Extreme fault tolerance control method for bionic underwater robot

Abstract

The invention discloses an extreme fault fault-tolerant control method for a bionic underwater robot, the bionic underwater robot is provided with a buoyancy control device and at least one bionic execution part for providing propulsive force, and the extreme fault fault-tolerant control method comprises the following steps: controlling the bionic underwater robot to float to the water surface through the buoyancy control device; acquiring a motion law of the bionic underwater robot under the extreme fault, controlling the bionic underwater robot to execute a first type of action and a second type of action based on a model prediction control strategy, and driving the bionic underwater robot to perform plane position motion on the water surface so as to reach a target recovery point; according to the model prediction control strategy, an optimization problem is constructed based on current state information, target state information and a motion dynamics model of the bionic underwater robot, and the optimization problem is solved through rolling optimization to generate an optimal control input sequence in a future time domain. The purpose of recycling the bionic underwater robot under the extreme fault condition is achieved.

Description

Technical Field

[0001] This invention relates to the field of underwater robot control technology, and in particular to a computational fault-tolerant control method for biomimetic underwater robots. Background Technology

[0002] With economic development and social progress, the importance of marine resources is increasing daily, attracting widespread attention. Because biomimetic underwater robots can perform reconnaissance, exploration, and data collection in harsh marine environments, even the deep sea, their research and development has garnered significant attention in recent years. However, the underwater environment is harsh, and the actuators of current biomimetic underwater robots are relatively immature and easily damaged. Due to the complexity of the underwater environment, the cost and risk of manually recovering an underwater robot after actuator failure are far higher than in conventional situations on land, and the robot is also highly susceptible to loss, leading to substantial economic losses.

[0003] Based on the aforementioned practical application needs, it is necessary to solve the fault-tolerant control problem of underwater robots. Existing technical literature only studies the fault-tolerant control problem of biomimetic underwater robots with a large number of remaining actuators or traditional underwater robots using propeller propulsion. For example, when there are many remaining actuators, thrust redistribution methods are used to achieve the desired driving effect; for traditional underwater robots, when only one propeller propulsion is available, the only publicly available literature presents a simple fault-tolerant control scheme using helical motion for position control. This simple fault-tolerant control scheme utilizes the characteristic of traditional underwater robots exhibiting circular motion when only one propeller propulsion is available. By changing the thrust of the propeller propulsion, the radius of the circular motion is changed, thereby creating displacement; its principle is as follows: Figure 1 As shown. However, this simplified fault-tolerant control scheme cannot be applied to fin-driven bionic underwater robots. Direct application to fin-driven bionic underwater robots would result in insufficient utilization of the bionic actuator's performance, leading to performance loss. Furthermore, this simplified fault-tolerant control scheme has strict requirements on the trajectory of circular motion, placing high demands on the controller; moreover, its robustness and anti-interference capabilities are poor.

[0004] The above background information is provided only to aid in understanding the concept and technical solution of this invention. It does not necessarily belong to the prior art of this patent application. In the absence of clear evidence that the above information was disclosed on the filing date of this patent application, the above background information should not be used to evaluate the novelty and inventiveness of this application. Summary of the Invention

[0005] To solve the above technical problems, the application provides an extreme fault tolerance control method for a bionic underwater robot, which fully utilizes the performance of a bionic execution unit and realizes the purpose of recovering the bionic underwater robot in an extreme fault condition.

[0006] To achieve the above purpose, the application adopts the following technical scheme: In a first aspect, the application discloses an extreme fault tolerance control method for a bionic underwater robot, wherein the bionic underwater robot is provided with a buoyancy control device and at least one bionic execution unit for providing propulsion force, and the extreme fault tolerance control method comprises the following steps: S1: controlling the bionic underwater robot to float to the water surface through the buoyancy control device; S2: obtaining the motion law of the bionic underwater robot in an extreme fault condition, and controlling the bionic underwater robot to perform a first type of action and a second type of action based on a model predictive control strategy, so as to drive the bionic underwater robot to perform planar position motion on the water surface to reach a target recovery point; wherein the first type of action is used to change the orientation of the bionic underwater robot, and the second type of action is used to change the position of the bionic underwater robot by generating a transient impulse; wherein the model predictive control strategy is based on the current state information, target state information and motion dynamics model of the bionic underwater robot to construct an optimization problem, and the optimization problem is solved by rolling optimization to generate an optimal control input sequence in a future time domain, and the optimal control sequence is configured to control the bionic underwater robot to perform the first type of action and the second type of action.

[0007] Preferably, the optimization problem constructed based on the current state information, target state information and motion dynamics model of the bionic underwater robot is as follows:

[0008] wherein, , is a prediction time domain; is an extended system state of a model taking the change of the input quantity as a first-order process; is a state sequence vector, wherein represents the system state at i time predicted at k time; is an input sequence vector, wherein represents the input at i time planned at k time; is k+i- the change rate of the input quantity at 1 time; is the minimum value of the input change rate, is the maximum value of the input rate of change, is the minimum value of the input quantity, is the maximum value of the input quantity; is a discrete motion dynamics model of the biomimetic underwater robot; is a terminal state set; is an objective function to be optimized.

