Robot control method and system based on ontology intelligence
By constructing the geometric reference path and perceived motion reference value of the robot's zero moment point, determining the balancing relationship of the subsystem and controlling the step movement of the humanoid robot, the problem of low dynamic balancing control accuracy in the existing technology is solved, and higher accuracy and Stable motion control.
Patent Information
- Application Number
- CN202411108502.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-13
- Publication Date
- 2024-12-10
AI Technical Summary
When controlling the dynamic balance of humanoid robots, the prior art cannot effectively avoid dynamic interference caused by external uncertainty, resulting in low motion control accuracy. Especially when there are model errors and environmental dynamic uncertainty, it is difficult to achieve synchronization and coordination.
By constructing the geometric reference path of the robot's zero moment point, the perceived motion reference value of the robot subsystem is obtained, the subsystem's balance relationship is determined, and the step movement is controlled based on these reference values. The perceived motion reference value is used for real-time balance control to avoid time advancement. Dynamic balance is performed using force, torque and inertia sensors.
It realizes higher precision dynamic balanced motion control, avoids interference caused by external uncertainty, ensures synchronization and coordination of the movements of the robot's legs and trunk, and improves movement stability.
Smart Images
Figure CN119105329A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of intelligent robot control technology, and in particular to a robot control method and system based on ontology intelligence. Background Art
[0002] Humanoid robots are similar in appearance to humans, and the gait trajectory of humanoid robots is to plan the trajectory of a certain type of ground reference point through a series of valid footholds. Then, based on a simplified dynamic model and some assumptions, such as perfect ZMP (zero moment point) tracking, flat terrain, etc., the CoG (center of gravity) and the synchronized feasible trajectory of the legs are calculated. When implementing this time-based synchronization plan, the trajectory planning of the actual limbs and torso needs to ensure that they coordinate and cooperate when executing the trajectory of the CoG and ground reference points (such as ZMP).
[0003] The related art controls the motion dynamics of humanoid robots based on the time motion reference of the global clock or the perceptual motion reference related to kinematics. However, for the time motion reference of the global clock, its simplified dynamic model cannot accurately describe the real dynamics at the time level. Therefore, as long as there is a model error, the time parameterized trajectory is easy to fail, and this error will even accumulate continuously. Secondly, when local interference occurs (such as one leg is blocked), other parts (such as the other leg and CoG) cannot simply slow down or stop moving to achieve synchronization. For the perceptual motion reference related to kinematics, the required movement is driven by perceptual information and robot status in real time, and the movement of many humanoid parts should adaptively handle dynamic uncertainties from configuration aspects (including errors in simplified dynamic models, various communication delays between motors or robots) and environmental aspects (unexpected slippery or soft ground, or unexpected external forces). Therefore, the time motion reference based on the global clock or the perceptual motion reference related to kinematics is not the most ideal motion reference, and the related art cannot effectively control the negative impact of external interference on the control of humanoid robots.
[0004] In summary, the technical problems existing in the relevant technologies need to be improved. Summary of the invention
[0005] The main purpose of the embodiments of the present application is to propose a robot control method and system based on ontological intelligence, which can avoid dynamic interference caused by external uncertainties, thereby achieving higher-precision control of the dynamic balance movement of the robot.
[0006] To achieve the above objectives, one aspect of an embodiment of the present application proposes a robot control method based on ontology intelligence, the method comprising:
[0007] Constructing a geometric reference path of a zero-moment point of the robot and obtaining a sensory motion reference value of a robot subsystem, wherein the robot includes a plurality of said subsystems;
[0008] Determining a balance relationship of several subsystems of the robot according to a sensed motion reference value of the robot subsystem;
[0009] The robot's stepping motion is controlled according to the balance relationships of the subsystems.
[0010] In some embodiments, the step of constructing a geometric reference path of a robot zero moment point and obtaining a sensory motion reference value of a robot subsystem includes:
[0011] Based on the two-dimensional plane of the robot movement, construct a geometric reference path of the robot's zero moment point;
[0012] Acquire a support point of the three-dimensional inverted pendulum system on the geometric reference path, wherein the support point represents a moving distance of the zero-moment point of the robot along the geometric reference path;
[0013] Projection mapping is performed on the support points to obtain a sensory motion reference value of the robot subsystem.
