An unmanned ship navigation control system quick refitting device and control method

Abstract

This invention discloses a rapid retrofit device and control method for an unmanned vessel navigation control system. The system includes: a main engine control module that converts the original ship's control signals into digital electronic control signals; a rudder control module that adjusts rudder angle and stabilizes course; a navigation control system module with a shipborne-base station dual-layer architecture supporting three modes: manual driving, base station remote control, and autonomous navigation; an environmental perception system module that collects information on the ship's position, attitude, environment, and targets; and a communication module responsible for command, status, and data exchange between the shipborne system and the base station. During system initialization or mode switching, the navigation control module outputs test signals, collects the original ship's response data, and constructs a nonlinear dead zone and hysteresis model. In unmanned navigation mode, this model is used to perform feedforward compensation on commands, eliminating control hysteresis and oscillations caused by mechanical coupling. The system also includes an energy management module and an additional module. This provides reliable technical support for upgrading unmanned ship technology.

Description

Technical Field

[0001] This invention belongs to the field of ship intelligent control technology, and more specifically, relates to a rapid retrofit device and control method for an unmanned vessel navigation control system. Background Technology

[0002] The global military is undergoing a rapid transformation towards intelligent warfare, leading to a surge in demand for various marine development and maritime operations. The conversion of ships to unmanned operation has become a crucial direction for transforming maritime warfare and upgrading the civilian aviation industry. For a long time, the navigation and control of various vessels have relied entirely on manual labor, requiring operators to adjust gears, throttle, and rudder angles in real time to maintain the vessel's navigation. In long-term operations or complex sea conditions, the drawbacks of manual control are amplified, not only increasing overall labor costs but also making manual errors unavoidable. Furthermore, the response speed of manual emergency response to unforeseen maritime situations is significantly insufficient, failing to meet the demands of high-intensity operations.

[0003] Currently, the industry's efforts to retrofit ships to unmanned operation still face numerous unresolved technical challenges. The electronic control systems of different ship types vary significantly; digital and non-digital ships cannot share a single retrofit solution. Communication protocols for various devices are incompatible, and the conversion from traditional physical control signals to standard digital electronic control signals is difficult. A universal retrofit path has long been lacking. Furthermore, most retrofit solutions isolate the main engine, steering gear, and navigation control systems, hindering efficient data exchange between units and resulting in low overall system data fusion capabilities and collaborative decision-making levels. Existing retrofit equipment generally lacks comprehensive hardware redundancy design, and its accompanying software functions are relatively limited, failing to achieve seamless switching between various operating modes of autonomous navigation remotely controlled by a base station, thus limiting its adaptability to various scenarios. In addition, facing complex and variable sea conditions, existing retrofit equipment performs poorly in autonomous navigation path tracking and full-process task management, failing to meet navigation stability and operational capabilities standards.

[0004] The continuous expansion of marine application scenarios has raised the bar for unmanned ship conversion technology. Various shortcomings of traditional conversion technologies severely hinder the large-scale implementation of unmanned ship technology and fail to adapt to the development trends of military applications and civilian shipping. Given this industry situation, the market urgently needs a highly modular and adaptable unmanned conversion solution for ship navigation control systems. Developing new conversion devices and control methods can effectively overcome existing technological bottlenecks, reduce the difficulty of unmanned upgrades for different ship types, and comprehensively improve ship handling precision and operational reliability. This has significant practical implications for supporting the development of military intelligence, empowering the transformation and upgrading of civilian shipping, and deeply exploring the potential for marine resource development. Summary of the Invention

[0005] This invention aims to address the problems of poor adaptability in the unmanned conversion of existing ships, difficulties in signal conversion, insufficient coordination capabilities among various systems, poor switching between multiple modes, and weak navigation performance in complex sea conditions. By building an integrated navigation control conversion device, it achieves standardized conversion of control signals for different ship types, strengthens the collaborative management and control capabilities of equipment, supports flexible switching between multiple driving modes, improves navigation control accuracy and operational reliability, and meets the actual needs of rapid unmanned upgrades for various types of ships.

[0006] In view of the above-mentioned defects or improvement needs of the existing technology, as a first aspect of the present invention, the present invention provides a rapid retrofit device for an unmanned vessel navigation control system, comprising: a main engine control module for converting the original control signals of the vessel into digital electronic control signals; a rudder control module for realizing rudder angle adjustment and heading stability control; a navigation control system module for adopting a shipborne-base station dual-layer architecture and supporting three working modes: manual driving, base station remote control, and autonomous navigation; an environmental perception system module for collecting information on the vessel's position, attitude, environment, and target; a communication module for realizing command interaction, status feedback, and data transmission between the base station and the shipborne terminal; and an energy management module and an installation module. During system initialization or mode switching, the navigation control system module outputs a step test signal to drive the modified actuators of the main control module and the servo control module, collects the actual response delay and steady-state error of the original ship's machinery, and constructs a nonlinear dead zone and hysteresis model of the original ship's machinery. In autonomous navigation or base station remote control mode, the navigation control system module will input the generated basic control commands into the hysteresis model for feedforward compensation, and generate corrective control commands that include dead zone crossing compensation and hysteresis advance to eliminate control hysteresis and oscillation caused by the mechanical coupling between the external modification mechanism and the original ship.

[0007] Furthermore, the main control module adopts a PLC and servo motor architecture to be compatible with the original control methods of pneumatic start, pneumatic reversing, pneumatic speed regulation, cable throttle, and hydraulic shifting, and to uniformly and standardize the physical control signals into digital electronic control signals; at the same time, the main control module is equipped with bypass priority control logic and an emergency stop mechanism, specifically including: An electromagnetic clutch and a torque sensor are connected in series between the power output end of the servo motor and the original throttle or shift lever. The machine-side priority control logic is as follows: The PLC collects the torque of the operating lever in real time through the torque sensor. When the reverse torque generated by manually operating the lever exceeds the set safety threshold, the PLC immediately outputs a disengagement command to de-energize and separate the electromagnetic clutch, and blocks the enable signal of the servo motor, thereby completing the physical decoupling between the modified mechanism and the original mechanical structure, and unconditionally transferring control to the machine-side manual operation. The emergency stop mechanism is as follows: a hardware safety relay emergency stop circuit is set up independently of the PLC's normal control program. When a serious system-level fault is triggered, the communication link times out, or an external emergency stop signal is received, the hardware safety relay circuit directly bypasses the PLC's control authority, forcibly drives the servo motor to pull the throttle lever to the zero position, and simultaneously cuts off the power supply to the original fuel solenoid valve, completing a dual physical shutdown of mechanical and electrical components.

