An underwater robot posture control method for suspended inclined operation

Abstract

This invention discloses an attitude control method for underwater robots performing buoyant tilting operations. The method includes the following steps: First, a dynamic model considering the coupling between depth and attitude angle during the underwater robot's motion is established. Next, based on the dynamic model, a desired-compensation adaptive robust attitude controller for the underwater robot is constructed. Then, the actual state of the underwater robot is continuously acquired, and the desired-compensation adaptive robust attitude controller continuously predicts the resultant thrust based on the desired state and the acquired actual state, and sends the predicted resultant thrust to the actuators of the underwater robot, thereby achieving stable control of the underwater robot's posture. This invention enables precise real-time control of the underwater robot's posture, solving the control difficulties and parameter uncertainties caused by the coupling of the dynamic model in the buoyant tilting state. This reduces the posture control error of the ROV, improves control performance, ensures system stability, and provides a foundation for underwater robots to perform tilting operations.

Description

Technical Field

[0001] This invention belongs to the field of motion control for underwater robots, and specifically relates to an attitude control method for underwater robots oriented towards suspended tilting operations. Background Technology

[0002] Remotely operated vehicles (ROVs) are widely used in various underwater exploration and operational tasks. Most ROVs are used as equipment platforms, but with industrial development and continuous human exploration, new demands are being placed on their operational capabilities. To meet the needs of ship inspection and exploration in complex waters, ROVs need to have the ability to control their own tilting.

[0003] In a tightly coupled motion system like an ROV, traditional open-loop control and proportional-integral-differential (PID) control often require decoupling and cannot account for the uncertainties in the ROV's dynamic model parameters, failing to meet the precise control requirements for tilting and hovering. In this context, developing a nonlinear controller based on the dynamic model of a fully free-degree-of-freedom underwater robot (ROV) is an effective solution. However, most existing ROV motion controller designs often neglect full-degree-of-freedom control and lack consideration for model parameter uncertainties, making it difficult for the ROV to actively achieve tilting attitudes, or to perform precise pose control in tilted attitudes, thus hindering its ability to handle tasks in a hovering and tilting state. Summary of the Invention

[0004] The purpose of this invention is to address the shortcomings of existing technologies by providing a posture control method for underwater robots (ROVs) operating in a hovering and tilting posture. Specifically, this invention is an adaptive robust motion control method that considers the pose coupling of the ROV in a hovering and tilting posture. This method effectively improves the response speed of the control system, overcomes the influence of the ROV's own nonlinear strong coupling and external disturbances, and ensures the stability of its multi-degree-of-freedom motion control in a tilting posture. Therefore, it provides control-level protection for hovering and tilting operations and enhances the ROV's motion capabilities.

[0005] The technical solution adopted in this invention is: In a first aspect, the present invention proposes an attitude control method for an underwater robot oriented towards suspended tilting operations, the method comprising the following steps: Step 1: Establish a dynamic model that considers the coupling between depth and attitude angle during the underwater robot's motion; Step 2: Based on the dynamic model established in Step 1, construct an expected compensation adaptive robust attitude controller for the underwater robot. The third step is to continuously acquire the actual state of the underwater robot. The expected compensation adaptive robust attitude controller continuously predicts the resultant thrust based on the expected state and the acquired actual state, and sends the predicted resultant thrust to the actuator of the underwater robot to achieve stable control of the underwater robot's posture.

[0006] Furthermore, in the first step, the dynamic model considering the coupling of depth and attitude angle in the underwater robot's motion is a four-degree-of-freedom dynamic model, with the four degrees of freedom being the underwater robot's depth, pitch, roll, and heading.

[0007] Furthermore, in the desired compensation adaptive robust attitude controller, a virtual error is calculated based on the desired state and the acquired actual state; after parameter adaptation based on the virtual error, estimated values ​​of the underwater robot's hydrodynamic-damping parameters and inertial dynamic parameters are output; after model compensation based on the acquired actual state, virtual error, and estimated values ​​of the underwater robot's hydrodynamic-damping parameters and inertial dynamic parameters, a model compensation term is obtained; after robust feedback based on the virtual error, a robust feedback term is obtained; finally, a predicted resultant thrust is generated based on the model compensation term and the robust feedback term.

