Dynamic decoupling control system and method of omnidirectional flight platform with multiple mechanical arms

Abstract

The application discloses a multi-mechanical arm-mounted omni-directional flight platform dynamic decoupling control system and method, and belongs to the technical field of flight robot control. The system comprises a generalized dynamics modeling and real-time parameter identification module, a force / torque multi-dimensional decomposition and distribution module, a vector propulsion active compensation control module, a mechanical arm cooperative contact management module, an energy optimization and endurance extension module and a safety monitoring and fault processing module. The system firstly establishes and online updates a generalized dynamics model; secondly orthogonally decomposes the generalized force / torque into multiple force / torque components, and distributes them to corresponding actuators according to priorities; then realizes active compensation through quadratic programming thrust distribution and extended state observer, and realizes double-arm cooperative operation through impedance adaptive control and role dynamic switching; finally dynamically optimizes the gravity sharing ratio to minimize energy consumption. The application realizes active decoupling from the root of dynamics, and improves the system operation accuracy, stability and endurance.

Description

Technical Field

[0001] This invention relates to the field of flight robot control technology, specifically to a dynamic decoupling control system and method for an omnidirectional flight platform equipped with multiple robotic arms, which is particularly suitable for real-time compensation and energy optimization control of multi-source coupling forces / torques between the flight platform and the robotic arms in complex high-altitude operation scenarios. Background Technology

[0002] Omnidirectional flying robots have become a research hotspot in the field of aerial work robots in recent years. Compared with traditional multi-rotor drones, omnidirectional flying platforms, by configuring tiltable or vectorable propulsion units, can generate thrust in any spatial direction and have independent control over six degrees of freedom of posture. After mounting a robotic arm, such platforms can perform high-risk and high-difficulty high-altitude operations such as wind turbine blade maintenance, power line inspection and repair, bridge structural flaw detection, and maintenance of large petrochemical equipment, and have significant engineering application value.

[0003] However, when an omnidirectional flight platform is equipped with two robotic arms to perform contact or interactive operations, the system faces extremely complex multi-source coupling problems: (1) Inertial coupling and dynamic disturbance: The Coriolis force, centrifugal force and inertial torque generated by the joint movement of the robotic arms are directly transmitted to the flight platform, causing attitude disturbances of the aircraft. Especially when the two arms move at the same time, the joint inertia matrix of the two arm systems has strong nonlinear time-varying characteristics, and traditional PID controllers are difficult to achieve effective compensation. (2) Contact force impact and transmission: When the end of the robotic arm establishes physical contact with the work object, the contact force / torque is transmitted to the flight platform through the robotic arm linkage. If there is no effective buffering mechanism, it will cause uncontrollable attitude changes of the aircraft, which may lead to a crash in severe cases. (3) High energy consumption due to gravity compensation: During flight operations, the propulsion system needs to continuously output a large amount of thrust to overcome the weight of the entire aircraft (including the arm load). The long-term high-power operation of the propulsion unit leads to a significant increase in energy consumption, which greatly limits the endurance and work efficiency. (4) Incomplete force / torque decoupling: Existing control methods based on centralized dynamic models often treat the flight platform and the robotic arm as a unified rigid body, ignoring the differences in their dynamic characteristics during operation. This results in the mixing of effective and ineffective components in the control commands, reducing control accuracy and safety.

[0004] Therefore, there is an urgent need for a new control scheme that can actively decouple the dynamics of the platform and the robotic arm, optimize system energy consumption, and ensure operational safety. Summary of the Invention

[0005] This invention aims to overcome the shortcomings of existing technologies and provide a dynamic decoupling control system and method for an omnidirectional flight platform equipped with multiple robotic arms. This method can orthogonally decompose the complex coupling forces / torques acting on the system and actively allocate and coordinate them based on the physical characteristics of each actuator, thereby fundamentally reducing coupling interference, improving operational accuracy and stability, and significantly reducing system energy consumption by optimizing gravity distribution.

[0006] In a first aspect, the present invention provides a dynamic decoupling control system for an omnidirectional flight platform equipped with multiple robotic arms, comprising: a force / torque multidimensional decomposition and allocation module, used to receive overall system status information, orthogonally decompose generalized force / torque into multiple types of force / torque components, and allocate them to corresponding actuators according to preset priorities; a vector propulsion active compensation control module, used to actively counteract the dynamic disturbances caused by the movement of the robotic arms to the flight platform by adopting a disturbance observation and compensation mechanism; and a robotic arm cooperative contact management module, used to dynamically manage the working state of the robotic arms.

[0007] Optionally, the force / torque multidimensional decomposition and allocation module orthogonally decomposes the generalized force / torque into multiple force / torque components using a pseudo-inverse projection algorithm of the task Jacobian matrix. These components include: effective working force / torque for performing end-effector tasks, inertial compensation force / torque generated by the movement of the robotic arm, gravity balancing force / torque for balancing the overall weight of the machine, and contact buffer force / torque generated by the contact between the robotic arm and the work object. The preset priority of the force / torque multidimensional decomposition and allocation module is as follows: contact buffer force / torque is preferentially borne by the fixed arm, effective working force / torque is output by the working arm, inertial compensation force / torque is compensated by the vector thruster array, and gravity balancing force / torque is jointly borne by the vector thruster array and the fixed arm.

[0008] Optionally, the disturbance observation and compensation mechanism employs a quadratic programming thrust allocation algorithm and an extended state observer based on active disturbance rejection control. The extended state observer based on active disturbance rejection control is used to calculate in real time high-frequency disturbances in the whole machine that are not covered by the generalized dynamic model, and obtain disturbance estimates. The quadratic programming thrust allocation algorithm takes the minimum total power consumption of the vector thruster array as the optimization objective. The constraints include the total equivalent force / torque generated by the vector thruster array must be equal to the sum of the inertial compensation force / torque component and the portion of the gravity balance force / torque component it shares, the upper and lower limits of the thrust amplitude of each vector thruster, and the limit of the thrust vector angle change rate. The disturbance estimates are input as feedforward compensation quantities into the quadratic programming thrust allocation algorithm to enhance the omnidirectional flight platform's ability to suppress high-frequency dynamic disturbances.

