Self-adaptive nonlinear robust control method and system of flight double-arm system

Abstract

The invention belongs to the field of nonlinear under-actuated electromechanical system automatic control, and provides a self-adaptive nonlinear robust control method and system for a flight double-arm system, and the method comprises the steps: building a double-arm kinematic model based on measured double-arm geometric parameters; according to the double-arm kinematics model, the influence of double-arm motion on the system mass center is analyzed, and the mass point system momentum theorem is utilized to construct an unmanned aerial vehicle dynamics model; a deep neural network feedforward item is designed, a robustness error integral feedback item is designed based on the unmanned aerial vehicle dynamics model, and a complete nonlinear robust controller is constructed; a state feedback signal of a flight double-arm system is acquired in real time, a real-time weight is provided for each layer of the deep neural network according to an adaptive update law, a control signal for realizing accurate trajectory tracking of the unmanned aerial vehicle is calculated by using a nonlinear robust controller, and the accurate trajectory tracking of the unmanned aerial vehicle is completed under the driving of the signal.

Description

Technical Field

[0001] The present invention belongs to the technical field of automatic control of nonlinear underactuated electromechanical systems, and in particular relates to an adaptive nonlinear robust control method and system for a flying dual-arm system. Background Art

[0002] The statements in this section merely provide background information related to the present invention and do not necessarily constitute prior art.

[0003] In recent years, the interest in the collaborative operation of dual-arm robots has exploded in the field of robotics. This interest is not only due to the significant performance advantages of dual-arm robots over single-arm robots, but also because of their multifunctional operation capabilities and larger workspace, which has opened up new paths for the further development and application of robotics technology.

[0004] The dual-arm robot not only greatly improves the flexibility and accuracy of operation through the collaborative work of two mechanical arms, but also significantly increases the robot's workspace and adaptability. This design concept enables the dual-arm robot to demonstrate higher efficiency and wider application prospects when handling complex tasks than single-arm robots.

[0005] In the research of dual-arm robots, an attractive direction is to integrate them with multi-rotor drones. This integration method takes into account the wide working space of dual-arm robots while retaining the vertical take-off and landing capabilities and high flexibility of multi-rotor drones. Combining these two technologies, we can obtain a new type of robot system that can both fly freely in the air and precisely manipulate objects.

[0006] The combination of drones and robotic arms not only has great research value, but also has broad application prospects. Compared with ground mobile robots, the flying robotic arm system composed of drones and robotic arms can expand the operating range to the three-dimensional space in the air, greatly expanding the operating capabilities and application scope of drones. The flying robotic arm system can expand the applicability of drones from traditional passive scenarios to more active scenarios, such as aerial construction at construction sites, inspection and maintenance of power facilities, and even emergency rescue at disaster sites. Chen et al. proposed a finite-time control strategy for drones using an adaptive sliding mode disturbance observer (ASMDO) to effectively deal with uncertainties and external disturbances; Kim et al. used a flying robotic arm to open and close unknown drawers by analyzing interactions.

[0007] As an important form of flying manipulator system, the flying dual-arm system has also developed rapidly in recent years. Compared with the single-arm system, the dual-arm system has higher flexibility and stronger operation ability, and can better meet the needs of complex tasks. In the research of the flying dual-arm system, how to design a light and flexible manipulator structure and how to achieve coordinated control between the two arms are the focus of the research. Through rigorous testing and verification processes, researchers continuously optimize the manipulator structure and control strategy to improve the performance and stability of the system. Suarez et al. designed a light and flexible dual-arm structure, combined with a rigorous testing and verification process suitable for scenes such as power lines and pipelines; Orsag et al. introduced a flying dual-arm system with multiple degrees of freedom and successfully applied it to valve rotation; Kong et al. introduced a new flying dual-arm system teleoperation platform, the manipulator of which has a unique joint structure including tendons and is supported by elastic components.