[0009] Preferably, the system state is constructed based on a planar simplified dynamics model of the biomimetic underwater robot, and an equation of the planar simplified dynamics model is:

[0010] In the formula, is a direction angle of the biomimetic underwater robot in a body coordinate system of the biomimetic underwater robot when the biomimetic execution part is installed in a planar direction; , , , are respectively an equivalent mass of the biomimetic underwater robot in an x-axis direction of the body coordinate system, an equivalent mass in a y-axis direction, and an equivalent mass of the Coriolis effect; , , respectively represent a velocity of the biomimetic underwater robot in the x-axis direction of the body coordinate system, a velocity in the y-axis direction, and a yaw angular velocity, , , are damping coefficients of motions in an x-axis direction, a y-axis direction, and a z-axis direction of the body coordinate system, is a derivative of an x-axis component of a position of the biomimetic underwater robot in a ground coordinate system, is a derivative of a y-axis component of the position of the biomimetic underwater robot in the ground coordinate system; is an Euler angle of a yaw order of the position of the biomimetic underwater robot in the ground coordinate system, , , , are derivatives of the , , , , , , , is an equivalent inertia of the biomimetic underwater robot in a z-axis direction of the body coordinate system, is a coefficient of a moment of force generated by a thrust, is a horizontally averaged force generated by fin vibration.

[0011] Preferably, the element in the input sequence vector is the average thrust of the expected bionic actuator, wherein the magnitude of the average thrust is a function of the amplitude, and the direction of the average thrust is consistent with the direction of the bias angle.

[0012] Preferably, the form of the objective function is:

[0013] wherein, is the state represented by the target recovery point, is the error weight matrix, is the control weight matrix, is the terminal cost matrix.

[0014] Preferably, in the optimization problem, the constraint condition includes a control rate of change for limiting the action frequency of the bionic actuator.

[0015] Preferably, the motion law of the bionic underwater robot under extreme failure in step S2 includes: S21: establishing a motion dynamics model of the whole bionic underwater robot; S22: establishing a vibration equation of the bionic actuator according to the fact that the thrust of the bionic actuator is generated by the periodic vibration of the bionic actuator; S23: obtaining the input quantity under the condition that all the bionic actuators in the bionic underwater robot are all effective according to the motion dynamics model of the whole bionic underwater robot, obtaining the relationship between the angle and the thrust of each bionic actuator according to the vibration equation, and then combining the input quantity under the condition that all the bionic actuators in the bionic underwater robot are all effective with the relationship between the angle and the thrust of each bionic actuator to establish a driving equation of the bionic underwater robot; S24: deleting the column corresponding to the bionic actuator invalid under extreme failure in the coefficient matrix of the driving equation, and deleting the component corresponding to the bionic actuator invalid under extreme failure in the input quantity, to obtain the driving equation of the bionic underwater robot under extreme failure; S25: extracting the equation related to the planar motion state in the motion dynamics model of the whole bionic underwater robot to establish a planar simplified dynamics model; S26: combining the driving equation of the bionic underwater robot under extreme failure with the planar simplified dynamics model to establish a steady-state equation of the bionic underwater robot, to obtain the motion law of the bionic underwater robot under extreme failure.

[0016] Preferably, the motion dynamics model of the whole bionic underwater robot in step S21 is established as a Lagrange-Euler equation; The vibration equation in step S22 is:

[0017] wherein, is the angle of the bionic actuator, is the amplitude, is the vibration frequency, is the phase, is the bias angle; The driving equation of the bionic underwater robot in step S23 is:

[0018] wherein, is the standardization of the input quantity, is the coefficient matrix, respectively represent ; is the yaw angle of the direction of the bionic actuator relative to the body coordinate system when installed; is the absolute value of the relative position of the action point of the resultant force generated by the bionic actuator to the origin of the body coordinate system; represents the yaw moment coefficient; six-dimensional input quantity represents the force or torque on the corresponding degree of freedom.

[0019] Preferably, the steady-state equation of the bionic underwater robot established in step S26 is:

[0020] wherein, , , respectively are the equivalent mass of the bionic underwater robot in the x-axis direction of the body coordinate system, the equivalent mass in the y-axis direction, and the equivalent mass of the Coriolis effect; is the steady-state velocity in the x-axis direction of the body coordinate system, is the steady-state velocity in the y-axis direction of the body coordinate system, is the steady-state yaw angular velocity, respectively are the damping coefficients of the motion in the x direction, direction and direction of the body coordinate system in the steady state, is the horizontal average force generated by the fin vibration, is the yaw angle of the direction of the bionic actuator relative to the body coordinate system when installed, is the coefficient of the moment generated by the thrust.

[0021] In a second aspect, the present application discloses a computer readable storage medium, and the computer readable storage medium stores a computer program, wherein the computer program is arranged to be run by a processor to execute the extreme fault tolerance control method for the bionic underwater robot.

