Flight double-arm adaptive optimization control method and system considering safety constraint

Abstract

The invention discloses a flight double-arm self-adaptive optimization control method and system considering safety constraints, and belongs to the technical field of nonlinear front drive electromechanical system automatic control, and the method comprises the steps: building a kinematic model of flight double arms according to the geometric parameters of the double arms and a flight body; carrying out system dynamics modeling by adopting a Lagrange modeling method, introducing a dynamics equation of a grabbed load, and jointly constructing an overall dynamics model of the system; designing a control barrier function, constructing a performance index function, and calculating optimal control input in combination with the evaluation-action network; introducing an RBF neural network as a feedforward item, and constructing a nonlinear safety optimization controller and a network adaptive updating law in combination with sliding mode control; and collecting a state feedback signal of the flight double-arm system in real time, adjusting a network weight according to an adaptive update law, and generating a control instruction by using a nonlinear safety optimization controller to control a flight body to perform trajectory tracking. According to the invention, balance between tracking control precision and safety can be realized.

Description

Technical Field

[0001] This invention belongs to the field of automatic control technology for nonlinear forward-drive electromechanical systems, and particularly relates to an adaptive optimization control method and system for flight dual arms that takes into account safety constraints. Background Technology

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

[0003] In recent years, aerial robots have gradually become a strategically significant research hotspot and have shown broad prospects in many economic fields. Currently, aerial robots are gradually upgrading from "passive perception" (such as monitoring and search and rescue) to "active interaction" (such as aerial assembly and construction operations). The Unmanned Aerial Vehicle (UAM) system possesses active operational capabilities, forming a composite system with dexterity and maneuverability by combining a multi-rotor aircraft with a multi-degree-of-freedom robotic arm. In this system, the unmanned aerial vehicle provides maneuverability and flexible deployment capabilities, while the robotic arm provides precise operational capabilities and can be further equipped with end effectors (such as grippers) to meet the needs of complex tasks.

[0004] Current research on most flying robotic arm systems still focuses on single-arm structures. However, this structure has inherent limitations when performing aerial tasks requiring dexterity, such as rotating valves, moving long objects, or simultaneously operating two independent targets. These tasks often require a dual-arm system to be effectively completed. To meet these practical needs, existing research has gradually developed a flying dual-arm operating system for complex aerial tasks. This system platform consists of a UAV body, two robotic arms, and a two-degree-of-freedom "head," while integrating a flexible gripper at the end of each robotic arm to enhance its maneuverability.

[0005] For the control of a dual-arm flight system, current research has proposed various control strategies, such as acquiring operator signals through the main arm and driving the slave arm to mimic execution, and designing a finite-time convergent ultratorsion sliding mode controller to ensure rapid convergence of system state errors. However, due to the nonlinear dynamic characteristics and coupling relationships of the dual-arm flight system itself, deriving an analytical solution to the Hamilton-Jacobi-Bellman (HJB) partial differential equation is extremely challenging. Therefore, current research has turned to Adaptive Dynamic Programming (ADP) to approximate the optimal control. Although traditional ADP methods have shown good potential in solving optimal control problems, they often treat the problem as a "unconstrained" performance optimization, i.e., minimizing the cost function through value iteration / policy iteration. This approach lacks rigorous feasibility proofs for time-by-time state / input / environment constraints, cannot guarantee the positive invariance of the safety set, and is still limited by the lack of theoretical guarantees for safety in practical applications. It is difficult to ensure the safety of the system during operation (such as avoiding collisions with obstacles), and thus difficult to achieve stable and reliable control of the system. In addition, the system not only needs to achieve optimal performance and security, but also needs to have robust response capabilities to internal and external disturbances. Summary of the Invention

[0006] To address the shortcomings of the existing technologies, this invention provides an adaptive optimization control method and system for dual-arm flight that considers safety constraints. The method incorporates the object or load to be grasped into the modeling framework of the dual-arm flight cooperative operating system, thereby effectively reducing the impact of the object's presence on the overall system performance. Simultaneously, based on adaptive dynamic programming (ADP) theory, a Critic-Actor neural network is designed to approximate the optimal control input, and a Control Barrier Function (CBF) is introduced into the control framework to handle the asymmetric safety constraints present in the system. This allows for precise tracking control while ensuring system safety, achieving a good balance between optimal performance, safety, robustness, and operability.

