Robot control method based on hybrid integral discrete neural network under nonlinear constraints

Abstract

The present application discloses a robot control method of a hybrid integral discrete neural network under nonlinear constraints, which relates to the field of robot control technology, including: constructing a robot control problem; the robot control problem includes an objective function and a constraint equation; constructing a deviation equation of the robot control problem according to Lagrange multipliers and nonlinear functions; constructing a hybrid integral neural network solver based on the deviation equation; converting the hybrid integral neural network solver into a matrix form under discrete time as a discrete time model; inputting the expected target of the robot control problem into the discrete time model, and solving to obtain a joint angle target solution; sending the joint angle target solution to the robot controller, and using the controller to control the robot motion according to the joint angle target solution. Combining the learning ability of the neural network and the solving characteristics of the integral equation, the robot's joint angle can be planned under discrete time and then the robot motion can be driven, which can realize real-time control and rapid response of the robot.

Description

Technical Field

[0001] The present application relates to the field of robot control technology, and in particular to a robot control method of a mixed integral discrete neural network under nonlinear constraints. Background Art

[0002] Robot control involves a variety of nonlinear problems and different dynamic environments, including obstacles in the workspace, motion constraints, and task requirements. In many practical applications, most robot control systems are based on digital controllers, that is, the controller itself is a discrete-time system, which makes continuous-time solvers inefficient in practical applications and may cause numerical errors and response delays.

[0003] Most existing solutions design solvers in continuous time, which usually involve complex nonlinear optimization problems, resulting in heavy computational load and poor real-time performance, especially in real-time control tasks, which is not conducive to actual operations. Summary of the invention

[0004] The main purpose of the embodiments of the present application is to propose a robot control scheme of a mixed integral discrete neural network under nonlinear constraints to improve the accuracy and computational efficiency of the robot control tasks and enhance the stability and adaptability of the robot control tasks.

[0005] To achieve the above object, an embodiment of the present application provides a robot control method of a hybrid integral discrete neural network under nonlinear constraints, the method comprising the following steps:

[0006] Constructing a robot control problem; wherein the robot control problem includes an objective function and a constraint equation;

[0007] Constructing the deviation equation of the robot control problem according to Lagrange multipliers and nonlinear functions;

[0008] Constructing a hybrid integral neural network solver based on the deviation equation;

[0009] Converting the hybrid integral neural network solver into a matrix form under discrete time as a discrete time model;

[0010] Inputting the expected target of the robot control problem into the discrete time model to obtain the joint angle target solution;

[0011] The joint angle target solution is sent to a controller of the robot, and the controller is used to control the movement of the robot according to the joint angle target solution.

[0012] In some embodiments, constructing a robot control problem comprises the following steps:

[0013] The objective function for constructing the robot control problem is:

[0014] ;

[0015] in, , Indicates that the robot is The angular velocity of each joint at the moment, is the robot's degrees of freedom;

[0016] The constraint equations for constructing the robot control problem are:

[0017] ;

[0018] in, is the equality constraint equation, is the inequality constraint equation.

[0019] In some embodiments, constructing the deviation equation of the robot control problem according to the Lagrange multiplier and the nonlinear function comprises the following steps:

[0020] The deviation equation of the robot control problem constructed according to Lagrange multipliers and nonlinear functions is:

[0021] ;

[0022] in, is the deviation equation, ; is the solution to the robot control problem, ; , are the Lagrange multipliers of the equality constraint equation and the inequality constraint equation respectively; is the nonlinear functional relationship of the inequality constraint equation.

[0023] In some embodiments, constructing a hybrid integral neural network solver based on the deviation equation comprises the following steps:

[0024] Construct the expression of hybrid integral neural network;

[0025] The expression of the hybrid integral neural network is:

[0026] ;

[0027] in, is the expression of the hybrid integral neural network; is a constant; is an integral variable, the integral lower limit of the integral variable is 0, the first integral upper limit is t, and the second integral upper limit is ;

[0028] Constructing the hybrid integral neural network solver according to the deviation equation and the expression of the hybrid integral neural network;

[0029] The hybrid integral neural network solver is:

[0030] ;

[0031] in, The hybrid integral neural network solver; is the interference item;

[0032] Specifically:

[0033] ;

[0034] in, , is the functional expression of the nonlinear functional relationship, is the dimension of the equality constraint equation, is the dimension of the inequality constraint equation;

[0035] Specifically:

[0036] ;

[0037] Specifically:

[0038] ;

[0039] in, represents a diagonal matrix.