[0008] Further, the calculation of the virtual error s based on the desired state and the acquired actual state includes: ; ; in, The derivative of the control error e is represented by k1; the virtual error gain coefficient is represented by k1; and the actual pose of the underwater robot is represented by η. d The desired pose of the underwater robot; Furthermore, the step of outputting estimated values ​​of the underwater robot's hydrodynamic-damping parameters and inertial dynamic parameters after parameter adaptation based on virtual errors includes:

[0009]

[0010] in, The desired linear regression vector of the system model; This is the parameter estimation vector for the hydrodynamics and damping of the underwater robot; Here, is the parameter adaptive gain matrix; s represents the virtual error; For projection functions; Estimates of hydrodynamic-damping parameters for underwater robots The i-th component in These are the i-th components of the upper and lower bounds of the hydrodynamic-damping parameters of the underwater robot, respectively.

[0011] Furthermore, after performing model compensation based on the obtained actual state, virtual error, and estimated values ​​of the underwater robot's hydrodynamic-damping parameters and inertial dynamic parameters, a model compensation term is obtained, including:

[0012] in, Hydrodynamic-damping parameters and inertial dynamic parameters of underwater robots The estimated value; These are estimated values ​​of the hydrodynamic-damping parameters and inertial dynamic parameters of the underwater robot. The corresponding regression matrix.

[0013] Furthermore, the robust feedback term obtained after performing robust feedback based on the virtual error includes:

[0014]

[0015] Among them, u s It is the robust feedback term; s is the virtual error; u s1 It is a linear robust feedback term; u s2 It is a nonlinear robust feedback term; k s1 k represents the first robust feedback gain coefficient. s2 d represents the second robust feedback gain coefficient; max d min These are the upper and lower bounds of the external disturbance; θ gmax θ gmin These are the upper and lower bounds of the inertial dynamics parameters of the underwater robot; θ amax θ amin ε represents the upper and lower bounds of the hydrodynamic-damping parameters of the underwater robot; ε is the control adjustment parameter of the nonlinear robust feedback term.

[0016] Furthermore, in the third step, the predicted combined thrust is sent to the actuators of the underwater robot, including: The thrust of each thruster in the underwater robot is generated based on the predicted combined thrust. Considering the nonlinearity of the thruster itself, the control signal input of each thruster is generated based on the thrust of each thruster, and then each thruster is driven based on the generated control signal input.

[0017] Secondly, the present invention proposes an attitude control device for an underwater robot oriented towards suspended tilting operations, the device comprising: The model building unit is used to establish a dynamic model that considers the coupling between depth and attitude angle during the motion of an underwater robot. The state acquisition unit is used to acquire the actual state of the underwater robot; The control unit is used to construct an expected compensation adaptive robust attitude controller for the underwater robot by combining the established dynamic model. The expected compensation adaptive robust attitude controller predicts the resultant thrust based on the expected state and the acquired actual state. The thrust conversion unit is used to send the predicted combined thrust to the actuators of the underwater robot.

[0018] Furthermore, the desired compensation adaptive robust attitude controller includes: The virtual error calculation module is used to calculate the virtual error based on the desired state and the acquired actual state. The parameter adaptive law is used to adapt parameters based on virtual errors and output estimated values ​​of the hydrodynamic-damping parameters and inertial dynamic parameters of the underwater robot. The robust feedback module is used to obtain robust feedback items after robust feedback is performed based on the virtual error; The model compensation module is used to perform model compensation based on the actual state, virtual error, and estimated values ​​of the underwater robot's hydrodynamic-damping parameters and inertial dynamic parameters, and then obtain the model compensation term. The combined thrust output module is used to generate the predicted combined thrust based on the model compensation term and the robust feedback term.

[0019] The beneficial effects of this invention are: 1. This invention solves the problem of actual ROV tilt attitude control by establishing a dynamic model that considers the coupling between depth and attitude angle in ROV motion and using a full-degree-of-freedom model to avoid model errors caused by decoupling; at the same time, it considers the influence of the thruster dead zone and its nonlinear characteristics on the control effect.