[0007] In a first aspect, the present invention provides an adaptive optimization control method for flight dual arms that takes into account safety constraints.

[0008] An adaptive optimization control method for a dual-arm flight system considering safety constraints includes: Based on the geometric parameters of the twin arms and the flight fuselage, a kinematic model of the flight twin arms is constructed. Based on the kinematic model, the Lagrange modeling method is used to model the system dynamics, and the dynamic equations of the grasped load are introduced to jointly construct the overall dynamic model of the system. Based on the overall dynamics model, a control obstacle function is designed, a performance index function is constructed, and the optimal control input is calculated by combining an evaluation-action network. Based on the optimal input, an RBF neural network is introduced as a feedforward term, and combined with sliding mode control, a nonlinear safety optimization controller and a network adaptive update law are constructed. The system collects the status feedback signals of the flight dual-arm system in real time, adjusts the network weights according to the adaptive update law, and generates control commands using a nonlinear safety optimization controller to control the flight body to track the trajectory.

[0009] Secondly, the present invention provides an adaptive optimization control system for a flight dual-arm that takes into account safety constraints.

[0010] An adaptive optimization control system for a dual-arm flight system considering safety constraints, comprising: The dual-arm flight model building module is used to build a kinematic model of the flight arms based on the geometric parameters of the dual arms and the flight body; The system overall model building module is used to perform system dynamics modeling based on kinematic models and using the Lagrange modeling method, and to introduce the dynamic equations of the grasped load to jointly construct the overall dynamics model of the system. An adaptive optimization controller building module is used to design control obstacle functions and construct performance index functions based on the overall dynamic model, and to calculate the optimal control input by combining an evaluation-action network. Based on the optimal input, an RBF neural network is introduced as a feedforward term, and combined with sliding mode control, a nonlinear safety optimization controller and a network adaptive update law are constructed. The flight tracking control module is used to collect the status feedback signals of the flight dual-arm system in real time, adjust the network weights according to the adaptive update law, and generate control commands using a nonlinear safety optimization controller to control the flight body to track the trajectory.

[0011] Thirdly, the present invention also provides an electronic device, comprising: a memory for storing executable instructions; and a processor for implementing the above-described adaptive optimization control method for flight dual arms considering safety constraints when executing the executable instructions stored in the memory.

[0012] Fourthly, the present invention also provides a computer-readable storage medium storing executable instructions for causing a processor to execute the executable instructions to implement the above-described adaptive optimization control method for flight dual arms considering safety constraints.

[0013] Fifthly, the present invention also provides a computer program product comprising executable instructions stored in a computer-readable storage medium; wherein, when the processor of an electronic device reads the executable instructions from the computer-readable storage medium and executes the executable instructions, it implements the above-mentioned adaptive optimization control method for flight dual arms considering safety constraints.

[0014] The above one or more technical solutions have the following beneficial effects: 1. This invention proposes an adaptive optimization control method and system for a dual-arm flight system that considers safety constraints. By studying the dual-arm flight system in the presence of a grasped object, an accurate kinematic model of the dual arms is established. In order to ensure the smooth and reliable control of the system and its robust response to internal and external disturbances, this invention incorporates the grasped object or load into the modeling framework of the dual-arm flight cooperative operating system and establishes a complete dynamic model of the system that includes load information, thereby effectively reducing the impact of the load on the overall performance of the system.

