A feedforward compensation control method and system for the release attitude of an underwater vehicle

By constructing an offline disturbance torque database and implementing feedforward compensation control, the underwater vehicle achieved active attitude stabilization when releasing the load, solving the attitude instability problem caused by transient impacts and improving the system's reliability and attitude control performance.

CN122861374BActive Publication Date: 2026-11-17NANJING UNIV OF SCI & TECH
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202611374979.8
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2026-09-07
Publication Date
2026-11-17
Estimated Expiration
2046-09-07

AI Technical Summary

Technical Problem

When underwater vehicles release their load, they become unstable due to transient strong disturbances. Existing technologies make it difficult to achieve effective real-time stable control, and external sensors are easily damaged or increase mechanical burden.

Method used

An offline disturbance torque database is constructed. The impact torque is calculated in advance and commands are sent to the attitude actuator through the feedforward compensation control method. Combined with the system response hysteresis time of the attitude actuator, the compensation torque and the impact torque are synchronously offset. Then, the control is switched to closed loop to correct the residual deviation.

Benefits of technology

This effectively overcomes feedback control hysteresis without increasing mechanical burden, improving the reliability and environmental adaptability of the underwater release system and ensuring attitude stability.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122861374B_ABST
    Figure CN122861374B_ABST
Patent Text Reader

Abstract

This invention discloses a feedforward compensation control method and system for the release attitude of an underwater vehicle, belonging to the field of vehicle attitude control technology. The method includes: constructing an offline disturbance torque database; extracting the transient impact torque characteristics generated on the underwater vehicle during load release under different operating conditions and preloading them into the main control module; acquiring current operating condition parameters and load model parameters; upon receiving the load release signal, calculating the estimated impact torque in real time in the database; before the estimated impact torque occurs, issuing a feedforward control command to the attitude actuator to establish a reverse compensation torque for active counterbalancing; after the impact force decays, eliminating residual attitude deviations through a closed-loop control algorithm. This invention overcomes the shortcomings of traditional feedback control in handling millisecond-level transient impacts due to physical hysteresis, leading to attitude instability, by combining feedforward compensation and closed-loop feedback control, thus improving the attitude stability of the underwater vehicle during load release.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention belongs to the field of vehicle attitude control technology, specifically relating to a feedforward compensation control method and system for the release attitude of an underwater vehicle. Background Technology

[0002] When underwater vehicles release loads, they face the problem of transient strong disturbances generated by load initiation / release. Due to the relatively small displacement and inertia of the underwater vehicle, the jet impact generated at the moment of load initiation during unmoored buoyancy creates a peak impact torque that bursts in milliseconds, severely disrupting the vehicle's attitude. Without effective control, this impact will seriously damage the underwater vehicle's motion attitude, leading to pitch, yaw, or roll instability, affecting subsequent operations and potentially causing the vehicle to lose its navigation control. Existing technologies typically employ two types of solutions to address the attitude stability problem during underwater load release.

[0003] One approach involves installing an array of physical pressure sensors external to the launch vehicle to detect changes in the external flow field and incorporate feedback control algorithms for attitude correction. However, under the high-pressure jet conditions generated during load startup, the physical sensors exposed on the outer shell surface are highly susceptible to damage and failure. Furthermore, feedback control relies on the generation of errors—that is, attitude deflection is detected first, followed by control correction. The entire feedback chain, from attitude sensing and controller calculation to the actuator establishing thrust, suffers from physical hysteresis. Under millisecond-level peak impact torque conditions, by the time the actuator establishes thrust, the impact has already caused severe attitude deviation, making effective real-time stable control impossible.

[0004] Another type involves adding mechanical shock absorbers to the vehicle to absorb impact energy, but for micro-sized underwater vehicles, such devices would impose an unacceptable weight and volume burden.