[0008] Furthermore, the servo control module is a parallel-installed unmanned steering system that supports switching between four control modes: manual, servo, automatic heading, and automatic trajectory. It also features dual redundant control channels that operate independently and incorporates an adaptive control method; specifically including: A torque sensor and an electromagnetic clutch are installed at the connection between the unmanned servo actuator and the original ship's rudder. When switching to manual steering mode, the electromagnetic clutch is de-energized and disengaged, eliminating the mechanical back pressure of the external mechanism on the original ship's rudder. When in unmanned control mode and the original ship's hydraulic servo is working simultaneously, the output thrust of the unmanned servo is dynamically adjusted by collecting the real-time pressure of the original ship's hydraulic pipeline, completing the torque coordination of the two power sources and preventing mechanical interference and structural damage. The dual-redundant control channels include a main control channel and a hot standby channel, which cross-compare rudder angle commands and feedback status in real time. When a fault is detected in the main control channel or the control deviation exceeds the safety threshold, the hot standby channel takes over control at the moment of taking over control. It uses the actual physical rudder angle collected at the moment of takeover as the initial integral state of the PID control algorithm to achieve a disturbance-free switching of control output and avoid sudden changes in rudder angle caused by the moment of switching, which would lead to a sudden change in the ship's course. A rudder gain scheduling model is constructed based on the ship's real-time speed and draft. Under high-speed or light-load conditions, the control proportional gain is automatically reduced and a rudder angle rate limit is introduced to prevent the ship from rolling and overshooting oscillations. Under low-speed or heavy-load conditions, the control proportional gain is automatically increased and the integral limit is relaxed.

[0009] Furthermore, the rudder effect gain scheduling model constructed based on the ship's real-time speed and draft specifically includes: Based on the hydrodynamic derivatives of the original ship or the actual ship maneuverability test data, the ship turning index under different speed and draft combinations is extracted, and a two-dimensional benchmark gain lookup table with speed and draft as two independent variables is constructed. During navigation, the surface speed and draft are collected in real time. A reference proportional gain is matched in the two-dimensional reference gain lookup table using a bilinear interpolation algorithm. Considering the hydrodynamic characteristics that the rudder force is proportional to the square of the speed, a speed compensation coefficient is introduced to dynamically correct the reference proportional gain. The speed compensation coefficient is the ratio of the square of the design cruise speed to the square of the current real-time surface speed. At the same time, a low-speed stall threshold and a high-speed saturation threshold are set to limit the amplitude of the corrected proportional gain. The ship's turning inertia correction factor is calculated based on the current draft. When an increase in draft or a shallow water channel is detected that causes an increase in the ship's turning resistance, the integral limit threshold is increased to allow a larger steady-state rudder angle to be output. The corrected proportional gain and integral limiting threshold are input into the adaptive control method. When the external unmanned servo actuator is detected to have reached the physical limit rudder angle or the maximum push stick speed, the anti-integral saturation mechanism is triggered to freeze or reverse the decay of the integral term in the PID control algorithm, so as to prevent the heading overshoot caused by the continuous accumulation of control error when the system goes out of the limit state.

[0010] Furthermore, the underlying software of the navigation control system module adopts a C++ / QT multi-threaded modular architecture, and its functional modules include at least: an electronic chart module, a comprehensive situation module, a route management module, an autonomous navigation management module, and a base station remote control navigation module; the navigation control system module is used to complete the full lifecycle management of editing, storing, issuing, executing, pausing, and resetting control tasks.

[0011] Furthermore, the environmental perception system module includes a combined navigation device, a navigation radar, an ultrasonic weather station, and optoelectronic equipment; wherein, the combined navigation device is used to provide centimeter-level position, speed, and attitude information; the navigation radar is used for water surface target identification and tracking; the ultrasonic weather station is used to collect wind speed, wind direction, temperature, humidity, and air pressure data; and the optoelectronic equipment is used for day and night video observation.

[0012] Furthermore, the communication module adopts an RS485 bus link. When issuing control tasks, it uses structured data frames for transmission. The data frame includes a header structure, a tail structure, and a coordinate point structure containing longitude, latitude, and elevation information. Batch interaction of task data is achieved by traversing the waypoint list in the task object.

[0013] Furthermore, the construction process of the nonlinear dead zone and hysteresis model of the original ship machinery specifically includes: During system initialization or when the ship is safely moored, a low-frequency multi-step triangular wave test excitation signal covering the entire stroke is injected into the external actuator, and the command input sequence of the external actuator and the actual physical response sequence of the original ship machinery are acquired at high frequency simultaneously. The collected physical response sequence is filtered by a sliding window to eliminate hydrodynamic and mechanical vibration interference noise. The zero-crossing point and steady-state dwell point of the velocity during forward and reverse motion are extracted. The forward and reverse motion deviation curves of the original ship's machinery at different stroke positions are calculated, and the static dead zone boundary and dynamic hysteresis loop width are identified accordingly. A piecewise linear function is used to construct a dead zone compensation sub-model, and a discrete Play operator with memory effect is used to construct a hysteresis compensation sub-model. The static dead zone boundary and the dynamic hysteresis loop width are used as initial parameter inputs to combine and generate the nonlinear dead zone and hysteresis model to adapt to the limited computing power of the shipboard controller. During normal navigation of the ship, steady-state error data of daily control commands and actual responses are extracted. The parameters of the nonlinear dead zone and hysteresis model are dynamically corrected online using the recursive least squares method with forgetting factor to compensate for the hysteresis drift caused by long-term wear and tear of the ship's machinery and changes in hydraulic oil temperature.

