Robot control method, system, robot and medium
By extracting and retaining the linear components of the nonlinear moment, an optimized linearized centroid motion model is generated, which solves the problem of large error in the rotational motion control of robots in the prior art, and achieves high precision and high robust motion control.
Patent Information
- Application Number
- CN202110632729.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2021-06-07
- Publication Date
- 2024-12-06
- Estimated Expiration
- 2041-06-07
AI Technical Summary
During the rotational motion, existing robot motion control directly applies nonlinear torque, which has large calculation amount and poor real-time performance. The nonlinear amount is omitted and linearized for linearization, resulting in large motion control errors and low accuracy and robustness.
By obtaining the robot's current rotational motion information and target rotational motion reference information, the linear component in the nonlinear torque is extracted and retained using a linear function to generate an optimized linearized centroid motion model for controlling the robot's target torque and contact force, thereby Achieve high precision and flexible motion control.
It improves the accuracy and robustness of robot motion control, reduces computational complexity, and enhances the real-time and stability of control.
Smart Images

Figure CN115502965B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the fields of artificial intelligence and robots, and more specifically to a robot control method, system, robot and medium. Background Art
[0002] With the widespread application of artificial intelligence and robotics technology in civil and commercial fields, robots based on artificial intelligence and robotics technology are playing an increasingly important role in smart transportation, smart home and other fields, and are also facing higher requirements.
[0003] At present, when the robot is controlled in motion, the target motion reference information of the robot is first generated through the motion planning information of the motion planner, and then the target motion estimation of the robot is determined based on the center of mass motion model of the robot. The optimal target contact force is obtained through the target motion reference information and the target motion estimation to realize the control of the robot motion, so that the actual motion of the robot accurately follows the planned target motion process. Among them, since the rotational motion of the robot (especially the angular motion therein) is a nonlinear motion, when the nonlinear rotational motion parameter (torque) is directly applied, the amount of calculation in the motion control process is large, and the real-time performance of the motion control is poor; in the process of linearizing the torque, in the current motion control, the nonlinear amount in the torque is directly omitted, and only the linear amount is used for calculation, which will lead to a large error in the motion control of the robot, and there is a large error between the actual motion of the robot and the planned motion process, and the motion control accuracy is poor and the robustness is low.
[0004] Therefore, there is a need for a method that can achieve robot motion control, especially real-time robot motion control, by better extracting and retaining the linear components in the nonlinear torque to improve the accuracy of the torque, thereby generating an optimized linearized center of mass motion model to flexibly and accurately control the robot's motion process. The method has good accuracy and stability and high robustness. Summary of the invention
[0005] In view of the above problems, the present invention provides a robot control method, a robot control system, a robot and a medium. The robot control method provided by the present invention can be used to perform good motion control on the robot, and by making the current rotational motion information include at least one of the current rotation angle of the robot and the current rotation angular velocity of the robot, and making the target rotational motion reference information include at least one of the target rotation reference angle of the robot, the target rotation reference angular velocity of the robot, and the target rotation reference angular acceleration of the robot, the robot can accurately realize the target motion process, realize flexible and high-precision control, and the method has good reliability, stability and robustness.
[0006] According to one aspect of the present disclosure, a robot control method is proposed, the method comprising: obtaining current rotational motion information of the robot and target rotational motion reference information of the robot; determining a target torque for controlling the target rotational motion of the robot based on a linear function according to the current rotational motion information of the robot; determining a target torque of each joint of the robot based on the target torque, the current rotational motion information of the robot and the target rotational motion reference information of the robot; wherein the current rotational motion information comprises at least one of the current rotation angle of the robot and the current rotation angular velocity of the robot; and the target rotational motion reference information comprises at least one of the target rotation reference angle of the robot, the target rotation reference angular velocity of the robot and the target rotation reference angular acceleration of the robot.
[0007] In some embodiments, the linear function includes: at least two of the robot target rotation angular acceleration parameter, the robot angular velocity gain parameter, the robot angle gain parameter, and the robot current rotation angular velocity parameter, wherein the robot angular velocity gain parameter is the difference between the robot target rotation angular velocity and the robot current rotation angular velocity; the angle gain parameter is the difference between the robot target rotation angle and the robot current rotation angle.
[0008] In some embodiments, the linear function includes: a robot target rotation angular acceleration parameter, a robot angular velocity gain parameter, a robot angle gain parameter, and a robot current rotation angular velocity parameter; wherein the robot angular velocity gain parameter is the difference between the robot target rotation angular velocity and the robot current rotation angular velocity; and the angle gain parameter is the difference between the robot target rotation angle and the robot current rotation angle.
[0009] In some embodiments, the method also includes: obtaining the current translational motion information of the robot and the target translational motion reference information of the robot; and wherein the target translational motion reference information of the robot includes: the robot target motion reference position, the robot target motion reference speed, and the robot target motion reference acceleration; the current translational motion information of the robot includes: the robot current motion position, the robot current motion speed, and the robot current motion acceleration.
[0010] In some embodiments, the linear function is:
[0011]
[0012] And among them, is the target time t k The target torque is I0, the moment of inertia at the current time t0, is the robot's target rotation angular acceleration, P and Q are parameters selected according to actual needs, and Δωk is the robot angular velocity gain parameter, where Δω k =ω k -ω0,ω k is the robot's target rotation angular velocity, ω0 is the robot's current rotation angular velocity, Δθ k is the robot angle gain parameter, where Δθ k =θ k -θ0,θ k is the target rotation angle of the robot, θ0 is the current rotation angle of the robot, It means converting the current angular velocity of the robot into a skew-symmetric matrix.
[0013] In some embodiments, determining the target torque of each joint of the robot based on the target torque, the current rotational motion information of the robot and the target rotational motion reference information of the robot includes: determining the target contact force of the robot according to the target torque, the current motion information of the robot and the target motion reference information of the robot; determining the target torque of each joint of the robot based on the target contact force, the skeletal structure of the robot and the posture information of the robot; wherein the current motion information of the robot includes: current rotational motion information and current translational motion information; the target motion reference information of the robot includes: target rotational motion reference information and target translational motion reference information.
[0014] In some embodiments, determining the target contact force of the robot according to the target torque, the current motion information of the robot and the target motion reference information of the robot includes: generating a target motion estimate of the robot based on a robot center of mass motion model according to the current motion information of the robot and the target torque, wherein the target motion estimate is a function of the contact force of the robot; generating an error function of the target motion reference information and the target motion estimate based on the target motion reference information and the target motion estimate; determining the target contact force based on the error function; wherein the target motion estimate includes: at least a portion of a target motion position estimate, a target motion velocity estimate, a target motion acceleration estimate, a target rotation angle estimate, a target rotation angular velocity estimate and a target rotation angular acceleration estimate.
[0015] In some embodiments, determining the target contact force based on the error function includes: optimizing the error function based on a quadratic optimization algorithm, and determining the contact force that minimizes the error function as the target contact force.
[0016] In some embodiments, determining the target torque of each joint of the robot based on the target contact force, the skeletal structure of the robot and the posture information of the robot includes: determining the main torque of each joint of the robot based on the target contact force and the skeletal structure of the robot; obtaining the target posture reference information and current posture information of each joint of the robot; determining the additional torque of each joint of the robot based on the target posture reference information of the robot and the current posture information of the robot; and determining the target torque of each joint of the robot based on the main torque and the additional torque.