[0022] Compared with the prior art, the extreme fault tolerance control method for the bionic underwater robot has the advantages that the bionic underwater robot is first floated to the water surface, the motion law of the bionic underwater robot under the extreme fault is obtained, and a model predictive control strategy is proposed based on the motion law, so that the performance of the bionic execution unit of the bionic underwater robot can be fully utilized, the motion ability under the extreme actuator fault condition is enhanced, and the safety is improved; through the combination of the model predictive control strategy and the specific action of the bionic execution unit, the effect that the motion trajectory can still be accurately controlled under the extreme fault is realized, so that the bionic underwater robot is pushed to move in the desired direction by performing the first type of action and the second type of action by the bionic execution unit, and the purpose of recovering the bionic underwater robot under the extreme fault condition is achieved.

[0023] Other advantages of the embodiments of the present application will be further described below. BRIEF DESCRIPTION OF DRAWINGS

[0024] Figure 1 is a schematic diagram of the principle of the existing fault tolerance control scheme for the traditional underwater robot; Figure 2 is a flowchart of the extreme fault tolerance control method for the bionic underwater robot of the preferred embodiment of the present application.

[0025] Figure 3 is the principle of the fault tolerance control scheme proposed by the specific embodiment of the present application; Figure 4 is the comparison result of the motion trajectory in the simulation experiment; Figure 5 is the change of the error with time in the simulation experiment. DETAILED DESCRIPTION

[0026] The embodiments of the present application will be described in detail below. It should be emphasized that the following description is only exemplary and is not intended to limit the scope of the present application and its applications.

[0027] It should be noted that when an element is referred to as being "fixed to" or "set on" another element, it can be directly on the other element or indirectly on the other element. When an element is referred to as being "connected to" another element, it can be directly connected to the other element or indirectly connected to the other element. In addition, the connection can be for fixing or for circuit / signal communication.

[0028] It should be understood that the terms "length", "width", "upper", "lower", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer" and the like indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the embodiments of the present application and simplifying the description, and do not indicate or imply that the devices or elements referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as a limitation of the present application.

[0029] In addition, the terms "first", "second" are only for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly indicating the number of technical features referred to. Therefore, the features defined with "first", "second" can explicitly or implicitly include one or more of the features. In the description of the embodiments of the present application, the meaning of "a plurality of" is two or more, unless otherwise explicitly and specifically limited.

[0030] As shown in Figure 2 The preferred embodiment of the present application discloses an extreme fault tolerant control method for a bionic underwater robot, the bionic underwater robot is provided with a buoyancy control device and at least one bionic actuator for providing propulsion force, and the extreme fault tolerant control method comprises the following steps: S1: controlling the bionic underwater robot to float to the water surface by the buoyancy control device; S2: obtaining the motion law of the bionic underwater robot under extreme fault, controlling the bionic underwater robot to perform a first type of action and a second type of action based on a model predictive control strategy, and driving the bionic underwater robot to perform planar position motion on the water surface to reach a target recovery point; Wherein, the first type of action is used to change the orientation of the bionic underwater robot, and the second type of action is used to change the position of the bionic underwater robot by generating a transient impulse; for example, the first type of action is a rotating action, and the second type of action is a flipping action.

[0031] Wherein, the model predictive control strategy is to construct an optimization problem based on the current state information, target state information and motion dynamics model of the bionic underwater robot, and to generate an optimal control input sequence in the future time domain by solving the optimization problem through rolling optimization, and the optimal control sequence is configured to control the bionic underwater robot to perform the first type of action and the second type of action.

[0032] Further, in the optimization problem, the constraint condition includes a control amount rate of change for limiting the action frequency of the bionic actuator.

[0033] Further, the step S2 of obtaining the motion law of the bionic underwater robot under extreme fault comprises: S21: establishing a motion dynamics model of the whole bionic underwater robot; S22: the thrust of the bionic actuator is generated by the periodic vibration of the bionic actuator, and a vibration equation of the bionic actuator is established; S23: the input quantity under the condition that all the bionic actuators in the bionic underwater robot are effective is obtained according to the overall motion dynamics model of the bionic underwater robot, the relationship between the angle and the thrust of each bionic actuator is obtained according to the vibration equation, and then the driving equation of the bionic underwater robot is established by combining the input quantity under the condition that all the bionic actuators in the bionic underwater robot are effective with the relationship between the angle and the thrust of each bionic actuator; S24: the column corresponding to the bionic actuator ineffective under the extreme fault is deleted in the coefficient matrix of the driving equation, and the component corresponding to the bionic actuator ineffective under the extreme fault is deleted in the input quantity, so as to obtain the driving equation of the bionic underwater robot under the extreme fault; S25: the equation related to the planar motion state in the overall motion dynamics model of the bionic underwater robot is extracted to establish a planar simplified dynamics model; S26: the steady-state equation of the bionic underwater robot is established by combining the driving equation of the bionic underwater robot under the extreme fault with the planar simplified dynamics model, and the motion law of the bionic underwater robot under the extreme fault is obtained.