[0008] However, the flying dual-arm system also faces some challenges, mainly because the presence of the robotic arm and its movement will affect the stable flight of the multirotor, and these effects cannot be directly compensated due to the presence of unmeasured signals related to joint velocity and acceleration. In addition, real-world factors such as unmodeled effects and external interference will bring unexpected risks to the system. Summary of the invention

[0009] In order to solve the above problems, the present invention proposes an adaptive nonlinear robust control method and system for a flying dual-arm system. The present invention uses the RISE feedback control strategy and combines it with the DNN feedforward control term to enhance the transient and steady-state performance. Considering using DNN as a feedforward component with universal function approximation capability, the RISE feedback term is combined with the DNN feedforward term to ensure the asymptotic tracking of the UAV.

[0010] According to some embodiments, a first solution of the present invention provides an adaptive nonlinear robust control method for a flight dual-arm system, which adopts the following technical solution:

[0011] An adaptive nonlinear robust control method for a flying dual-arm system, comprising:

[0012] Based on the measured geometric parameters of the arms, a kinematic model of the arms is established;

[0013] According to the dual-arm kinematic model, the influence of dual-arm motion on the center of mass of the system is analyzed, and the momentum theorem of mass point system is used to construct the UAV dynamics model;

[0014] Design a deep neural network feedforward term and a robust error integral feedback term based on the UAV dynamics model to build a complete nonlinear robust controller;

[0015] The state feedback signal of the flight dual-arm system is obtained in real time, and real-time weights are provided for each layer of the deep neural network according to the adaptive update law. The nonlinear robust controller is used to calculate the control signal that enables the UAV to achieve precise trajectory tracking, and the precise trajectory tracking of the UAV is completed under the drive of this signal.

[0016] Furthermore, the kinematic model of the double arms is established based on the measured geometric parameters of the double arms, specifically:

[0017] Get the distances between adjacent joints in the double-arm configuration;

[0018] The joint angles in the robot arm are expressed as in, represents the angle of the i-th joint from top to bottom, i.e., shoulder-pitch, elbow-pitch, elbow-roll, k=1,2 represents the right arm and left arm respectively;

[0019] According to the distance between adjacent joints in the double-arm configuration, taking the right arm as an example, the DH parameter table is calculated to determine the following double-arm kinematic model, specifically:

[0020]

[0021] in, represents the speed and angular velocity of the right arm end under the machine system, is the joint position vector, J(η 1 ) is the Jacobian matrix of the robot arm.

[0022] Furthermore, the UAV dynamics model is specifically:

[0023]

[0024] Among them, m t is the total mass of the whole system, is the position of the drone in the world coordinate system, g is the acceleration due to gravity, is a unit vector, F m is the force applied by the robotic arm to the drone, F f represents the friction force, F d represents the general nonlinear perturbation force, Represents the control input of the drone.

[0025] Furthermore, the deep neural network feedforward term is specifically:

[0026]

[0027] Among them, f d =F f is an auxiliary function related to the derivative of the desired trajectory, It is ford The estimate of , i.e. the deep neural network feedforward term, is the input vector, Represent the target constant weight matrix and its estimated value respectively, represents the activation function, ε(x d ) represents the bounded function reconstruction error.

[0028] Furthermore, the projection operator is used to design a real-time update law for the weights of the deep neural network:

[0029]

[0030] Among them, Γ i , i=0,1,...,k is a positive definite diagonal gain matrix, is the derivative of the activation function.

[0031] Furthermore, the nonlinear robust controller includes a deep neural network feedforward term and a robust error integral feedback term, specifically:

[0032]

[0033]

[0034] Among them, K s =diag([k sx ,k sy ,k sz ]),B1=diag([β 1x ,β 2x ,β 3x ]) is a positive definite diagonal matrix, I is the identity matrix, the robust error integral feedback term μ and the deep neural network feedforward term

[0035] According to some embodiments, a second solution of the present invention provides an adaptive nonlinear robust control system for a flight dual-arm system, which adopts the following technical solution:

[0036] An adaptive nonlinear robust control system for a flying dual-arm system, comprising:

[0037] The dual-arm model building module is specifically configured as follows: based on the measured dual-arm geometric parameters, a dual-arm kinematic model is established;

[0038] The specific configuration of the UAV model building module is as follows: based on the dual-arm kinematic model, the influence of the dual-arm movement on the center of mass of the system is analyzed, and the momentum theorem of the mass point system is used to build the UAV dynamics model;