[0014] Furthermore, in autonomous navigation or base station remote control mode, the flight control system module inputs the generated basic control commands into the hysteresis model for feedforward compensation, specifically including: Based on the nonlinear dead zone and hysteresis model, a series-parallel inverse compensator is constructed. The expected original ship mechanical physical response, i.e. the basic control command, is used as the input of the inverse compensator to calculate the theoretical driving command. When the change trajectory of the basic control command is detected to cross the static dead zone boundary, a step compensation amount proportional to the width of the original ship's mechanical dead zone is superimposed on the theoretical drive command to overcome the static friction and mechanical clearance of the original ship's machinery and eliminate the response dead zone. Calculate the first-order rate of change of the basic control command, and based on the width of the dynamic hysteresis loop, generate a hysteresis advance compensation amount that is proportional to the rate of change of the command and superimpose it onto the theoretical drive command to counteract the phase lag caused by mechanical transmission and hydraulic response. Apply speed change rate limiting and low-pass filtering to the corrected control command after superimposed compensation to suppress instantaneous destructive torque output by the external actuator caused by dead zone boundary jumps or high-frequency control commands, and prevent rigid impact between the modified mechanism and the original ship machinery. The smoothed and corrected control command is output to the external actuator, and the actual physical feedback of the original ship machinery is collected. The closed-loop residual between the actual physical feedback and the basic control command is calculated. The residual is then corrected a second time by a proportional-integral controller to generate the final control signal.

[0015] As a second aspect of the present invention, a control method for a rapid retrofit device for an unmanned vessel navigation control system is also provided, applicable to any one of the rapid retrofit devices for an unmanned vessel navigation control system, comprising the following steps: S1. Inject test excitation signals into the modified actuator under safe conditions, synchronously collect the command input and the actual physical response of the original ship machinery, extract motion deviations to identify the static dead zone boundary and dynamic hysteresis loop width, and construct the nonlinear dead zone and hysteresis model of the original ship machinery. S2. Switch between manual driving, base station remote control, and autonomous navigation modes according to mission requirements; in unmanned control mode, calculate and generate basic control commands for the original ship's main engine and rudder angle based on the ship's perception information. S3. Input the basic control command into the hysteresis model for feedforward compensation, superimpose the dead zone crossing compensation and hysteresis advance, and apply the speed change rate limit to generate correction commands for execution; during the execution process, collect the original ship hydraulic pipeline pressure to perform torque coordination of the dual power sources, and calculate the closed-loop residual based on the actual physical feedback for secondary correction and online model update; S4. Real-time monitoring of the control channel, communication link, and operating lever stress status; dual-redundancy seamless switching when a channel fails; when manual intervention torque exceeds the limit is detected, the electromagnetic clutch is disconnected to transfer control to the operator on-site; when a serious fault is triggered, the original machine is forced to stop through an independent hardware circuit.

[0016] In summary, compared with the prior art, the above-described technical solutions conceived by this invention can achieve the following beneficial effects: 1. The rapid retrofit device for the unmanned vessel navigation control system of this invention integrates a main engine control module, a servo control module, a navigation control system module, an environmental perception system module, and a communication module to form an overall working system. Utilizing the PLC and servo motor architecture of the main engine control module, it unifies and converts various existing physical control signals of the vessel into standard digital electronic control signals. This module is compatible with various traditional control methods such as pneumatic start, pneumatic reversing, cable throttle, and hydraulic shifting. Simultaneously, it incorporates an electromagnetic clutch and torque sensor in the transmission structure, along with engine-side priority control logic. When the reverse torque generated by manually operating the control lever exceeds a set threshold, the device can disconnect the physical connection between the retrofit mechanism and the original engine, transferring control to on-site personnel. It is also equipped with an independent hardware safety relay emergency stop circuit, which forces the throttle to zero and fuel cut-off operations in the event of system failure or communication link timeout, achieving a dual physical shutdown. This design not only completes the digital transformation of control signals for different vessel types but also establishes a safety protection mechanism at the hardware level, ensuring basic safety during vessel operation.

[0017] 2. The rapid retrofit device for the unmanned vessel navigation control system of the present invention, during system initialization or working mode switching, drives the retrofitted actuator by outputting a step test signal from the navigation control system module, simultaneously collecting the actual response delay and steady-state error of the original ship's machinery, and thereby constructing a nonlinear dead zone and hysteresis model of the original ship's machinery. When the device operates in autonomous navigation or base station remote control mode, the navigation control system inputs the generated basic control commands into the model for feedforward compensation, and generates correction commands including dead zone crossing compensation and hysteresis lead based on model calculations. This processing method can effectively eliminate the control hysteresis caused by the mutual coupling between the retrofitted mechanism and the original ship's machinery, reducing the control oscillations during ship navigation. The device also continuously collects operating data during ship navigation, performs online correction of the parameters of the nonlinear model, compensates for the wear caused by long-term use of the original ship's machinery and the characteristic deviation caused by changes in hydraulic oil temperature, ensuring that the model always matches the actual operating conditions of the ship and stably maintains the execution effect of control commands.