[0034] The following further describes the extreme fault tolerance control method for the bionic underwater robot disclosed in the above preferred embodiment of the application by means of a specific embodiment. In the specific embodiment, the bionic actuator is a fin, and the extreme fault tolerance control method comprises the following steps: A1: modeling the fin-driven bionic underwater robot; The step A1 comprises: A11: establishing a six-degree-of-freedom model of the fin-driven bionic underwater robot According to the Fossen hydrodynamic formula (wherein the Fossen model is a dynamics model for describing the motion characteristics of a ship or an underwater robot, and is widely used in the fields of navigation, underwater robots, etc.), the overall motion dynamics model of the fin-driven bionic underwater robot can be established as a Lagrange-Euler equation:

[0035] wherein, is a generalized velocity, each component of which respectively represents the velocity in the x-axis direction, the velocity in the y-axis direction, the velocity in the z-axis direction, the roll angular velocity, the pitch angular velocity, and the yaw angular velocity in the body coordinate system; is a generalized position, each component of which respectively represents the x-axis component, the y-axis component, and the z-axis component of the position in the ground coordinate system, and the Euler angles in the order of roll, pitch, and yaw; derivative of generalized velocity; derivative of generalized position; represents the six-degree-of-freedom input generated by the fins through the generalized velocity derivative of generalized position transformation matrix; and respectively represent the inertia matrix and the additional inertia matrix of the biomimetic underwater robot; represents the Coriolis-Centrifugal matrix of the biomimetic underwater robot; represents the drag coefficient matrix of the biomimetic underwater robot; represents the gravity and buoyancy acting on the biomimetic underwater robot; represents the six-degree-of-freedom input generated by the fins; represents the equivalent representation of the disturbance acting on the biomimetic underwater robot on the six-degree-of-freedom input.

[0036] wherein the body coordinate system refers to a right-handed coordinate system with the center of mass of the robot as the origin and the three directions of front, left and up as the three fixed directions, and the ground coordinate system refers to a right-handed coordinate system with the initial position of the center of mass of the robot as the origin and the three directions of front, left and up at the initial time of the robot as the three fixed directions.

[0037] The thrust generated by the fins is generated by the periodic vibration of the fins, and the vibration equation is

[0038] wherein is the angle of the fin, is the amplitude, is the vibration frequency, is the phase, is the bias angle. The size and direction of the thrust generated by the fins are determined by the hydrodynamic characteristics of the fins, which are related to the amplitude and bias angle of the fins. In the case of a fixed frequency, the size has a functional relationship with the amplitude , and the direction is consistent with the direction of the bias angle .

[0039] The relationship between the six-degree-of-freedom input and the angle and thrust of each fin under the condition of all fins being healthy is taken as the driving equation. In the case of failure, the six-degree-of-freedom input ​The number and position of the remaining fins are relevant. Therefore, the driving equations under fault conditions can be obtained by modifying the driving equations under all healthy fins conditions. Therefore, the driving equations under all healthy fins conditions are given first here. This part of the modeling is related to the actuator configuration of the fin-driven bionic underwater robot. The following example uses a four-fin driven bionic underwater robot, with the four fins numbered 1, 2, 3, and 4 according to right front, right rear, left rear, and left front. Let the corresponding _ i The average angle of each fin vibration is The average thrust generated is The driving equations for all healthy fins are shown below:

[0040] In the above formula, ; This is the absolute value of the relative position of the point of application of the resultant force generated by the fins in the design with respect to the origin of the body coordinate system; The yaw angle of the fin relative to the airframe coordinate system when it is installed in the design; Indicates the yaw moment coefficient; It involves standardizing the input quantities to obtain a coefficient matrix that is easier to represent. ; Each element in the equation represents the horizontal and vertical components of the force generated by each fin. This represents the horizontal component of force generated by the i-th fin. This represents the vertical component of the force generated by the i-th fin; Six-dimensional input It represents the force or torque at the corresponding degree of freedom.

[0041] Based on the aforementioned hydrodynamic model and driving equations, when the fins completely fail, they can be considered to have no effect on the driving equations. Therefore, by removing the coefficient matrix of the driving equations for all fin-healthy cases... The columns corresponding to these fins are in the input vector. By deleting the components corresponding to these fins and retaining only the columns and components corresponding to the remaining healthy fins, the driving equations and dynamic model of the fin-driven biomimetic underwater robot under this fault condition can be obtained.

[0042] The advantage of using this method to build a robot model after a failure is that it can more clearly show the role of motion characteristics in the model, and it is easier to analyze the relationship between input and motion characteristics from the model's perspective, thus facilitating the use of the corresponding motion characteristics in motion planning and fault-tolerant control.

[0043] A12: Establish a simplified planar model Since conventional biomimetic underwater robots are equipped with buoyancy control devices, the fault-tolerant control scheme proposed in this invention first uses fins to drive the buoyancy device of the biomimetic underwater robot to float to the surface, and then controls the fins to drive the biomimetic underwater robot to move in a planar position on the water surface, thereby achieving the goal of reaching the target recovery point. Therefore, it is necessary to establish its planar motion model to analyze its motion dynamics and design the fault-tolerant control scheme.