[0020] 2. The method proposed in this invention achieves model compensation through adaptive model parameter adjustment and designs robust feedback terms, thereby ensuring the overall stability of the control system, shortening the response time, reducing control error, and improving control performance. Attached Figure Description

[0021] Figure 1 This is a control block diagram of the present invention; Figure 2 This is the experimental environment and experimental hardware of the present invention; wherein, (a) is a schematic diagram of the experimental hardware, and (b) is a schematic diagram of the experimental environment; Figure 3 This invention presents the control objectives and attitude control results when the four degrees of freedom of the ROV (depth, roll, pitch, and yaw) are controlled simultaneously in the experiment. Among them, (a) is the expected target and control result for depth; (b) is the expected target and control result for roll; (c) is the expected target and control result for pitch; and (d) is the expected target and control result for yaw. Figure 4 This invention relates to the control objectives and attitude control results of simultaneously controlling the four degrees of freedom of attitude angle and depth during planar motion along a desired trajectory in the experiment of this invention; where (a) is the desired objective and control result for depth; (b) is the desired objective and control result for roll; (c) is the desired objective and control result for pitch; and (d) is the desired objective and control result for yaw. Figure 5 These are the model parameter estimation results of the experiment of this invention when the ROV moves horizontally in an inclined posture; where (a) is the estimated external disturbance and (b) is the estimated model parameter. Detailed Implementation

[0022] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. The specific embodiments described herein are merely illustrative and are not intended to limit the scope of the invention.

[0023] like Figure 1 As shown, the underwater robot attitude control method for levitation and tilting operations proposed in this invention includes the following steps: Step 1: Establish a dynamic model that considers the coupling between depth and attitude angle during the underwater robot's motion; This invention considers the difference in ROV state information between the volume coordinate system and the ground coordinate system. The specific dynamic model of the pose coupling effect during ROV motion and the hydrodynamic characteristics of ROV is as follows: To address the four-degree-of-freedom control problem of ROV depth and three attitude angles, this invention balances the deployability of the controller with the accuracy of model compensation. Based on the traditional six-degree-of-freedom ROV dynamics model, a simplified and modified four-degree-of-freedom dynamics model is established:

[0024] in, It is the inertia matrix. It is the matrix of Coriolis force and centripetal force. Here is the fluid damping matrix. It is the gravity vector. Indicates time-varying external disturbances. This indicates the uncertainty of the model. This represents the desired resultant thrust output by the controller. These are the ROV's longitudinal, pitch, roll, and bow velocities in the body coordinate system. This represents the four-degree-of-freedom pose of the ROV in the ground coordinate system, satisfying... , These represent the ROV's longitudinal, pitch, roll, and bow attitudes in the ground coordinate system. Represents four-degree-of-freedom velocity The differential, T represents the transpose operation.

[0025] Inertia matrix M and fluid damping matrix They are diagonalized because their off-diagonal coupling terms are small enough to be incorporated into the model's uncertainty handling. The four-degree-of-freedom inertia matrix M is as follows:

[0026] Where m is the mass of the ROV. Let be the moments of inertia about the x-axis, y-axis, and z-axis. It is the additional mass generated underwater by the four degrees of freedom: longitudinal, pitch, roll, and bow.

[0027] Four-degree-of-freedom fluid damping matrix Specifically as follows:

[0028] in, These are the primary and secondary fluid damping coefficients for the degrees of freedom of depth, pitch, roll, and bow, respectively.

[0029] Based on the Coriolis force and centripetal force matrices in six degrees of freedom, the coupling terms of the x-axis, y-axis, and z-axis are added. , Incorporating the matrix, the coupling terms of the x-axis and y-axis By incorporating the external perturbation d, we obtain the Coriolis force and centripetal force matrices under four degrees of freedom. The formula is as follows:

[0030] Vector of restoring force and torque The formula is as follows:

[0031] Where W is the magnitude of the ROV's weight, and B is the magnitude of the buoyancy force acting on the ROV. , , Here are the coordinates of the ROV's center of gravity. , , Here are the coordinates of the ROV's center of buoyancy.