[0009] Optionally, the robotic arm collaborative contact management module includes a fixed arm control unit and a working arm control unit; the fixed arm control unit is used to receive the contact buffer force / torque allocated by the force / torque multidimensional decomposition and allocation module, and control at least one robotic arm as a fixed arm to maintain contact with the work object in a structural constraint manner to absorb impact and provide support; the working arm control unit is used to receive the effective working force / torque allocated by the force / torque multidimensional decomposition and allocation module, and control the remaining robotic arms as working arms to perform the target operation task.

[0010] Optionally, the robotic arm in the collaborative contact management module adopts an impedance adaptive control strategy. The desired end stiffness of the robotic arm decreases exponentially as the real-time contact force between the end effector and the work object increases. The desired end damping of the robotic arm is automatically adjusted according to the critical damping condition.

[0011] Optionally, the robotic arm collaborative contact management module further includes a role dynamic switching unit, used to determine and execute the role swap between the fixed arm and the working arm according to preset trigger conditions; the trigger conditions include: a forced switching condition triggered when the torque of any joint of the fixed arm exceeds the rated preset threshold, a task switching condition triggered according to the upper-level task planning instruction, and a predictive switching condition triggered when the current fixed arm end pose is predicted to exceed the effective workspace within a preset time based on kinematic prediction; during the switching transition, a dual-fixed short-time bridging strategy is executed, controlling the robotic arm to maintain contact with the work object simultaneously and each share part of the nominal contact force, and after the center displacement and attitude angle deviation of the omnidirectional flight platform are both less than the preset safety threshold, the contact of the original fixed arm is released, and the role swap is completed.

[0012] Optionally, it also includes a safety monitoring and fault handling module, which includes a generalized momentum observer for real-time calculation of the weighted L2 norm of the generalized momentum residual; when the weighted L2 norm of the residual continuously exceeds the residual preset threshold determined by Monte Carlo simulation, a safety response sequence including robotic arm joint locking, vector thrust derating, and guided autonomous landing is triggered.

[0013] Optionally, it also includes: a generalized dynamics modeling and real-time parameter identification module, which is used to establish a generalized dynamics model including an omnidirectional flight platform and two robotic arms, and to identify inertia parameters online in real time using a recursive least squares method with a forgetting factor.

[0014] Optionally, it also includes an energy optimization and range extension module. The energy optimization and range extension module adopts a single-variable optimization algorithm based on the golden section search method. With the minimization of the total power consumption of the vector thruster array as the objective function, and under the premise of satisfying the safety constraints of the fixed arm joint torque, it solves the optimal gravity sharing ratio coefficient online in real time to dynamically optimize the gravity sharing ratio between the fixed arm and the vector thruster array.

[0015] A second aspect of the present invention provides a dynamic decoupling control method for an omnidirectional flight platform equipped with a robotic arm, applied to the aforementioned dynamic decoupling control system for an omnidirectional flight platform equipped with multiple robotic arms, comprising the following steps: receiving overall system status information through a force / torque multidimensional decomposition and allocation module, orthogonally decomposing generalized forces / torques into multiple types of force / torque components, and allocating them to corresponding actuators according to preset priorities; actively offsetting dynamic disturbances caused to the flight platform by the robotic arm movement through a vector propulsion active compensation control module using a disturbance observation and compensation mechanism; and dynamically managing the working state of the robotic arm through a robotic arm collaborative contact management module.

[0016] Compared with the prior art, the beneficial effects of the present invention are as follows.

[0017] 1. Active decoupling enhances control precision and stability: This invention decomposes generalized forces / torques into four types of force / torque components based on their physical origins: effective working force / torque, inertial compensation force / torque, gravity balance force / torque, and contact buffer force / torque. Precise separation is achieved through a pseudo-inverse projection algorithm of the task Jacobian matrix. This design fundamentally severs the coupling path between the robotic arm's motion and the omnidirectional flight platform's attitude, allowing the omnidirectional flight platform to focus solely on its own stability control while the robotic arm can independently perform precise operations. This significantly improves flight attitude stability and end-effector accuracy. Simulation results demonstrate that the decoupling accuracy is approximately 40% better than existing centralized control methods.

[0018] 2. Multi-source disturbance collaborative suppression significantly enhances system robustness: For inertial disturbances generated by the robotic arm's motion, this invention utilizes a vector propulsion active compensation control module and a quadratic programming thrust allocation algorithm to generate optimal thrust commands, actively offsetting their impact on the omnidirectional flight platform. For impact disturbances generated by the robotic arm's contact with the work object, the fixed arm absorbs and buffers them through structural constraints, preventing their transmission to the omnidirectional flight platform. The integrated extended state observer based on active disturbance rejection control effectively improves system robustness and operational safety through real-time estimation and feedforward compensation. Simulation verification shows that the quadratic programming thrust allocation algorithm based on the interior point method has a single solution time of <0.5 ms, meeting the 200 Hz control frequency requirement, and exhibits superior power allocation performance compared to existing pseudo-inverse methods.

[0019] 3. Dynamic Optimization of Gravity Distribution, Significantly Extending Endurance: This invention fully utilizes the structural advantages of dual robotic arms, employing a combined gravity distribution strategy of "fixed arm and vector thruster array." By leveraging the mechanical connection of the fixed arm to the work object, it provides additional support torque to the flight system, reducing the gravity compensation burden on the vector thrusters and significantly extending endurance. Simultaneously, through energy optimization and endurance extension modules, the optimal gravity distribution ratio is solved online using the golden section search method. While meeting the safety constraints of the fixed arm joint torque, it maximizes the use of the fixed arm's mechanical connection to the work object to distribute the overall weight of the aircraft, thereby minimizing system energy consumption. Simulation verification shows that under effective fixed arm constraints, system propulsion power consumption is reduced by 18%~35%, and endurance is extended by 22%~51%, depending on the mission type and load conditions.