[0018] 3. The rapid retrofit device for the unmanned vessel navigation control system of this invention optimizes the ship's rudder angle adjustment and heading control effects by configuring dual redundant control channels, a torque coordination structure, and a rudder effect gain scheduling model for the rudder control module. The independent main control channel and hot standby channel can compare operating data in real time. When the main control channel malfunctions, the hot standby channel can seamlessly switch based on the current actual rudder angle, avoiding sudden changes in heading. The sensors and electromagnetic clutch at the rudder connection point enable smooth switching between manual and unmanned steering, and dynamically adjust the output thrust in conjunction with hydraulic pipeline pressure to prevent mechanical interference between different power sources. The device constructs a gain scheduling model based on the ship's speed and draft, adaptively adjusting control parameters according to navigation conditions to meet the control requirements of different navigation states. The entire device adopts a shipborne-base station dual-layer architecture, relying on a sensing system to collect environmental and ship status information, and using a communication module to complete command and data interaction, smoothly achieving switching between three working modes: manual driving, base station remote control, and autonomous navigation. Attached Figure Description

[0019] Figure 1 This is a schematic diagram of a rapid retrofit device for an unmanned vessel navigation control system according to an embodiment of the present invention; Figure 2 This is a schematic diagram of the overall modification scheme of an embodiment of the present invention; Figure 3 This is a schematic diagram of the method flow according to an embodiment of the present invention. Detailed Implementation

[0020] 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. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention. Furthermore, the technical features involved in the various embodiments of this invention described below can be combined with each other as long as they do not conflict with each other.

[0021] Example 1 Please refer to Figure 1 This embodiment 1 provides a rapid retrofit device for an unmanned vessel navigation control system, including: a main engine control module for converting the original ship control signals into digital electronic control signals; a rudder control module for realizing rudder angle adjustment and heading stability control; a navigation control system module for adopting a ship-base station dual-layer architecture and supporting three working modes: manual driving, base station remote control, and autonomous navigation; an environmental perception system module for collecting ship position, attitude, environment, and target information; a communication module for realizing command interaction, status feedback, and data transmission between the base station and the ship-mounted terminal; and an energy management module and an installation module. During system initialization or mode switching, the navigation control system module outputs a step test signal to drive the modified actuators of the main control module and the servo control module, collects the actual response delay and steady-state error of the original ship's machinery, and constructs a nonlinear dead zone and hysteresis model of the original ship's machinery. In autonomous navigation or base station remote control mode, the navigation control system module will input the generated basic control commands into the hysteresis model for feedforward compensation, and generate corrective control commands that include dead zone crossing compensation and hysteresis advance to eliminate control hysteresis and oscillation caused by the mechanical coupling between the external modification mechanism and the original ship.

[0022] Please refer to Figure 2 This embodiment 1 will further elaborate on the above system modules.

[0023] (1) Host control module The core architecture of the main control module adopts a combination of PLC and servo motor, designed to be compatible with various original ship mechanical control methods such as pneumatic start, pneumatic reversing, pneumatic speed regulation, cable-operated throttle, and hydraulic shifting. In terms of physical connection, an electromagnetic clutch and torque sensor are connected in series between the power output end of the servo motor and the original ship's throttle or shift lever. The PLC, as the control center, is responsible for collecting the original machine's status signals and driving the servo motor, thereby standardizing the physical control signals into digital electronic control signals, realizing the drive and feedback of the original ship's power system.

[0024] During operation, the host control module is configured with bypass priority control logic. The PLC collects the torque applied to the operating lever in real time via a torque sensor. and with the preset safe torque threshold A comparison is performed. When a reverse torque generated by manually operating the lever is detected and the following conditions are met... At that moment, the PLC immediately outputs a disengagement command, de-energizing and disengaging the electromagnetic clutch, and blocking the enable signal of the servo motor. This mechanism forcibly decouples the modified mechanism from the original mechanical structure, ensuring that control is unconditionally and without delay transferred to manual operation at the machine site, thus guaranteeing operational safety in emergency situations.

[0025] Meanwhile, the main control module is also equipped with an emergency stop mechanism independent of the PLC's regular control program to deal with extreme situations such as severe system-level failures or communication link timeouts. This mechanism uses hardware safety relays to construct an emergency stop circuit. When an external emergency stop signal is received or a dangerous condition is detected, the hardware safety relays directly bypass the PLC's control authority, forcibly drive the servo motor to pull the throttle lever to the zero position, and simultaneously cut off the power supply to the original engine's fuel solenoid valve. This dual physical shutdown method, combining mechanical and electrical components, constitutes the last line of defense for system safety, ensuring that the ship can quickly stop power output under any abnormal conditions.

[0026] Furthermore, this module supports smooth switching between three modes: manual control, base station remote control, and autonomous navigation. In unmanned mode, the servo motor adjusts the throttle opening and gear according to digital commands issued by the navigation control system; while in manual mode, the disengagement of the electromagnetic clutch restores the original ship's control levers to their freedom of movement, unaffected by the back pressure of the retrofitted mechanism. The entire control process communicates with the host computer via an RS485 bus, enabling real-time command interaction and dynamic status feedback.

[0027] (2) Servo control module The servo control module adopts a parallel-mounted architecture, supporting switching between four operation modes: manual, follow-up, automatic heading, and automatic trajectory. Through built-in adaptive control methods and a dual redundancy mechanism, it achieves rudder angle adjustment and heading stability control. At the hardware connection level, a torque sensor and an electromagnetic clutch are connected in series at the connection point between the unmanned servo actuator and the original ship's rudder. When the system is in manual steering mode, the electromagnetic clutch is de-energized and disengaged, eliminating the mechanical back pressure exerted by the external mechanism on the original ship's rudder, ensuring ease of manual operation. When in unmanned control mode and the original ship's hydraulic servo is operating simultaneously, the output thrust of the unmanned servo is dynamically adjusted by collecting real-time pressure data from the original ship's hydraulic lines, achieving torque coordination between the two power sources and preventing mechanical interference and structural damage.

[0028] In terms of control logic safety, a dual-redundant control architecture consisting of a main control channel and a hot standby channel that operate independently is configured. The two channels cross-compare rudder angle commands and feedback status in real time. Once a fault is detected in the main control channel or the control deviation exceeds the safety threshold, the hot standby channel will immediately take over control. To ensure the smoothness of the switching process, the hot standby channel uses the actual physical rudder angle collected at the moment of takeover as the initial integral state of the PID control algorithm, thereby achieving a disturbance-free switching of control output and avoiding sudden changes in the ship's course caused by a step change in rudder angle during the switching instant.

