which the robot completes a single-step rolling process, corresponding to the minimum angle, is selected as the rolling axis of the tensegrity robot in the single-step rolling process. The nodes on the opposite side of the rolling axis of the tensegrity robot in single-step rolling are selected as driving points.

2. A tensegrity robot rolling motion method based on closed-loop control according to claim 1, characterized in that: The method of planning the motion trajectory of the roll angle and the roll radius of the tensegrity robot according to the roll axis and the drive point, and calculating the expected trajectory of the center of mass motion based on the planned motion trajectory of the roll angle and the roll radius of the tensegrity robot, comprises: Plan the rolling angular motion trajectory of the tensegrity robot during the initial rolling phase based on the rolling axis and driving point and rolling radius trajectory ; The rolling angle motion trajectory of the tensegrity robot in the initial rolling stage and rolling radius trajectory , converted into the expected motion trajectory of the robot's center of mass ; in, is the rolling angular displacement of the robot's center of mass at time k; is the rolling radius of the robot at time k; is the forward motion trajectory in the initial stage of rolling; It is the vertical motion trajectory in the initial stage of rolling.

3. The method for rolling motion of a tensegrity robot based on closed-loop control according to claim 1, characterized in that: By using a dynamic model of a tensegrity robot that considers contact and collision conditions, a desired trajectory of the center of mass motion of the tensegrity robot is tracked to obtain a desired center of mass motion instruction. The tensegrity robot is controlled to achieve a desired rolling motion with a desired rolling radius and rolling angle according to the desired center of mass motion instruction, including: Using the tensegrity robot dynamics model considering contact and collision conditions, and based on the neural network adaptive sliding mode closed-loop control method, the center of mass trajectory tracking control law of the tensegrity robot in the initial rolling stage is designed to track the desired center of mass motion trajectory of the tensegrity robot and obtain the desired center of mass motion command; The tensegrity robot is controlled to achieve a desired rolling motion with a desired rolling radius and rolling angle according to the desired center of mass motion instruction.

4. The method for rolling motion of a tensegrity robot based on closed-loop control according to claim 3, characterized in that: The method for designing the center of mass trajectory tracking control law of the tensegrity robot in the initial rolling stage based on the neural network adaptive sliding mode closed-loop control method includes: Where, is the control input; Input for prestress balance control of tensegrity structures; is the robot motion torque instruction; represents the pseudo-inverse of the robot's center-of-mass task space manipulation matrix; , , , represents the Young's modulus of the cable member, represents the cross-sectional area of ​​the cable member, The prestressed cable member balances the tension and satisfies the prestressed balance condition. ; represents the parameter vector of the robot's center of mass task space model; represents the Jacobian matrix of the robot's center of mass task space; represents the mass matrix of the tensegrity robot model after the projection transformation in the constraint space; The null space projection matrix representing the robot's constraint space; represents the equilibrium matrix of the tensegrity structure; Represents the estimated value of the neural network weight; Represents the neural network radial basis function vector; represents the expected trajectory of the center of mass motion; , is the control law parameter matrix; for estimated value of; represents the generalized coordinate vector of the tensegrity robot, indicating the position and orientation of the robot rod components; is a positive definite matrix; is the tracking error.

5. The method for rolling motion of a tensegrity robot based on closed-loop control according to claim 1, characterized in that: When the tensegrity robot in the desired rolling motion rolls into the falling phase, a closed-loop control law for the rolling falling phase of the tensegrity robot is designed based on a partial feedback linearization control method to control the tensegrity robot to land, including: When the tensegrity robot in the desired rolling motion rolls into the falling phase, a closed-loop control law for the rolling falling phase of the tensegrity robot is designed based on the partial feedback linearization control method, including the closed-loop control law for the landing shock absorption control of the tensegrity robot in the rolling falling phase and the closed-loop control law for the landing shock absorption control of the tensegrity robot in the rolling falling phase; The closed-loop control law of the tensegrity robot's landing shock absorption control during the rolling falling stage is used to control the tensegrity robot's landing shock absorption control during the rolling falling stage. The closed-loop control law for the configuration recovery control of the tensegrity robot during the rolling drop phase is used to control the configuration recovery stability control of the tensegrity robot during the rolling drop phase.