[0005] Therefore, providing an underwater stable release attitude control method and system that does not rely on external physical sensors, can overcome feedback control physical hysteresis, and does not increase additional mechanical burden has become a technical problem that urgently needs to be solved in this field. Summary of the Invention

[0006] The purpose of this invention is to provide a feedforward compensation control method for the release attitude of an underwater vehicle, comprising the following steps: Step S100: Construct an offline interference torque database, extract the transient impact torque characteristics of the starting jet generated by the load on the underwater vehicle under different operating conditions and model parameters, and preload it into the main control module; Step S200: Obtain the current operating parameters of the underwater vehicle and the model parameters of the load to be released; Step S300: After receiving the load release signal, calculate the estimated impact torque in real time in the offline interference torque database; Step S400: Before the estimated start time of the load arrives, send a feedforward command to the attitude actuator of the underwater vehicle to drive the attitude actuator to establish a reverse compensation torque in advance; Step S500: After the load impact force decays, dynamically adjust the output thrust of the attitude actuator through a closed-loop control algorithm until the underwater vehicle recovers the target attitude.

[0007] Further, step S100 includes: step S110, establishing a three-dimensional flow field model of the load and the underwater vehicle using computational fluid dynamics, setting different operating parameters and model parameters as boundary conditions, simulating the self-propulsion wake generated at the moment of start-up of the load self-propulsion system and the impact of the self-propulsion wake on the underwater vehicle; step S120, extracting the triaxial impact torque characteristics at the center of mass of the underwater vehicle, constructing an offline disturbance torque database and preloading it into the storage unit of the main control module.

[0008] Further, step S300 includes: step S310, when the main control module receives the load release signal, recording the current release time, where Δt represents the time when the main control module receives the load release signal; calculating the precise start time of the load underwater based on the load's set start delay time Δt, where Δt represents the delay time from the release signal to the actual start time of the load's underwater self-propulsion system; step S320, using the operating condition parameters and model parameters obtained in step S200 as indexes, performing real-time interpolation calculations in the offline interference torque database to extract the time-series curve of the estimated impact torque changing with time.

[0009] Further, step S400 includes: step S410, the main control module pre-stores the system dynamic parameters of the attitude actuator, including the system response hysteresis time τ required for the attitude actuator to output the target thrust from receiving the electrical signal command, where τ represents the time delay between the attitude actuator receiving the electrical signal command issued by the main control module and actually establishing the target thrust; step S420, before the precise start time arrives, the main control module issues a feedforward command to the attitude actuator at a certain time, driving the attitude actuator to start in advance, establishing a compensation torque to offset the estimated impact torque, so that the compensation torque and the actual self-propulsion wake impact torque are synchronously offset on the time axis.

[0010] Further, step S500 includes: step S510, after the load impact force attenuates, the main control module switches to closed-loop control mode, and collects the pitch angle and yaw angle attitude deviation data of the underwater vehicle through the status module, with the goal of maintaining the initial floating pitch angle, yaw angle and roll angle of the underwater vehicle, and dynamically adjusts the output thrust of the attitude actuator using a PID control algorithm; step S520, dynamically attenuates the feedforward compensation thrust proportionally, and superimposes the closed-loop feedback thrust based on the error signal obtained by the status module to correct the residual attitude deviation of the underwater vehicle until the underwater vehicle eliminates the disturbance and returns to the initial target attitude.

[0011] Further, it includes: an underwater vehicle, divided sequentially along its longitudinal axis into a bow section, a control section, and a stern section; a release device, located in the bow section, for loading the load to be released; an attitude actuator, located on the underwater vehicle, comprising a vertically penetrating channel propeller and a horizontally penetrating channel propeller penetrating the bow and stern section hulls respectively; a status module, located inside the control section, for acquiring the underwater vehicle's navigation attitude and operating condition data; a main control module, located inside the control section, communicating with the attitude actuator and the status module, the main control module storing an offline disturbance torque database; the main control module includes a calculation unit and a timing control unit; the calculation unit, upon receiving the load release signal, calculates the estimated impact torque in real time from the offline disturbance torque database based on operating condition parameters and model parameters; the timing control unit, before the estimated load start time arrives, sends a feedforward command to the attitude actuator to drive the attitude actuator to establish a reverse compensation torque.