[0029] In addition, a rudder gain scheduling model is constructed based on the ship's real-time speed and draft. Under high-speed or light-load conditions, the control proportional gain is automatically reduced and a rudder angle rate limit is introduced to prevent increased rolling. Under low-speed or heavy-load conditions, the control proportional gain is automatically increased and the integral limit is relaxed to provide sufficient turning torque to overcome water flow interference and ensure heading response speed.

[0030] The process of constructing a rudder effect gain scheduling model based on the ship's real-time speed and draft begins with the offline data modeling stage. First, relying on the original ship's hydrodynamic derivatives or actual ship maneuverability test data, the ship's turning index under different operating conditions is extracted. With speed as the independent variable and draft independent variable Construct a two-dimensional reference gain lookup table This two-dimensional reference gain lookup table Internally, discrete storage contains the reference gain data required for the ship to maintain course stability under various steady-state conditions.

[0031] Once the ship enters actual navigation mode, the sensor network continuously collects the current real-time water speed. With current draft The current real-time speed of the waterway. With current draft As input coordinates, in a two-dimensional reference gain lookup table The four adjacent discrete nodes surrounding the current operating point are located, and the speed boundaries corresponding to these four nodes are respectively... and The draft boundaries are respectively and Using the gain values ​​stored in these four nodes, the baseline proportional gain under the current operating condition is calculated through a bilinear interpolation algorithm. The calculation process is expressed as follows: In the formula , , and These are the gain values ​​stored at four adjacent discrete nodes.

[0032] To address the hydrodynamic physical characteristic that rudder force is proportional to the square of speed, a speed compensation coefficient is introduced. For reference proportional gain Dynamic corrections are performed. Speed ​​compensation coefficient. The calculation is expressed as ,in For the design cruising speed of the ship, The effective speed used in the calculation. To prevent the compensation coefficient from approaching infinity at extremely low speeds, leading to system loss of control, and to prevent excessively rapid gain decay at high speeds, resulting in sluggish response, a low-speed stall threshold is set. With high speed saturation threshold For effective speed Implement amplitude limiting protection. Effective speed The value selection logic is expressed as a piecewise function: when hour, ;when hour, ;when hour, The speed compensation coefficient is calculated using this limiting logic. Then, compare it with the reference scaling gain. Multiplying these yields the final dynamic correction proportional gain input to the adaptive control method. ,Right now .

[0033] During the closed-loop control execution phase, the current draft depth will be... With the preset shallow water draft threshold Compare the current draft. Greater than or equal to the shallow water draft threshold The system determines that the vessel is in a shallow waterway or under heavy load, at which point the vessel's turning resistance increases significantly, and the system accordingly increases the integral limit threshold in the PID control algorithm. This allows the controller to output a larger steady-state rudder angle to overcome the sluggish heading response caused by large inertia and shallow water effects. The physical operating status of the external unmanned servo actuator is continuously monitored; when the actual output rudder angle is detected to reach the physical limit rudder angle... Or the push rod speed reaches the maximum push rod speed. When this occurs, it indicates that the actuator has entered a physical saturation state. At this point, the anti-integral saturation mechanism is immediately triggered, forcibly freezing or reverse-decreasing the integral term in the PID control algorithm. Integral term The update logic is expressed as ,in The heading control error at the current moment. This is the integral decay factor. When the actuator is in a state of physical saturation, the integral decay factor... Forced to zero or set to a negative value, by blocking the integral term. The continuous accumulation of this effectively prevents the excessive accumulation of control errors during the saturation period, and avoids severe heading overshoot and reverse oscillation of the ship due to residual integral action when the external unmanned servo actuator goes out of the physical limit state.

[0034] (3) Flight control system module The navigation control system module adopts a C++ / QT multi-threaded modular architecture at its software level, integrating functional modules such as electronic charts, comprehensive situational awareness, route management, autonomous navigation management, and base station remote-controlled navigation. It supports full lifecycle management of three working modes: manual control, base station remote control, and autonomous navigation. At critical nodes such as system initialization or mode switching, the navigation control system module actively outputs step test signals to drive the modified actuators of the main engine and servo control modules. By collecting the actual response delay and steady-state error of the original ship's machinery, it constructs and updates the nonlinear dead zone and hysteresis model of the original ship's machinery in real time. In subsequent autonomous navigation or base station remote control modes, the generated basic control commands are input into this model for feedforward compensation, generating corrective control commands that include dead zone crossing compensation and hysteresis lead, thereby effectively eliminating control hysteresis and oscillations caused by the coupling between the modified mechanism and the original ship's machinery, ensuring high-precision control.

[0035] In this embodiment, the construction process of the nonlinear dead zone and hysteresis model of the original ship's machinery begins with the offline testing phase during system initialization or when the ship is safely moored. The control unit injects a low-frequency multi-step triangular wave test excitation signal covering the entire stroke into the external actuator. ,in For discrete-time indexing. During the injection of the test excitation signal. At the same time, with a fixed sampling period Synchronous high-frequency acquisition of the instruction input sequence of the external actuator The actual physical response sequence of the original ship's machinery Instruction input sequence The theoretical control inputs and actual physical response sequences sent from the controller to the actuators were recorded. The actual physical displacement or angle of the original ship's mechanical rudder handle or push rod was recorded, and the two together constituted the original dataset for subsequent model identification.

[0036] After obtaining the raw dataset, the actual physical response sequence A sliding window filter is used to eliminate hydrodynamic and mechanical vibration noise. The sliding window length is set. The filtered physical response sequence is output by calculating the arithmetic mean of the data within the window. Based on the filtered physical response sequence With sampling period Extracting the physical response velocity sequence through differential operations By traversing the physical response velocity sequence Extracting the set of points where the velocity changes from positive to negative or from negative to positive. And extract the physical response speed sequence. The set of steady-state dwell points whose absolute value is less than a preset static threshold Using the set of velocity zero-crossing points With steady-state dwell point set Input the instruction sequence With the filtered physical response sequence By performing alignment and comparison, the positive motion deviation curves of the original ship's machinery at different stroke positions were calculated. Deviation curve with opposite motion Based on the positive motion deviation curve Deviation curve with opposite motion Intersection features near zero input identify the lower boundary of the static dead zone. Upper boundary of static dead zone And based on the positive motion deviation curve Deviation curve with opposite motion The dynamic hysteresis loop width is calculated by taking the difference in command input at the same physical response location. .