[0020] 4. Intelligent Collaborative Operation of Robotic Arms, Expanding Task Execution Boundaries: The robotic arm collaborative contact management module designed in this invention supports dynamic role switching between the fixed arm and the working arm, covering three types of triggering conditions: forced switching, task switching, and predictive switching. A dual-fixed short-time bridging strategy ensures a smooth and undisturbed switching process. The working arm employs an impedance adaptive control strategy, where the desired end-effector stiffness decreases exponentially with increasing real-time contact force, and the desired end-effector damping is automatically adjusted according to critical damping conditions to achieve compliant operation on objects with varying stiffness. This impedance adaptive control strategy reduces the peak contact impact force by approximately 60% compared to fixed-parameter impedance control, significantly improving operational safety. This design enables the system to perform complex tasks that a single arm cannot achieve, such as heavy-load handling and high-precision assembly, significantly expanding the task execution boundaries.

[0021] 5. Full-process safety monitoring ensures zero operational risk: This invention introduces a safety monitoring and fault handling module based on a generalized momentum observer. By calculating the weighted L2 norm of the generalized momentum residual in real time, it achieves accurate monitoring of the system status. When the weighted L2 norm of the residual continuously exceeds the preset threshold determined by Monte Carlo simulation, a safety response sequence including robotic arm joint locking, thruster derating, and autonomous descent is immediately triggered, providing reliable assurance for the safe operation of the system in high-altitude working environments. The fault detection response time is <50 ms, and the false alarm rate is <0.1%, meeting the safety requirements for high-altitude operations.

[0022] 6. The system has good scalability: the control framework has a modular design, which can be easily expanded to omnidirectional flight platforms with different degrees of freedom robotic arm configurations and different numbers of vector thrusters. Attached Figure Description

[0023] Figure 1 This is a system framework diagram of the present invention.

[0024] Figure 2 This is the overall flowchart of the present invention.

[0025] Figure 3 This is a schematic diagram of the omnidirectional flight platform equipped with dual robotic arms according to the present invention.

[0026] Figure 4 This is a schematic diagram of an electronic device.

[0027] Figure reference numerals: 1. Omnidirectional flight platform with dual robotic arms; 2. Vector thruster; 3. Fixed arm; 4. Working arm; 5. Monitoring mechanism; 6. Main control cabin. Detailed Implementation

[0028] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0029] Example 1

[0030] See Figure 1This application provides a dynamic decoupling control system for an omnidirectional flight platform equipped with multiple robotic arms. It mainly includes a generalized dynamics modeling and real-time parameter identification module, a force / torque multidimensional decomposition and allocation module, a vector propulsion active compensation control module, a robotic arm collaborative contact management module, an energy optimization and endurance extension module, and a safety monitoring and fault handling module. The core of this system lies in constructing a complete closed loop from "perception-decision-execution-optimization-safety". First, the system identifies inertia parameters online in real-time through the generalized dynamics modeling and real-time parameter identification module. The force / torque multidimensional decomposition and allocation module decomposes the total generalized force / torque of the system into four orthogonal components based on different physical meanings (operation, inertia, gravity, contact), and allocates them to the corresponding actuators according to priority. The vector propulsion active compensation control module employs a disturbance observation and compensation mechanism, utilizing the vector thruster array of the omnidirectional flight platform to actively generate or cancel specific force / torque components to compensate for and isolate the dynamic disturbances caused by the robotic arm movement to the omnidirectional flight platform, thereby achieving decoupling of system dynamics at the aircraft level. The contact management module, which works in conjunction with the robotic arm, absorbs the contact buffer force / torque through the fixed arm, which acts as a "mechanical anchor." The effective working force / torque is executed by the working arm, which employs an impedance adaptive control strategy for compliant operation and supports dynamic role switching based on multiple triggering conditions. The energy optimization and endurance extension module is responsible for distributing the continuous load (gravity) collaboratively between the vector thruster array and the fixed arm, and optimizes the gravity distribution ratio online using the golden ratio search method to minimize the total power consumption of the vector thruster array. Furthermore, the system is equipped with a safety monitoring and fault handling module, which monitors the overall system status in real time through a generalized momentum observer and triggers a safety protection sequence when an anomaly is detected.

[0031] The generalized dynamics modeling and real-time parameter identification module is used to establish a generalized dynamics model that includes an omnidirectional flight platform and two robotic arms, and to identify inertia parameters online in real time using a recursive least squares method with a forgetting factor.

[0032] In this embodiment, taking an omnidirectional flight platform equipped with dual robotic arms as an example (it should be noted that the number of robotic arms involved in this invention is ≥2, while the number of robotic arms in this embodiment is 2), the specific implementation process of the generalized dynamics modeling and real-time parameter identification module for online real-time identification of inertia parameters is described, including the following steps.

[0033] 1. Establish a generalized dynamic model including the omnidirectional flight platform and the two robotic arms, with the center of mass of the omnidirectional flight platform as the origin of the coordinate system. Let the generalized coordinate vector of the system be... Where q is the generalized coordinate vector of the omnidirectional flight platform, and T is the matrix rank transformation symbol. The center of gravity of the omnidirectional flight platform. For Euler angle orientation, These are the joint angle vectors of the two robotic arms, respectively. These represent the degrees of freedom of the two robotic arms.

[0034] 2. The generalized dynamic model is as follows: ,in, For the generalized inertia matrix, The Coriolis force / centrifugal force matrix, The gravity vector To access the Jacobian matrix, In order to access generalized force, This is the generalized control torque vector. , Let be the velocity vector in the generalized coordinate system of the omnidirectional flight platform. The acceleration vector in the generalized coordinates of the omnidirectional flight platform.

[0035] 3. Targeting To address the problem of unknown mass and inertia parameters due to load variations, a method based on recursive least squares combined with a forgetting factor is employed. Online parameter identification algorithm:

[0036] ,in, This is the inertia parameter estimation vector at the current time k. This is the inertia parameter estimation vector for the previous time step k-1. Let k be the gain matrix at the current time. These are joint torque measurements. The regression vector is constructed from joint acceleration and angular velocity, and T is the matrix rank transformation symbol. Let be the covariance matrix of the inertia parameter estimation vector at the current time k. This is the covariance matrix of the inertia parameter estimation vector at the previous time k-1, where I is the identity matrix (e.g., a square matrix with 1s on the main diagonal and 0s elsewhere), and λ is the forgetting factor, typically between 0.95 and 1. The identification frequency should be no less than 500. This ensures that the parameter estimation lag does not exceed a single control cycle.