[0012] Furthermore, the main control module is connected to the communication module that receives remote release control commands to synchronously acquire the load release signal.

[0013] Furthermore, the outer shell of the underwater vehicle exposed to the external water environment is not equipped with external physical sensors for sensing the pressure of the self-propelled wake flow field.

[0014] Compared with the prior art, the present invention has the following advantages: (1) By constructing an offline interference torque database for real-time lookup and calculation, the external physical sensors exposed to harsh working conditions are replaced, avoiding direct impact damage to the measuring elements caused by the high temperature, high pressure and overpressure disturbance waves generated by the load start-up jet, effectively improving the operational reliability, environmental adaptability and overall reliability of the underwater release system under complex working conditions.

[0015] (2) By introducing a timing compensation mechanism that includes system response lag time, a feedforward command is sent to the attitude actuator in advance before the estimated start time arrives, so that the compensation torque established by the attitude actuator and the transient impact torque generated by the load self-propulsion wake are precisely synchronized and offset on the time axis, changing "passive response" to "active offset", and overcoming the inherent defect of traditional feedback control that cannot cope with millisecond-level peak impact due to the physical lag of the perception-decision-execution link.

[0016] (3) By directly incorporating the transient disturbances of the extreme multiphase flow field into the attitude control equation, the existing through-type attitude actuator of the underwater vehicle is called to carry out high-power shock suppression. Without adding additional mechanical shock absorption devices or recoil reduction mechanisms, the active defense upgrade is completed, and the system optimization with zero additional dead weight and zero additional volume occupation is achieved.

[0017] The present invention will now be further described with reference to the accompanying drawings. Attached Figure Description

[0018] Figure 1 This is a schematic diagram of the control logic and timing flow of the underwater stable release method in this invention.

[0019] Figure 2 This is a schematic diagram of the underwater vehicle attitude stabilization control system based on feedforward compensation control in this invention.

[0020] In the figure, the underwater vehicle is 1, the bow section is 2, the control section is 3, the stern section is 4, the load is 5, the release device is 6, the attitude actuator is 7, the vertical through-channel thruster is 8, the horizontal through-channel thruster is 9, the status module is 10, and the main control module is 11. Detailed Implementation

[0021] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0022] This invention provides a feedforward compensation control method and system for the release attitude of an underwater vehicle, which is used to solve the problem of attitude instability caused by the transient impact of the starting jet when the underwater vehicle releases the load while suspended underwater.

[0023] Reference Figure 1 and Figure 2 The entire system is mounted on the underwater vehicle 1. The underwater vehicle 1 is divided into the bow section 2, the control section 3, and the stern section 4 along its longitudinal axis.

[0024] Reference Figure 2The bow section 2 is equipped with a release device 6 for loading the load 5 to be released. The underwater vehicle 1 is equipped with an attitude actuator 7 for attitude adjustment. The attitude actuator 7 includes a vertical through-slot thruster 8 and a horizontal through-slot thruster 9 that penetrate the bow section 2 and stern section 4 respectively, to provide lateral and vertical compensation torques with high dynamic response. The control section 3 houses a main control module 11 and a status module 10, which are sealed together. The status module 10 is used to acquire the current navigation status data of the underwater vehicle, such as depth, pitch angle, and yaw angle. The main control module 11 is communicatively connected to the attitude actuator 7 and the status module 10, and is also communicatively connected to a communication module that receives remote release control commands to synchronously acquire the release signal of the load 5. To adapt to extreme release conditions, the outer shell of the underwater vehicle 1 exposed to the external water environment is not equipped with external physical sensors for sensing the pressure of the self-propelled wake flow field.

[0025] Reference Figure 1 The main control module 11 executes a combined feedforward and feedback attitude control method, including the following steps.

[0026] Step S100: Construct an offline disturbance torque database and preload it into the main control module 11. This includes steps S110 and S120.