6. The method for rolling motion of a tensegrity robot based on closed-loop control according to claim 5, characterized in that: The closed-loop control law for the landing shock absorption control of the tensegrity robot during the rolling descent phase includes: in, is the control input; Input for prestress balance control of tensegrity structures; The pseudo-inverse of the space dynamics manipulation matrix for the robot's rolling and falling phase task; Represents the dynamic parameter vector of the robot during the rolling and falling phase; Represents the Jacobian matrix of the robot's task space during the rolling and falling phase; represents the equilibrium matrix of the tensegrity structure; represents the expected vertical displacement of the robot's landing contact point, is a positive definite matrix representing the desired damping matrix, is a positive definite matrix representing the expected stiffness matrix; is the inverse of the tensegrity robot's dynamic mass matrix after the projection transformation of the constraint space during the rolling drop phase; is the null space projection matrix of the robot constraint space during the rolling and falling phase; is the vertical displacement of the robot’s landing contact point; The closed-loop control law for the configuration recovery control of the tensegrity robot during the rolling and falling phase includes: in, represents the expected length of the cable member, is a positive definite matrix representing the desired damping matrix, is a positive definite matrix representing the expected stiffness matrix; The pseudo-inverse of the spatial dynamic manipulation matrix for the robot's rolling drop phase configuration recovery control task; represents the spatial dynamic parameter vector of the configuration recovery control task during the robot's rolling and falling phase; represents the Jacobian matrix of the robot configuration recovery control task space; is the expected length of the cable member; is a positive definite matrix representing the expected damping matrix; is a positive definite matrix representing the expected stiffness matrix; is the length vector of the cable member of the tensegrity robot.

7. The method for rolling motion of a tensegrity robot based on closed-loop control according to claim 1, characterized in that: The determining whether the tensegrity robot reaches a set target point when landing includes: Determine whether there are three nodes of the tensegrity robot in contact with the ground at the same time when it lands; If there are no three nodes in contact with the ground at the same time, continue execution; If there are three nodes that are in contact with the ground at the same time, determine whether the target point is within the triangle formed by the three nodes on the bottom surface; If the target point is within the triangle, the tensegrity robot is judged to have successfully completed a rolling motion; If the target point position is not within the triangle, the rolling axis and driving point are reselected based on the minimum direction principle according to the target point position of the tensegrity robot rolling, and the rolling motion is continued.

8. A system for a tensegrity robot rolling motion method based on closed-loop control according to any one of claims 1 to 7, characterized in that: include: Roll axis and driving point selection module: used to select the rolling axis and driving point based on the minimum direction principle according to the target point position of the tensegrity robot; Obtaining the expected trajectory of the center of mass motion: a module for planning the motion trajectory of the roll angle and roll radius of the tensegrity robot according to the roll axis and the driving point, and calculating the expected trajectory of the center of mass motion based on the planned motion trajectory of the roll angle and roll radius of the tensegrity robot; The desired rolling motion module is used to track the desired trajectory of the tensegrity robot's center of mass motion using a tensegrity robot dynamics model that considers contact and collision conditions, and obtain the desired center of mass motion instructions. Based on the desired center of mass motion instructions, the tensegrity robot is controlled to achieve the desired rolling motion with the desired rolling radius and rolling angle. Robot landing module: When the tensegrity robot in the desired rolling motion enters the falling phase, a closed-loop control law for the rolling falling phase of the tensegrity robot is designed based on the partial feedback linearization control method to control the landing of the tensegrity robot; Rolling completion judgment module: It is used to judge whether the tensegrity robot reaches the set target point when landing. If it reaches the target point, it is determined that the tensegrity robot has successfully completed the continuous rolling task.

9. An electronic device, characterized in that: The method comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the method realizes the steps of a rolling motion method of a tensegrity robot based on closed-loop control as described in any one of claims 1 to 7.

Images