[0037] After feature parameter identification, the system constructs a dead-zone compensation sub-model using piecewise linear functions and a hysteresis compensation sub-model using a discrete Play operator with memory effect. The output of the dead-zone compensation sub-model is shown below. Expressed as a piecewise function: when the instruction is input Less than the lower boundary of the static dead zone hour, When the command is input Between the lower boundary of the static dead zone Upper boundary of static dead zone In between, When the command is input Greater than the upper boundary of the static dead zone hour, ,in Input the original command for the current moment. This represents the compensation slope for the linear segment outside the dead zone. The output of the hysteresis compensation sub-model. It is calculated by the discrete Play operator and is expressed as follows: ,in For the input command at the current moment, The hysteresis threshold of the Play operator is equal to the width of the dynamic hysteresis loop. Half of This outputs the operator state from the previous time step. The system will output the dead-zone compensation sub-model. Output of the hysteresis compensation sub-model The values ​​are superimposed to generate the total compensation control quantity. Its calculation is expressed as The total compensation control amount The output of the final generated nonlinear dead zone and hysteresis model is superimposed on the main control loop through feedforward compensation to adapt to the limited computing power of the shipboard controller and eliminate the effects of mechanical nonlinearity.

[0038] After the ship enters normal navigation, steady-state error data of daily control commands and actual responses are continuously extracted. A recursive least squares method with a forgetting factor is used to dynamically correct the parameters of the nonlinear dead zone and hysteresis model online. Subsequently, a vector of parameters to be identified is constructed. This vector contains the slope of the dead zone out-of-zone linear segment compensation that needs to be corrected online. Static dead zone lower boundary upper boundary of static dead zone and hysteresis threshold Within each control cycle, calculate the steady-state error at the current moment. And construct a parameter vector to be identified. Corresponding observation regression vector The parameter vector update process is expressed as follows: ,in This is the parameter vector from the previous time step. This is the gain vector at the current moment. (Gain vector) The calculation is expressed as ,in Let be the covariance matrix of the previous time step. This is a forgetting factor, ranging from zero to one, used to reduce the weight of historical data on the current parameter estimate. The updated covariance matrix is ​​expressed as follows: ,in The matrix is ​​the identity matrix. Through the above recursive iterative calculation, the system updates the vector of parameters to be identified in real time. The various elements in the model are used to dynamically correct the parameters of the nonlinear dead zone and hysteresis model online, effectively compensating for the drift of hysteresis characteristics caused by long-term wear and tear of the original ship's machinery and changes in hydraulic oil temperature, and ensuring the compensation accuracy of the model throughout its entire life cycle.

[0039] In autonomous navigation or base station remote control mode, the process of the navigation control system module inputting the generated basic control commands into the hysteresis model for feedforward compensation begins with the construction of a series-parallel inverse compensator. First, the desired original ship mechanical-physical response, i.e., the basic control commands, is input... As the input to the inverse compensator, This is the index for the current discrete control cycle. The inverse compensator performs inverse mapping calculations based on a nonlinear dead-zone and hysteresis model, outputting theoretical drive commands. This theory drives the instructions. This represents driving the ship's machinery to achieve basic control commands under ideal, undisturbed conditions. The theoretical control input required for the corresponding physical displacement or angle.

[0040] In generating theoretical driving instructions Based on this, monitor basic control commands in real time. The trajectory of change. When a basic control command is detected. The trajectory of change crosses the lower boundary of the static dead zone. Or the upper boundary of the static dead zone At that time, in the theoretical driving instruction Superimposed step compensation amount Step compensation amount The calculation is expressed as ,in This is the step compensation proportional coefficient. The original ship's mechanical dead zone width and meets the requirements , The change in the basic control command in the current cycle is , This is the sign function. The step compensation amount is then superimposed. It can instantly provide sufficient driving force to overcome the static friction and mechanical clearance of the original ship's machinery, thereby eliminating the response dead zone.

[0041] To address the phase lag issue caused by mechanical transmission and hydraulic response, the system calculates the basic control commands. First rate of change Its discrete computation expression is as follows ,in The sampling period of the control system. The width of the dynamic hysteresis loop identified in the preceding steps. Generate hysteresis-leadership compensation amount Its calculation is expressed as ,in This is the hysteresis-lead compensation ratio coefficient. The step compensation amount... With delay and advance compensation Together superimposed on the theoretical driving command The corrected control command after superimposed compensation is obtained. Its calculation is expressed as This corrected control command By introducing a leading term that is proportional to the rate of change of the command, the phase lag caused by mechanical transmission and hydraulic response is effectively offset.

[0042] To prevent sudden changes in dead zone boundaries or high-frequency control commands from causing instantaneous destructive torque in the external actuator output, a corrective control command with superimposed compensation is applied. Apply speed change rate limiting and low-pass filtering. In the speed change rate limiting stage, the system calculates the desired change in the corrected control command. ,in This is the smoothing control command for the final output of the previous cycle. It sets the maximum permissible rate of change. The expected change is limited to obtain the limited change. Its calculation is expressed as This leads to the command after limiting the rate of change of speed. In the low-pass filtering stage, the system uses a first-order discrete low-pass filter. Perform smoothing processing and output the smoothed correction control command. Its calculation is expressed as ,in It is a low-pass filter coefficient with a value between zero and one.