[0044] Since the robot floats to the water surface via a buoyancy device, its roll, pitch, and heave motions will not affect the robot's retrieval under a given horizontal motion scheme, and the depth is fixed. Therefore, in the water surface retrieval task that this scheme focuses on, the three degrees of freedom of surge, sway, and yaw can be mainly considered. The simplified planar model can be obtained by fixing the state variables related to roll, pitch, and heave in the complete motion dynamics model (Equations (1.1) and (1.2)) to zero and extracting the equations for the remaining three degrees of freedom. That is, the equations related to the planar motion state in the Lagrange-Euler equations (Equations (1.1) and (1.2)) are extracted as follows:

[0045] in, When installing fins on the body Direction angle in a plane; , , , , , , To account for the equivalent mass and inertia of hydrodynamic effects, we define the equivalent mass along the x-axis, y-axis, and z-axis of the body coordinate system, the equivalent mass of the Coriolis effect, the equivalent inertia along the x-axis, the equivalent inertia along the y-axis, and the equivalent inertia along the z-axis, respectively. The average horizontal force generated by fin vibration; for , and The damping coefficient for directional motion is of the form:

[0046] in, All are with , and The coefficients in the damping formula corresponding to directional motion, in the subscript Represents the coefficient of the linear term. Represents the coefficients of nonlinear terms.

[0047] Let be the coefficient of the torque generated by the thrust, in planar motion. Therefore, we have the simplified planar model equation:

[0048] This model is the foundation for motion law analysis and is also the target for designing fault-tolerant control schemes.

[0049] A2: Analysis of motion patterns under extreme fault conditions When only one healthy fin remains, the biomimetic underwater robot can only perform curved movements.

[0050] First, we analyze its steady-state motion under constant input. Under constant input, let its steady-state motion be... ,in Represents steady-state generalized velocity. The steady-state velocities in the x-axis and y-axis directions are respectively... For the steady-state angular velocity in the direction, the steady-state equation is:

[0051] Actual motion analysis revealed that the equation has a solution. Further analysis showed that the steady-state solution obtained from the steady-state equation almost represents uniform circular motion, meaning that only a finite number of isolated points satisfy the linear velocity requirement. or angular velocity . The steady-state velocity is in the x-axis direction. The steady-state velocity is in the y-axis direction. For steady-state yaw rate, These represent the x-axis and y-axis directions in the body coordinate system when in steady state, respectively. Damping coefficient of directional motion, The average horizontal force generated by fin vibration.

[0052] First, let's set This means Then there is

[0053] Then there is , that is Let's assume again... Then there is

[0054] After further sorting, there is

[0055] in, and All of these are polynomial functions corresponding to (5.1)-(5.3). According to the properties of polynomials, at most a finite number of isolated zeros satisfy the above equations. In summary, at most a finite number of isolated steady-state points represent non-uniform circular motion states.

[0056] Under the aforementioned extreme actuator failure conditions, the steady-state point of the biomimetic underwater robot's motion is determined by the input magnitude, that is: the parameters of the uniform circular motion and... The magnitude of the equations exhibits a functional relationship, which can be obtained through actual experiments or by solving the steady-state equations. Furthermore, due to the hydrodynamic effects experienced by the fin-driven biomimetic underwater robot, the equilibrium points represented by its steady state are all locally stable.

[0057] A3: Fault-tolerant control schemes under extreme failure conditions This step includes: A31: Proposal of Fault-Tolerant Control Scheme Principles Based on the motion laws of the biomimetic underwater robot analyzed in step A2, it can be seen that when the fin-driven biomimetic underwater robot is in a steady state, its orientation can change continuously and stably due to its uniform circular motion, which is called "rotation" motion. Furthermore, according to the motion dynamics model, when the fin undergoes a "flip," that is, when the average thrust generated by the fin suddenly reverses, a transition process will occur because the fin's state is continuously changing. According to mechanical principles, this transition process is actually caused by the impulse brought about by the fin's "flip," which propels the underwater robot to move in the flipped direction; this action is called "flip" motion. From the above analysis, it can be seen that by changing the robot's orientation through rotation, and then continuously performing fin "flip" motions at appropriate times determined by the controller to propel the robot closer to the target point, the biomimetic underwater robot can be made to move in a certain direction, thereby reaching the set target recovery point (i.e.,...). Figure 3 The "target point" shown is also the basic principle of the fault-tolerant control scheme proposed in this embodiment of the invention, and the process is as follows: Figure 3 As shown.