[0039] The DNN-based adaptive RISE controller building module is specifically configured as follows: designing a deep neural network feedforward term and designing a robust error integral feedback term based on the UAV dynamics model to build a complete nonlinear robust controller;

[0040] The specific configuration of the UAV positioning control module is as follows: real-time acquisition of the flight dual-arm system state feedback signal, providing real-time weights for each layer of the deep neural network according to the adaptive update law, using a nonlinear robust controller to calculate the control signal that enables the UAV to achieve precise trajectory tracking, and driven by this signal, completes the precise trajectory tracking of the UAV.

[0041] Furthermore, the flying dual-arm operating system comprises a rotary-wing UAV, dual arms composed of a steering gear and a connecting piece, and each mechanical arm in the dual arms comprises three joints, namely, shoulder-pitch, elbow-pitch, and elbow-roll;

[0042] The rotary wing UAV includes an onboard computer and a flight control unit, and the onboard computer and the flight control unit communicate with each other;

[0043] The onboard computer adopts the adaptive nonlinear robust control method of the flying dual-arm system described in the first scheme to calculate the control signal of the rotorcraft UAV and transmit the control signal to the driver of the flight control unit, thereby realizing accurate trajectory tracking of the UAV.

[0044] According to some embodiments, a third aspect of the present invention provides a computer-readable storage medium.

[0045] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps in the adaptive nonlinear robust control method for a flying dual-arm system as described in the first aspect above.

[0046] According to some embodiments, a fourth aspect of the present invention provides a computer device.

[0047] A computer device comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps in the adaptive nonlinear robust control method for a flying dual-arm system as described in the first aspect above are implemented.

[0048] Compared with the prior art, the present invention has the following beneficial effects:

[0049] The current flight robotic arm system is mainly operated by a single arm, but when faced with difficult tasks such as grasping large objects, it cannot achieve the intended goal well. The present invention studies the flight dual-arm system and designs a dual-arm structure that can greatly expand the flight operation capability and workspace, and establishes precise kinematic and dynamic models.

[0050] Compared with the flying single-arm operating system, the movement of the two arms in the flying dual-arm system has a more significant impact on the UAV, which can easily cause the system to crash. To address this problem, the present invention proposes a DNN-based adaptive RISE controller for the flying dual-arm system; the controller can not only achieve accurate trajectory tracking of the UAV, but also ensure the safety and smooth operation of the system when performing tasks; it is expected to be further applied to large unmanned helicopters, which has very important research value and practical significance. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] The accompanying drawings in the specification, which constitute a part of the present invention, are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute improper limitations on the present invention.

[0052] Figure 1 This is a structural diagram of a flying dual-arm system in Embodiment 1 of the present invention;

[0053] Figure 2 This is a model diagram of a flying dual-arm system in Embodiment 1 of the present invention;

[0054] Figure 3 This is a control block diagram in Embodiment 2 of the present invention;

[0055] Figure 4 This is a diagram of the hardware platform in the second embodiment of the present invention;

[0056] FIG5( a ) is a graph showing the position variation over time of the UAV tracking experiment in the second embodiment of the present invention;

[0057] FIG5( b ) is a graph showing the thrust variation over time in the UAV tracking experiment in the second embodiment of the present invention;

[0058] FIG5( c ) is a graph showing the network output variation over time in the UAV tracking experiment in the second embodiment of the present invention;

[0059] Figure 6 This is a system structure diagram of embodiment 3 of the present invention. DETAILED DESCRIPTION

[0060] The present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0061] It should be noted that the following detailed descriptions are all illustrative and intended to provide further explanation of the present invention. Unless otherwise specified, all technical and scientific terms used herein have the same meanings as those commonly understood by those skilled in the art to which the present invention belongs.

[0062] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit exemplary embodiments according to the present invention. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that when the terms "comprising" and / or "including" are used in this specification, it indicates the presence of features, steps, operations, devices, components and / or combinations thereof.

[0063] In the absence of conflict, the embodiments of the present invention and the features of the embodiments may be combined with each other.