[0017] In some embodiments, obtaining the target posture reference information and current posture information of each joint of the robot includes: obtaining the motion planning information of the robot, and generating the target posture reference information of the robot based on the motion planning information, the target posture reference information including the joint target reference angle and the joint target reference angular velocity of each joint of the robot; obtaining the current posture information of the robot, the current posture information including the joint current angle and the joint current angular velocity of each joint.
[0018] According to another aspect of the present disclosure, a robot control system is proposed, which includes: a rotational motion information acquisition module, which is configured to acquire the current rotational motion information of the robot and the target rotational motion reference information of the robot; a target torque generation module, which is configured to determine the target torque for controlling the target rotational motion of the robot based on a linear function according to the current rotational motion information of the robot; a joint torque generation module, which is configured to determine the target torque of each joint of the robot based on the target torque, the current rotational motion information of the robot and the target rotational motion reference information of the robot; wherein the current rotational motion information includes: at least one of the current rotation angle of the robot and the current rotation angular velocity of the robot; and the target rotation motion reference information includes: at least one of the target rotation reference angle of the robot, the target rotation reference angular velocity of the robot and the target rotation reference angular acceleration of the robot.
[0019] In some embodiments, the linear function includes: at least two of the robot target rotation angular acceleration parameter, the robot angular velocity gain parameter, the robot angle gain parameter, and the robot current rotation angular velocity parameter, wherein the robot angular velocity gain parameter is the difference between the robot target rotation angular velocity and the robot current rotation angular velocity; the angle gain parameter is the difference between the robot target rotation angle and the robot current rotation angle.
[0020] In some embodiments, the linear function includes: a robot target rotation angular acceleration parameter, a robot angular velocity gain parameter, a robot angle gain parameter, and a robot current rotation angular velocity parameter; wherein the robot angular velocity gain parameter is the difference between the robot target rotation angular velocity and the robot current rotation angular velocity; and the angle gain parameter is the difference between the robot target rotation angle and the robot current rotation angle.
[0021] According to another aspect of the present disclosure, a robot is proposed. The robot includes the robot control system as described above, and the robot controls the motion of the robot through the robot control method as described above.
[0022] According to another aspect of the present disclosure, a computer-readable storage medium is provided, wherein computer-readable instructions are stored thereon, and when the instructions are executed by a computer, the above-mentioned method is executed.
[0023] By using the robot control method, system, robot and medium provided by the present invention, the accuracy of the generated target torque can be improved by making the current rotational motion information include at least one of the current rotation angle of the robot and the current rotation angular velocity of the robot. By making the target rotational motion reference information include at least one of the target rotation reference angle of the robot, the target rotation reference angular velocity of the robot and the target rotation reference angular acceleration of the robot, it is beneficial to subsequently realize the generation of target torques of various joints of the robot based on the target rotational motion reference information, improve the accuracy of the generated target torque, and thus realize the motion control of the robot well, especially, it has high control accuracy and good robustness. BRIEF DESCRIPTION OF THE DRAWINGS
[0024] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. The following drawings are not deliberately scaled according to the actual size, and the focus is on illustrating the main purpose of the present invention.
[0025] Figure 1 An exemplary flow chart of a robot control method 100 according to an embodiment of the present disclosure is shown;
[0026] Figure 2 An exemplary flow chart of a process S103 of determining the target torque of each joint of the robot according to an embodiment of the present disclosure is shown;
[0027] Figure 3FIG. 4 shows an exemplary flow chart of a process S1031 of determining the target contact force of the robot according to an embodiment of the present disclosure;
[0028] Figure 4 An exemplary flow chart of a process S1032 for determining a target torque of each joint of the robot according to an embodiment of the present disclosure is shown;
[0029] Figure 5 A schematic diagram of a robot control process according to an embodiment of the present disclosure is shown;
[0030] Figure 6 An exemplary block diagram of a robot control system 600 according to an embodiment of the present invention is shown. DETAILED DESCRIPTION
[0031] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work also fall within the scope of protection of the present invention.
[0032] As shown in this application and claims, unless the context clearly indicates an exception, the words "a", "an", "an" and / or "the" do not refer to the singular and may also include the plural. Generally speaking, the terms "include" and "comprise" only indicate the inclusion of the steps and elements that have been clearly identified, and these steps and elements do not constitute an exclusive list. The method or device may also include other steps or elements.
[0033] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, any number of different modules can be used and run on the user terminal and / or server. The modules are only illustrative, and different aspects of the system and method can use different modules.
[0034] Flowcharts are used in the present application to illustrate the operations performed by the system according to the embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed accurately in order. On the contrary, various steps may be processed in reverse order or simultaneously as required. Meanwhile, other operations may also be added to these processes, or a certain step or several steps of operations may be removed from these processes.
[0035] Artificial Intelligence (AI) is the theory, method, technology and application system that uses digital computers or machines controlled by digital computers to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use knowledge to obtain the best results. In other words, artificial intelligence is a comprehensive technology in computer science that attempts to understand the essence of intelligence and produce a new intelligent machine that can respond in a similar way to human intelligence. Artificial intelligence is to study the design principles and implementation methods of various intelligent machines so that machines have the functions of perception, reasoning and decision-making.
[0036] The present application relates to an application of artificial intelligence in motion control. Specifically, the present application proposes a robot control method based on artificial intelligence. By utilizing exponential coordinate quantities to represent relevant nonlinear parameter quantities (target torque parameters) of the robot during rotational motion, the linear component in the nonlinear parameter quantity can be further obtained and retained. Subsequently, the linearization of the center of mass motion model of the robot can be better achieved based on the target torque parameter, thereby improving the accuracy and reliability of the robot motion control on the basis of achieving real-time control based on the linearized model.
[0037] The robot described in the present application is a robot capable of realizing autonomous motion control, and it can have various forms according to actual needs. For example, the robot can be a scooter, a bicycle, or other types of robotic devices. The embodiments of the present disclosure are not limited by the specific type and composition of the robot.
[0038] At present, when the robot is controlled in motion, the target motion reference information of the robot is first generated through the motion planning information of the motion planner, and then the target motion estimation of the robot is determined based on the center of mass motion model of the robot. The optimal target contact force is obtained through the target motion reference information and the target motion estimation to realize the control of the robot motion, so that the actual motion of the robot accurately follows the planned target motion process. Among them, since the rotational motion of the robot (especially the angular motion therein) is a nonlinear motion, when the nonlinear rotational motion parameter (torque) is directly applied, the amount of calculation in the motion control process is large, and the real-time performance of the motion control is poor; in the process of linearizing the torque, in the current motion control, the nonlinear amount in the torque is directly omitted, and only the linear amount is used for calculation, which will lead to a large error in the motion control of the robot, and there is a large error between the actual motion of the robot and the planned motion process, and the motion control accuracy is poor and the robustness is low.
[0039] Based on the above, a robot control method is proposed in the present application. The method is suitable for realizing robot motion control, especially real-time robot motion control, by further extracting the linear component in the nonlinear parameter quantity, so that the motion process of the robot can be flexibly and highly accurately controlled via an optimized linear centroid motion model. Figure 1 An exemplary flow chart of a robot control method 100 according to an embodiment of the present disclosure is shown.
[0040] Reference Figure 1 , first, in step S101, the current rotational motion information of the robot and the target rotational motion reference information of the robot are obtained.
[0041] The current rotational motion information refers to data information used to represent the rotational motion state of the robot at the current moment. In the present application, the current rotational motion information includes at least one of the current rotation angle of the robot and the current rotational angular velocity of the robot. However, it should be understood that, according to actual needs, the current rotational motion information may also include, for example, the current rotational angular acceleration of the robot.