[0058] A32: Introduction to Fault-Tolerant Control Schemes and Design of Fault-Tolerant MPC (Model Predictive Control) Controllers Based on the above analysis, the fault-tolerant control scheme proposed in this embodiment of the invention is as follows: (1) First, the buoyancy control device of the fin-driven bionic underwater robot is used to slowly make it float to the surface of the water; this process does not require the use of a separately designed controller; (2) Using a fault-tolerant controller designed based on the principle of fault-tolerant control scheme, the rotation of the fin is controlled to change its orientation, and the flipping of the fin is controlled to move it. This drives the bionic underwater robot to move in the desired direction, thereby moving to the target recovery point and achieving the purpose of safe recovery.

[0059] Since the transition process of a biomimetic underwater robot is actually determined by its motion dynamics and is quite complex, step A31 only describes the principle of a simplified fault-tolerant control scheme. The actual motion dynamics model needs to be considered in the controller design. Based on this consideration, this embodiment of the invention selects a model-based MPC controller that can explicitly consider state and control constraints as the fault-tolerant controller. Simultaneously, considering the principle of fin drive, flipping should not be too frequent in practical applications to prevent large oscillations in the vertical direction. Therefore, the MPC controller designed in this embodiment of the invention also constrains the rate of change of the control variable. The optimization problem to be solved by the proposed MPC controller is:

[0060] in, , To predict the time domain, To treat changes in input as an extension of the system state in a model that is modeled as a first-order process, Let be a state sequence vector, where This represents the system state predicted at time i from time k. Given an input sequence vector, where The input at time i represents the planned input at time k. The elements in the input sequence vector are the expected average thrust of the bionic actuator. The magnitude of the average thrust is a function of the amplitude, and the direction of the average thrust is consistent with the direction of the offset angle.

[0061] This represents the rate of change of the input quantity. It is the minimum rate of change of the input quantity. It is the maximum rate of change of the input quantity. It is the minimum value of the average horizontal force. It is the maximum value of the average horizontal force input.

[0062] Based on the simplified planar model and input change model established in step A12, and using The discrete motion dynamics model obtained as the sampling time. A set of terminal states; Let be the objective function to be optimized, and its form is:

[0063] in, The state represented by the target safe point. It is the error weight matrix. It is the control weight matrix. These are terminal cost matrices, all designed as diagonal matrices.

[0064] The designed MPC controller obtains the optimal control input sequence by solving the optimization problem (Equations (11.1) to (11.5)). And the first element of the optimal control sequence The control input is fed into the system, and the optimization problem is solved by rolling optimization at each subsequent time step to obtain the desired control law. This is achieved by appropriately adjusting the weight matrix. and prediction time domain By adjusting the size, a better control effect can be achieved.

[0065] Optimal control input sequence This is a horizontally averaged force sequence, determined by the offset angle and amplitude information. During actual driving, the offset angle determines the average direction of the thrust. Tracking is achieved by causing a step change (i.e., "flipping") in the offset angle. The instructions in the MPC generate the required transient impulse. Variations in amplitude are then used to adjust the thrust magnitude. Rate of change constraints in MPC and This limits the maximum rate of change of the offset angle, thereby indirectly limiting the frequency of the "flip" action.

[0066] The fault-tolerant control scheme proposed in the specific embodiments of the present invention can fully utilize the performance of the actuator of the fin-driven bionic underwater robot, enhance its motion capability under extreme actuator failure conditions, and improve its safety; moreover, by using the MPC fault-tolerant controller proposed in the specific embodiments of the present invention, it is possible to enable the fin-driven bionic underwater robot to reach the designated target position on the water surface under extreme actuator failure conditions, so as to achieve the purpose of safe recovery.

[0067] This invention specifically addresses a fin-driven bionic underwater robot, analyzing its motion characteristics under extreme failure conditions where only one fin remains operational. Based on these characteristics, a fault-tolerant control scheme is proposed. Compared to existing technologies, this scheme more fully utilizes the performance of the fin-driven bionic underwater robot's actuators, achieving better control. Specifically, this scheme achieves the following: fully utilizing the performance of the bionic actuators, it uses the impulse generated by the fin's flipping motion to produce displacement, propelling the robot in the desired direction. Compared to existing methods that utilize the radius difference of a steady-state circle for displacement, this scheme achieves better fault-tolerant control. Specifically, the rotational motion of the fins allows the robot to perform uniform circular motion, thus changing its orientation; the flipping motion of the fins propels the robot to generate displacement.

[0068] In addition, the specific embodiments of the present invention use the MPC strategy for controller design, taking into account the transition process of the model, thereby alleviating the strict requirements on the circular motion trajectory; compared with the prior art, since the displacement generated by the flipping action is more stable and easier to control, and MPC itself can be specifically optimized for the robustness and anti-interference of the controller, the proposed solution has better robustness and anti-interference.

[0069] The following simulation experiment demonstrates the extreme fault-tolerant control method for biomimetic underwater robots proposed in this invention, using a four-finned biomimetic turtle underwater robot as an example. The model parameters of the robot system are as follows:

[0070] The control variable designed for the controller is the one in the simplified planar model in step A12. .