[0014] In some embodiments, the expression of the geometric reference path of the robot's zero moment point is specifically as follows:
[0015]
[0016] In the above formula, y represents the geometric reference path of the robot's zero torque point, i represents the i-th path, s represents the sensory motion reference value of the robot subsystem, and f i ′(·) represents the derivative of the trajectory function, f i (·) represents the trajectory function on the xy plane, x(·) represents the inverse function of the sensory motion reference value of the robot subsystem, and R represents a real number.
[0017] In some embodiments, the step of obtaining a support point of the three-dimensional inverted pendulum system on the geometric reference path includes:
[0018] Determining the center of gravity of the robot and the zero moment point of the robot;
[0019] Based on the center of gravity of the robot and the zero moment point of the robot, a dynamic relationship between the center of gravity of the robot and the zero moment point of the robot is determined by a three-dimensional linear inverted pendulum model;
[0020] The supporting point of the three-dimensional inverted pendulum system on the geometric reference path is determined according to the dynamic relationship.
[0021] In some embodiments, the expression of the dynamic relationship between the center of gravity of the robot and the zero moment point of the robot is specifically as follows:
[0022]
[0023] In the above formula, M represents the mass of the robot, g represents the acceleration due to gravity, and x ZMP Indicates the horizontal position of the robot's zero torque point, y ZMP represents the vertical position of the robot's zero torque point, x CMP ,y CMP Indicates the positional relationship between the robot's center of gravity and the robot's zero moment point, represents the centroid vector, τ x represents the moment of the center of mass around the x-axis, τ y represents the moment of the center of mass around the y-axis, F z Represents the force in the z-axis direction.
[0024] In some embodiments, the expression for obtaining the sensory motion reference value of the robot subsystem is specifically as follows:
[0025]
[0026] In the above formula, s represents the sensory motion reference value of the robot subsystem, m represents the number of path segments at the robot's zero torque point, i represents the i-th path segment, and x i represents the horizontal coordinate corresponding to the i-th path, f i ′(x) represents the derivative of the trajectory function, x p Indicates the horizontal coordinate corresponding to the newly inserted p-segment path, x m Indicates the horizontal coordinate corresponding to the mth path segment.
[0027] In some embodiments, determining the balance relationship of several subsystems of the robot according to the sensory motion reference value of the robot subsystem includes:
[0028] Obtain the reference value of the first subsystem of the robot for sensing the movement at the start of a swing, the reference value of the first subsystem of the robot for sensing the movement at the end of a swing, the reference value of the second subsystem of the robot for sensing the movement at the start of a swing, and the reference value of the second subsystem of the robot for sensing the movement at the end of a swing, perform difference calculation processing, and determine the balance relationship between the several subsystems of the robot.
[0029] In some embodiments, the expressions of the balance relations of several subsystems of the robot are specifically as follows:
[0030]
[0031] In the above formula, s lfRepresents the balance relationship of several subsystems of the robot, Indicates the reference value of the first subsystem starting to perceive motion, represents the end perceived motion reference value of the first subsystem, Indicates the reference value of the second subsystem starting to perceive motion, represents the end sensory motion reference value of the second subsystem, and s represents the sensory motion reference value of the robot subsystem.
[0032] In some embodiments, controlling the robot to take steps according to the balance relationship of the subsystems includes:
[0033] According to the balance relationships of the subsystems, obtaining a sensory motion reference value of the robot subsystem;
[0034] Setting the expected landing point of the robot subsystem;
[0035] Controlling the robot's stepping motion according to the sensed motion reference value to obtain the actual landing point of the robot subsystem;
[0036] The position difference between the expected landing point and the actual landing point is obtained through an encoder, and the perceived motion reference value is compensated, and the robot's stepping motion is controlled based on the compensated perceived motion reference value.