[0027] Step S110: A three-dimensional flow field model of the load 5 and the underwater vehicle 1 is established using computational fluid dynamics. Different operating parameters, model parameters, and other boundary conditions are set to simulate the self-propulsion wake generated at the moment of startup of the load 5's self-propulsion system, and the impact of this process on the underwater vehicle 1.

[0028] Step S120: Extract the triaxial impact torque characteristics at the center of mass of the underwater vehicle 1, construct an offline disturbance torque database and preload it into the storage unit of the main control module 11.

[0029] Step S200: Obtain the initial release parameters. This includes steps S210 and S220.

[0030] In step S210, after the underwater vehicle 1 enters the standby buoyancy state, the main control module 11 reads the current operating parameters through the status module 10.

[0031] In step S220, the control system confirms the model parameters of the load 5 to be released.

[0032] Step S300: Real-time calculation of the predicted impact torque. This includes steps S310 and S320.

[0033] Step S310: When the main control module 11 synchronously obtains the remote release signal of the load 5, record the current release time. Based on the start-up delay time Δt set for load 5, the main control module 11 calculates the precise start-up time of load 5 underwater. ; In step S320, the main control module 11 uses the operating condition parameters and model parameters obtained in step S200 as multi-dimensional indexes to perform real-time interpolation calculations in the offline interference torque database and extract the time-series curve of the corresponding estimated impact torque changing with time.

[0034] Step S400, feedforward compensation control. This includes steps S410 and S420.

[0035] In step S410, the main control module 11 pre-stores the system dynamic parameters of the attitude actuator 7, including the system response hysteresis time τ required from receiving the electrical signal command to the attitude actuator 7 outputting the target thrust.

[0036] Step S420, at the precise start-up time Before arrival, the main control module 11 at the specified time... Forced intervention involves sending feedforward commands to specific vertical through-channel thrusters 8 and horizontal through-channel thrusters 9, driving the attitude actuator 7 to start in advance and establishing a compensating torque to counteract the estimated impact torque. Through timing compensation, the compensating torque and the actual self-propulsion wake impact torque are synchronously offset on the time axis.

[0037] The technical principle of this step is explained below. Traditional feedback control relies on the generation of errors—that is, first sensing the attitude deflection, and then performing control correction. In the specific operating condition of underwater load starting jet, the impact torque explodes to its peak within milliseconds. However, the entire process of the feedback control link, from attitude sensing and controller calculation to the propeller establishing thrust, has physical lag. By the time the compensation torque is established, the impact has already caused a severe attitude deviation of the underwater vehicle, and the control cannot play a stabilizing role. This step pre-calculates the characteristics of the impact torque through an offline database and combines it with the system response lag time τ of the attitude actuator 7 to issue the action command in advance before the impact occurs. This ensures that the establishment of the compensation torque and the explosion of the impact torque are precisely synchronized on the time axis, changing the passive response to active counterbalancing, and overcoming the lag defect of feedback control at its root.

[0038] Step S500: Closed-loop feedback fine-tuning. This includes steps S510 and S520.

[0039] In step S510, as the load 5 moves away, the impact force from the self-propulsion wake decreases. After the feedforward compensation phase ends, the main control module 11 switches to closed-loop control mode. The attitude angle deviation data such as pitch angle and yaw angle of the underwater vehicle 1 are collected by the status module 10. With the goal of maintaining the initial hovering pitch angle, yaw angle and roll angle of the underwater vehicle 1, the output thrust of the attitude actuator 7 is dynamically adjusted using a PID control algorithm.

[0040] In step S520, as the impact force decreases, the feedforward compensation thrust is dynamically and proportionally reduced, and the closed-loop feedback thrust based on the error signal obtained by the state module 10 is superimposed to correct the residual attitude deviation of the underwater vehicle until the underwater vehicle 1 eliminates the disturbance and returns to the initial target attitude.