[0032] Step 2: Based on the dynamic model established in Step 1, construct an expected compensation adaptive robust attitude controller for the underwater robot. The third step is to continuously acquire the actual state of the underwater robot. The expected compensation adaptive robust attitude controller continuously predicts the resultant thrust based on the expected state and the acquired actual state, and sends the predicted resultant thrust to the actuator of the underwater robot to achieve stable control of the underwater robot's posture.

[0033] In the desired compensation adaptive robust attitude controller, the virtual error s is calculated based on the desired state and the acquired actual state; after parameter adaptation based on the virtual error s, the hydrodynamic-damping parameters and inertial dynamic parameters θ of the underwater robot are output. a θ g The estimated value , Based on the obtained actual state, virtual error s, and underwater robot hydrodynamic-damping parameters and inertial dynamic parameters θ a θ g The estimated value , After model compensation, the model compensation term is obtained; after robust feedback based on the virtual error s, the robust feedback term is obtained; finally, the predicted combined thrust is generated based on the model compensation term and the robust feedback term.

[0034] The predicted combined thrust is generated based on the model compensation term and the robust feedback term, as shown in the following formula:

[0035] Where u is the predicted resultant thrust output by the controller, u m It is the model compensation term output by the controller, u s It is the robust feedback term output by the controller. m Feedback linearization is achieved based on the model parameters estimated by the parameter adaptive rate, thereby reducing the influence of nonlinear dynamics. s Robust control of ROVs is achieved based on control errors.

[0036] For calculating the compensation term u in the model m Define the virtual error s and design the parameter adaptive rate.

[0037] The virtual error s is calculated based on the desired state and the acquired actual state, including: ;

[0038] in, η represents the derivative of the control error e; k1 represents the virtual error gain coefficient, which is a positive number to ensure that when the virtual error s approaches 0, the control error e also approaches 0; η is the actual pose of the underwater robot; η d The desired pose of the underwater robot; Differentiating the virtual error s yields... Combining the established four-degree-of-freedom dynamic model and kinematic model, the following is derived:

[0039] Further refinement and definition Represents unknown model parameters. It is the corresponding regression matrix. Let the derivative of the desired pose of the underwater robot be expressed as:

[0040] in, This indicates the hydrodynamic-damping parameters of the underwater robot. This represents the inertial dynamics parameters of the underwater robot. The details are as follows:

[0041] Where J is the kinematic transformation matrix of the ROV. Represents the differential of the kinematic transformation matrix; used to describe the velocity in the body coordinate system. Transformation relationship between pose η in ground coordinate system and ground coordinate system:

[0042]

[0043] The model compensation term is derived from the formula as follows:

[0044] in, These are the hydrodynamic-damping parameters and inertial dynamic parameters θ of the underwater robot. a θ g The estimated value is obtained through the adaptive rate of the designed parameters; These are estimated values ​​of the hydrodynamic-damping parameters and inertial dynamic parameters of the underwater robot. The corresponding regression matrix. It's important to note that the model compensation terms use the desired attitude and velocity, not the measured values.

[0045] The specific adaptive rate of the designed parameters is as follows:

[0046]

[0047] in, The desired linear regression vector of the system model; This is the parameter estimation vector for the hydrodynamics and damping of the underwater robot; Here, is the parameter adaptive gain matrix; s represents the virtual error; For projection functions; Estimates of hydrodynamic-damping parameters for underwater robots The i-th component in These are the i-th components of the upper and lower bounds of the hydrodynamic-damping parameters of the underwater robot, respectively. For the inertial dynamic parameters... The calculation of its estimated value is the same as .

[0048] After robust feedback is performed based on the virtual error s, robust feedback terms are obtained, including:

[0049]

[0050] Among them, u s It is the robust feedback term output by the controller; s is the virtual error; u s1 It is a linear robust feedback term; u s2 It is a nonlinear robust feedback term; k s1 k represents the first robust feedback gain coefficient. s2 d represents the second robust feedback gain coefficient; max d min These are the upper and lower bounds of the external disturbance; θ gmax θ gmin These are the upper and lower bounds of the inertial dynamics parameters of the underwater robot; θ amax θ amin ε represents the upper and lower bounds of the hydrodynamic-damping parameters of the underwater robot; ε is the control adjustment parameter of the nonlinear robust feedback term.