[0042] The target rotational motion reference information refers to data information used to characterize the expected state of the robot's rotational motion at a target moment (which may be, for example, the next moment of the current moment, depending on actual needs). In the present application, the target rotational motion reference information includes at least one of: a robot target rotation reference angle, a robot target rotation reference angular velocity, and a robot target rotation reference angular acceleration. However, it should be understood that, depending on actual needs, the target rotational motion reference information may also include other parameter quantities, for example. The embodiments of the present disclosure are not limited by the specific composition of the target rotational motion expected information.
[0043] For example, the current rotational motion information of the robot can be obtained through torque sensors arranged at each joint of the robot and visual sensors arranged in the surrounding environment of the robot; and the target rotational motion reference information of the robot can be obtained based on processing the motion planning information generated by the motion planner of the robot.
[0044] However, it should be understood that the above only provides an exemplary method of obtaining the target rotational motion reference information and current rotational motion information of the robot, and the embodiments of the present disclosure are not limited to the specific method of obtaining the target rotational motion reference information and current rotational motion information.
[0045] Thereafter, in step S102 , according to the current rotational motion information of the robot, a target torque for controlling the target rotational motion of the robot is determined based on a linear function.
[0046] The linear function is, for example, a linear expression of the target torque. Specifically, for example, the rotational motion process of the robot can be approximated by exponential coordinates, thereby further expanding the nonlinear parameter quantity in the nonlinear expression of the existing torque parameter quantity, and extracting the optimized linear expression obtained by linear components therein.
[0047] After the target torque is obtained, in step S103, the target torque of each joint of the robot is determined based on the target torque, the current rotational motion information of the robot and the target rotational motion reference information of the robot.
[0048] Specifically, for example, the target contact force of the robot can be determined first according to the target torque, the current rotational motion information of the robot, and the target rotational motion reference information of the robot. Then, for example, the target torque of each joint of the robot can be determined based on the target contact force, the skeletal structure of the robot, and the posture information of the robot.
[0049] It should be understood that the above only provides an exemplary method for achieving target torque determination, and the embodiments of the present disclosure are not limited by the specific process and method of determining the target torque.
[0050] Based on the above, in the present application, on the basis of obtaining the current rotational motion information of the robot and the target rotational motion reference information of the robot, the target torque is obtained based on the current rotational motion information by applying the optimized linear function, and the target torque of each joint of the robot is determined according to the obtained target torque, so that the robot can be accurately controlled to perform the target motion process on the basis of achieving good real-time control of the robot, and the control accuracy and robustness of the robot are significantly improved on the basis of taking into account real-time performance. Specifically, by making the current rotational motion information include at least one of the current rotation angle of the robot and the current rotation angular velocity of the robot, when the linear function is applied to realize the calculation of the target torque, the target torque can be well characterized based on the current multi-dimensional and multi-level motion information, and the accuracy of the generated target torque can be improved. In addition, by making the target rotational motion reference information include at least one of the target rotation reference angle of the robot, the target rotation reference angular velocity of the robot, and the target rotation reference angular acceleration of the robot, the target rotational motion process can be described more comprehensively, which is conducive to the subsequent generation of the target torque of each joint of the robot based on the target rotational motion reference information, and the accuracy of the generated target torque can be improved.
[0051] In some embodiments, the linear function includes at least two of: a robot target rotation angular acceleration parameter, a robot angular velocity gain parameter, a robot angle gain parameter, and a robot current rotation angular velocity parameter.
[0052] Among them, the robot angular velocity gain parameter is the difference between the robot's target rotation angular velocity and the robot's current rotation angular velocity; the angle gain parameter is the difference between the robot's target rotation angle and the robot's current rotation angle, which will be described in more detail below in conjunction with specific embodiments.
[0053] By making the linear function include at least two of the robot's target rotation angular acceleration parameter, the robot's angular velocity gain parameter, the robot's angle gain parameter, and the robot's current rotation angular velocity parameter, compared to currently using only one item of the robot's target rotation angular acceleration parameter when calculating the target torque, by further expanding the linear function expression of the target torque, more linear components that can reflect the target torque are retained, so that while realizing linear calculation to improve the calculation speed and simplify the amount of calculation, the accuracy of the calculated target torque is significantly improved, which is conducive to the subsequent good motion control of the robot based on the target torque.
[0054] In some embodiments, the linear function includes: a robot target rotation angular acceleration parameter, a robot angular velocity gain parameter, a robot angle gain parameter, and a robot current rotation angular velocity parameter.
[0055] Among them, the robot angular velocity gain parameter is the difference between the robot's target rotation angular velocity and the robot's current rotation angular velocity; the angle gain parameter is the difference between the robot's target rotation angle and the robot's current rotation angle.
[0056] For example, the linear function will be described in more detail below in combination with the robot's center of mass dynamics equation and an example expression of the linear function.
[0057] For example, for the motion process of a robot, it is known that the center of mass dynamics equation of the robot is as follows:
[0058]
[0059]
[0060] Among them, equation 1a) represents the translational motion process of the robot at the current moment, and equation 1b) is the Euler equation of the robot, which is used to represent the rotational motion process of the robot at the current moment.
[0061] And among them, f i is the contact force received by the robot at the ith contact point at that moment, m is the mass of the robot’s center of mass, g is the acceleration of gravity, p is the position of the robot at that moment, is the speed of the robot at that moment, is the acceleration of the robot at that moment. And wherein i is a positive integer greater than 0 and less than N, and N is the total number of contact points of the robot. It should be understood that the contact point is, for example, the point of action where the robot contacts the surrounding environment, which is predetermined according to the actual structure of the robot and its mode of action. For example, when the robot is a quadruped robot and it interacts with the surrounding environment through the quadruped feet, the total number of contact points of the robot is, for example, 4, and each contact point is the part of the robot's foot that contacts the surrounding environment. And wherein, The position of the robot's i-th contact point at this moment is converted into a skew-symmetric matrix, is the position of the robot's center of mass at this moment converted into a skew-symmetric matrix, L is the robot's angular momentum, is the torque of the robot.
[0062] And wherein, the torque of the robot can be further expressed as:
[0063]
[0064] in, is the torque parameter of the robot, is the angular acceleration vector of the robot at the current moment, I is the moment of inertia of the robot, ω is the angular velocity of the robot, The angular velocity of the robot is converted into a skew-symmetric matrix for calculating the cross product of the angular velocity vector and the moment of inertia.
[0065] Since the rotational motion of the robot is nonlinear, the torque parameter expression is Item All terms are nonlinear. At this time, when seeking the target time t k When the target torque is , then for example, the target moment t k The formula for the lower moment parameter is as follows:
[0066]
[0067] The subscript k represents t k The parameter quantity at the moment, and the meaning of each parameter quantity is as described above. At this time, in the current motion control process, the torque parameter quantity is usually The term is approximately Among them, I0 is the moment of inertia value at the current time t0, is the target time t k The estimated value of the target angular acceleration in . The target time t is calculated by k The expression of the target torque under However, since the nonlinear quantity in the torque parameter is ignored, the accuracy of the calculated torque parameter is greatly reduced, which is not conducive to achieving good motion control.
[0068] Based on this, in this application, the rotational motion process of the robot is approximated by exponential coordinates. Specifically, the angular motion process (rotational motion process) during the current robot motion process can be expanded, for example, by the following formula:
[0069]
[0070] Among them, R k Representation at the target time t k The angular orientation information of the robot (i.e., its Euler angle information, such as roll angle, pitch angle, and yaw angle), R k-1 Characterized in t k-1 The robot’s angle orientation information at the moment, Characterized in t k-1 Time to t k The exponential coordinate of the rotation amount at time.