[0041] Steps S100 to S500 constitute a complete feedforward and feedback combined attitude control closed loop. Steps S100 to S300 provide accurate impact torque prediction data for feedforward control through offline pre-calculation and real-time table lookup; step S400 utilizes the timing compensation of system response hysteresis time to achieve synchronous offsetting of compensation torque and impact torque, which is a core innovation that distinguishes it from traditional feedback control; step S500 takes over control after impact attenuation to eliminate residual deviation. This scheme completes the active defense upgrade entirely using the existing through-type attitude actuator 7 of the underwater vehicle 1 without adding additional mechanical shock absorption devices or recoil reduction mechanisms, achieving system optimization with zero additional dead weight and zero additional volume occupation.

[0042] The method will be described below with reference to specific embodiments.

[0043] In this embodiment, the application scenario is a certain type of underwater vehicle releasing a light load while suspended at a depth of 30 meters underwater. The bow section 2 and stern section 4 of the underwater vehicle 1 are each equipped with a vertical through-slot thruster 8 and a horizontal through-slot thruster 9, and the system response hysteresis time τ of the attitude actuator 7 is 50 milliseconds.

[0044] During the system calibration phase, computational fluid dynamics was used to numerically simulate the underwater startup jet process of the load 5 in the depth range of 20 to 50 meters. The triaxial impact torque characteristics at the center of mass of the underwater vehicle 1 under each working condition were extracted, and an offline disturbance torque database was constructed.

[0045] During the release mission, the underwater vehicle 1 enters a hovering, ready-to-go state at a depth of 30 meters. The main control module 11 reads the current operating parameters as 30 meters from the status module 10, and the control system confirms that the load 5 to be released is a lightweight model. The main control module 11 simultaneously acquires the remote release signal from the load 5 and records the current release moment. Based on the start-up delay time Δt set to 100 milliseconds according to load 5, the main control module 11 calculates the precise underwater start-up time. for Add 100 milliseconds.

[0046] The main control module 11 uses the 30-meter depth and lightweight model parameters as indexes to perform real-time interpolation calculations in the offline interference torque database, extracting the corresponding time-series curve of the estimated impact torque changing over time. At startup... 50 milliseconds before arrival, the main control module 11 sends feedforward commands to the vertical through-channel thruster 8 and the horizontal through-channel thruster 9, driving the attitude actuator 7 to establish a reverse compensation torque in advance. At the moment the load 5 self-propulsion system starts, the self-propulsion wake impact torque and the compensation torque established by the attitude actuator 7 are synchronously offset on the time axis, and the attitude fluctuation of the underwater vehicle 1 is controlled within 0.5 degrees of pitch angle deviation and within 0.3 degrees of yaw angle deviation.

[0047] After the load 5 leaves the work area, the impact force decreases, and the main control module 11 switches to PID closed-loop control mode to dynamically adjust the output thrust of the attitude actuator 7. Within 2 seconds, the attitude of the underwater vehicle 1 is corrected to the initial floating state, and the stable release task is completed.

Claims

1. A feedforward compensation control method for the release attitude of an underwater vehicle, characterized in that, Includes the following steps: Step S100: Construct an offline interference torque database, extract the transient impact torque characteristics of the starting jet on the underwater vehicle under different operating conditions and model parameters, and preload it into the main control module; Step S200: Obtain the current operating parameters of the underwater vehicle and the model parameters of the load to be released; Step S300: After receiving the load release signal, calculate and estimate the impact torque in real time in the offline interference torque database; Step S400: Before the precise start-up time of the load arrives, a feedforward command is sent to the attitude actuator of the underwater vehicle to drive the attitude actuator to establish a reverse compensation torque in advance. In step S500, after the load impact force decays, the output thrust of the attitude actuator is dynamically adjusted through a closed-loop control algorithm until the underwater vehicle recovers its target attitude. Step S300 includes: Step S310: When the main control module receives the load release signal, record the current release time. ,in This indicates the moment when the main control module receives the load release signal; based on the load's set start-up delay time Δt, the precise start-up moment of the load underwater is calculated. Where Δt represents the delay time from the issuance of the release signal to the actual activation of the load's underwater self-propulsion system. Indicates the actual start-up time of the underwater self-propulsion system under load; Step S320: Using the working condition parameters and model parameters obtained in step S200 as indexes, perform real-time interpolation calculation in the offline disturbance torque database to extract the time series curve of the estimated impact torque changing with time. Step S400 includes: Step S410: The main control module has pre-stored the system dynamic parameters of the attitude actuator, including the system response hysteresis time τ required for the attitude actuator to output the target thrust from receiving the electrical signal command. Here, τ represents the time delay between the attitude actuator receiving the electrical signal command issued by the main control module and actually establishing the target thrust. Step S420, at the precise start-up time Before arrival, the main control module at time A feedforward command is sent to the attitude actuator to drive it to start in advance and establish a compensation torque to offset the estimated impact torque, so that the compensation torque and the actual self-propulsion wake impact torque are synchronously offset on the time axis.