[0037] To achieve the above objectives, another aspect of the embodiment of the present application provides a robot control system based on ontology intelligence, the system comprising:
[0038] An acquisition module, used to construct a geometric reference path of a zero-moment point of the robot and acquire a sensory motion reference value of a robot subsystem, wherein the robot includes several of the subsystems;
[0039] A balancing module, used to determine the balancing relationship of several subsystems of the robot according to the sensed motion reference value of the robot subsystem;
[0040] The control module is used to control the stepping motion of the robot according to the balance relationship of the subsystems.
[0041] The embodiments of the present application include at least the following beneficial effects: The present application provides a robot control method and system based on ontological intelligence, which performs robot path tracking by acquiring sensory motion reference values of the robot subsystems, which is used as a reference for planning the synchronous motion of the robot legs and the motion of the center of mass, and determines the balance relationship of several subsystems of the robot according to the sensory motion reference values of the robot subsystems, controls the robot's stepping motion, and dynamically balances based on the real-time output of force, torque and inertial sensors. Balance control is recommended by information rather than time advancement. Continuous sensory information is used as a common motion reference in the path tracking of the robot's legs and body, which can avoid dynamic interference caused by external uncertainties, thereby achieving higher-precision control of the robot's dynamic balance motion. BRIEF DESCRIPTION OF THE DRAWINGS
[0042] Figure 1 is a flow chart of a robot control method based on ontology intelligence provided in an embodiment of the present application;
[0043] Figure 2 It is a schematic diagram of the structure of a robot control system based on ontology intelligence provided in an embodiment of the present application;
[0044] Figure 3 It is a structural schematic diagram of the robot stepping motion control provided in an embodiment of the present application. DETAILED DESCRIPTION
[0045] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application is further described in detail below in conjunction with the accompanying drawings and examples. It should be understood that the specific embodiments described herein are only used to explain the present application and are not intended to limit the present application. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the embodiments of the present application. They are only examples of systems and methods consistent with some aspects of the embodiments of the present application as detailed in the attached claims.
[0046] It is understood that the terms "first", "second", etc. used in this application can be used to describe various concepts in this article, but unless otherwise specified, these concepts are not limited by these terms. These terms are only used to distinguish one concept from another concept. For example, without departing from the scope of the embodiment of the present application, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Depending on the context, the words "if" and "if" as used herein can be interpreted as "at the time of" or "when" or "in response to determination".
[0047] The terms "at least one", "multiple", "each", "any", etc. used in this application, at least one includes one, two or more, multiple includes two or more, each refers to each of the corresponding multiple, and any refers to any one of the multiple.
[0048] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which this application belongs. The terms used herein are only for the purpose of describing the embodiments of this application and are not intended to limit this application.
[0049] Reference Figure 1 , Figure 1 A flowchart of a robot control method based on ontology intelligence provided by an embodiment of the present invention, referring to Figure 1 , the method comprises the following steps:
[0050] S100, constructing a geometric reference path of a zero-torque point of the robot, and obtaining a sensory motion reference value of a robot subsystem, wherein the robot includes several of the subsystems;
[0051] It should be noted that, in some embodiments, step S100 may include: S110, constructing a geometric reference path of the robot's zero moment point based on the two-dimensional plane of the robot's movement; S120, obtaining the support points of the three-dimensional inverted pendulum system on the geometric reference path, the support points representing the moving distance of the robot's zero moment point along the geometric reference path; S130, performing projection mapping on the support points to obtain the perceived motion reference value of the robot subsystem.
[0052] In some embodiments, step S120 may include: S121, determining the center of gravity of the robot and the zero moment point of the robot; S122, based on the center of gravity of the robot and the zero moment point of the robot, determining the dynamic relationship between the center of gravity of the robot and the zero moment point of the robot through a three-dimensional linear inverted pendulum model; S123, determining the support point of the three-dimensional inverted pendulum system on the geometric reference path according to the dynamic relationship.