[0043] Next, the smoothed correction control command will be... The output is sent to an external actuator to drive the original ship's machinery to produce actual physical motion. Simultaneously, a sensor network collects real-time physical feedback from the original ship's machinery. Calculate the actual physical feedback. With basic control commands Closed-loop residuals between Its calculation is expressed as To eliminate the residual nonlinear error and external hydrodynamic disturbances that feedforward compensation could not completely overcome, a proportional-integral controller was used to control the closed-loop residual. Perform a second correction to generate a feedback correction amount. Its calculation is expressed as ,in For proportional control gain, For integral control gain, Add an index to the integral. Then, add the feedback correction amount. With the smoothed correction control command The signals are superimposed to generate the final control signal. Its calculation is expressed as The final control signal As the final drive input for the external actuators, it ensures that the original ship's machinery can still track the basic control commands under complex nonlinear characteristics and external disturbances. .

[0044] (4) Environmental perception system module The environmental perception system module integrates a combined navigation device, navigation radar, ultrasonic weather station, and optoelectronic equipment. It connects to the flight control unit via a standard interface and outputs fused data after spatiotemporal synchronization. The combined navigation device employs a tightly coupled GNSS and INS architecture, providing real-time centimeter-level position information, velocity vectors, and three-dimensional attitude angles, providing a benchmark for path planning and attitude control. The navigation radar acquires target distance, azimuth, and radial velocity through electromagnetic wave scanning, and calculates collision risk parameters using a multi-frame tracking algorithm to achieve dynamic obstacle avoidance.

[0045] Ultrasonic weather stations measure wind speed, wind direction, temperature, humidity, and air pressure non-contactly. The data, after motion compensation, is used for wind disturbance correction and environmental adaptability adjustment. All meteorological data is uploaded to the flight control host at a fixed sampling rate and spatiotemporally aligned with navigation data to construct a complete environmental perception map.

[0046] In visual perception and day / night monitoring, optoelectronic equipment serves as an effective supplement to radar perception, providing intuitive video observation capabilities. This equipment typically integrates a visible light camera and an infrared thermal imager, supporting all-weather, all-time target identification and situational awareness. Real-time analysis of the video stream is performed using image processing algorithms to extract target contour features, movement direction, and category attributes, and this data is fused with radar tracking results to improve the confidence level of target identification. At night or in low-visibility conditions, the infrared thermal imager can effectively detect heated targets on the water surface, compensating for the limitations of visible light cameras. The gimbal control system of the optoelectronic equipment can automatically adjust the pitch and azimuth angles according to flight control commands, enabling continuous tracking and observation of specific areas or targets, providing reliable visual basis for remote control and autonomous decision-making. Data from all sensors is uniformly registered by the flight control host to form a complete situational awareness map, supporting autonomous navigation and remote monitoring.

[0047] (5) Communication module The communication module establishes a bidirectional data channel between the base station and the shipborne terminal based on an RS485 bus link. The physical layer employs a differential signal transmission mechanism to enhance anti-interference capabilities and ensure the reliability of command interaction, status feedback, and data transmission in complex electromagnetic environments. When issuing control tasks, the data is encapsulated and parsed according to a structured data frame protocol. The data frame consists of a header structure, a tail structure, and a coordinate point structure containing longitude, latitude, and elevation information. The header structure includes a frame synchronization code and a function identifier, used by the receiver to identify the data start position and service type. The tail structure includes a checksum field to verify the integrity and accuracy of data transmission, preventing erroneous command execution due to noise interference.

[0048] To enable batch interaction of task data, the sending program dynamically generates a continuous sequence of data frames by traversing the waypoint list within the task object. The waypoint list is an ordered set that stores all critical path points of a preset route. During transmission, elements in the waypoint list are read sequentially according to their indexes, and the longitude, latitude, and elevation information of each element are filled into the coordinate point structure and encapsulated into a frame, until the list traversal is complete. The receiving end then reassembles the data based on the sequence number in the header structure to reconstruct the complete route data. Simultaneously, this link supports reverse status feedback; the shipborne terminal encapsulates its own operating parameters into a response frame with a similar structure and transmits it back to the base station, thus forming a closed-loop monitoring and control system.

[0049] (6) Energy management module and additional modules The energy management module provides a stable and uninterrupted power supply to the entire unmanned vessel navigation control system, and features energy distribution, status monitoring, and emergency shutdown functions. Considering the continuity and safety of unmanned navigation operations, this embodiment employs a dual-power redundant power supply architecture, including a main battery pack and a backup battery pack. The core control unit of the energy management module communicates in real-time with the navigation control system module via an RS485 bus, establishing a dynamic monitoring model that includes voltage, current, remaining power, and charging / discharging status.

[0050] In terms of specific control logic, the energy management module is deeply coupled with the aforementioned main engine control module and servo control module. When the main power supply voltage drops to a set threshold or a communication link malfunctions, the energy management module automatically triggers a seamless power switching logic, smoothly switching the load power supply from the main battery pack to the backup battery pack, ensuring the continuous operation of the servo motors, PLCs, and sensor networks. Simultaneously, this module is equipped with an independent hardware safety relay control loop, which shares safety logic with the emergency stop mechanism in the main engine control module. When a system-level serious fault is triggered, the energy management module will prioritize cutting off power to non-core loads (such as optoelectronic equipment and unnecessary lighting), prioritizing the power supply to the servo control module and communication module until the ship is safely moored.

[0051] The add-on module enables rapid physical integration and mechanical connection of the aforementioned functional modules onto the original ship structure. To address the differences in deck layout and mechanical structure among different types of vessels, the add-on module employs standardized flange interfaces and a modular assembly structure. Specifically, between the servo motor of the main engine control module and the original ship's throttle / gear control lever, the add-on module includes a coupling and a quick-release robotic arm to ensure rigidity and responsiveness in power transmission; at the connection between the steering gear control module and the original ship's rudder handle, the add-on module is designed with an adaptive centering mechanism and a torque sensor mounting bracket to eliminate additional frictional torque caused by installation deviations.