2. The feedforward compensation control method for the release attitude of an underwater vehicle according to claim 1, characterized in that, Step S100 includes: Step S110: A three-dimensional flow field model of the load and the underwater vehicle is established using computational fluid dynamics. Different operating parameters and model parameters are set as boundary conditions to simulate the self-propulsion wake generated at the moment of start-up of the load's self-propulsion system and the impact of the self-propulsion wake on the underwater vehicle. Step S120: Extract the triaxial impact torque characteristics at the center of mass of the underwater vehicle, construct an offline disturbance torque database, and preload it into the storage unit of the main control module.

3. The feedforward compensation control method for the release attitude of an underwater vehicle according to claim 1, characterized in that, Step S500 includes: Step S510: After the load impact force decays, the main control module switches to closed-loop control mode. The status module collects pitch and yaw attitude deviation data of the underwater vehicle. With the goal of maintaining the initial floating pitch, yaw and roll angles of the underwater vehicle, the PID control algorithm is used to dynamically adjust the output thrust of the attitude actuator. Step S520: Dynamically and proportionally attenuate the feedforward compensation thrust and superimpose the closed-loop feedback thrust based on the error signal obtained by the state module to correct the residual attitude deviation of the underwater vehicle until the underwater vehicle eliminates the disturbance and returns to the initial target attitude.

4. A feedforward compensation control system for the release attitude of an underwater vehicle, used to implement the feedforward compensation control method for the release attitude of an underwater vehicle as described in claim 1, characterized in that, include: The underwater vehicle (1) is divided into a bow section (2), a control section (3) and a stern section (4) along its longitudinal axis. Release device (6), located in the bow section (2), is used to load the load (5) to be released; Attitude actuator (7) is installed on the underwater vehicle (1). The attitude actuator (7) includes a vertical through-channel thruster (8) and a horizontal through-channel thruster (9) that penetrate the bow section (2) and stern section (4) respectively. The status module (10) is located inside the control section (3) and is used to acquire the navigation attitude and operating condition data of the underwater vehicle. The main control module (11) is located inside the control section (3) and is connected to the attitude actuator (7) and the status module (10). The main control module (11) stores an offline disturbance torque database. The main control module (11) includes a calculation unit and a timing control unit. The calculation unit is used to calculate the estimated impact torque in real time in the offline interference torque database according to the working condition parameters and model parameters after receiving the load (5) release signal. The timing control unit is used to send a feedforward command to the attitude actuator (7) in advance before the precise start time of the load (5) arrives, and drive the attitude actuator (7) to establish a reverse compensation torque.

5. The feedforward compensation control system for the release attitude of an underwater vehicle according to claim 4, characterized in that, The main control module (11) is connected to the communication module that receives remote release control commands, and is used to synchronously acquire the release signal of the load (5).

6. The feedforward compensation control system for the release attitude of an underwater vehicle according to claim 4, characterized in that, The outer shell of the underwater vehicle (1) exposed to the external water environment is not equipped with external physical sensors for sensing the pressure of the self-propelled wake flow field.

Citation Information

Patent Citations

  • Self-adaptive advanced pitch control method, system and equipment for wind turbine generator and medium

    CN122649953A

  • A flight robot control method for an over-the-air release task

    CN122691280A