[0015] 2. Considering that existing research on dual-arm flying systems largely focuses on control accuracy while neglecting system optimality and safety, this invention, based on Adaptive Dynamic Programming (ADP) theory, designs a Critic-Actor neural network to approximate the optimal control input. Furthermore, a Control Barrier Function (CBF) is introduced into the control framework to handle asymmetric safety constraints. This results in a safety-optimized controller for dual-arm flying systems. This controller not only achieves high-precision trajectory tracking of flying bodies such as UAVs but also ensures system optimality and safety during flight, achieving a good balance between performance optimization, safety, robustness, and operability. Moreover, this control method can be extended to the field of large unmanned aerial vehicles, possessing significant research value and application prospects.

[0016] Advantages of additional aspects of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0017] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an improper limitation of the invention.

[0018] Figure 1 This is a model diagram of the flight dual-arm system in an embodiment of the present invention; Figure 2 This is a hardware platform diagram of the adaptive optimization control of the dual-arm flight configuration in this embodiment of the invention. Figure 3This is a flowchart illustrating the adaptive optimization control method for dual-arm flight proposed in this embodiment of the invention. Figure 4 This is a schematic diagram illustrating the change of the drone's position over time in a tracking experiment according to an embodiment of the present invention; Figure 5 This is a schematic diagram illustrating the change of UAV position error over time in a tracking experiment according to an embodiment of the present invention. Figure 6 This is a schematic diagram illustrating the change of the average / maximum position error of the UAV in the tracking experiment of this invention over time. Figure 7 This is a schematic diagram illustrating the change of load position over time in a tracking experiment according to an embodiment of the present invention; Figure 8 This is a schematic diagram illustrating the change in the attitude of a drone over time during a tracking experiment in an embodiment of the present invention. Figure 9 This is a schematic diagram illustrating the change of UAV control input over time in a tracking experiment according to an embodiment of the present invention; Figure 10 This is a structural block diagram of the adaptive optimization control system for flight dual arms proposed in an embodiment of the present invention. Detailed Implementation

[0019] It should be noted that the following detailed descriptions are exemplary and are intended only to describe specific embodiments and to provide further explanation of the invention, and are not intended to limit the scope of exemplary embodiments of the invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. Furthermore, it should be understood that when the terms "comprising" and / or "including" are used in this specification, they indicate the presence of features, steps, operations, devices, components, and / or combinations thereof.

[0020] Example 1 like Figure 1 As shown, this embodiment first proposes a dual-arm flying system. Its hardware structure includes a quadcopter drone, two robotic arms composed of servo motors and connectors, and a grasped object, enabling flexible operation and transportation of the load. To improve operational capabilities, the single-arm structure in existing research is replaced with a dual-arm structure. These two arms and grippers are composed of eight servo motors and connectors. The joint layout of the robotic arms mimics the joint configuration of a human arm, with each robotic arm having three joints, or three degrees of freedom, from top to bottom: shoulder-pitch joint, elbow-pitch joint, and elbow-roll joint. The servo motor models and their parameters corresponding to the joints are shown in Table 1 below.

[0021] Table 1 Joint and Servo Parameter Table

[0022] In this embodiment, the UAV in the dual-arm flight system is a quadcopter with a wheelbase of 550mm, equipped with an onboard computer NUC and a PixHawk flight control unit. The onboard computer and the flight control unit establish a communication connection through the MAVLINK protocol.

[0023] For the aforementioned dual-arm flight system, this embodiment proposes an adaptive optimization control method for dual-arm flight that considers safety constraints. The onboard computer in the UAV executes this adaptive optimization control method, calculates the control signal to achieve precise trajectory tracking of the UAV, and inputs the control signal into the driver of the flight control unit to drive the UAV to complete high-precision trajectory tracking.

[0024] The adaptive optimization control method for dual-arm flight that considers safety constraints proposed in this embodiment, such as... Figure 3 As shown, the specific steps include: Step S1: Based on the geometric parameters of the two arms and the flight body, construct the kinematic model of the flight arms.