[0037] The force / torque multidimensional decomposition and allocation module is used to receive the overall machine status information, orthogonally decompose the generalized force / torque into multiple types of force / torque components, and allocate them to the corresponding actuators according to preset priorities.

[0038] The force / torque multidimensional decomposition and allocation module orthogonally decomposes the generalized force / torque into multiple force / torque components using a pseudo-inverse projection algorithm of the task Jacobian matrix. These components include: effective working force / torque for performing end-effector tasks, inertial compensation force / torque generated by the movement of the robotic arm, gravity balancing force / torque for balancing the overall weight of the machine, and contact buffer force / torque generated by the contact between the robotic arm and the work object. The preset priority of the force / torque multidimensional decomposition and allocation module is as follows: contact buffer force / torque is preferentially borne by the fixed arm, effective working force / torque is output by the working arm, inertial compensation force / torque is compensated by the vector thruster array, and gravity balancing force / torque is jointly borne by the vector thruster array and the fixed arm.

[0039] The force / torque multidimensional decomposition and allocation module decomposes the generalized force / torque acting on the system into four types of force / torque components, forming an orthogonal decomposition framework: (1) effective working force / torque (1) The actual output force / torque required for the end effector to complete the task, which is the target component; (2) Inertial compensation force / torque : The inertial coupling disturbance caused by the movement of the robotic arm needs to be actively canceled; (3) Gravity balance force / torque The component required to counteract the overall weight of the machine is jointly borne by the vector thruster array and the structural constraints of the robotic arm; (4) Contact buffer force / torque The impact component generated by the contact between the robotic arm and the work object is preferentially absorbed by the constraints of the fixed arm structure.

[0040] The decomposition process employs a pseudo-inverse projection algorithm of the task Jacobian matrix, defining the task space force spinor. Decoupling is achieved through the projection matrices of each subspace: ,in, For the mission space force spinor, For the mission space force vector, The mission space torque vector, For the task Jacobian matrix, the Moore-Penrose pseudoinverse is given. and These are the velocity and acceleration vectors of the robotic arm, respectively. It is the gravity vector. Contact force relative nominal value The deviation amount.

[0041] The above four types of force / torque components are allocated to the execution layer according to the following priorities: Priority 1 (highest): Contact buffer force / torque is borne by the fixed arm structure constraint; Priority 2: Effective working force / torque is actively controlled and output by the working arm; Priority 3: Inertial compensation force / torque is compensated by the vector thruster array; Priority 4: Gravity balance force / torque is jointly borne by the vector thruster array and the fixed arm.

[0042] The vector propulsion active compensation control module is used to actively counteract the dynamic disturbances caused by the movement of the robotic arm to the flight platform by employing a disturbance observation and compensation mechanism.

[0043] The disturbance observation and compensation mechanism employs a quadratic programming thrust allocation algorithm and an extended state observer based on active disturbance rejection control (ADRC). The extended state observer based on ADRC is used to calculate in real time high-frequency disturbances in the entire system that are not covered by the generalized dynamic model, and obtain disturbance estimates. The quadratic programming thrust allocation algorithm aims to minimize the total power consumption of the vector thruster array. The constraints include the total equivalent force / torque generated by the vector thruster array being equal to the sum of the inertial compensation force / torque component and the portion of the gravity balance force / torque component it shares, the upper and lower limits of the thrust amplitude of each vector thruster, and the limit on the rate of change of the thrust vector angle. The disturbance estimates are input as feedforward compensation quantities into the quadratic programming thrust allocation algorithm to enhance the omnidirectional flight platform's ability to suppress high-frequency dynamic disturbances.

[0044] The inertial disturbances caused by the movement of the robotic arm are actively counteracted by the vector propulsion active compensation control module. The specific implementation process includes the following steps.

[0045] The omnidirectional flight platform is equipped with a vector thruster array, such as m vector thrusters (m≥4), and the thrust direction of each vector thruster can be continuously adjusted within a certain range. The thrust spin of each vector thruster is: ,in, The thrust spin of each rotor, For scalar thrust magnitude, The unit direction vector is determined by the inclination angle and azimuth angle. Vectoring the installation position of the thrust vectoring unit. Total thrust spin of the system. for: ,in, Configure matrices for each thruster direction, where m is the number of vector thrusters. This is the thrust amplitude vector.

[0046] A thrust allocation algorithm based on quadratic programming (QP) is adopted, with the objective function of minimizing propulsion power consumption, while simultaneously satisfying force / torque compensation requirements. Where α is the rotor tilt angle and β is the rotor azimuth angle. This represents the maximum change in the rotor's tilt angle. This represents the maximum change in the rotor's azimuth angle. Let be the power consumption weighting coefficient for the i-th vector thruster. Let i be the thrust of the i-th propulsion unit. The compensating force spindle that needs to be borne by the vector thruster The gravity sharing ratio coefficient (dynamically adjusted by the energy optimization and range extension module). , These are the lower and upper limits of scalar thrust, respectively. , These are the tilt angle and azimuth angle of the rotor, respectively.

[0047] The quadratic programming problem is relatively small in scale (3m variables and 6+2m constraints). It is solved using the interior point method, and the single solution time on the embedded processor is less than 0.5 ms, which meets the real-time control requirements (control frequency ≥ 200 Hz).

[0048] Furthermore, for rapidly varying disturbances (bandwidth > 20 Hz), an extended state observer (ESO) based on active disturbance rejection control (ADRC) is superimposed on the quadratic programming compensation to estimate and feedforward compensate for the residual disturbance, further improving the high-frequency suppression performance. ,in, These are the derivatives of the flight state x and the state derivative, respectively. And the estimation of total disturbance, All are the gains of the extended state observer (and are positive scalars). The unit is S -1 This determines the strength of the extended state observer's correction to the state estimation error, thus affecting the convergence speed of the state estimation. The unit is S -2 This is used to determine the strength of the correction for the state derivative estimation error by the extended state observer. The unit is S -3 This is used to determine the strength of the correction for the total disturbance estimation error by the extended state observer, directly affecting the speed and accuracy of disturbance observation and compensation. Tuning according to the bandwidth parameter of the extended state observer It is a nonlinear function. and Let be a nonlinear factor, where satisfies 0 < < The condition is less than 1, where δ is the width of the linear interval. To control the input, For input gain, , and They are respectively The corresponding rate of change.