[0071] Based on the above, by setting the target time t k The angular orientation information under is approximated by the exponential coordinates to obtain formula 4), and then the expression is further Taylor expanded, the higher-order terms are ignored, and the linear components are retained. For example, the linear function in the present application can be obtained, which is, for example:
[0072]
[0073] The subscript k represents the target time t k The subscript 0 represents the parameters at the current time t0. And among them, is the target torque, I0 is the moment of inertia at the current time t0, is the robot's target rotation angular acceleration, P and Q are parameters selected according to actual needs, and Δω k is the robot angular velocity gain parameter, where Δω k =ω k -ω0,ω k is the robot's target rotation angular velocity, ω0 is the robot's current rotation angular velocity, Δθ k is the robot angle gain parameter, where Δθ k =θ k -θ0,θ k is the target rotation angle of the robot, θ0 is the current rotation angle of the robot, It refers to converting the current angular velocity of the robot into a skew-symmetric matrix for calculating the cross product of the angular velocity vector and the moment of inertia.
[0074] And among them, The term is the robot's target rotation angular acceleration parameter term, QΔω k The term is the robot angular velocity gain parameter term, PΔθ k That is the robot angle gain parameter item, The item is the current rotation angular velocity parameter item of the robot.
[0075] Based on the above, in the present application, by setting the linear function to include: robot target rotation angular acceleration parameter, robot angular velocity gain parameter, robot angle gain parameter, robot current rotation angular velocity parameter, so that compared with the prior art, only one robot target rotation angular acceleration parameter is included and the nonlinear term is directly omitted (as mentioned above By optimizing the linear function (via exponential coordinate approximation), more linear components in the nonlinear term are further extracted and retained, which is conducive to more accurate generation of the target torque, thereby being able to more comprehensively and accurately reflect the target motion state of the robot, improving the accuracy of motion control and taking into account the good real-time performance of motion control.
[0076] In some embodiments, the robot control method further includes: acquiring current translation motion information of the robot and target translation motion reference information of the robot.
[0077] The current translational motion information of the robot refers to data information used to characterize the motion state of the robot at the current moment. For example, the current translational motion information of the robot includes: the current motion position of the robot, the current motion speed of the robot, and the current motion acceleration of the robot.
[0078] The target translation motion reference information of the robot refers to data information used to characterize the expected state of the translation motion of the robot at a target moment (which can be, for example, the next moment of the current moment according to actual needs). According to actual needs, the target translation motion reference information of the robot includes: a target motion reference position of the robot, a target motion reference speed of the robot, and a target motion reference acceleration of the robot.
[0079] For example, the current translational motion information of the robot can be obtained by processing data from a motion sensor disposed inside the robot or a visual sensor disposed in the robot's surrounding environment; and the target translational motion reference information of the robot can be obtained by processing based on the motion planning information generated by the robot's motion planner. It should be understood that the embodiments of the present disclosure are not limited by the specific sources and methods of obtaining the target translational motion reference information and the current translational motion information.
[0080] Based on the above, in the present application, on the basis of obtaining the rotational motion information of the robot (including current rotational motion information and target rotational motion reference information), by further collecting and acquiring the translational motion information of the robot (including current translational motion information and target translational motion reference information), it is possible to obtain the current state of the robot's translational motion process and rotational motion process and the expected state at the target moment, which is conducive to further obtaining the contact force between the robot and the environment based on the robot's center of mass motion model and realizing its motion control.
[0081] In some embodiments, the above-mentioned step S103 may be described in more detail, for example. Figure 2 An exemplary flow chart of the process S103 of determining the target torque of each joint of the robot according to an embodiment of the present disclosure is shown.
[0082] Reference Figure 2 First, in step S1031, the target contact force of the robot is determined according to the target torque, the current motion information of the robot and the target motion reference information of the robot.
[0083] The current motion information of the robot includes: current rotational motion information and current translational motion information; the target motion reference information of the robot includes: target rotational motion reference information and target translational motion reference information. The specific meaning of each information is as described above and will not be repeated here.
[0084] The target contact force refers to the interaction force between the robot and the surrounding environment at the target motion moment when the robot is regarded as a rigid body. The target contact force is associated with the center of mass motion state of the robot at the target moment. Therefore, based on the determined target contact force, the torque of each joint can be further determined and the motion control of the robot can be achieved. It should be understood that, as described above in conjunction with formula 1b), the target contact force corresponds to the contact point between the robot and the surrounding environment.
[0085] For example, according to actual conditions, the target contact force may be a single contact force. For example, when the robot is a unicycle, the robot has only one contact point with the surrounding environment and has only one contact force. Alternatively, the target contact force may also be multiple contact forces. Specifically, for example, if the robot is a quadruped robot and the robot contacts the surrounding environment (here, for example, a tabletop) via the four feet, the robot has four contact points with the surrounding environment, and for each contact point, there is a contact force corresponding to the contact point. The embodiments of the present disclosure are not limited by the specific number of the target contact forces.
[0086] For example, in some embodiments, in the process of obtaining the target contact force, the target motion estimate of the robot can be generated based on the robot's center of mass motion model according to the robot's current motion information and the target torque, wherein the target motion estimate is a function of the robot's contact force; based on the target motion reference information and the target motion estimate, an error function between the target motion reference information and the target motion estimate is generated; and based on the error function, the target contact force is determined.
[0087] It should be understood that the above only provides a specific example of determining the target contact force, and the embodiments of the present disclosure are not limited by the specific method of determining the target contact force.
[0088] Thereafter, in step S1032, the target torque of each joint of the robot is determined based on the target contact force, the skeletal structure of the robot and the posture information of the robot.
[0089] For example, the torque of each joint of the robot can be generated directly based on the skeletal structure of the robot and the current posture information of the robot according to the target contact force. Alternatively, the target posture reference information (the target posture reference information refers to the data information used to characterize the expected state of the posture of each joint of the robot at the target moment) can be comprehensively considered on the basis of the above, and the target posture reference information, the current posture information, and the skeletal structure of the robot can be comprehensively considered to generate the target torque of each joint of the robot based on the target contact force using a preset algorithm.
[0090] Based on the above, in the present application, the target contact force of the robot is determined based on the target torque, the current motion information of the robot and the target motion reference information of the robot; then the target torque of each joint of the robot is determined based on the target contact force, the skeletal structure of the robot and the posture information of the robot, so that the target torque of each joint of the robot can be generated simply and efficiently based on the calculated target torque, which is conducive to realizing real-time and high-precision motion control of the robot.
[0091] In some embodiments, the above process S1031 of determining the target contact force of the robot according to the target torque, the current motion information of the robot and the target motion reference information of the robot can be described in more detail, for example. Figure 3 An exemplary flowchart of process S1031 of determining the target contact force of the robot according to an embodiment of the present disclosure is shown in FIG.
[0092] Reference Figure 3 First, in step S1031-1, according to the current motion information of the robot and the target torque, the target motion estimation of the robot is generated based on the robot center of mass motion model, wherein the target motion estimation is a function of the contact force of the robot.
[0093] The target motion estimation includes: at least a part of a target motion position estimation, a target motion speed estimation, a target motion acceleration estimation, a target rotation angle estimation, a target rotation angular velocity estimation, and a target rotation angular acceleration estimation.