[0025] Specifically, based on the geometric parameters of the system hardware structure, a multi-coordinate system is established to facilitate the subsequent model building. This multi-coordinate system includes: an inertial frame. Body coordinate system The joints of both arms and the coordinate system of the manipulated object (i.e., the load being grasped). The system's dual-arm configuration employs a three-degree-of-freedom serial structure of "shoulder-pitch, elbow-pitch, and elbow-roll"; simultaneously, a generalized coordinate vector is constructed. ,in Location of the drone. The UAV's Euler angle attitude. For the first The joint angles of a robotic arm.

[0026] Furthermore, by transforming the Euler angle derivative with angular velocity, the angular velocity of the aircraft is defined. With linear velocity Based on the Denavit-Hartenberg parametric method, the Jacobian matrix of each joint of the two arms is constructed, and the mapping relationship between the flight body (hereinafter referred to as the body) and joint velocities and generalized velocities is established as follows: ; in, This is the corresponding mapping matrix.

[0027] Step S2: Based on the kinematic model, the system dynamics model is modeled using the Lagrange modeling method, and the dynamic equations of the grasped load are introduced to jointly construct the overall dynamic model of the system.

[0028] Specifically, using the Lagrange modeling method and combining d'Alembert's principle, the total kinetic and potential energy of the system are calculated, and the dynamic equations of the system are established as follows: ; in, , This represents the total kinetic energy of the system. This represents the total potential energy of the system. To control the input, External disturbances For Jacobian matrices, This is the end force.

[0029] Based on this, the pose state of the manipulated object is defined as follows: The dynamic equation of the load is constructed as follows: ; in, These represent the inertia matrix, Coriolis force matrix, and gravity vector, respectively. This is the torque mapping matrix of the load end relative to the gripping point.

[0030] To ensure stable and reliable system control and robust response to internal and external disturbances, rigid grasping constraints are considered. The grasped object is further incorporated into the modeling framework of the flight dual-arm cooperative operating system, and the load is dynamically mapped into the system dynamics, resulting in the overall dynamic equations of the joint system: ; in, These represent the inertia matrix, Coriolis force matrix, and gravity vector of the complete system, including the load, respectively.

[0031] Step S3: Based on the overall dynamic model, design the control obstacle function, construct the performance index function, and calculate the optimal control input by combining the evaluation-action (Critic-Actor) network.

[0032] Specifically, based on the overall dynamic model constructed in the previous step, the following error vector is defined: ; in, Indicates the desired state. This indicates the system speed and the desired speed.

[0033] Furthermore, the definition of sorting is: ; in, It is a positive definite symmetric gain matrix.

[0034] Based on the error vector defined above, design a control barrier function. ,in Regarding the error The function, Given a positive definite diagonal gain matrix, the performance index function is constructed as follows: ; in, It is a positive definite symmetric matrix. For error The filtered signal, , This represents a positive definite function used to handle input saturation. For positive constant parameters, It is a positive definite matrix.

[0035] Furthermore, the Hamiltonian function is defined as follows: ; in, express gradient, For the drift vector field of error dynamics, For the input gain matrix, This is a disturbance term.

[0036] Furthermore, based on the theory of adaptive dynamic programming (ADP), a dual neural network with a Critic-Actor structure is used to approximate the cost function and gradient (i.e., approximate the performance index function constructed above), and the optimal control input is solved using the HJB equation to obtain the result. ; in, It is an estimate of the action network weight matrix. It is the activation function of the action network.

[0037] Step S4: Based on the optimal input, an RBF neural network is introduced as a feedforward term, and combined with sliding mode control, a nonlinear safety optimization controller and a network adaptive update law are constructed.