[0049] The robotic arm collaborative contact management module is used to dynamically manage the working status of the robotic arm.

[0050] The robotic arm collaborative contact management module includes a fixed arm control unit and a working arm control unit. The fixed arm control unit receives the contact buffer force / torque allocated by the force / torque multidimensional decomposition and allocation module, and controls at least one robotic arm as a fixed arm to maintain contact with the work object in a structural constraint manner to absorb impact and provide support. The working arm control unit receives the effective working force / torque allocated by the force / torque multidimensional decomposition and allocation module, and controls the remaining robotic arms as working arms to perform the target operation task.

[0051] In this embodiment, an omnidirectional flight platform equipped with two robotic arms is used as an example. When the end effector of one of the robotic arms (designated as the fixed arm) is fixed to the work object by joint locking, a closed-loop constraint chain is formed. Let the velocity spin of the work object in the world coordinate system be... Then the constraint equation at the end of the fixed arm is: ,in, For a fixed-arm Jacobian array, To fix the joint angles and angular velocities of the arm, For the Jacobian movement of the omnidirectional flight platform to its end point, For the velocity spinor of the flight platform.

[0052] The upper bound of the spin of the support force that the fixed arm can provide to the omnidirectional flight platform under constrained conditions is: ,in, For the maximum support force spindle, To convert the fixed-arm Jacobian matrix to a transpose, the operational space force spinor is mapped back to the joint space torque. This represents the maximum output torque vector for each joint. Here is the joint space inertia matrix of the fixed arm. Let be the angular acceleration of each joint. The Coriolis force / centrifugal force matrix, This is the gravity term vector of the fixed arm. This support force spinor is used to offset part of the gravity and buffer contact impacts, and coordinates with the vector thrusters through the energy optimization and range extension module to share the overall weight of the aircraft.

[0053] The robotic arm in the collaborative contact management module adopts an impedance adaptive control strategy. The desired end stiffness of the robotic arm decreases exponentially as the real-time contact force between the end effector and the work object increases. The desired end damping of the robotic arm is automatically adjusted according to the critical damping condition.

[0054] For the end effector of the boom, an impedance adaptive control strategy is used to establish a virtual spring-damped-mass model: ,in, For end position error, , These represent the acceleration and rate of change of the end-position error, respectively. These represent the expected virtual mass, damping, and stiffness matrix, respectively. It is an external contact force.

[0055] This invention proposes a parameter adaptive mechanism that dynamically adjusts the impedance parameters based on the magnitude of the contact force and the stage of operation. ,in, Let be the stiffness of the end effector of the work arm at time t. For nominal stiffness, This is the stiffness attenuation coefficient (the greater the contact force, the lower the stiffness, achieving compliant contact). Let ζ be the expected damping matrix at time t at the end of the operation, and let ζ be the damping ratio (taken as 0.7~1.0 to ensure stability). This represents the desired virtual mass matrix of the end effector. This mechanism enables the system to maintain high stiffness and accurate tracking during free-space motion, and automatically softens upon contact, effectively suppressing peak contact impact forces.

[0056] The robotic arm collaborative contact management module also includes a role dynamic switching unit, which is used to determine and execute the role swap between the fixed arm and the working arm according to preset trigger conditions. The trigger conditions include: a forced switching condition triggered when the torque of any joint of the fixed arm exceeds the rated preset threshold, a task switching condition triggered according to the upper-level task planning instructions, and a predictive switching condition triggered when the current fixed arm end pose is predicted to exceed the effective workspace within a preset time based on kinematic prediction. During the switching transition, a dual-fixed short-time bridging strategy is executed to control the robotic arm to maintain contact with the work object at the same time and share part of the nominal contact force. After the center displacement and attitude angle deviation of the omnidirectional flight platform are both less than the preset safety threshold, the contact of the original fixed arm is released, and the role swap is completed.

[0057] For the dual-arm role switching protocol, a set of robotic arm system working states is defined, including the robotic arm's fixed state, working state, and transition state. The switching decision is based on the following conditional logic:

[0058] Triggering condition 1 (forced switching): When any component of the current fixed boom joint torque exceeds 85% of the rated torque, active switching is triggered, and the original boom switches to the fixed pre-contact control sequence;

[0059] Triggering condition 2 (task switching): The job task planning layer issues a switching command and executes according to the preset switching sequence;

[0060] Triggering condition 3 (prediction switch): Based on the robot arm's kinematics prediction, the current fixed position will... If the reachability constraint is not met within the time limit, the pre-switching action is initiated in advance. The time is determined by the system's safety margin, with a default of 2 seconds.

[0061] During the switching transition, the controller executes a dual fixed short-time bridging strategy: both robotic arms are simultaneously in a light contact constraint state with the work object, each applying 50% of the nominal contact force, ensuring that the system's center of gravity displacement is less than 5 cm and the attitude deviation is less than 2° throughout the switching process.

[0062] The energy optimization and range extension module adopts a single-variable optimization algorithm based on the golden section search method. With the minimization of the total power consumption of the vector thruster array as the objective function, and under the premise of satisfying the safety constraints of the fixed arm joint torque, it solves the optimal gravity sharing ratio coefficient online in real time to dynamically optimize the gravity sharing ratio between the fixed arm and the vector thruster array.

[0063] This invention proposes an energy collaborative optimization framework based on Lyapunov stability assurance. The core idea is to maximize the proportion of gravity sharing by the fixed arm constraint while satisfying flight stability constraints, thereby reducing the energy consumption of the vector thruster array.