[0040] In some embodiments, converting the hybrid integral neural network solver into a matrix form under discrete time as a discrete time model comprises the following steps:

[0041] Converting the hybrid integral neural network solver into a matrix form;

[0042] The hybrid integral neural network solver in matrix form is:

[0043] ;

[0044] in, is the integral term of the deviation equation, ; is the double integral term of the deviation equation, ; is an intermediate parameter used to simplify the expression. ); , for The inverse matrix of

[0045] Construct discrete-time expressions;

[0046] The discrete time expression is:

[0047] ;

[0048] in, is the derivative of the discrete time series, is the discrete time series; is the sampling period;

[0049] According to the discrete time expression, the hybrid integral neural network solver in matrix form is converted into a matrix form under discrete time as the discrete time model;

[0050] The discrete time model is:

[0051] ;

[0052] in, for The value at time step k, for The value at time step k, for The value at time step k, for The value at time step k, for The value at time step k, is the time value corresponding to the kth time step, is the value of the interference term at time step k; k is an integer, representing the index of the discrete time step, and the value range of k is 0 to 6;

[0053] ), .

[0054] In some embodiments, the step of inputting the expected target of the robot control problem into the discrete time model and solving the model to obtain the joint angle target solution comprises the following steps:

[0055] Inputting the vector of expected targets into the discrete-time model;

[0056] The discrete-time model is used to solve the deviation equation based on the vector of the expected target, and the solution when the deviation equation is 0 is used as the joint angle target solution.

[0057] In some embodiments, the method further comprises the following steps:

[0058] Obtaining the current position of the end effector of the robot;

[0059] Calculating a deviation value between a current position of the end effector and a corresponding position of the current joint angle target solution;

[0060] If the deviation value is greater than a preset threshold, returning to input the expected target of the robot control problem into the discrete time model to solve and obtain the latest joint angle target solution;

[0061] Based on the latest target joint angle solution, the step of sending the target joint angle solution to a controller of the robot is performed, and the controller is used to control the movement of the robot according to the target joint angle solution.

[0062] To achieve the above purpose, another aspect of the embodiment of the present application provides a robot control device of a hybrid integral discrete neural network under nonlinear constraints, the device comprising:

[0063] A control problem construction unit, used to construct a robot control problem; wherein the robot control problem includes an objective function and a constraint equation;

[0064] A deviation equation construction unit, used for constructing the deviation equation of the robot control problem according to Lagrange multipliers and nonlinear functions;

[0065] A solver construction unit, used for constructing a hybrid integral neural network solver based on the deviation equation;

[0066] A solver conversion unit, used for converting the hybrid integral neural network solver into a matrix form under discrete time as a discrete time model;

[0067] A joint angle solving unit, used for inputting the expected target of the robot control problem into the discrete time model to solve and obtain the joint angle target solution;

[0068] A robot control unit is used to send the joint angle target solution to a controller of the robot, and use the controller to control the movement of the robot according to the joint angle target solution.

[0069] To achieve the above objective, another aspect of an embodiment of the present application provides an electronic device, the electronic device comprising a memory and a processor, the memory storing a computer program, and the processor implementing the above method when executing the computer program.

[0070] To achieve the above objective, another aspect of an embodiment of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and the computer program implements the above method when executed by a processor.

[0071] The embodiments of the present application include at least the following beneficial effects:

[0072] The present application can construct a robot control problem; wherein the robot control problem includes an objective function and a constraint equation; construct a deviation equation of the robot control problem based on Lagrange multipliers and nonlinear functions; construct a hybrid integral neural network solver based on the deviation equation; convert the hybrid integral neural network solver into a matrix form under discrete time as a discrete time model; input the expected goal of the robot control problem into the discrete time model, and solve to obtain the joint angle target solution; send the joint angle target solution to the robot controller, and use the controller to control the robot motion according to the joint angle target solution. By combining the learning ability of the neural network and the solving characteristics of the integral equation, the joint angle of the robot can be planned in discrete time and then the robot motion can be driven, which can realize the real-time control and rapid response of the robot, and adapt to changes in the dynamic environment, while improving the robustness, stability and adaptability. BRIEF DESCRIPTION OF THE DRAWINGS

[0073] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0074] Figure 1 A schematic flow chart of a robot control method of a hybrid integral discrete neural network under nonlinear constraints provided in an embodiment of the present application;

[0075] Figure 2 A schematic diagram of the structure of a robot control device of a mixed integral discrete neural network under nonlinear constraints provided in an embodiment of the present application;

[0076] Figure 3 A schematic diagram of the hardware structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0077] 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 devices and methods consistent with some aspects of the embodiments of the present application as detailed in the attached claims.