[0064] Terminology explanation:

[0065] RISE: It is a method in control theory. Its full name is “Robust Integral of the Sign of the Error”. It is a method for designing robust controllers to improve the robust performance of control systems against uncertainties and disturbances.

[0066] Embodiment 1

[0067] This embodiment provides an adaptive nonlinear robust control method for a flying dual-arm system. In this embodiment, the method comprises the following steps:

[0068] This embodiment discloses a flying dual-arm system, and its hardware structure is as follows: Figure 1 and Figure 2 As shown, Figure 1 The clamp is not shown in the figure. Figure 2 The system includes a drone and a double arm consisting of a servo and a connector, which can achieve flexible operation.

[0069] In order to improve the operational capability of the flight robotic arm system and expand the workspace, it is considered to replace the single arm with a double arm. The left and right arms and the gripper are composed of eight servos and connectors. The robotic arm imitates the joint configuration of the human arm. Each arm contains three joints, that is, three degrees of freedom, from top to bottom: shoulder-pitch, elbow-pitch, elbow-roll. The joints, servo models and corresponding parameters are shown in Table 1.

[0070] Table 1 Joint and servo parameters

[0071] joint Servo Maximum torque (N·m) Weight(g) Shoulder-Pitch XM540-W270 12.9 165 Elbow-Pitch XM430-W350 4.8 82 Elbow-roll XM430-W350 4.8 82 Gripper AX-12A 1.5 54.6

[0072] The drone uses a rotorcraft, including but not limited to a quad-rotor, and a hexa-rotor or an octa-rotor of similar size can also be selected. In this embodiment, a quad-rotor drone with a wheelbase of 550 mm is used. The drone is equipped with an onboard computer NUC and a PixHawk flight control unit, and the NUC communicates with the flight control unit through the MAVLINK protocol.

[0073] The onboard computer uses the adaptive nonlinear robust control method of the flying dual-arm system disclosed in Example 2 to calculate the control signal that enables the UAV to achieve precise trajectory tracking, and inputs the control signal into the driver of the flight control unit to complete the precise trajectory tracking of the UAV.

[0074] Embodiment 2

[0075] like Figure 3 As shown, this embodiment provides an adaptive nonlinear robust control method for a flying dual-arm system. In this embodiment, the method includes the following steps:

[0076] Step S201: establishing a dual-arm kinematic model based on the measured dual-arm geometric parameters;

[0077] Step S202: Analyze the influence of the movement of the two arms on the center of mass of the system according to the kinematic model of the two arms, and construct the dynamic model of the UAV by using the momentum theorem of the mass point system;

[0078] Step S203: Design a deep neural network feedforward term, and design a robust error integral feedback term based on the UAV dynamics model to construct a complete nonlinear robust controller;

[0079] Step 204: Acquire the flight dual-arm system state feedback signal in real time, provide real-time weights for each layer of the deep neural network according to the adaptive update law, use the nonlinear robust controller to calculate the control signal that enables the UAV to achieve precise trajectory tracking, and complete the precise trajectory tracking of the UAV under the drive of this signal.

[0080] Step S201: Establishing a dual-arm kinematic model based on the measured dual-arm geometric parameters.

[0081] The joint angles in the robot arm are expressed as in, represents the angle of the i-th joint from top to bottom, namely shoulder-pitch, elbow-pitch, elbow-roll, and k=1,2 represent the right arm and left arm respectively.

[0082] Measure the geometric parameters of the dual arms, that is, the distance between two adjacent joints in the robotic arm, such as Figure 2 L1, L2, L3 shown in . Taking the right arm as an example, the following double-arm kinematic model can be obtained according to the DH parameter table:

[0083]

[0084] in, represents the speed and angular velocity of the right arm end under the machine system, is the joint position vector, J(η 1 ) is the Jacobian matrix of the robot arm.

[0085] Step S202: According to the dual-arm kinematic model, the influence of the dual-arm movement on the center of mass of the system is analyzed, and the momentum theorem of the mass point system is used to construct the UAV dynamics model.