[0094] For example, the center of mass dynamic equations of the robot are as shown in formula 1a) and formula 1b), then for example, for the target time t k In terms of the center of mass motion of the robot, the center of mass motion should satisfy the following center of mass motion equation:
[0095]
[0096] The subscript k in this formula represents the target time t k The parameters below, the meanings of the other parameters are as mentioned above, and will not be repeated here. By combining the two equations of this system, we can get the following center of mass motion model:
[0097]
[0098] in, is the target time t k The robot's acceleration is converted into a skew-symmetric matrix, Δp k is the position gain, Δp k =p k -p0, where p k is the target time t k The position of the robot at the current moment t0, p0 is the position of the robot at the current moment t0, To convert the gravitational acceleration into a skew-symmetric matrix. The meanings of other parameters are as described above.
[0099] If the expression of the target torque calculated by the linear function is as shown in formula 5), then for example, by substituting formula 5) into the motion expression 6) and processing it, the expression of the target rotational angular acceleration can be obtained:
[0100]
[0101] And among them, and is a constant term calculated based on the current motion information of the robot, it can be known that the target rotational angular acceleration is a function of the contact force f. Based on this, when the current motion information is substituted into the expression, it can be obtained for the target time t k The target rotation angular acceleration estimate of the robot
[0102] And through the relationship between the rotational angular acceleration, the rotation angle and the angular velocity, the target rotational angular velocity ω can be further obtained.k and the target rotation angle Δθ k The expression is:
[0103] ω k =A ω,k f+b ω,k 8)
[0104] Δθ k =A θ,k f+b θ,k 9)
[0105] And among them, A ω,k f, b ω,k , A θ,k f, b θ,k is also a constant term calculated based on the current motion information of the robot. The target rotation angular velocity and target rotation angle are also functions of the contact force f. Therefore, when the current motion information is substituted into the expression, the target time t can be obtained. k The target rotation angular velocity estimate of the robot And the target rotation angle estimate
[0106] Based on the same solution method, the target motion position estimation, target motion speed estimation, and target motion acceleration estimation can also be solved, which will not be repeated here.
[0107] Based on the above, the target motion estimation of the robot is generated based on the robot center of mass motion model and the target torque, and each estimated quantity in the target motion estimation is an expression of the contact force f, and except for the contact force f which is an unknown number, the remaining parameter quantities can be calculated through the current motion information.
[0108] Thereafter, in step S1031 - 2 , an error function of the target motion reference information and the target motion estimation is generated based on the target motion reference information and the target motion estimation.
[0109] For example, the error function e k For example, it can have the following form:
[0110]
[0111] in, Characterizing the target motion reference position and target motion position estimate The difference, Characterize the reference speed of target movement Target velocity estimation The difference, Characterize the target rotation reference angle Estimator of the rotation angle with the target The difference, Characterize the target rotation reference angular velocity Estimator of the target rotation angular velocity The difference.
[0112] After obtaining the error function, in step S1031 - 3 , the target contact force is determined based on the error function.
[0113] For example, the error function may be processed based on an optimization algorithm to determine the optimal contact force value and direction, and the target contact force may be determined based on the optimal contact force value and direction. Alternatively, the target contact force may be generated based on other methods. The embodiments of the present disclosure are not limited by the specific method of generating the target contact force.
[0114] Based on the above, in the present application, the target motion estimate of the robot is generated based on the robot's center of mass motion model according to the robot's current motion information and the target torque; thereafter, based on the target motion reference information and the target motion estimate, an error function of the target motion reference information and the target motion estimate is generated and the target contact force is determined, so that in the process of generating the target contact force, the robot's center of mass motion model, current and target motion states are comprehensively considered, which is conducive to generating high-precision target contact force, so that the robot can perform the desired motion well and improve the reliability and accuracy of motion control.
[0115] For example, in some embodiments, determining the target contact force based on the error function includes: optimizing the error function based on a quadratic optimization algorithm, and determining the contact force that minimizes the error function as the target contact force. By adjusting the contact force so that the error function minimizes (i.e., the target motion reference information and the target motion estimate have a minimum error value), the optimal contact force for achieving motion control of the robot is determined as the target contact force.
[0116] It should be understood that the secondary optimization algorithm may be selected based on actual needs, for example, and the embodiments of the present disclosure are not limited by the specific algorithm type of the secondary optimization algorithm used.
[0117] Based on the above, in this application, the optimal solution of the error function is solved by applying a quadratic optimization algorithm, and the contact force that makes the error function obtain the optimal solution is determined as the target contact force, so that the target contact force can be easily solved through the optimization process, and the obtained target contact force can enable the robot to perform the desired motion process well, thereby improving the flexibility and robustness of motion control.
[0118] In some embodiments, when there are multiple contact points between the robot and the surrounding environment, that is, when the robot has multiple contact forces, when there are multiple groups of contact force combinations that make the error function reach the minimum value, then at this time, each group of contact force combinations will be further weighted and summed to calculate the total contact force corresponding to the contact force combination, and the contact force combination that calculates the minimum total contact force will be determined as the target contact force combination.
[0119] When there are multiple contact points and thus multiple contact forces, if multiple groups of contact force combinations can achieve the optimal solution of the error function, then by further comparing the total contact force sizes of the multiple groups of contact force combinations, it is possible to take into account both the execution of the desired motion process and the minimization of the total contact force, which is beneficial to further optimize the robot's motion control process.
[0120] In some embodiments, the process S1032 of determining the target torque of each joint of the robot based on the target contact force, the skeletal structure of the robot, and the posture information of the robot can be described in more detail, for example. Figure 4 An exemplary flow chart of the process S1032 of determining the target torque of each joint of the robot according to an embodiment of the present disclosure is shown.
[0121] Reference Figure 4 First, in step S1032-1, based on the target contact force and the skeletal structure of the robot, the main torque of each joint of the robot is determined.
[0122] The main torque refers to the torque that each joint of the robot should have corresponding to the target contact force based on the target contact force and the skeletal structure of the robot. The torque can be used to achieve the target contact force of the robot, that is, to achieve the desired motion process.
[0123] Thereafter, in step S1032-2, the target posture reference information and current posture information of each joint of the robot are obtained.
[0124] The target posture reference information refers to data information used to characterize the desired posture of each joint of the robot at the target moment, which is, for example, the joint reference angle of each joint of the robot at the target moment.
[0125] The current posture information refers to data information used to characterize the posture of each joint of the robot at the current moment, which is, for example, the current angle of each joint of the robot at the current moment.
[0126] It should be understood that, for example, the motion planning information of the robot can be obtained, and the target posture reference information of the robot can be generated based on the motion planning information; and the current posture information of the robot can be obtained via torque sensors arranged at each joint of the robot.
[0127] However, the above only provides an exemplary method for obtaining the target posture reference information and current posture information of each joint of the robot. According to actual needs, other methods can also be selected to obtain the information. The embodiments of the present disclosure are not limited to the specific method of obtaining the target posture reference information and current posture information of each joint of the robot.
[0128] Thereafter, in step S1032-3, the additional torque amount of each joint of the robot is determined based on the target posture reference information of the robot and the current posture information of the robot.
[0129] The additional torque is used to adjust the main torque so that the robot can achieve the desired motion process while each joint of the robot can also have the desired joint posture.
[0130] For example, an additional torque can be generated based on the target posture information and the current posture information through a preset algorithm, or the target posture reference information, the current posture information of the robot, the skeletal structure information of the robot and the main torque of each joint of the robot can be input into a preset algorithm to comprehensively consider various information to generate an additional torque.