[0064] In this embodiment, an omnidirectional flight platform equipped with dual robotic arms is used as an example for explanation, and the total power consumption of the system is defined as follows: To minimize the total system power, the gravity contribution ratio coefficient is solved online. .in, The total power of the system, The total power consumption of the vector thruster array. and The total power of the two robotic arms. For efficiency coefficient, Let m be the thrust of the i-th propulsion unit, and m be the number of vector thrusters. include and , Let j be the transpose of the angular velocity vector of the j-th robotic arm joint. Let J be the actual output torque vector of each joint of the j-th robotic arm. Let β be the gravity sharing ratio coefficient, and let β be the domain of the optimization variable. Let be the total system power under the current β. The total generalized force spinor actually generated by the system. To maintain the stable hovering and operational performance of the omnidirectional flight platform, the target generalized force spinor is required, where ε is the upper bound of the allowable error of the force balance constraint. This represents the actual torque vector that each joint of the fixed arm needs to output under the current β. This represents the rated maximum output torque vector of each joint actuator in the fixed arm. The above single-variable constrained optimization problem is solved online using the golden section search method, which has extremely low computational cost and can be solved within 100... Completed within the specified time.

[0065] Theoretical analysis and simulation verification show that in typical high-altitude welding operation scenarios (when the fixed arm joint torque margin is sufficient), this optimization strategy can reduce the total power consumption of the thruster by 18% to 35%, and correspondingly extend the endurance by about 22% to 51% (depending on the task load).

[0066] The system also includes a safety monitoring and fault handling module, which contains a generalized momentum observer for real-time calculation of the weighted L2 norm of the generalized momentum residual. When the weighted L2 norm of the residual continuously exceeds the preset threshold of the residual determined by Monte Carlo simulation, a safety response sequence is triggered, including robotic arm joint locking, vector thrust derating, and guidance for autonomous landing.

[0067] To detect accidental collisions and malfunctions in robotic arms or omnidirectional flight platforms, this invention introduces a Generalized Momentum Observer (GMO): ,in, Let be the residual vector at time t. The positive definite observation gain matrix is... Let be the generalized momentum vector at time t. for The generalized momentum vector at time t. This represents the total generalized control torque vector of the system. For the generalized momentum form of the Coriolis force / centrifugal force, Let p be the gravity vector, p be the generalized momentum vector, and r be the residual vector. During normal operation, →0; When an unexpected collision or parameter mutation occurs... It will deviate rapidly from zero.

[0068] The weighted L2 norm of the residual vector is used as a fault indicator: ,in, This is a fault detection metric, specifically the weighted L2 norm of the residual vector. The residual weighting matrix, This is the transpose of the residual vector of the generalized momentum observer at time t. A threshold is set. (Determined by Monte Carlo simulation, guaranteeing a false alarm rate of <0.1%) > If the duration exceeds 20ms, a safety protection sequence is triggered (arm joint lock-up → thrust reduction → autonomous descent).

[0069] By cyclically executing the above steps, this invention achieves dynamic decoupling of the flight platform and the dual robotic arm system, high-precision operation, and high-energy-efficiency operation, providing a reliable technical solution for complex aerial operation tasks.

[0070] Example 2

[0071] This embodiment, based on Embodiment 1 above, provides a dynamic decoupling control method for an omnidirectional flight platform equipped with a robotic arm. Please refer to [link to previous document]. Figure 2 The following is a specific implementation step of a dynamic decoupling control system for an omnidirectional flight platform equipped with multiple robotic arms, applied in Embodiment 1.

[0072] S1. Receive the overall machine status information through the force / torque multidimensional decomposition and allocation module, orthogonally decompose the generalized force / torque into multiple types of force / torque components, and allocate them to the corresponding actuators according to the preset priority.

[0073] S2. The vector propulsion active compensation control module adopts a disturbance observation and compensation mechanism to actively counteract the dynamic disturbances caused by the movement of the robotic arm to the flight platform.

[0074] S3. Dynamically manage the working status of the robotic arm through the robotic arm collaborative contact management module.

[0075] In this embodiment, taking an omnidirectional flight platform equipped with dual robotic arms as an example, the dynamic decoupling control method of an omnidirectional flight platform equipped with robotic arms is further explained. The specific application scenario is the replacement operation of insulators for power transmission lines.

[0076] In one embodiment of this application, the method further includes: establishing a generalized dynamic model including an omnidirectional flight platform and dual robotic arms through a generalized dynamic modeling and real-time parameter identification module, and identifying inertia parameters online in real time using a recursive least squares method with a forgetting factor; and using a single-variable optimization algorithm based on the golden section search method through an energy optimization and endurance extension module, with the goal of minimizing the total power consumption of the vector thruster array, and under the premise of satisfying the safety constraints of the fixed arm joint torque, solving the optimal gravity sharing ratio coefficient online in real time, so as to dynamically optimize the gravity sharing ratio between the fixed arm and the vector thruster array.

[0077] Work Scenario: Insulator replacement work on a high-voltage transmission line with a voltage of 500kV, located 30 meters above the ground. The work requires an omnidirectional flying platform equipped with dual robotic arms to remove the old insulators and install the new ones.

[0078] Please see Figure 3System configuration: An omnidirectional flight platform 1 equipped with dual robotic arms, featuring six independently controllable vector thrusters 2. Two robotic arms with at least six degrees of freedom are mounted on the omnidirectional flight platform, each equipped with joint torque sensors, encoders, and a six-dimensional force / torque sensor (not shown) at the end effector. The flight controller and robotic arm controller are integrated into a high-performance embedded computing unit. The end effector of the fixed arm 3 is equipped with an insulated suction cup (not shown), and the end effector of the working arm 4 is equipped with an insulated gripper and a six-dimensional force / torque sensor (not shown). The embedded computing unit has a main frequency of 1.8 GHz and a control cycle of 5 ms. The front end of the omnidirectional flight platform is equipped with a monitoring mechanism 5 containing a camera and lidar; the omnidirectional flight platform also houses a main control cabin 6, within which the battery, flight controller, robotic arm controller, and high-performance embedded computing unit are all located.