[0086] The dynamic model of the drone is as follows:

[0087]

[0088] Among them, m t is the total mass of the whole system, is the position of the drone in the world coordinate system, is a unit vector, F f represents the friction force, F d represents the general nonlinear disturbance force (including unmodeled effects and system parameter deviations, etc.), represents the control input of the UAV, F m It is the force applied by the robotic arm to the drone, which is expressed as follows:

[0089]

[0090] Among them, ω represents the angular velocity of the drone in the machine system, R represents the rotation matrix of the drone in the world system, and r oc It represents the position of the center of mass of the double arms under the machine system.

[0091] Furthermore, the system model can be rewritten as follows:

[0092]

[0093] U=U c -F c

[0094] F c =m t ge3+m t R[ω×(ω×r oc )]

[0095]

[0096] in, represents the virtual control input to be designed, F crepresents the force effect that can be directly feedforward compensated, and F represents the force effect that is difficult to obtain directly.

[0097] Step S203: Design a DNN (deep neural network) feedforward term, and design a RISE (robust integral error) feedback term based on the UAV dynamics model to construct a complete nonlinear robust controller.

[0098] The specific expression of DNN is:

[0099]

[0100] Among them, f d =F f is an auxiliary function related to the derivative of the desired trajectory, It is for d The estimate, is the input vector, Represent the target constant weight matrix and its estimated value respectively, represents the activation function, ε(x d ) represents the bounded function reconstruction error.

[0101] Using the projection operator, we design a real-time update law for network weights:

[0102]

[0103] Among them, Γ i , i=0,1,...,k is a positive definite diagonal gain matrix, is the derivative of the activation function.

[0104] Define the UAV tracking error e1 and the filtered signal e2,r:

[0105] e1=pp d

[0106]

[0107] in, represents the expected time-varying trajectory of the system, k1 and k2 are positive constants.

[0108] Furthermore, a complete nonlinear robust controller is designed, including the RISE feedback control term μ and the DNN feedforward control term The specific form is as follows:

[0109]

[0110] Among them, K s =diag([k sx ,k sy ,k sz ]),B1=diag([β1x ,β 2x ,β 3x ]) is a positive definite diagonal matrix and I is the identity matrix.

[0111] Step S204: Acquire the flight system state feedback signal in real time, provide real-time weights for each layer of the DNN according to the adaptive update law, use the nonlinear robust controller to calculate the control signal that enables the UAV to achieve precise trajectory tracking, and complete the precise trajectory tracking of the UAV under the drive of this signal.

[0112] Specifically, the real-time posture of the UAV and the center of mass position of the two arms are obtained, and the final control signal U is calculated based on the nonlinear robust tracking controller shown in formula (8). The control signal U is input into the driver of the flight control unit to complete the high-precision trajectory tracking of the UAV.

[0113] In order to verify the effectiveness of the method, a test experiment was conducted on the nonlinear robust tracking control method of this embodiment; a NUC airborne computer running a 64-bit Ubuntu 18.04 operating system was installed on the quadcopter, which was connected to the ground station via 5G band WIFI. The experimental platform is shown in the attached figure. Figure 4 As shown; the main physical parameters of the flying dual-arm system are

[0114] m t =4.85kg, g=9.8kg·m / s 2 ,L1=0.13m, L2=0.04m, L3=0.20m

[0115] The control gain of the nonlinear robust control method of this embodiment is:

[0116] K s =diag([10.0,10.0,14.5]),B1=diag([4.0,4.0,4.0]),k1=0.69,k2=0.5

[0117] DNN consists of four layers, the first layer contains three neurons, and the remaining three layers have four neurons each. The weight adaptive update law is set as:

[0118] Γ0=diag([2,2,2,2])×10 6 ,Γ i =diag([2,2,2,2,2])×10 6 ,i=1,2,3

[0119] Figure 5 shows the relevant results of the UAV tracking the “figure 8” trajectory. The UAV position corresponds to p x ,p y ,p z , the input system control signal is Ucx ,U cy ,U cz , that is, the thrust in the x-axis, y-axis and z-axis directions, the blue dotted line represents the method of this embodiment, the red dotted line represents the traditional method, and the black solid line is the desired trajectory; it can be seen from the figure that the error between the blue dotted line and the black solid line is smaller than the error between the red dotted line and the black solid line. Therefore, the nonlinear robust control method of this embodiment can successfully drive the UAV to track the desired trajectory with higher tracking accuracy.