[0051] In one feasible implementation, the predicted combined thrust is sent to the actuator of the underwater robot, comprising: The thrust of each thruster in the underwater robot is generated based on the predicted combined thrust. Considering the nonlinearity of the thruster itself, the control signal input of each thruster is generated based on the thrust of each thruster, and then each thruster is driven based on the generated control signal input.

[0052] Specifically, the desired resultant thrust output by the controller Thrust vector assigned to the desired individual thruster The specific conversion relationships are as follows:

[0053] Where B is the configuration matrix related to the propulsion system layout. It is the pseudo-inverse of the configuration matrix B.

[0054] The thruster's inherent nonlinearity includes the input dead zone and the speed-thrust nonlinearity, and the thruster thrust. With thruster control signal input u p The relationship between them can be specifically described as follows:

[0055] Among them, u d It corresponds to the thrust dead zone boundary Value, k c It is the second-order factor when the thruster is rotating forward, k ac The second-order factor when the thruster reverses.

[0056] This invention employs a closed-loop structure for precise control of the depth and attitude angles of an ROV. First, the system receives externally set desired attitude and depth commands. Simultaneously, the system measures and acquires the ROV's actual attitude and depth using sensors. The desired state is compared with the actual state to generate an error signal. This error signal serves as the controller input. Model parameters are estimated using an adaptive parameter law, and model compensation and robust feedback terms are calculated. These two factors are combined to obtain the controller output. Based on the ROV's thrust matrix, thrust is allocated using a pseudo-inverse method to generate the desired thrust for each thruster, which is then calculated based on the nonlinear relationship between thrust and control signals. Finally, the control input directly acts on the ROV, driving its thrusters to adjust its attitude and depth. The ROV's motion state is continuously monitored and fed back by sensors, thus forming a control closed loop.

[0057] A control experiment based on a fully free-roaming ROV was conducted to test the above control method, and the results were compared with those of a simple decoupled PID controller to verify the control effect of the proposed control method. The experimental hardware included... Figure 2 As shown in (a), the experimental environment is as follows Figure 2 As shown in (b).

[0058] The following comparative experiments were conducted: Experiment 1 implemented four-degree-of-freedom joint control of the ROV, setting the yaw angle control target to -1.5 rad, the roll and pitch angle control targets to -0.2 rad, and the depth control target to 0.4 m; Experiment 2, based on the four-degree-of-freedom joint control, allowed the ROV to move horizontally in a plane along a predetermined trajectory, setting the yaw angle control target to 0 rad, the roll and pitch angle control targets to -0.2 rad, and the depth control target to 0.4 m.

[0059] The PID controller is denoted as C1, and the controller proposed in this invention is denoted as C2. During verification, the parameter selections for controllers C1 and C2 are shown in Table 1.

[0060] Table 1 Controller Parameter Selection

[0061] The experimental results of Experiment 1 are as follows Figure 3 As shown, Figure 3 (a) represents the desired target and control result for depth; Figure 3 (b) represents the expected target and control result for the roll angle; Figure 3 (c) represents the desired target and control result for the pitch angle; Figure 3 (d) represents the desired target and control result for the yaw angle; curve Ref is the desired target, curve C1 is the control result of the PID controller, and curve C2 is the control result of the desired compensation adaptive robust controller; the C1 controller has a slower response speed to the control target, a larger overshoot, and low control stability, while the C2 controller can effectively suppress oscillations and has significant control performance.

[0062] The experimental results of Experiment 2 are as follows Figure 4 As shown, Figure 4 (a) represents the desired target and control result for depth; Figure 4 (b) represents the expected target and control result for the roll angle; Figure 3 (c) represents the desired target and control result for the pitch angle; Figure 4 (d) represents the desired target and control result for the yaw angle; curve Ref is the desired target, curve C1 is the control result of the PID controller, and curve C2 is the control result of the desired compensated adaptive robust controller. From Figure 4 As can be seen, under the condition of maintaining a tilted attitude and performing planar motion, the C1 controller exhibits obvious oscillations, while the C2 controller can effectively suppress these disturbances caused by motion and maintain stable high-precision tracking. Its root mean square error of depth and roll angle is significantly reduced compared to the C1 controller.