[0131] After obtaining the main torque and the additional torque, in step S1032-4, the target torque of each joint of the robot is determined based on the main torque and the additional torque.
[0132] For example, for each joint, the main torque and the additional torque can be superimposed to generate the target torque of the joint. This process is shown by the following formula:
[0133] τ m =τ m ff +τ m fb 10)
[0134] Among them, τ m is the target torque of the mth joint of the robot. And where τ m ff is the main torque of the mth joint, τ m fb is the attachment torque of the mth joint. And m is a positive integer greater than 0 and less than or equal to M, where M is the total number of joints of the robot.
[0135] Based on the above, by comprehensively considering the overall expected movement of the robot and the expected joint posture of each joint of the robot in the process of generating the target torque of each joint, the generated target torque of each joint can enable the robot to perform the expected movement well, and further enable each joint of the robot to be in the expected joint posture angle during the overall movement process, thereby achieving more comprehensive and precise motion control.
[0136] In some embodiments, obtaining the target posture reference information and current posture information of each joint of the robot includes: obtaining the motion planning information of the robot, and generating the target posture reference information of the robot based on the motion planning information, the target posture reference information including the joint target reference angle and the joint target reference angular velocity of each joint of the robot at the target moment; obtaining the current posture information of the robot, the current posture information including the joint current angle and the joint current angular velocity of each joint of the robot at the current moment.
[0137] The motion planning information is the global planning information of the overall motion of the robot, and may include, for example, the starting position and ending position of the robot, the starting angle and ending angle, the total time of the robot's motion process, and the average speed of the robot's motion process, etc. By processing the motion planning information, the expected motion information of the robot at each motion moment can be obtained, for example, the expected joint angle and expected joint angular velocity of each joint of the robot at each motion moment, that is, the target posture reference information.
[0138] Based on the above, by utilizing the motion planning information of the robot to generate target posture reference information, and further setting the target posture reference information to include the joint reference angles of each joint of the robot at the target moment, it is possible to comprehensively consider the global motion planning to determine the joint reference angles and joint target reference angular velocity of each joint, which is conducive to achieving accurate and real-time motion control at multiple levels such as global motion and local motion.
[0139] Next, the aforementioned robot control method will be described in more detail in combination with specific application scenarios. Figure 5 A schematic diagram of a robot control process according to an embodiment of the present disclosure is shown.
[0140] For example, when a quadruped robot is used to realize linear acceleration motion on a plane, the robot has, for example, four feet for contacting the ground, and accordingly it has four corresponding contact forces acting on the four feet respectively.
[0141] The robot is provided with a target planner, a motion controller and a detector, for example. The target planner is used to generate the motion planning information of the robot based on the user input and the preset information of the system. As mentioned above, it is, for example, the global motion planning of the robot, for example, including the global position planning information Global speed planning information Global acceleration planning information
[0142] The detector includes, for example, torque sensors arranged at each joint of the robot, speed sensors arranged on the robot, displacement sensors, and visual sensors arranged in the surrounding environment of the robot, etc., which are used to detect the current motion information of the robot (current rotational motion information and current translational motion information), such as the current position p0, the current speed The current rotation Euler angle data R0 (based on the Euler angle data, the angle data θ0 can be calculated), the current rotation angular velocity ω0, the current contact point data r between the robot and the environment I , where I is, for example, the total number of contact points of the robot, and the value of I in the quadruped robot is 4.
[0143] The motion controller obtains the motion planning information of the robot, and generates the target motion reference information (target rotation motion reference information and target translation motion reference information) of the robot based on the motion planning information through the target motion reference information generation module, for example, including: target motion reference position Target motion reference speed Target rotation reference angle Target rotation reference angular velocity wait.
[0144] The motion controller also obtains the current motion information from the detector via the current motion information acquisition module, and generates the target torque based on the current motion information and the linear function. Based on the target moment and the center of mass motion model of the robot, a target motion estimation of the robot is generated, wherein the estimated values of each target motion information are functions of the contact force. Subsequently, an error function is generated based on the target motion estimation and the aforementioned target motion reference information, and the target contact force is determined by solving the optimal solution of the error function using a quadratic optimization algorithm.
[0145] In addition, the motion controller also receives target posture reference information from the target planner, which includes, for example, the joint target reference angles of each joint of the robot at the target time. and joint target reference angular velocity Where M is the total number of joints of the robot. The motion controller also receives current posture information from the detector, which includes the current joint angle q of each joint of the robot at the current moment. 0,M And the current joint angular velocity
[0146] Finally, based on the target contact force, the target posture reference information, the current posture information, and taking into account the skeletal structure of the robot in the torque generation module, the motion controller will generate a target torque for controlling each joint of the robot to perform the desired motion.
[0147] Based on the above, in the present application, on the basis of obtaining the current rotational motion information of the robot and the target rotational motion reference information of the robot, the target torque is obtained based on the current rotational motion information by applying an optimized linear function, and the target torque of each joint of the robot is determined according to the obtained target torque, thereby enabling the robot to be accurately controlled to execute the target motion process on the basis of achieving good real-time control of the robot, and significantly improving the control accuracy and robustness of the robot while taking into account real-time performance.
[0148] According to another aspect of the present disclosure, a robot control system is provided, wherein the robot comprises a front handle and a front handle controller, wherein the front handle controller provides a steering torque for the front handle. Figure 6 An exemplary block diagram of a robot control system 600 according to an embodiment of the present invention is shown.
[0149] like Figure 6 The robot control system 600 shown includes a rotational motion information acquisition module 610 , a target torque generation module 620 and a joint torque generation module 630 .
[0150] The rotational motion information acquisition module 610 is configured to execute Figure 1 In the process of step S101, the current rotational motion information of the robot and the target rotational motion reference information of the robot are obtained.
[0151] The current rotational motion information refers to data information used to represent the rotational motion state of the robot at the current moment. In the present application, the current rotational motion information includes at least one of the current rotation angle of the robot and the current rotational angular velocity of the robot. However, it should be understood that, according to actual needs, the current rotational motion information may also include, for example, the current rotational angular acceleration of the robot.
[0152] The target rotational motion reference information refers to data information used to characterize the expected state of the robot's rotational motion at a target moment (which may be, for example, the next moment of the current moment, depending on actual needs). In the present application, the target rotational motion reference information includes at least one of: a robot target rotation reference angle, a robot target rotation reference angular velocity, and a robot target rotation reference angular acceleration. However, it should be understood that, depending on actual needs, the target rotational motion reference information may also include other parameter quantities, for example. The embodiments of the present disclosure are not limited by the specific composition of the target rotational motion expected information.
[0153] For example, the current rotational motion information of the robot can be obtained through torque sensors disposed at each joint of the robot and visual sensors disposed in the surrounding environment of the robot; and the target rotational motion reference information of the robot is obtained by processing the motion planning information generated by the motion planner of the robot. The embodiments of the present disclosure are not limited by the specific method of obtaining the target rotational motion reference information and the current rotational motion information.
[0154] The target torque generation module 620 is configured to perform Figure 1 In the process of step S102, according to the current rotational motion information of the robot, a target torque for controlling the target rotational motion of the robot is determined based on a linear function.
[0155] The linear function is, for example, a linear expression of the target torque. Specifically, for example, the rotational motion process of the robot can be approximated by exponential coordinates, thereby further expanding the nonlinear parameter quantity in the nonlinear expression of the existing torque parameter quantity, and extracting the optimized linear expression obtained by linear components therein.