[0120] In summary, the present invention not only proposes a physical structure of a flexibly operable flying dual-arm operating system, but also the proposed controller can significantly reduce internal and external disturbances including the influence of the robotic arm, thereby achieving high-precision tracking of the UAV.

[0121] Embodiment 3

[0122] like Figure 6 As shown, this embodiment provides an adaptive nonlinear robust control system for a flight dual-arm system, including:

[0123] The dual-arm model building module is specifically configured as follows: based on the measured dual-arm geometric parameters, a dual-arm kinematic model is established;

[0124] The specific configuration of the UAV model building module is as follows: based on the dual-arm kinematic model, the influence of the dual-arm movement on the center of mass of the system is analyzed, and the momentum theorem of the mass point system is used to build the UAV dynamics model;

[0125] The nonlinear robust controller building module is specifically configured as follows: designing a deep neural network feedforward term and designing a robust error integral feedback term based on the UAV dynamics model to build a complete nonlinear robust controller;

[0126] The specific configuration of the UAV positioning control module is as follows: real-time acquisition of the flight dual-arm system state feedback signal, providing real-time weights for each layer of the deep neural network according to the adaptive update law, using a nonlinear robust controller to calculate the control signal that enables the UAV to achieve precise trajectory tracking, and driven by this signal, completes the precise trajectory tracking of the UAV.

[0127] The examples and application scenarios implemented by the above modules and corresponding steps are the same, but are not limited to the contents disclosed in the above embodiment 1. It should be noted that the above modules as part of the system can be executed in a computer system such as a set of computer executable instructions.

[0128] The description of each embodiment in the above embodiments has different emphases. For parts not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0129] The proposed system can be implemented in other ways. For example, the system embodiment described above is only illustrative, and the division of the modules is only a logical function division. In actual implementation, there may be other division methods, such as multiple modules can be combined or integrated into another system, or some features can be ignored or not executed.

[0130] Embodiment 4

[0131] This embodiment provides a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, the steps in the adaptive nonlinear robust control method of a flight dual-arm system as described in the above-mentioned embodiment 2 are implemented.

[0132] Embodiment 5

[0133] This embodiment provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, the steps in the adaptive nonlinear robust control method for a flight dual-arm system as described in the above-mentioned embodiment 2 are implemented.

[0134] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of hardware embodiments, software embodiments, or embodiments combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage and optical storage, etc.) containing computer-usable program code.

[0135] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0136] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0137] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0138] A person skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program, and the program can be stored in a computer-readable storage medium, and when the program is executed, it can include the processes of the embodiments of the above-mentioned methods. The storage medium can be a disk, an optical disk, a read-only memory (ROM) or a random access memory (RAM), etc.

[0139] Although the above describes the specific implementation mode of the present invention in conjunction with the accompanying drawings, it is not intended to limit the scope of protection of the present invention. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art on the basis of the technical solution of the present invention without creative work are still within the scope of protection of the present invention.

Claims

1. An adaptive nonlinear robust control method for a flying dual-arm system, characterized in that: include: Based on the measured geometric parameters of the arms, a kinematic model of the arms is established; According to the dual-arm kinematic model, the influence of dual-arm motion on the center of mass of the system is analyzed, and the momentum theorem of mass point system is used to construct the UAV dynamics model; Design a deep neural network feedforward term and a robust error integral feedback term based on the UAV dynamics model to build a complete nonlinear robust controller; The state feedback signal of the flying dual-arm system is obtained in real time, and real-time weights are provided for each layer of the deep neural network according to the adaptive update law. The nonlinear robust controller is used to calculate the control signal that enables the UAV to achieve precise trajectory tracking, and the precise trajectory tracking of the UAV is completed under the drive of this signal.