[0053] In some specific embodiments, first, a given ZMP geometric reference path on a two-dimensional xy plane can be described as a fragment function y=f i ′(x(s))=f i (s)∈R, where i=1,2,…,m, m is the number of ZMP path segments, s∈R, s∈[0-s semd ]. s is a scalar parameter in the ZMP path tracking problem, which is calculated based on the real-time P on the ZMP geometric reference path. ZMP The projection point P P (x P ,x P). Secondly, according to the expression of the reference value of the perceived motion of propulsion, the moving distance of ZMP along the reference path represents the support point of the three-dimensional inverted pendulum system. The support point of the three-dimensional inverted pendulum system represents a key dynamic state that can continuously guide the movement of other parts of the body.
[0054] The functional expression of the reference value s of the perceived motion of propulsion is as follows:
[0055]
[0056] In the above formula, s represents the sensory motion reference value of the robot subsystem, m represents the number of path segments at the robot's zero torque point, i represents the i-th path segment, and x i represents the horizontal coordinate corresponding to the i-th path, f i ′(x) represents the derivative of the trajectory function, x p Indicates the horizontal coordinate corresponding to the newly inserted p-segment path, x m Indicates the horizontal coordinate corresponding to the mth path segment.
[0057] The dynamics and relationship between the robot center of gravity (CoG) and the robot zero moment point (ZMP) in this system are described by a three-dimensional linear inverted pendulum (LIP) model, that is, the dynamic relationship between the two is obtained. The expression of the dynamic relationship between the robot center of gravity and the robot zero moment point is as follows:
[0058]
[0059] In the above formula, M represents the mass of the robot, g represents the acceleration due to gravity, and x ZMP Indicates the horizontal position of the robot's zero torque point, y ZMP represents the vertical position of the robot's zero torque point, x CMP ,y CMP Indicates the positional relationship between the robot's center of gravity and the robot's zero moment point, represents the centroid vector, τ x represents the moment of the center of mass around the x-axis, τ y represents the moment of the center of mass around the y-axis, F z Represents the force in the z-axis direction.
[0060] S200, determining a balance relationship of several subsystems of the robot according to a sensed motion reference value of the robot subsystem;
[0061] It should be noted that, in some embodiments, the reference value of the first subsystem of the robot at the beginning of the swing perception movement, the reference value of the first subsystem of the robot at the end of the swing perception movement, the reference value of the second subsystem of the robot at the beginning of the swing perception movement and the reference value of the second subsystem of the robot at the end of the swing perception movement are obtained, and the difference calculation is performed to determine the balance relationship of several subsystems of the robot.
[0062] In some specific embodiments, in order to establish the relationship between the leg motion reference value and the actual value of the main motion reference value, we define the expressions of the balance relationship of several subsystems of the robot to describe the balance relationship between the subsystems, and the expressions are:
[0063]
[0064] In the above formula, s lf Represents the balance relationship of several subsystems of the robot, Indicates the reference value of the first subsystem starting to perceive motion, represents the end perceived motion reference value of the first subsystem, Indicates the reference value of the second subsystem starting to perceive motion, represents the end sensory motion reference value of the second subsystem, and s represents the sensory motion reference value of the robot subsystem.
[0065] Specifically, the exemplary ones are as follows:
[0066]
[0067] In the above formula, represents the path coordinates, represents the planning path of the first subsystem, Represents the planned path of the second subsystem.
[0068] S300, controlling the robot to move according to the balance relationship of the subsystems;
[0069] It should be noted that, in some embodiments, step S300 may include: S310, obtaining a perceived motion reference value of the robot subsystem based on the balance relationship of several subsystems; S320, setting an expected landing point of the robot subsystem; S330, controlling the robot's stepping motion according to the perceived motion reference value, and obtaining the actual landing point of the robot subsystem; S340, obtaining the position difference between the expected landing point and the actual landing point through an encoder, and compensating the perceived motion reference value, and controlling the robot's stepping motion based on the compensated perceived motion reference value.