[0078] 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".

[0079] 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.

[0080] 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.

[0081] Reference Figure 1 , the embodiment of the present application provides a robot control method of a hybrid integral discrete neural network under nonlinear constraints, the method may include but is not limited to including S100 to S150, specifically as follows:

[0082] S100: Constructing a robot control problem; wherein the robot control problem includes an objective function and a constraint equation.

[0083] Specifically, this embodiment can construct a corresponding control problem according to the actual control task; wherein the objective function can be used to quantify the design goal, such as various parameters required to control the robot.

[0084] Furthermore, S100 may include:

[0085] The objective function for constructing the robot control problem is:

[0086] ;

[0087] in, , Indicates that the robot is The angular velocity of each joint at the moment, is the robot's degrees of freedom;

[0088] The constraint equations for constructing the robot control problem are:

[0089] ;

[0090] in, is the equality constraint equation, is the inequality constraint equation.

[0091] S110: Constructing a deviation equation for the robot control problem according to Lagrange multipliers and nonlinear functions.

[0092] Specifically, the Lagrange multiplier can be used to find the extreme value of the objective function under the constraint equation, that is, it can be used to solve the optimal value of the control robot parameters.

[0093] Furthermore, S111 may include:

[0094] The deviation equation of the robot control problem constructed according to Lagrange multipliers and nonlinear functions is:

[0095] ;

[0096] in, is the deviation equation, ; is the solution to the robot control problem, ; , are the Lagrange multipliers of the equality constraint equation and the inequality constraint equation respectively; is the nonlinear functional relationship of the inequality constraint equation.

[0097] S120: Constructing a hybrid integral neural network solver based on the deviation equation.

[0098] It can be understood that, in this embodiment, a solver for the robot control task can be constructed based on the deviation equation and combined with a hybrid integral neural network, and the solver can be a mathematical model.

[0099] Further, S120 may include:

[0100] Construct the expression of hybrid integral neural network;

[0101] The expression of the hybrid integral neural network is:

[0102] ;

[0103] in, is the expression of the hybrid integral neural network; is a constant. In some optional embodiments, Can be set to 1; is an integral variable, the integral lower limit of the integral variable is 0, the first integral upper limit is t, and the second integral upper limit is ;

[0104] Constructing the hybrid integral neural network solver according to the deviation equation and the expression of the hybrid integral neural network;

[0105] The hybrid integral neural network solver is:

[0106] ;

[0107] in, The hybrid integral neural network solver; is the interference term; it should be noted that the interference term can be replaced by a constant term, a linear term, a quadratic term, etc.

[0108] Specifically:

[0109] ;

[0110] in, , is the functional expression of the nonlinear functional relationship, is the dimension of the equality constraint equation, is the dimension of the inequality constraint equation;

[0111] Specifically:

[0112] ;

[0113] Specifically:

[0114] ;

[0115] in, represents a diagonal matrix.

[0116] Since there are physical limitations on the joints when controlling the robot, it is necessary to consider the influence of multiple constraints and external interferences in the control to adapt to the control tasks under discrete time states, thereby improving the robustness, stability and adaptability of the control.

[0117] S130: Convert the hybrid integral neural network solver into a matrix form under discrete time as a discrete time model.

[0118] Planning in discrete time can achieve real-time control and rapid response, and adapt to changes in dynamic environments. This embodiment can be used in the fields of industrial robots, service robots, etc. The hybrid integral neural network combines the learning ability of the neural network and the solving characteristics of the integral equation, which enables the solver of this embodiment to better handle complex nonlinear problems and improve the robustness and accuracy of robot control.

[0119] Further, S130 may include:

[0120] Converting the hybrid integral neural network solver into a matrix form;

[0121] The hybrid integral neural network solver in matrix form is:

[0122] ;

[0123] in, is the integral term of the deviation equation, ; is the double integral term of the deviation equation, ; is an intermediate parameter used to simplify the expression. ); , for The inverse matrix of

[0124] Construct discrete-time expressions;

[0125] The discrete time expression is:

[0126] ;

[0127] in, is the derivative of the discrete time series, is the discrete time series; is the sampling period;

[0128] According to the discrete time expression, the hybrid integral neural network solver in matrix form is converted into a matrix form under discrete time as the discrete time model;

[0129] The discrete time model is:

[0130] ;

[0131] in, for The value at time step k, for The value at time step k, for The value at time step k, for The value at time step k, for The value at time step k, is the time value corresponding to the kth time step, is the value of the interference term at time step k; it should be noted that k is an integer, representing the index of the discrete time step, and optionally, the value range of k is 0 to 6, which is used to mark the discrete sequence of time, similar to the time variable t in the continuous time system;

[0132] ), .