[0038] Specifically, based on the optimal input, a radial basis function neural network (RBF NN) is introduced as a feedforward term, and combined with sliding mode control, a complete nonlinear robust controller is designed, with the following specific form: ; in, It is a positive definite diagonal matrix. and These are the weight estimates and activation function of the RBF network, respectively.

[0039] Simultaneously, the adaptive laws for the Critic-Actor network and the RBF network are designed as follows: ; in, For positive definite parameter matrices / vectors, As an auxiliary vector, This is the nominal model matrix.

[0040] Step S5: Real-time acquisition of the state feedback signal of the flight dual-arm system, adjustment of network weights according to the adaptive update law, and generation of control commands using the nonlinear safety optimization controller to control the flight body to perform trajectory tracking.

[0041] Specifically, after completing the design of the above control law, the state feedback signals of the flight system (including position, attitude, linear velocity, angular velocity, joint angle, joint angular velocity, etc.) are acquired in real time. The weights of the evaluation-action network and the radial basis function network are adjusted according to the adaptive update law. Then, the nonlinear safety optimization controller is used to calculate the control signal for realizing high-precision trajectory tracking of the UAV. Under the drive of the control signal, the UAV's accurate trajectory tracking is completed.

[0042] To verify the effectiveness of the method proposed in this embodiment, a test experiment was conducted on this nonlinear robust tracking control method. The quadcopter UAV was equipped with a NUC onboard computer running a 64-bit Ubuntu 20.04 operating system. This onboard computer established a wireless connection with the ground station via 5G band Wi-Fi. The experimental platform is as follows: Figure 2 As shown. The main physical parameters of the flight dual-arm system are: ; .

[0043] Based on the above nonlinear safety optimization control method, the corresponding control gain is set as follows:

[0044] Figures 4-9 The results of the drone tracking a "spiral" trajectory are shown, with the drone's position corresponding to... The positional error is The location of the load is The drone's attitude is The control input is the resultant force. torque The blue dashed line represents the method of this embodiment, the yellow dotted line and the red dashed line represent two traditional methods, the black solid line is the desired trajectory, and the black dashed line is the upper and lower bounds of the error. As can be seen from the figure, the blue dashed line is closer to the black solid line, and in the error graph, the blue dashed line is completely enclosed within the black dashed line. Therefore, the nonlinear safety optimization control method proposed in this embodiment can successfully drive the UAV to track the desired trajectory and has higher tracking accuracy.

[0045] In summary, this embodiment proposes a physical structure for a flexible dual-arm flight operating system. By utilizing the designed controller, it can effectively suppress internal and external disturbances, including those from the load and the robotic arm, to achieve high-precision trajectory tracking of the UAV, while also possessing optimality, safety, robustness, and operability.

[0046] Example 2 This embodiment proposes an adaptive safety optimization control system for a flight dual-arm system, such as... Figure 10 As shown, it includes: The dual-arm flight model building module is used to build a kinematic model of the flight arms based on the geometric parameters of the dual arms and the flight body; The system overall model building module is used to perform system dynamics modeling based on kinematic models and using the Lagrange modeling method, and to introduce the dynamic equations of the grasped load to jointly construct the overall dynamics model of the system. An adaptive optimization controller building module is used to design control obstacle functions and construct performance index functions based on the overall dynamic model, and to calculate the optimal control input by combining an evaluation-action network. Based on the optimal input, an RBF neural network is introduced as a feedforward term, and combined with sliding mode control, a nonlinear safety optimization controller and a network adaptive update law are constructed. The flight tracking control module is used to collect the status feedback signals of the flight dual-arm system in real time, adjust the network weights according to the adaptive update law, and generate control commands using a nonlinear safety optimization controller to control the flight body to track the trajectory.

[0047] The examples and application scenarios implemented by the above modules and corresponding steps are the same, but are not limited to the content disclosed in Embodiment 1 above. 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.