[0071] The weight matrices Q, R, and P in the MPC controller are selected as positive definite diagonal matrices and adjusted through preliminary simulation experiments to achieve a balance between response speed and control smoothness. Input rate of change constraint. and The value is determined based on the physical actuation frequency limit of the bionic actuator (fin). Specifically, the parameters of the fault-tolerant MPC controller are set as follows:

[0072] In the experiment, by selecting an appropriate terminal cost matrix, the constraint on the terminal state set is ignored; that is, the terminal state set is set as follows: This is the full state space.

[0073] Simultaneously, a comparison is made with existing technologies, namely, the helical motion scheme. Let the control law of the helical motion scheme be...

[0074] in Let be the error vector between the origin of the body coordinate system and the target point. Let the target point be... The changes in the planar motion trajectory and the error over time obtained by comparing the two are as follows: Figure 4 and Figure 5 As shown.

[0075] from Figure 4 As can be seen from the motion trajectory diagram, the underwater robot's trajectory under the control of the MPC controller exhibits a sawtooth pattern, consistent with the principle of the proposed fault-tolerant control scheme. Meanwhile, from... Figure 5 As can be seen from the above, compared with the prior art, the fault-tolerant control scheme proposed in this invention has a faster error convergence speed, a more direct trajectory, and a more accurate final position. This shows that the fault-tolerant control scheme proposed in the embodiments of this invention can more effectively utilize the characteristics and motion performance of the fin actuator, enabling it to achieve faster and more precise motion, thereby making the recovery efficiency of the fin-driven bionic underwater robot higher and safer in extreme actuator failure situations.

[0076] Another preferred embodiment of the present invention discloses a computer-readable storage medium storing a computer program, wherein the computer program is configured to be run by a processor to perform the computational fault-tolerant control method for a biomimetic underwater robot described in the preferred embodiment above.

[0077] Optionally, the aforementioned storage media may include, but are not limited to, USB flash drives, read-only memory (ROM), random access memory (RAM), portable hard drives, magnetic disks, or optical disks, and other media capable of storing computer programs.

[0078] The background section of this invention may include background information about the problems or circumstances surrounding the invention, rather than a description of prior art by others. Therefore, the content included in the background section is not an admission of prior art by the applicant.

[0079] The above description provides a further detailed explanation of the present invention in conjunction with specific / preferred embodiments, and 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 substitutions or modifications can be made to these described embodiments without departing from the concept of the present invention, and all such substitutions or modifications should be considered within the scope of protection of the present invention. In the description of this specification, the reference to terms such as "an embodiment," "some embodiments," "preferred embodiment," "example," "specific example," or "some examples," etc., indicates that the specific features, structures, materials, or characteristics described in connection with that embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described can be combined in any suitable manner in one or more embodiments or examples. Furthermore, those skilled in the art can combine and integrate different embodiments or examples and features of different embodiments or examples described in this specification without contradiction. Although the embodiments of the present invention and their advantages have been described in detail, it should be understood that various changes, substitutions, and modifications can be made herein without departing from the scope defined by the appended claims.

Claims

1. An extreme fault-tolerant control method for biomimetic underwater robots, characterized in that, The biomimetic underwater robot is equipped with a buoyancy control device and at least one biomimetic actuator for providing propulsion. The extreme fault-tolerant control method includes the following steps: S1: The bionic underwater robot is controlled to rise to the surface of the water by the buoyancy control device; S2: Obtain the motion law of the bionic underwater robot under extreme fault conditions, and combine it with the model predictive control strategy to control the bionic underwater robot to perform a first type of action and a second type of action, driving the bionic underwater robot to perform planar position movement on the water surface to reach the target recovery point; wherein, the first type of action is used to change the orientation of the bionic underwater robot, and the second type of action is used to change the position of the bionic underwater robot by generating a transient impulse; The model predictive control strategy is based on the current state information, target state information and motion dynamics model of the biomimetic underwater robot to construct an optimization problem, and generates the optimal control input sequence in the future time domain by solving the optimization problem through rolling optimization. The optimal control sequence is configured to control the biomimetic underwater robot to perform a first type of action and a second type of action.

2. The extreme fault-tolerant control method for biomimetic underwater robots according to claim 1, characterized in that, The optimization problem constructed based on the current state information, target state information, and motion dynamics model of the aforementioned biomimetic underwater robot is as follows: ； In the formula, , For prediction in the time domain; To extend the system state by modeling changes in input quantities as a first-order process; Let be a state sequence vector, where This represents the system state at time i predicted at time k; Given an input sequence vector, where This represents the input at time i for the plan at time k; for k+i- The rate of change of input at time 1; It is the minimum rate of change of the input. It is the maximum value of the input rate of change. It is the minimum input value. It is the maximum value of the input; It is a discrete motion dynamics model of a biomimetic underwater robot; A set of terminal states; Let be the objective function to be optimized.