[0079] Work process: (1) Initial positioning and modeling: The omnidirectional flight platform autonomously navigates to hover near the transmission tower, starts the parameter identification module, and updates the inertia parameters online (such as the mass, inertia, center of mass, etc. of the omnidirectional flight platform and the robotic arm). (2) Contact establishment and role allocation: The omnidirectional flight platform slowly approaches the transmission tower, controls the suction cup at the end of the fixed arm to adhere to the tower, and establishes stable contact. At this time, the force / torque multidimensional decomposition and allocation module detects the contact buffer force / torque and allocates it to the fixed arm. The working arm enters the standby state. (3) Insulator removal: The working arm receives the insulator positioning command and moves to the old insulator position. The force / torque multidimensional decomposition and allocation module identifies the removal force generated by the gripper as the effective working force / torque and allocates it to the working arm. The fixed arm continuously bears the contact buffer force / torque and 40% of the gravity balance component. The vector thruster array compensates in real time for the inertial compensation force / torque generated by the movement of the robotic arm and 60% of the gravity balance component. (4) Attitude stabilization control: During the dismantling process, the extended state observer based on active disturbance rejection control estimates and compensates for high-frequency disturbances caused by wind in real time, ensuring that the attitude deviation of the flight platform is less than 0.8°. (5) Role switching: After the working arm completes the removal of the old insulator, it receives the instruction to install the new insulator and triggers the task switching. The system starts the dual fixed short-time bridging strategy, the fixed arm maintains contact, and the working arm moves to the position of the new insulator to complete the task switching. (6) Energy consumption optimization: The energy optimization and endurance extension module adjusts the gravity sharing ratio in real time according to the joint torque of the fixed arm, so that the total power consumption of the vector thruster matrix is ​​reduced by 32% and the endurance is extended by 42%. (7) Safety monitoring: The generalized momentum observer continuously monitors the system status and no abnormalities are detected, ensuring the safety of the operation.

[0080] Example 3

[0081] Based on Embodiment 2 described above, this embodiment also provides an electronic device, please refer to the appendix. Figure 4 , Figure 4The electronic device shown is merely an example and should not be construed as limiting the functionality and scope of the embodiments of this disclosure.

[0082] like Figure 4 As shown, an electronic device may include a processing unit (such as a central processing unit, graphics processing unit, etc.) that can perform various appropriate actions and processes based on a program stored in read-only memory (ROM) or a program loaded from a storage device into random access memory (RAM). The RAM also stores various programs and data required for the operation of the electronic device. The processing unit, ROM, and RAM are interconnected via a bus. Input / output (I / O) interfaces are also connected to the bus.

[0083] Typically, the following devices can be connected to an I / O interface: input devices such as touchscreens, touchpads, keyboards, mice, and cameras; output devices such as liquid crystal displays (LCDs) and speakers; storage devices such as magnetic tapes and hard drives; and communication devices. Communication devices allow electronic devices to communicate wirelessly or wiredly with other devices to exchange data. Although Figure 4 Electronic devices with various devices are shown, but it should be understood that it is not required to implement or have all of the devices shown. More or fewer devices may be implemented or have instead. Figure 4 Each box shown can represent a device or multiple devices as needed.

[0084] In particular, according to some embodiments of this disclosure, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, some embodiments of this disclosure include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device, or installed from a storage device, or installed from a ROM. When the computer program is executed by a processing device, it performs the functions defined above in the methods of some embodiments of this disclosure.

[0085] Example 4

[0086] Based on Embodiment 2 above, this embodiment also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the above method.

[0087] It should be noted that, in some embodiments of this disclosure, the computer-readable medium described above may be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. A computer-readable storage medium may be, for example,—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In some embodiments of this disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In some embodiments of this disclosure, a computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium may be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to: wires, optical fibers, RF (radio frequency), or any suitable combination thereof.

[0088] In some embodiments, the client and server may communicate using any currently known or future-developed network protocol such as HTTP (Hypertext Transfer Protocol) and may interconnect with digital data communication (e.g., communication networks) of any form or medium. Examples of communication networks include local area networks (“LANs”), wide area networks (“WANs”), the Internet (e.g., the Internet of Things), and peer-to-peer networks (e.g., ad hoc peer-to-peer networks), as well as any currently known or future-developed networks.

[0089] The aforementioned computer-readable medium may be included in the aforementioned device or may exist independently without being assembled into the electronic device. The aforementioned computer-readable medium carries one or more programs, which, when executed by the electronic device, cause the electronic device to: receive overall system status information through a force / torque multidimensional decomposition and allocation module, orthogonally decompose generalized forces / torques into multiple types of force / torque components, and allocate them to corresponding actuators according to preset priorities; actively counteract the dynamic disturbances caused to the flight platform by the robotic arm's movement through a disturbance observation and compensation mechanism via a vector propulsion active compensation control module; and dynamically manage the working state of the robotic arm through a robotic arm collaborative contact management module.

[0090] Computer program code for performing operations of some embodiments of this disclosure can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Python and C++, and conventional procedural programming languages ​​such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0091] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0092] The units described in some embodiments of this disclosure can be implemented in software or hardware. The described units can also be housed in a processor; for example, a processor may be described as including a generalized dynamics modeling and real-time parameter identification unit, a force / torque multidimensional decomposition and allocation unit, a vector propulsion active compensation control unit, a dual-arm cooperative contact management unit, an energy optimization and endurance extension unit, and a safety monitoring and fault handling unit. The names of these units do not necessarily limit the unit itself; for example, the force / torque multidimensional decomposition and allocation unit may also be described as "receiving overall machine status information, orthogonally decomposing generalized forces / torques into multiple types of force / torque components, and allocating them to the corresponding actuators according to a preset priority."

[0093] The functions described above in this document can be performed at least in part by one or more hardware logic components. For example, exemplary types of hardware logic components that can be used, without limitation, include: field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), system-on-a-chip (SoCs), complex programmable logic devices (CPLDs), etc.

[0094] Obviously, those skilled in the art will understand that the various steps of the present invention described above can be performed in a manner different from that described above, and the simulation methods and experimental equipment include, but are not limited to, the above description. The steps of the present invention described above can be performed in a different order in certain circumstances, and the steps shown or described above can be performed separately. Therefore, the present invention is not limited to any particular combination of hardware and software.

[0095] The above description, in conjunction with specific embodiments, provides a further detailed explanation of the present invention. It should not be construed that the specific implementation of the present invention is limited to these descriptions. For those skilled in the art, various simple deductions or substitutions can be made without departing from the concept of the present invention, and all such deductions or substitutions should be considered within the scope of protection of the present invention.