[0063] The parameter estimation results of Experiment 2 are as follows Figure 5 As shown, Figure 5 (a) demonstrates the controller's efficient estimation of the lumped external disturbances experienced by the system. Figure 5 (b) shows the online estimation process of the ROV model parameters by the adaptive law. It can be seen that the parameter estimates can converge quickly and stabilize around a specific value. The parameters that jump in some places also reflect the uncertainty of the model parameters under different motion states.

[0064] In summary, the adaptive robust controller proposed in this invention significantly improves upon traditional PID controllers in both transient response speed and steady-state tracking accuracy. Especially when facing multi-degree-of-freedom coupling and strong external dynamic disturbances, this invention, through parameter-adaptive model compensation and robust feedback mechanisms, exhibits superior control performance and robustness, effectively enhancing the operational capability and reliability of ROVs in tilted postures.

[0065] The present invention proposes an underwater robot attitude control device for levitation and tilting operations, comprising: The model building unit is used to establish a dynamic model that considers the coupling between depth and attitude angle during the motion of an underwater robot. The state acquisition unit is used to acquire the actual state of the underwater robot; The control unit is used to construct an expected compensation adaptive robust attitude controller for the underwater robot by combining the established dynamic model. The expected compensation adaptive robust attitude controller predicts the resultant thrust based on the expected state and the acquired actual state. The thrust conversion unit is used to send the predicted combined thrust to the actuators of the underwater robot.

[0066] In one feasible implementation, the desired compensation adaptive robust attitude controller includes: The virtual error calculation module is used to calculate the virtual error based on the desired state and the acquired actual state. The parameter adaptive law is used to adapt parameters based on the virtual error s, and outputs the hydrodynamic-damping parameters and inertial dynamic parameters θ of the underwater robot. a θ g The estimated value , ; The robust feedback module is used to obtain the robust feedback item after robust feedback is performed based on the virtual error s; The model compensation module is used to calculate the actual state, virtual error s, and hydrodynamic-damping parameters and inertial dynamic parameters θ of the underwater robot. a θ g The estimated value , After performing model compensation, the model compensation term is obtained; The combined thrust output module is used to generate the predicted combined thrust based on the model compensation term and the robust feedback term.

[0067] The above content is merely a technical concept of the present invention and should not be construed as limiting the scope of protection of the present invention. Any modifications made to the technical solution based on the technical concept proposed in this invention shall fall within the scope of protection of the claims of this invention.

Claims

1. A method for attitude control of an underwater robot for suspended tilting operations, characterized in that, Includes the following steps: Step 1: Establish a dynamic model that considers the coupling between depth and attitude angle during the underwater robot's motion; Step 2: Based on the dynamic model established in Step 1, construct the desired compensation adaptive robust attitude controller for the underwater robot. The third step is to continuously acquire the actual state of the underwater robot. The expected compensation adaptive robust attitude controller continuously predicts the resultant thrust based on the expected state and the acquired actual state, and sends the predicted resultant thrust to the actuator of the underwater robot to achieve stable control of the underwater robot's posture.

2. The attitude control method for an underwater robot oriented towards suspended tilting operations according to claim 1, characterized in that, In the first step, the dynamic model considering the coupling of depth and attitude angle in the underwater robot's motion is a four-degree-of-freedom dynamic model, with the four degrees of freedom being the underwater robot's depth, pitch, roll, and heading.

3. The attitude control method for an underwater robot oriented towards suspended tilting operations according to claim 1, characterized in that, In the desired compensation adaptive robust attitude controller, a virtual error is calculated based on the desired state and the acquired actual state; after parameter adaptation based on the virtual error, estimated values ​​of the underwater robot's hydrodynamic-damping parameters and inertial dynamic parameters are output; after model compensation based on the acquired actual state, virtual error, and estimated values ​​of the underwater robot's hydrodynamic-damping parameters and inertial dynamic parameters, a model compensation term is obtained; after robust feedback based on the virtual error, a robust feedback term is obtained; finally, the predicted resultant thrust is generated based on the model compensation term and the robust feedback term.