3. The extreme fault-tolerant control method for biomimetic underwater robots according to claim 2, characterized in that, The system state is constructed based on a simplified planar dynamics model of a biomimetic underwater robot, and the equations of the simplified planar dynamics model are: ； In the formula, When installing the bionic actuator, the bionic underwater robot in the body coordinate system Direction angle in a plane; , , These are the equivalent mass of the biomimetic underwater robot in the x-axis direction, the equivalent mass in the y-axis direction, and the equivalent mass due to the Coriolis effect in the body coordinate system, respectively. , , These represent the x-axis velocity, y-axis velocity, and yaw rate of the biomimetic underwater robot in the body coordinate system, respectively. , , For the body coordinate system direction, direction and Damping coefficient of directional motion, Let be the derivative of the x-axis component of the position of the biomimetic underwater robot in the terrestrial coordinate system. Let be the derivative of the y-axis component of the position of the biomimetic underwater robot in the terrestrial coordinate system; Euler angles for sorting the yaw of the biomimetic underwater robot in a terrestrial coordinate system. , , , They are respectively , , , The derivative of The equivalent inertia of the biomimetic underwater robot along the z-axis in the body coordinate system. The coefficient of the torque generated by the thrust. The average horizontal force generated by fin vibration.

4. The extreme fault-tolerant control method for biomimetic underwater robots according to claim 2, characterized in that, The elements in the input sequence vector represent the desired average thrust of the bionic actuator, wherein the magnitude of the average thrust is a function of the amplitude, and the direction of the average thrust is consistent with the direction of the offset angle.

5. The extreme fault-tolerant control method for biomimetic underwater robots according to claim 2, characterized in that, The objective function is in the form of: ； In the formula, The state represented by the target recycling point. It is the error weight matrix. It is the control weight matrix. It is the terminal cost matrix.

6. The extreme fault-tolerant control method for biomimetic underwater robots according to claim 1, characterized in that, In the optimization problem, the constraints include the rate of change of the control quantity used to limit the frequency of action of the bionic actuator.

7. The extreme fault-tolerant control method for biomimetic underwater robots according to claim 1, characterized in that, Step S2 involves obtaining the motion laws of the biomimetic underwater robot under extreme fault conditions, including: S21: Establish the overall motion dynamics model of the biomimetic underwater robot; S22: Based on the fact that the thrust of the bionic actuator is generated by the periodic vibration of the bionic actuator, establish the vibration equation of the bionic actuator; S23: Based on the overall motion dynamics model of the bionic underwater robot, obtain the input quantity when all bionic actuators of the bionic underwater robot are effective; obtain the relationship between the angle and thrust of each bionic actuator based on the vibration equation; and then, combine the input quantity when all bionic actuators of the bionic underwater robot are effective with the relationship between the angle and thrust of each bionic actuator to establish the driving equation of the bionic underwater robot. S24: Delete the column in the coefficient matrix of the driving equation that corresponds to the bionic actuator that is invalid under extreme faults, and delete the component in the input that corresponds to the bionic actuator that is invalid under extreme faults, to obtain the driving equation of the bionic underwater robot under extreme faults. S25: Extract the equations related to the planar motion state from the overall motion dynamics model of the biomimetic underwater robot to establish a simplified planar dynamics model; S26: By combining the driving equations of the biomimetic underwater robot under extreme faults with the simplified planar dynamics model, the steady-state equations of the biomimetic underwater robot are established, and the motion law of the biomimetic underwater robot under extreme faults is obtained.

8. The extreme fault-tolerant control method for biomimetic underwater robots according to claim 1, characterized in that, In step S21, the overall motion dynamics model of the biomimetic underwater robot is established as the Lagrange-Euler equation; The vibration equation in step S22 is: ； in, From the perspective of the bionic execution unit, For amplitude, The vibration frequency, For phase, This is the offset angle; The driving equation for the biomimetic underwater robot described in step S23 is: ； In the formula, It is the standardization of input quantities. The coefficient matrix, They represent ; The yaw angle of the direction in which the bionic actuator is installed relative to the body coordinate system; The absolute value of the relative position of the point of application of the resultant force generated by the bionic actuator with respect to the origin of the body coordinate system; Indicates the yaw moment coefficient; six-dimensional input quantity This represents the force or torque at the corresponding degree of freedom. This represents the matrix composed of the horizontal and vertical components of the force generated by the bionic actuator.

9. The extreme fault-tolerant control method for biomimetic underwater robots according to claim 1, characterized in that, The steady-state equation of the biomimetic underwater robot established in step S26 is as follows: ； In the formula, , , These are the equivalent mass of the biomimetic underwater robot in the x-axis direction, the equivalent mass in the y-axis direction, and the equivalent mass due to the Coriolis effect in the body coordinate system, respectively. The steady-state velocity along the x-axis in the body coordinate system. The steady-state velocity along the y-axis in the body coordinate system. For steady-state yaw rate, These represent the x-direction in the body coordinate system when in steady state, direction and Damping coefficient of directional motion, The average horizontal force generated by fin vibration. The yaw angle of the direction in which the bionic actuator is installed relative to the body coordinate system. This is the coefficient of the torque generated by the thrust.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, wherein the computer program is configured to be run by a processor to perform the extreme fault-tolerant control method for a biomimetic underwater robot as described in any one of claims 1 to 9.

Images