[0070] In some specific embodiments, after the reference of the system's forward movement is determined, the system can be controlled by the reference. The humanoid robot control system is divided into multiple subsystems (limbs) and the subsystems are interconnected to ensure that the planning of the legs and hands is synchronized. Figure 3 This paper describes the perception planning of the humanoid robot and the synchronous control of each subsystem. It mainly revolves around the perception motion reference point s. First, the reference value s of each subsystem is calculated by obtaining the expression of the perception motion reference value of the robot subsystem. i , i = 1…n, and then the expected landing point is calculated by the time planner of the subsystem, and then the humanoid robot subsystem moves to the landing point through the gait controller, and the position difference between the expected landing point and the actual landing point is calculated through sensors such as encoders, and then the next step of planning and control is carried out. This is a cyclic tracking process.
[0071] Further, it should be noted that the key point of the embodiment of the present invention is to introduce a sensory motion reference s, which is a scalar, an effective abstraction of the system state based on sensory information, and plans the required subsystem (limb) motion trajectory as a function of s. The key difference from traditional reference tracking is that the sensory motion reference s is used for path tracking in the motion task, rather than time-based reference tracking. Specifically, the leg and CoG paths are described using time-independent path variables, which act as synchronizers between the parts, and "path following" is used not only in ZMP control, but also in coordinating the synchronization of double-leg motion. In order to directly guarantee the dynamic balance standard without time constraints, the sensory reference value will be continuously calculated during the execution of the motion plan, and the impact of interference such as model inaccuracy or signal transmission delay on balance will be included. All expected values, including force, direction and speed, are described as functions of the sensory motion reference value s. s will replace time and become a reference for planning leg synchronization and center of mass motion.
[0072] The robot motion structure of the embodiment of the present invention adopts the ZMP balance criterion, that is, at any time during the walking process, the ZMP should be kept at the center of the support polygon. Therefore, the current walking distance of the ZMP along the planned ZMP reference path is selected as the sensory motion reference value, and the variable support polygon created by the support foot is used as the boundary.
[0073] See also Figure 2 The embodiment of the present application also provides a robot control system based on ontology intelligence, which can implement the above-mentioned robot control method based on ontology intelligence, and the system includes:
[0074] An acquisition module 201 is used to construct a geometric reference path of a zero-moment point of the robot and acquire a sensory motion reference value of a robot subsystem, wherein the robot includes several of the subsystems;
[0075] A balancing module 202, configured to determine a balance relationship of several subsystems of the robot according to a sensed motion reference value of the robot subsystem;
[0076] The control module 203 is used to control the robot's stepping motion according to the balance relationship of the subsystems.
[0077] It can be understood that the contents of the above method embodiments are all applicable to the present system embodiments, the functions specifically implemented by the present system embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.
[0078] The preferred embodiments of the present invention are described above with reference to the accompanying drawings, but the scope of the rights of the present invention is not limited thereto. Any modification, equivalent substitution and improvement made by a person skilled in the art without departing from the scope and essence of the present invention should be within the scope of the rights of the present invention.
Claims
1. A robot control method based on ontology intelligence, It is characterized in that The method comprises the following steps: Constructing a geometric reference path of a zero-moment point of the robot and obtaining a sensory motion reference value of a robot subsystem, wherein the robot includes a plurality of said subsystems; Determining a balance relationship of several subsystems of the robot according to a sensed motion reference value of the robot subsystem; The robot's stepping motion is controlled according to the balance relationships of the subsystems.
2. The method according to claim 1, It is characterized in that The step of constructing a geometric reference path of the robot's zero torque point and obtaining a sensory motion reference value of the robot subsystem includes: Based on the two-dimensional plane of the robot movement, construct a geometric reference path of the robot's zero moment point; Acquire a support point of the three-dimensional inverted pendulum system on the geometric reference path, wherein the support point represents a moving distance of the zero-moment point of the robot along the geometric reference path; Projection mapping is performed on the support points to obtain a sensory motion reference value of the robot subsystem.