Claims

1. A dynamic decoupling control system for an omnidirectional flight platform equipped with multiple robotic arms, characterized in that, include: The force / torque multidimensional decomposition and allocation module is used to receive the overall machine status information, orthogonally decompose the generalized force / torque into multiple types of force / torque components, and allocate them to the corresponding actuators according to preset priorities; The vector propulsion active compensation control module is used to actively counteract the dynamic disturbances caused by the movement of the robotic arm to the flight platform by adopting a disturbance observation and compensation mechanism. The robotic arm collaborative contact management module is used to dynamically manage the working status of the robotic arm.

2. The dynamic decoupling control system for an omnidirectional flight platform equipped with multiple robotic arms according to claim 1, characterized in that, The force / torque multidimensional decomposition and allocation module uses a pseudo-inverse projection algorithm of the task Jacobian matrix to orthogonally decompose generalized force / torque into multiple types of force / torque components, including: effective working force / torque for performing end-effector tasks, inertial compensation force / torque generated by the movement of the robotic arm, gravity balancing force / torque for balancing the overall weight of the machine, and contact buffer force / torque generated by the contact between the robotic arm and the work object. The preset priority of the force / torque multidimensional decomposition and allocation module is: contact buffer force / torque is preferentially borne by the fixed arm, effective working force / torque is output by the working arm, inertial compensation force / torque is compensated by the vector thruster array, and gravity balancing force / torque is jointly shared by the vector thruster array and the fixed arm.

3. The dynamic decoupling control system for an omnidirectional flight platform equipped with multiple robotic arms according to claim 1, characterized in that, The disturbance observation and compensation mechanism employs a quadratic programming thrust allocation algorithm and an extended state observer based on active disturbance rejection control (ADRC). The extended state observer based on ADRC is used to calculate in real time high-frequency disturbances in the entire system that are not covered by the generalized dynamic model, obtaining disturbance estimates. The quadratic programming thrust allocation algorithm aims to minimize the total power consumption of the vector thruster array. Constraints include the total equivalent force / torque generated by the vector thruster array being equal to the sum of the inertial compensation force / torque component and the portion of the gravity balance force / torque component it shares, upper and lower limits of the thrust amplitude of each vector thruster, and a limit on the rate of change of the thrust vector angle. The disturbance estimates are input as feedforward compensation quantities into the quadratic programming thrust allocation algorithm to enhance the omnidirectional flight platform's ability to suppress high-frequency dynamic disturbances.

4. The dynamic decoupling control system for an omnidirectional flight platform equipped with multiple robotic arms according to claim 1, characterized in that, The robotic arm collaborative contact management module includes a fixed arm control unit and a working arm control unit. The fixed arm control unit receives the contact buffer force / torque allocated by the force / torque multidimensional decomposition and allocation module, and controls at least one robotic arm as a fixed arm to maintain contact with the work object in a structural constraint manner to absorb impact and provide support. The working arm control unit receives the effective working force / torque allocated by the force / torque multidimensional decomposition and allocation module, and controls the remaining robotic arms as working arms to perform the target operation task.

5. The dynamic decoupling control system for an omnidirectional flight platform equipped with multiple robotic arms according to claim 4, characterized in that, The robotic arm in the collaborative contact management module adopts an impedance adaptive control strategy. The desired end stiffness of the robotic arm decreases exponentially as the real-time contact force between the end effector and the work object increases. The desired end damping of the robotic arm is automatically adjusted according to the critical damping condition.

6. The dynamic decoupling control system for an omnidirectional flight platform equipped with multiple robotic arms according to claim 1, characterized in that, The robotic arm collaborative contact management module also includes a role dynamic switching unit, which is used to determine and execute the role swap between the fixed arm and the working arm according to preset trigger conditions. The trigger conditions include: a forced switching condition triggered when the torque of any joint of the fixed arm exceeds the rated preset threshold, a task switching condition triggered according to the upper-level task planning instructions, and a predictive switching condition triggered when the current fixed arm end pose is predicted to exceed the effective workspace within a preset time based on kinematic prediction. During the switching transition, a dual-fixed short-time bridging strategy is executed to control the robotic arm to maintain contact with the work object at the same time and share part of the nominal contact force. After the center displacement and attitude angle deviation of the omnidirectional flight platform are both less than the preset safety threshold, the contact of the original fixed arm is released, and the role swap is completed.

7. The dynamic decoupling control system for an omnidirectional flight platform equipped with multiple robotic arms according to claim 1, characterized in that, Also includes: The safety monitoring and fault handling module includes a generalized momentum observer for real-time calculation of the weighted L2 norm of the generalized momentum residual. When the weighted L2 norm of the residual continuously exceeds the preset threshold of the residual determined by Monte Carlo simulation, a safety response sequence is triggered, including robotic arm joint locking, vector thrust derating, and guidance for autonomous landing.

8. The dynamic decoupling control system for an omnidirectional flight platform equipped with multiple robotic arms according to claim 1, characterized in that, Also includes: The generalized dynamics modeling and real-time parameter identification module is used to establish a generalized dynamics model including an omnidirectional flight platform and two robotic arms, and to identify inertia parameters online in real time using a recursive least squares method with a forgetting factor.

9. The dynamic decoupling control system for an omnidirectional flight platform equipped with multiple robotic arms according to claim 1, characterized in that, Also includes: The energy optimization and range extension module employs a single-variable optimization algorithm based on the golden section search method. With the goal of minimizing the total power consumption of the vector thruster array, and under the premise of satisfying the safety constraints of the fixed arm joint torque, it solves the optimal gravity sharing ratio coefficient online in real time to dynamically optimize the gravity sharing ratio between the fixed arm and the vector thruster array.

10. A dynamic decoupling control method for an omnidirectional flight platform equipped with a robotic arm, applied to the dynamic decoupling control system for an omnidirectional flight platform equipped with multiple robotic arms as described in any one of claims 1 to 9, characterized in that, Includes the following steps: The system receives overall machine status information through a force / torque multidimensional decomposition and allocation module, orthogonally decomposes generalized force / torque into multiple types of force / torque components, and allocates them to the corresponding actuators according to preset priorities. The vector propulsion active compensation control module employs a disturbance observation and compensation mechanism to actively counteract the dynamic disturbances caused by the movement of the robotic arm to the flight platform. The working status of the robotic arm is dynamically managed through the robotic arm collaborative contact management module.

Images