[0133] S140: Inputting the expected target of the robot control problem into the discrete time model to obtain the joint angle target solution.

[0134] Specifically, this embodiment inputs an expected target, which may be the angle of each joint of the robot, and then solves a solution matching the expected target based on a discrete-time model, and uses the solved joint angle as the joint angle target solution.

[0135] Furthermore, S140 may include:

[0136] Inputting the vector of expected targets into the discrete-time model;

[0137] The discrete-time model is used to solve the deviation equation based on the vector of the expected target, and the solution when the deviation equation is 0 is used as the joint angle target solution.

[0138] For example, the expected target vector of the robot control task is Input discrete time model, discrete time model solves the deviation function The joint angle at this time is taken as the joint target solution .

[0139] S150: Sending the joint angle target solution to a controller of the robot, and utilizing the controller to control the movement of the robot according to the joint angle target solution.

[0140] Specifically, in this embodiment, the obtained joint angle target solution can be sent to the controller of the robot through the network, and then the controller of the robot uses the joint angle target solution as a control reference value to control the movement of the robot.

[0141] In some embodiments, the robot of this embodiment may be a multi-axis robotic arm, such as a three-axis, five-axis, or seven-axis robotic arm.

[0142] In order to further improve the control accuracy of the robot, the embodiment of the present application may further include S161~S164:

[0143] S161: Acquire the current position of the end effector of the robot;

[0144] S162: Calculating a deviation value between the current position of the end effector and the corresponding position of the current joint angle target solution;

[0145] S163: If the deviation value is greater than a preset threshold, returning to the step of inputting the expected target of the robot control problem into the discrete time model to obtain the latest joint angle target solution;

[0146] S164: Based on the latest joint angle target solution, executing the step of sending the joint angle target solution to the controller of the robot, and utilizing the controller to control the movement of the robot according to the joint angle target solution.

[0147] It is understandable that this embodiment can dynamically obtain the current position of the robot end effector and receive the corresponding feedback signal, and use the robot controller to correct the joint angle target solution according to the feedback signal. Then, it is determined whether the current position of the end effector matches the corresponding position of the joint angle target solution. If the deviation value of the above two is less than or equal to the preset threshold, it can be considered to match. If the deviation value is greater than the preset threshold, the joint angle target solution is re-solved to improve the accuracy of robot control.

[0148] In some embodiments of the present application, the optimized robot control problem matrix is ​​solved by a hybrid integral neural network solver of discretization design, and the obtained target solution of each joint of the robot is input into the controller of the robot as a reference control value, and the robot is controlled in real time to complete the task under the feedback mechanism. The problem solving method is simplified by discretization, and the robustness of the robot under various interference conditions and the stability in complex environments are improved. By discretizing the continuous time optimization model into a discrete time step model, the computational complexity is significantly reduced. Compared with the continuous time solution method in the prior art, some embodiments of the present application are optimized in step selection and numerical stability, especially when dealing with interference and system errors, the control accuracy is improved through specific numerical discrete formulas.

[0149] Reference Figure 2 The embodiment of the present application also provides a robot control device of a hybrid integral discrete neural network under nonlinear constraints, which can implement the robot control method of a hybrid integral discrete neural network under nonlinear constraints. The device includes:

[0150] A control problem construction unit, used to construct a robot control problem; wherein the robot control problem includes an objective function and a constraint equation;

[0151] A deviation equation construction unit, used for constructing the deviation equation of the robot control problem according to Lagrange multipliers and nonlinear functions;

[0152] A solver construction unit, used for constructing a hybrid integral neural network solver based on the deviation equation;

[0153] A solver conversion unit, used for converting the hybrid integral neural network solver into a matrix form under discrete time as a discrete time model;

[0154] A joint angle solving unit, used for inputting the expected target of the robot control problem into the discrete time model to solve and obtain the joint angle target solution;

[0155] A robot control unit is used to send the joint angle target solution to a controller of the robot, and use the controller to control the movement of the robot according to the joint angle target solution.

[0156] It can be understood that the contents of the above method embodiments are all applicable to the present device embodiments, the functions specifically implemented by the present device 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.