[0156] The joint torque generation module 630 is configured to perform Figure 1 In the process of step S103, the target torque of each joint of the robot is determined based on the target torque, the current rotational motion information of the robot and the target rotational motion reference information of the robot.
[0157] Based on the above, in the present application, on the basis of obtaining the current rotational motion information of the robot and the target rotational motion reference information of the robot, the target torque is obtained based on the current rotational motion information by applying the optimized linear function, and the target torque of each joint of the robot is determined according to the obtained target torque, so that the robot can be accurately controlled to perform the target motion process on the basis of achieving good real-time control of the robot, and the control accuracy and robustness of the robot are significantly improved on the basis of taking into account real-time performance. Specifically, by making the current rotational motion information include at least one of the current rotation angle of the robot and the current rotation angular velocity of the robot, when the linear function is applied to realize the calculation of the target torque, the target torque can be well characterized based on the current multi-dimensional and multi-level motion information, and the accuracy of the generated target torque can be improved. In addition, by making the target rotational motion reference information include at least one of the target rotation reference angle of the robot, the target rotation reference angular velocity of the robot, and the target rotation reference angular acceleration of the robot, the target rotational motion process can be described more comprehensively, which is conducive to the subsequent generation of the target torque of each joint of the robot based on the target rotational motion reference information, and the accuracy of the generated target torque can be improved.
[0158] In some embodiments, the linear function includes at least two of: a robot target rotation angular acceleration parameter, a robot angular velocity gain parameter, a robot angle gain parameter, and a robot current rotation angular velocity parameter.
[0159] Among them, the robot angular velocity gain parameter is the difference between the robot's target rotation angular velocity and the robot's current rotation angular velocity; the angle gain parameter is the difference between the robot's target rotation angle and the robot's current rotation angle, which will be described in more detail below in conjunction with specific embodiments.
[0160] By making the linear function include at least two of the robot's target rotation angular acceleration parameter, the robot's angular velocity gain parameter, the robot's angle gain parameter, and the robot's current rotation angular velocity parameter, compared to currently using only one item of the robot's target rotation angular acceleration parameter when calculating the target torque, by further expanding the linear function expression of the target torque, more linear components that can reflect the target torque are retained, so that while realizing linear calculation to improve the calculation speed and simplify the amount of calculation, the accuracy of the calculated target torque is significantly improved, which is conducive to the subsequent good motion control of the robot based on the target torque.
[0161] In some embodiments, the linear function includes: a robot target rotation angular acceleration parameter, a robot angular velocity gain parameter, a robot angle gain parameter, and a robot current rotation angular velocity parameter.
[0162] Among them, the robot angular velocity gain parameter is the difference between the robot's target rotation angular velocity and the robot's current rotation angular velocity; the angle gain parameter is the difference between the robot's target rotation angle and the robot's current rotation angle.
[0163] Based on the above, in the present application, by setting the linear function to include: robot target rotation angular acceleration parameter, robot angular velocity gain parameter, robot angle gain parameter, robot current rotation angular velocity parameter, so that compared with the prior art, only one robot target rotation angular acceleration parameter is included and the nonlinear term is directly omitted (as mentioned above By optimizing the linear function (via exponential coordinate approximation), more linear components in the nonlinear term are further extracted and retained, which is conducive to more accurate generation of the target torque, thereby being able to more comprehensively and accurately reflect the target motion state of the robot, improving the accuracy of motion control and taking into account the good real-time performance of motion control.
[0164] In some embodiments, the robot control system is capable of executing the method described above and has the functions described above.
[0165] According to another aspect of the present disclosure, a robot is provided, wherein the robot has a control system as described above, and can execute the robot control method as described above, and realize the robot control function as described above.
[0166] In addition, the robot may also include a bus, a memory, a sensor component, a controller, a communication module, an input and output device, etc.
[0167] The bus may be a circuit that interconnects the components of the robot and passes communication information (eg, control messages or data) among the components.
[0168] The sensor component can be used to perceive the physical world, such as a camera, an infrared sensor, an ultrasonic sensor, etc. In addition, the sensor component can also include a device for measuring the current operation and motion state of the robot, such as a Hall sensor, a laser position sensor, or a strain sensor.
[0169] The controller is used to control the operation of the robot, for example, in an artificial intelligence control manner.
[0170] The controller includes, for example, a processing device. The processing device may include a microprocessor, a digital signal processor ("DSP"), an application specific integrated circuit ("ASIC"), a field programmable gate array, a state machine, or other processing device for processing electrical signals received from the sensor line. Such a processing device may include a programmable electronic device, such as a PLC, a programmable interrupt controller ("PIC"), a programmable logic device ("PLD"), a programmable read-only memory ("PROM"), an electronic programmable read-only memory, etc.
[0171] The communication module can be connected to a network, for example, by wire or invalid, so as to communicate with the physical world (e.g., a server). The communication module can be wireless and can include a wireless interface, such as an IEEE 802.11, Bluetooth, wireless local area network ("WLAN") transceiver, or a radio interface for accessing a cellular telephone network (e.g., a transceiver / antenna for accessing a CDMA, GSM, UMTS or other mobile communication network). In another example, the communication module can be wired and can include an interface such as Ethernet, USB or IEEE 1394.
[0172] The input-output device may transmit commands or data input, for example, from a user or any other external device to one or more other components of the robot, or may output commands or data received from one or more other components of the robot to the user or other external device.
[0173] A plurality of robots may form a robot system to collaboratively complete a task, the plurality of robots being communicatively connected to a server and receiving collaborative robot instructions from the server.
[0174] According to another aspect of the present invention, a non-volatile computer-readable storage medium is provided, on which computer-readable instructions are stored. When the instructions are executed by a computer, the above-mentioned method can be executed.
[0175] The program part of the technology can be considered as a "product" or "manufactured product" in the form of executable code and / or related data, which is participated in or realized by computer-readable media. Tangible and permanent storage media can include any memory or storage used by any computer, processor, or similar device or related module. For example, various semiconductor memories, tape drives, disk drives or any similar devices that can provide storage functions for software.
[0176] All software or parts of it may sometimes be communicated over a network, such as the Internet or other communications network. Such communications can load software from one computer device or processor to another. Therefore, another medium capable of transmitting software elements may also be used as a physical connection between local devices, such as light waves, radio waves, electromagnetic waves, etc., transmitted through cables, optical cables or air. Physical media used to carry carriers, such as cables, wireless connections or optical cables and the like, can also be considered as media that carry software. As used herein, unless limited to tangible "storage" media, other terms referring to computer or machine "readable media" refer to media that participate in the process of executing any instructions by the processor.
[0177] The present application uses specific words to describe the embodiments of the present application. For example, "first / second embodiment", "one embodiment", and / or "some embodiments" refer to a certain feature, structure or characteristic related to at least one embodiment of the present application. Therefore, it should be emphasized and noted that "one embodiment" or "an embodiment" or "an alternative embodiment" mentioned twice or multiple times in different positions in this specification does not necessarily refer to the same embodiment. In addition, some features, structures or characteristics in one or more embodiments of the present application can be appropriately combined.
[0178] In addition, it will be appreciated by those skilled in the art that various aspects of the present application may be illustrated and described by a number of patentable categories or situations, including any new and useful process, machine, product or combination of substances, or any new and useful improvements thereto. Accordingly, various aspects of the present application may be performed entirely by hardware, entirely by software (including firmware, resident software, microcode, etc.), or by a combination of hardware and software. The above hardware or software may all be referred to as "data blocks", "modules", "engines", "units", "components" or "systems". In addition, various aspects of the present application may be represented as a computer product located in one or more computer-readable media, which includes computer-readable program code.