[0048] The proposed system can be implemented in other ways. For example, the system embodiments described above are merely illustrative, and the division of modules described above is only a logical functional division. In actual implementation, there may be other division methods. For example, multiple modules may be combined or integrated into another system, or some features may be ignored or not executed.

[0049] Example 3 This embodiment provides an electronic device, including: a memory for storing executable instructions; and a processor for executing the executable instructions stored in the memory to implement the method provided in this embodiment.

[0050] Example 4 This embodiment also provides a computer-readable storage medium storing executable instructions, which, when executed by a processor, will cause the processor to execute the method described above in this embodiment.

[0051] Example 5 This embodiment provides a computer program product including executable instructions, which are computer instructions; the executable instructions are stored in a computer-readable storage medium. When the processor of an electronic device reads the executable instructions from the computer-readable storage medium and executes the executable instructions, the electronic device performs the method described in this embodiment.

[0052] The steps and methods involved in Embodiments 2 to 5 above correspond to those in Embodiment 1. For specific implementation details, please refer to the relevant description section of Embodiment 1. The term "computer-readable storage medium" should be understood as a single medium or multiple media including one or more instruction sets; it should also be understood as including any medium capable of storing, encoding, or carrying an instruction set for execution by a processor and enabling the processor to perform any of the methods in this invention.

[0053] Those skilled in the art will understand that the modules or steps of the present invention described above can be implemented using general-purpose computer devices. Optionally, they can be implemented using computer-executable program code, thereby allowing them to be stored in a storage device for execution by a computer device, or they can be fabricated as separate integrated circuit modules, or multiple modules or steps can be fabricated as a single integrated circuit module. The present invention is not limited to any particular combination of hardware and software.

[0054] The above description is only a preferred embodiment of the present invention. Although the specific implementation of the present invention has been described 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, based on the technical solution of the present invention, various modifications or variations that can be made by those skilled in the art without creative effort are still within the scope of protection of the present invention.

Claims

1. A flight dual-arm adaptive optimization control method considering safety constraints, characterized in that, include: Based on the geometric parameters of the twin arms and the flight fuselage, a kinematic model of the flight twin arms is constructed. Based on the kinematic model, the Lagrange modeling method is used to model the system dynamics, and the dynamic equations of the grasped load are introduced to jointly construct the overall dynamic model of the system. Based on the overall dynamics model, a control obstacle function is designed, a performance index function is constructed, and the optimal control input is calculated by combining an evaluation-action network. Based on the optimal input, an RBF neural network is introduced as a feedforward term, and combined with sliding mode control, a nonlinear safety optimization controller and a network adaptive update law are constructed. The system collects the status feedback signals of the flight dual-arm system in real time, adjusts the network weights according to the adaptive update law, and generates control commands using a nonlinear safety optimization controller to control the flight body to track the trajectory.

2. The adaptive optimization control method for dual-arm flight considering safety constraints as described in claim 1, characterized in that, The construction of the kinematic model of the flight arms includes: Establish a multi-coordinate system, including inertial frames. Body coordinate system The joints of both arms The coordinate system of the captured load ; The dual-arm configuration adopts a three-degree-of-freedom serial structure of shoulder-pitch, elbow-pitch, and elbow-roll; Constructing generalized coordinate vectors ,in Location of the drone. The UAV's Euler angle attitude. For the first The joint angles of a robotic arm; Based on the transformation relationship between the Euler angle derivative and angular velocity, the angular velocity of an aircraft is defined. With linear velocity ; Based on the Denavit-Hartenberg parameter method, the Jacobian matrix of each joint of the two arms is constructed, and the mapping relationship between the velocity of the flight body and joints and the generalized velocity is established.