[0157] The embodiment of the present application also provides an electronic device, the electronic device includes a memory and a processor, the memory stores a computer program, and the processor implements the robot control method of the mixed integral discrete neural network under nonlinear constraints when executing the computer program. The electronic device can be any intelligent terminal including a tablet computer, a car computer, etc.

[0158] It can be understood that the contents of the above method embodiments are all applicable to the present device embodiments, the functions specifically implemented by the present device 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.

[0159] See also Figure 3 , Figure 3 The hardware structure of an electronic device of another embodiment is illustrated, and the electronic device includes:

[0160] The processor 301 may be implemented by a general-purpose CPU (Central Processing Unit), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of the present application;

[0161] The memory 302 can be implemented in the form of a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM). The memory 302 can store an operating system and other application programs. When the technical solution provided in the embodiment of this specification is implemented by software or firmware, the relevant program code is stored in the memory 302, and the processor 301 calls and executes the robot control method of the mixed integral discrete neural network under nonlinear constraints of the embodiment of the present application;

[0162] Input / output interface 303, used to implement information input and output;

[0163] Communication interface 304, used to realize communication interaction between the device and other devices, which can be realized through wired mode (such as USB, network cable, etc.) or wireless mode (such as mobile network, WIFI, Bluetooth, etc.);

[0164] A bus 305 that transmits information between the various components of the device (e.g., the processor 301, the memory 302, the input / output interface 303, and the communication interface 304);

[0165] The processor 301 , the memory 302 , the input / output interface 303 and the communication interface 304 are connected to each other in communication within the device via the bus 305 .

[0166] An embodiment of the present application also provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, it implements the robot control method of the mixed integral discrete neural network under the above-mentioned nonlinear constraints.

[0167] It can be understood that the contents of the above method embodiments are all applicable to the present storage medium embodiments, the functions specifically implemented by the present storage medium 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.

[0168] The memory, as a non-transient computer-readable storage medium, can be used to store non-transient software programs and non-transient computer executable programs. In addition, the memory may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some embodiments, the memory may optionally include a memory remotely disposed relative to the processor, and these remote memories may be connected to the processor via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0169] The embodiments described in the embodiments of the present application are intended to more clearly illustrate the technical solutions of the embodiments of the present application and do not constitute a limitation on the technical solutions provided in the embodiments of the present application. Those skilled in the art will appreciate that with the evolution of technology and the emergence of new application scenarios, the technical solutions provided in the embodiments of the present application are also applicable to similar technical problems.

[0170] Those skilled in the art will appreciate that the technical solutions shown in the figures do not constitute a limitation on the embodiments of the present application, and may include more or fewer steps than shown in the figures, or a combination of certain steps, or different steps.

[0171] The device embodiments described above are merely illustrative, and the units described as separate components may or may not be physically separated, that is, they may be located in one place or distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0172] Those skilled in the art will appreciate that all or some of the steps in the methods disclosed above, and the functional modules / units in the systems and devices may be implemented as software, firmware, hardware, or a suitable combination thereof.

[0173] The terms "first", "second", "third", "fourth", etc. (if any) in the specification of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0174] It should be understood that in the present application, "at least one (item)" means one or more, and "plurality" means two or more. "And / or" is used to describe the association relationship of associated objects, indicating that three relationships may exist. For example, "A and / or B" can mean: only A exists, only B exists, and A and B exist at the same time, where A and B can be singular or plural. The character " / " generally indicates that the objects associated before and after are in an "or" relationship. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, at least one of a, b or c can mean: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, c can be single or multiple.

[0175] In the several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the above units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0176] The units described above as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0177] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0178] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including multiple instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of various embodiments of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory, referred to as ROM), random access memory (Random Access Memory, referred to as RAM), disk or optical disk and other media that can store programs.