[0179] Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the invention belongs. It should also be understood that terms such as those defined in common dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the relevant technology and should not be interpreted in an idealized or extremely formal sense, unless explicitly defined as such herein.
[0180] The above is an explanation of the present invention and should not be considered as a limitation thereof. Although several exemplary embodiments of the present invention have been described, it will be readily appreciated by those skilled in the art that many modifications may be made to the exemplary embodiments without departing from the novel teachings and advantages of the present invention. Therefore, all such modifications are intended to be included within the scope of the present invention as defined in the claims. It should be understood that the above is an explanation of the present invention and should not be considered as being limited to the specific embodiments disclosed, and modifications to the disclosed embodiments and other embodiments are intended to be included within the scope of the appended claims. The present invention is defined by the claims and their equivalents.
Claims
1. A robot control method, the method comprising: Obtaining the current rotational motion information of the robot and the target rotational motion reference information of the robot; According to the current rotational motion information of the robot, a nonlinear amount in an ideal torque for controlling the target rotational motion of the robot is approximated by a linear function to obtain a target torque for controlling the target rotational motion of the robot; Determining a target torque of each joint of the robot based on the target torque, the current rotational motion information of the robot and the target rotational motion reference information of the robot; Among them, the current rotational motion information includes: at least one of the robot's current rotation angle and the robot's current rotation angular velocity; the target rotational motion reference information includes: at least one of the robot's target rotation reference angle, the robot's target rotation reference angular velocity, and the robot's target rotation reference angular acceleration.
2. The robot control method according to claim 1, wherein: The linear function includes at least two of the robot target rotation angular acceleration parameter, the robot angular velocity gain parameter, the robot angle gain parameter, and the robot current rotation angular velocity parameter. Among them, the robot angular velocity gain parameter is the difference between the robot's target rotation angular velocity and the robot's current rotation angular velocity; the angle gain parameter is the difference between the robot's target rotation angle and the robot's current rotation angle.
3. The robot control method according to claim 1 or 2, wherein: The linear function includes: a robot target rotation angular acceleration parameter, a robot angular velocity gain parameter, a robot angle gain parameter, and a robot current rotation angular velocity parameter.
4. The robot control method according to claim 1, further comprising: Obtaining the current translational motion information of the robot and the target translational motion reference information of the robot; And wherein, the target translation motion reference information of the robot includes: the target motion reference position of the robot, the target motion reference speed of the robot, and the target motion reference acceleration of the robot; the current translation motion information of the robot includes: the current motion position of the robot, the current motion speed of the robot, and the current motion acceleration of the robot.
5. The robot control method according to claim 3, wherein the linear function is: And among them, is the target time t k The target torque is I0, the moment of inertia at the current time t0, is the robot's target rotation angular acceleration, P and Q are parameters selected according to actual needs, and Δω k is the robot angular velocity gain parameter, where Δω k =ω k -ω0,ω k is the robot's target rotation angular velocity, ω0 is the robot's current rotation angular velocity, Δθ k is the robot angle gain parameter, where Δθ k =θ k -θ0,θ k is the target rotation angle of the robot, θ0 is the current rotation angle of the robot, It means converting the current angular velocity of the robot into a skew-symmetric matrix.
6. The robot control method according to claim 4, wherein: Based on the target torque, the current rotational motion information of the robot and the target rotational motion reference information of the robot, determining the target torque of each joint of the robot includes: Determining a target contact force of the robot according to the target torque, current motion information of the robot and target motion reference information of the robot; Determining target torques of various joints of the robot based on the target contact force, the skeletal structure of the robot, and posture information of the robot; The current motion information of the robot includes: current rotational motion information and current translational motion information; the target motion reference information of the robot includes: target rotational motion reference information and target translational motion reference information.
7. The robot control method according to claim 6, wherein: Determining the target contact force of the robot according to the target torque, the current motion information of the robot and the target motion reference information of the robot includes: According to the current motion information of the robot and the target torque, a target motion estimate of the robot is generated based on a robot center of mass motion model, wherein the target motion estimate is a function of a contact force of the robot; Based on the target motion reference information and the target motion estimate, generating an error function between the target motion reference information and the target motion estimate; Based on the error function, a target contact force is determined; The target motion estimation includes at least a part of a target motion position estimation, a target motion speed estimation, a target motion acceleration estimation, a target rotation angle estimation, a target rotation angular velocity estimation, and a target rotation angular acceleration estimation.
8. The robot control method according to claim 7, wherein: Determining the target contact force based on the error function includes: optimizing the error function based on a quadratic optimization algorithm, and determining the contact force that minimizes the error function as the target contact force.
9. The robot control method according to claim 6, wherein: Based on the target contact force, the skeletal structure of the robot and the posture information of the robot, determining the target torque of each joint of the robot includes: Determine the main torque of each joint of the robot based on the target contact force and the skeletal structure of the robot; Obtain target posture reference information and current posture information of each joint of the robot; Determining the additional torque of each joint of the robot based on the target posture reference information of the robot and the current posture information of the robot; Based on the main torque and the additional torque, the target torque of each joint of the robot is determined.
10. The robot control method according to claim 9, wherein: Obtaining the target posture reference information and current posture information of each joint of the robot includes: Acquire motion planning information of the robot, and generate target posture reference information of the robot based on the motion planning information, wherein the target posture reference information includes joint target reference angles and joint target reference angular velocities of each joint of the robot; The current posture information of the robot is obtained, wherein the current posture information includes the current angle of each joint and the current angular velocity of each joint.
11. A robot control system, the system comprising: A rotational motion information acquisition module, configured to acquire current rotational motion information of the robot and target rotational motion reference information of the robot; a target torque generating module, which is configured to approximate the nonlinear amount in the ideal torque for controlling the target rotational motion of the robot using a linear function according to the current rotational motion information of the robot, so as to obtain the target torque for controlling the target rotational motion of the robot; a joint torque generation module, configured to determine a target torque of each joint of the robot based on the target torque, current rotational motion information of the robot and target rotational motion reference information of the robot; Among them, the current rotational motion information includes: at least one of the robot's current rotation angle and the robot's current rotation angular velocity; the target rotational motion reference information includes: at least one of the robot's target rotation reference angle, the robot's target rotation reference angular velocity, and the robot's target rotation reference angular acceleration.
12. The robot control system according to claim 11, wherein: The linear function includes at least two of the robot target rotation angular acceleration parameter, the robot angular velocity gain parameter, the robot angle gain parameter, and the robot current rotation angular velocity parameter. Among them, the robot angular velocity gain parameter is the difference between the robot's target rotation angular velocity and the robot's current rotation angular velocity; the angle gain parameter is the difference between the robot's target rotation angle and the robot's current rotation angle.
13. The robot control system according to claim 11, wherein: The linear function includes: robot target rotation angular acceleration parameter, robot angular velocity gain parameter, robot angle gain parameter, robot current rotation angular velocity parameter; Among them, the robot angular velocity gain parameter is the difference between the robot's target rotation angular velocity and the robot's current rotation angular velocity; the angle gain parameter is the difference between the robot's target rotation angle and the robot's current rotation angle.
14. A robot, comprising the robot control system according to any one of claims 11 to 13, and realizing motion control of the robot through the robot control method according to any one of claims 1 to 10.
15. A computer-readable storage medium, characterized in that: Computer-readable instructions are stored thereon, and when the instructions are executed by a computer, the method described in any one of claims 1 to 10 is performed.