3. The adaptive optimization control method for dual-arm flight considering safety constraints as described in claim 1, characterized in that, The construction of the overall dynamic model of the system includes: Using the Lagrange modeling method and d'Alembert's principle, the total kinetic and potential energy of the system are calculated, and the dynamic equations of the system are established as follows: ； in, , This represents the total kinetic energy of the system. This represents the total potential energy of the system. To control the input, External disturbances For Jacobian matrices, It is the end force; Define the pose state of the manipulated object as The dynamic equation of the load is constructed as follows: ； The object being manipulated refers to the load being captured. These represent the inertia matrix, Coriolis force matrix, and gravity vector, respectively. This is the torque mapping matrix of the load end relative to the gripping point; Considering rigid grasping constraints, the load is dynamically mapped to the system's dynamic equations, constructing the overall dynamic equations of the joint system as follows: ； in, These represent the inertia matrix, Coriolis force matrix, and gravity vector of the complete system, including the load, respectively.

4. The adaptive optimization control method for dual-arm flight considering safety constraints as described in claim 1, characterized in that, Design the optimal control input, including: Design control barrier function based on overall dynamics model Therefore, the performance index function is constructed as follows: ； in, Regarding the error The function, It is a positive definite diagonal gain matrix; It is a positive definite symmetric matrix. For error The filtered signal, , This represents a positive definite function used to handle input saturation. For positive constant parameters, It is a positive definite matrix; By employing an evaluation-action network to approximate the cost function and gradient, and combining it with an adaptive dynamic programming framework, the optimal control input is solved as follows: ； in, It is an estimate of the action network weight matrix. It is the activation function of the action network.

5. The adaptive optimization control method for dual-arm flight considering safety constraints as described in claim 1, characterized in that, The complete nonlinear safety optimization controller is represented in the form of a control law, as follows: ； in, For positive constant parameters, It is a positive definite matrix. It is an estimate of the action network weight matrix. It is the activation function of the action network. It is a positive definite diagonal matrix. and These are the weight estimates and activation function of the RBF network, respectively. For error The filtered signal.

6. A flight dual-arm adaptive optimization control system considering safety constraints, characterized in that, include: The dual-arm flight model building module is used to build a kinematic model of the flight arms based on the geometric parameters of the dual arms and the flight body; The system overall model building module is used to perform system dynamics modeling based on kinematic models and using the Lagrange modeling method, and to introduce the dynamic equations of the grasped load to jointly construct the overall dynamics model of the system. An adaptive optimization controller building module is used to design control obstacle functions and construct performance index functions based on the overall dynamic model, and to calculate the optimal control input by combining an evaluation-action network. Based on the optimal input, an RBF neural network is introduced as a feedforward term, and combined with sliding mode control, a nonlinear safety optimization controller and a network adaptive update law are constructed. The flight tracking control module is used to collect the status feedback signals of the flight dual-arm system in real time, adjust the network weights according to the adaptive update law, and generate control commands using a nonlinear safety optimization controller to control the flight body to track the trajectory.

7. The adaptive optimization control system for dual-arm flight considering safety constraints as described in claim 6, characterized in that, The complete nonlinear safety optimization controller is represented in the form of a control law, as follows: ； in, For positive constant parameters, It is a positive definite matrix. It is an estimate of the action network weight matrix. It is the activation function of the action network. It is a positive definite diagonal matrix. and These are the weight estimates and activation function of the RBF network, respectively. For error The filtered signal.

8. An electronic device, characterized in that, include: Memory, used to store executable instructions; The processor, when executing executable instructions stored in the memory, implements the adaptive optimization control method for flight dual arms considering safety constraints as described in any one of claims 1-5.

9. A computer-readable storage medium, characterized in that, The system stores executable instructions that, when executed by a processor, implement the adaptive optimization control method for flight dual arms considering safety constraints as described in any one of claims 1-5.

10. A computer program product, characterized in that, The computer program product includes executable instructions stored in a computer-readable storage medium; When the processor of the electronic device reads the executable instructions from the computer-readable storage medium and executes the executable instructions, it implements the adaptive optimization control method for flight dual arms considering safety constraints as described in any one of claims 1-5.

Images