[0179] 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 a hybrid integral discrete neural network under nonlinear constraints, characterized in that: The method comprises the following steps: Constructing a robot control problem; wherein the robot control problem includes an objective function and a constraint equation; Constructing the deviation equation of the robot control problem according to Lagrange multipliers and nonlinear functions; Constructing a hybrid integral neural network solver based on the deviation equation; Converting the hybrid integral neural network solver into a matrix form under discrete time as a discrete time model; Inputting the expected target of the robot control problem into the discrete time model to obtain the joint angle target solution; Sending the joint angle target solution to a controller of the robot, and using the controller to control the movement of the robot according to the joint angle target solution; The construction of the robot control problem includes the following steps: The objective function for constructing the robot control problem is: ； in, , Indicates that the robot is The angular velocity of each joint at the moment, is the robot's degrees of freedom; The constraint equations for constructing the robot control problem are: ； in, is the equality constraint equation, is the inequality constraint equation; The method of constructing the deviation equation of the robot control problem according to the Lagrange multiplier and the nonlinear function comprises the following steps: The deviation equation of the robot control problem constructed according to Lagrange multipliers and nonlinear functions is: ； in, is the deviation equation, ; is the solution to the robot control problem, ; , are the Lagrange multipliers of the equality constraint equation and the inequality constraint equation respectively; is a nonlinear functional relationship of the inequality constraint equation; The method of constructing a hybrid integral neural network solver based on the deviation equation comprises the following steps: Construct the expression of hybrid integral neural network; The expression of the hybrid integral neural network is: ； in, is the expression of the hybrid integral neural network; is a constant; is an integral variable, the integral lower limit of the integral variable is 0, the first integral upper limit is t, and the second integral upper limit is ; Constructing the hybrid integral neural network solver according to the deviation equation and the expression of the hybrid integral neural network; The hybrid integral neural network solver is: ； in, The hybrid integral neural network solver; is the interference item; Specifically: ； in, , is the functional expression of the nonlinear functional relationship, is the dimension of the equality constraint equation, is the dimension of the inequality constraint equation; Specifically: ； Specifically: ； in, represents a diagonal matrix; The step of converting the hybrid integral neural network solver into a matrix form under discrete time as a discrete time model comprises the following steps: Converting the hybrid integral neural network solver into a matrix form; The hybrid integral neural network solver in matrix form is: ； in, is the integral term of the deviation equation, ; is the double integral term of the deviation equation, ; is an intermediate parameter used to simplify the expression. ); , for The inverse matrix of Construct discrete-time expressions; The discrete time expression is: ； in, is the derivative of the discrete time series, is the discrete time series; is the sampling period; According to the discrete time expression, the hybrid integral neural network solver in matrix form is converted into a matrix form under discrete time as the discrete time model; The discrete time model is: ； in, for The value at time step k, for The value at time step k, for The value at time step k, for The value at time step k, for The value at time step k, is the time value corresponding to the kth time step, is the value of the interference term at time step k; k is an integer, representing the index of the discrete time step, and the value range of k is 0 to 6; )， 。 2. The robot control method of the hybrid integral discrete neural network under nonlinear constraints according to claim 1 is characterized in that: The method of inputting the expected target of the robot control problem into the discrete time model and solving the joint angle target solution comprises the following steps: Inputting the vector of expected targets into the discrete-time model; The discrete-time model is used to solve the deviation equation based on the vector of the expected target, and the solution when the deviation equation is 0 is used as the joint angle target solution.

3. The robot control method of a hybrid integral discrete neural network under nonlinear constraints according to any one of claims 1 to 2, characterized in that: The method further comprises the following steps: Obtaining the current position of the end effector of the robot; Calculating a deviation value between a current position of the end effector and a corresponding position of the current joint angle target solution; If the deviation value is greater than a preset threshold, returning to input the expected target of the robot control problem into the discrete time model to solve and obtain the latest joint angle target solution; Based on the latest target joint angle solution, the target joint angle solution is sent to a controller of the robot, and the controller is used to control the movement of the robot according to the target joint angle solution.

4. A robot control device based on a hybrid integral discrete neural network under nonlinear constraints, characterized in that: The device is used to implement the robot control method of the mixed integral discrete neural network under nonlinear constraints as claimed in claim 1, comprising: A control problem construction unit, used to construct a robot control problem; wherein the robot control problem includes an objective function and a constraint equation; A deviation equation construction unit, used for constructing the deviation equation of the robot control problem according to Lagrange multipliers and nonlinear functions; A solver construction unit, used for constructing a hybrid integral neural network solver based on the deviation equation; A solver conversion unit, used for converting the hybrid integral neural network solver into a matrix form under discrete time as a discrete time model; A joint angle solving unit, used for inputting the expected target of the robot control problem into the discrete time model to solve and obtain the joint angle target solution; A robot control unit is used to send the joint angle target solution to a controller of the robot, and use the controller to control the movement of the robot according to the joint angle target solution.

5. An electronic device, characterized in that: The electronic device comprises a memory and a processor, the memory stores a computer program, and the processor implements the method according to any one of claims 1 to 3 when executing the computer program.

6. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 3 is implemented.

Images