2. The adaptive nonlinear robust control method of a flight dual-arm system according to claim 1, characterized in that: The kinematic model of the double arms is established based on the measured geometric parameters of the double arms, specifically: Get the distances between adjacent joints in the double-arm configuration; The joint angles in the robot arm are expressed as in, represents the angle of the i-th joint from top to bottom, i.e., shoulder-pitch, elbow-pitch, elbow-roll, k=1,2 represents the right arm and left arm respectively; According to the distance between adjacent joints in the double-arm configuration, taking the right arm as an example, the DH parameter table is calculated to determine the following double-arm kinematic model, specifically: in, represents the speed and angular velocity of the right arm end under the machine system, is the joint position vector, J(η 1 ) is the Jacobian matrix of the robot.

3. The adaptive nonlinear robust control method of a flying dual-arm system according to claim 1, characterized in that: The UAV dynamics model is specifically: Among them, m t is the total mass of the whole system, p=[p x ,p y ,p z ] T is the position of the drone in the world coordinate system, g is the acceleration of gravity, e3 = [0,0,1] T is a unit vector, F m is the force applied by the robotic arm to the drone, F f represents the friction force, F d represents the general nonlinear disturbance force, U c =[U cx ,U cy ,U cz ] T Represents the control input of the drone.

4. The adaptive nonlinear robust control method of a flying dual-arm system according to claim 1, characterized in that: The deep neural network feedforward term is specifically: Among them, f d =F f is an auxiliary function related to the derivative of the desired trajectory, It is for d The estimate of , i.e. the deep neural network feedforward term, is the input vector, Represent the target constant weight matrix and its estimated value respectively, represents the activation function, ε(x d ) represents the bounded function reconstruction error.

5. The adaptive nonlinear robust control method of a flying dual-arm system according to claim 4, characterized in that: Using the projection operator, we design a real-time update law for the weights of a deep neural network: Among them, Γ i , i=0,1,...,k is a positive definite diagonal gain matrix, is the derivative of the activation function.

6. The adaptive nonlinear robust control method of a flying dual-arm system according to claim 1, characterized in that: The nonlinear robust controller includes a deep neural network feedforward term and a robust error integral feedback term, specifically: Among them, K s =diag([k sx ,k sy ,k sz ]),B1=diag([β 1x ,β 2x ,β 3x ]) is a positive definite diagonal matrix, I is the identity matrix, the robust error integral feedback term μ and the deep neural network feedforward term 7. An adaptive nonlinear robust control system for a flying dual-arm system, characterized in that: include: The dual-arm model building module is specifically configured as follows: based on the measured dual-arm geometric parameters, a dual-arm kinematic model is established; The specific configuration of the UAV model building module is as follows: based on the dual-arm kinematic model, the influence of the dual-arm movement on the center of mass of the system is analyzed, and the momentum theorem of the mass point system is used to build the UAV dynamics model; The nonlinear robust controller building module is specifically configured as follows: designing a deep neural network feedforward term and designing a robust error integral feedback term based on the UAV dynamics model to build a complete nonlinear robust controller; The specific configuration of the UAV positioning control module is as follows: real-time acquisition of the flight dual-arm system state feedback signal, providing real-time weights for each layer of the deep neural network according to the adaptive update law, using a nonlinear robust controller to calculate the control signal that enables the UAV to achieve precise trajectory tracking, and driven by this signal, completes the precise trajectory tracking of the UAV.

8. The adaptive nonlinear robust control system of a flying dual-arm system according to claim 7, characterized in that: The flying dual-arm operating system comprises a rotary-wing UAV, dual arms composed of a steering gear and a connecting piece, and each mechanical arm of the dual arms comprises three joints, namely, shoulder-pitch, elbow-pitch, and elbow-roll; The rotary wing UAV includes an onboard computer and a flight control unit, and the onboard computer and the flight control unit communicate with each other; The onboard computer adopts an adaptive nonlinear robust control method of a flying dual-arm system as described in any one of claims 1 to 6 to calculate the control signal of the rotorcraft UAV and transmit the control signal to the driver of the flight control unit, thereby realizing accurate trajectory tracking of the UAV.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps in the adaptive nonlinear robust control method of a flying dual-arm system as described in any one of claims 1-6 are implemented.

10. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the steps in the adaptive nonlinear robust control method of a flight dual-arm system as described in any one of claims 1-6 are implemented.

Images