4. The attitude control method for an underwater robot oriented towards suspended tilting operations according to claim 1, characterized in that, The calculation of the virtual error s based on the desired state and the acquired actual state includes: ； ； in, The derivative of the control error e is represented by k1; the virtual error gain coefficient is represented by k1; and η is the actual pose of the underwater robot. d This represents the desired pose of the underwater robot.

5. The attitude control method for an underwater robot oriented towards suspended tilting operations according to claim 1, characterized in that, The process of adapting parameters based on virtual errors and outputting estimated values ​​of the underwater robot's hydrodynamic-damping parameters and inertial dynamic parameters includes: in, The desired linear regression vector of the system model; This is the parameter estimation vector for the hydrodynamics and damping of the underwater robot; Here, is the parameter adaptive gain matrix; s represents the virtual error; For projection functions; Estimates of hydrodynamic-damping parameters for underwater robots The i-th component in These are the i-th components of the upper and lower bounds of the hydrodynamic-damping parameters of the underwater robot, respectively.

6. The attitude control method for an underwater robot oriented towards suspended tilting operations according to claim 1, characterized in that, After performing model compensation based on the obtained actual state, virtual error, and estimated values ​​of the underwater robot's hydrodynamic-damping parameters and inertial dynamic parameters, the model compensation term is obtained, including: in, Hydrodynamic-damping parameters and inertial dynamic parameters of underwater robots The estimated value; These are estimated values ​​of the hydrodynamic-damping parameters and inertial dynamic parameters of the underwater robot. The corresponding regression matrix.

7. The attitude control method for an underwater robot oriented towards suspended tilting operations according to claim 1, characterized in that, After performing robust feedback based on the virtual error, the robust feedback item is obtained, including: Among them, u s It is the robust feedback term; s is the virtual error; u s1 It is a linear robust feedback term; u s2 It is a nonlinear robust feedback term; k s1 k represents the first robust feedback gain coefficient. s2 d represents the second robust feedback gain coefficient; max d min These are the upper and lower bounds of the external disturbance; θ gmax θ gmin These are the upper and lower bounds of the inertial dynamics parameters of the underwater robot; θ amax θ amin ε represents the upper and lower bounds of the hydrodynamic-damping parameters of the underwater robot; ε is the control adjustment parameter of the nonlinear robust feedback term.

8. The attitude control method for an underwater robot oriented towards suspended tilting operations according to claim 1, characterized in that, In the third step, the predicted combined thrust is sent to the actuators of the underwater robot, including: The thrust of each thruster in the underwater robot is generated based on the predicted combined thrust. Considering the nonlinearity of the thruster itself, the control signal input of each thruster is generated based on the thrust of each thruster, and then each thruster is driven based on the generated control signal input.

9. An attitude control device for an underwater robot designed for suspended tilting operations, characterized in that, include: The model building unit is used to establish a dynamic model that considers the coupling between depth and attitude angle during the motion of an underwater robot. The state acquisition unit is used to acquire the actual state of the underwater robot. The control unit is used to construct an expected compensation adaptive robust attitude controller for the underwater robot by combining the established dynamic model. The expected compensation adaptive robust attitude controller predicts the resultant thrust based on the expected state and the acquired actual state. The thrust conversion unit is used to send the predicted combined thrust to the actuators of the underwater robot.

10. The underwater robot attitude control device for suspended tilting operations according to claim 9, characterized in that, The desired compensation adaptive robust attitude controller includes: The virtual error calculation module is used to calculate the virtual error based on the desired state and the acquired actual state. The parameter adaptive law is used to adapt parameters based on virtual errors and output estimated values ​​of the hydrodynamic-damping parameters and inertial dynamic parameters of the underwater robot. The robust feedback module is used to obtain robust feedback items after robust feedback is performed based on the virtual error; The model compensation module is used to perform model compensation based on the actual state, virtual error, and estimated values ​​of the underwater robot's hydrodynamic-damping parameters and inertial dynamic parameters, and then obtain the model compensation term. The combined thrust output module is used to generate the predicted combined thrust based on the model compensation term and the robust feedback term.

Images