3. The method according to claim 2, It is characterized in that The expression of the geometric reference path of the robot's zero moment point is specifically as follows: y=f i ′(x(s))=f i (s)∈R In the above formula, y represents the geometric reference path of the robot's zero torque point, i represents the i-th path, s represents the sensory motion reference value of the robot subsystem, and f i ′(·) represents the derivative of the trajectory function, f i (·) represents the trajectory function on the xy plane, x(·) represents the inverse function of the sensory motion reference value of the robot subsystem, and R represents a real number.
4. The method according to claim 2, It is characterized in that The obtaining of the support point of the three-dimensional inverted pendulum system on the geometric reference path comprises: Determining the center of gravity of the robot and the zero moment point of the robot; Based on the center of gravity of the robot and the zero moment point of the robot, a dynamic relationship between the center of gravity of the robot and the zero moment point of the robot is determined by a three-dimensional linear inverted pendulum model; The supporting point of the three-dimensional inverted pendulum system on the geometric reference path is determined according to the dynamic relationship.
5. The method according to claim 4, It is characterized in that The expression of the dynamic relationship between the center of gravity of the robot and the zero moment point of the robot is specifically as follows: In the above formula, M represents the mass of the robot, g represents the acceleration due to gravity, and x ZMP Indicates the horizontal position of the robot's zero torque point, y ZMP represents the vertical position of the robot's zero torque point, x CMP ,y CMP Indicates the positional relationship between the robot's center of gravity and the robot's zero moment point, represents the centroid vector, τ x represents the moment of the center of mass around the x-axis, τ y represents the moment of the center of mass around the y-axis, F z Represents the force in the z-axis direction.
6. The method according to claim 2, It is characterized in that The expression for obtaining the reference value of the perceived motion of the robot subsystem is specifically as follows: In the above formula, s represents the sensory motion reference value of the robot subsystem, m represents the number of path segments at the robot's zero torque point, i represents the i-th path segment, and x i represents the horizontal coordinate corresponding to the i-th path, f i ′(x) represents the derivative of the trajectory function, x p Indicates the horizontal coordinate corresponding to the newly inserted p-segment path, x m Indicates the horizontal coordinate corresponding to the mth path segment.
7. The method according to claim 1, It is characterized in that Determining the balance relationship of several subsystems of the robot according to the sensed motion reference value of the robot subsystem includes: Obtain the reference value of the first subsystem of the robot for sensing the movement at the start of a swing, the reference value of the first subsystem of the robot for sensing the movement at the end of a swing, the reference value of the second subsystem of the robot for sensing the movement at the start of a swing, and the reference value of the second subsystem of the robot for sensing the movement at the end of a swing, perform difference calculation processing, and determine the balance relationship between the several subsystems of the robot.
8. The method according to claim 7, It is characterized in that The expressions of the balance relations of several subsystems of the robot are specifically as follows: In the above formula, s lf Represents the balance relationship of several subsystems of the robot, Indicates the reference value of the first subsystem starting to perceive motion, represents the end perceived motion reference value of the first subsystem, Indicates the reference value of the second subsystem starting to perceive motion, represents the end sensory motion reference value of the second subsystem, and s represents the sensory motion reference value of the robot subsystem.
9. The method according to claim 1, It is characterized in that The stepping motion of the robot is controlled according to the balance relationship of the plurality of subsystems, including: According to the balance relationships of the subsystems, obtaining a sensory motion reference value of the robot subsystem; Setting the expected landing point of the robot subsystem; Controlling the robot's stepping motion according to the sensed motion reference value to obtain the actual landing point of the robot subsystem; The position difference between the expected landing point and the actual landing point is obtained through an encoder, and the perceived motion reference value is compensated, and the robot's stepping motion is controlled based on the compensated perceived motion reference value.
10. A robot control system based on ontology intelligence, It is characterized in that The system comprises: An acquisition module, used to construct a geometric reference path of a zero-moment point of the robot and acquire a sensory motion reference value of a robot subsystem, wherein the robot includes several of the subsystems; A balancing module, used to determine the balancing relationship of several subsystems of the robot according to the sensed motion reference value of the robot subsystem; The control module is used to control the stepping motion of the robot according to the balance relationship of the subsystems.