[0052] Furthermore, the added module integrates an electromagnetic compatibility (EMC) protection structure, including shielded cable trays and grounding copper busbars, to isolate high-frequency harmonics generated by the servo motor driver from interfering with sensing system modules (such as integrated navigation and radar), thereby ensuring the accuracy of the sensing data. The entire installation process requires no welding or large-scale modification of the original ship hull; it can be completed through bolt tightening and mechanical snap-fitting, enabling rapid deployment and plug-and-play functionality of the unmanned navigation control system, perfectly aligning with the invention's technical objective of "rapid retrofitting."

[0053] Example 2 Please refer to Figure 3 This embodiment 2 provides a control method for a rapid retrofit device for an unmanned vessel navigation control system, applicable to any of the rapid retrofit devices for an unmanned vessel navigation control system, including the following steps: S1. Inject test excitation signals into the modified actuator under safe conditions, synchronously collect the command input and the actual physical response of the original ship machinery, extract motion deviations to identify the static dead zone boundary and dynamic hysteresis loop width, and construct the nonlinear dead zone and hysteresis model of the original ship machinery. S2. Switch between manual driving, base station remote control, and autonomous navigation modes according to mission requirements; in unmanned control mode, calculate and generate basic control commands for the original ship's main engine and rudder angle based on the ship's perception information. S3. Input the basic control command into the hysteresis model for feedforward compensation, superimpose the dead zone crossing compensation and hysteresis advance, and apply the speed change rate limit to generate correction commands for execution; during the execution process, collect the original ship hydraulic pipeline pressure to perform torque coordination of the dual power sources, and calculate the closed-loop residual based on the actual physical feedback for secondary correction and online model update; S4. Real-time monitoring of the control channel, communication link, and operating lever stress status; dual-redundancy seamless switching when a channel fails; when manual intervention torque exceeds the limit is detected, the electromagnetic clutch is disconnected to transfer control to the operator on-site; when a serious fault is triggered, the original machine is forced to stop through an independent hardware circuit.

[0054] Those skilled in the art will readily understand that the above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A rapid retrofit device for an unmanned vessel navigation control system, characterized in that, include: The system includes a main engine control module for converting the ship's original engine control signals into digital electronic control signals; a rudder control module for adjusting rudder angle and stabilizing course; a navigation control system module that adopts a shipborne-base station dual-layer architecture and supports three working modes: manual driving, base station remote control, and autonomous navigation; an environmental perception system module for collecting information on the ship's position, attitude, environment, and targets; a communication module for enabling command interaction, status feedback, and data transmission between the base station and the shipborne terminal; and an energy management module and an add-on module. During system initialization or mode switching, the navigation control system module outputs a step test signal to drive the modified actuators of the main control module and the servo control module, collects the actual response delay and steady-state error of the original ship's machinery, and constructs a nonlinear dead zone and hysteresis model of the original ship's machinery. In autonomous navigation or base station remote control mode, the navigation control system module will input the generated basic control commands into the hysteresis model for feedforward compensation, and generate corrective control commands that include dead zone crossing compensation and hysteresis advance to eliminate control hysteresis and oscillation caused by the mechanical coupling between the external modification mechanism and the original ship.

2. The rapid retrofit device for an unmanned vessel navigation control system according to claim 1, characterized in that, The main control module adopts a PLC and servo motor architecture to be compatible with the original control methods of air start, air reversing, air speed regulation, cable throttle, and hydraulic shifting, and to uniformly and standardize the physical control signals into digital electronic control signals. Simultaneously, the main control module is equipped with bypass priority control logic and an emergency stop mechanism, specifically including: An electromagnetic clutch and a torque sensor are connected in series between the power output end of the servo motor and the original throttle or shift lever. The machine-side priority control logic is as follows: The PLC collects the torque of the operating lever in real time through the torque sensor. When the reverse torque generated by manually operating the lever exceeds the set safety threshold, the PLC immediately outputs a disengagement command to de-energize and separate the electromagnetic clutch, and blocks the enable signal of the servo motor, thereby completing the physical decoupling between the modified mechanism and the original mechanical structure, and unconditionally transferring control to the machine-side manual operation. The emergency stop mechanism is as follows: a hardware safety relay emergency stop circuit is set up independently of the PLC's normal control program. When a serious system-level fault is triggered, the communication link times out, or an external emergency stop signal is received, the hardware safety relay circuit directly bypasses the PLC's control authority, forcibly drives the servo motor to pull the throttle lever to the zero position, and simultaneously cuts off the power supply to the original fuel solenoid valve, completing a dual physical shutdown of mechanical and electrical components.

3. The rapid retrofit device for an unmanned vessel navigation control system according to claim 1, characterized in that, The servo control module is a parallel-mounted unmanned steering system that supports switching between four control modes: manual, servo, automatic heading, and automatic trajectory. It also features dual redundant control channels that operate independently and incorporates an adaptive control method. Specifically, it includes: A torque sensor and an electromagnetic clutch are installed at the connection between the unmanned servo actuator and the original ship's rudder. When switching to manual steering mode, the electromagnetic clutch is de-energized and disengaged, eliminating the mechanical back pressure of the external mechanism on the original ship's rudder. When in unmanned control mode and the original ship's hydraulic servo is working simultaneously, the output thrust of the unmanned servo is dynamically adjusted by collecting the real-time pressure of the original ship's hydraulic pipeline, completing the torque coordination of the two power sources and preventing mechanical interference and structural damage. The dual-redundant control channels include a main control channel and a hot standby channel, which cross-compare rudder angle commands and feedback status in real time. When a fault is detected in the main control channel or the control deviation exceeds the safety threshold, the hot standby channel takes over control at the moment of taking over control. It uses the actual physical rudder angle collected at the moment of takeover as the initial integral state of the PID control algorithm to achieve a disturbance-free switching of control output and avoid sudden changes in rudder angle caused by the moment of switching, which would lead to a sudden change in the ship's course. A rudder gain scheduling model is constructed based on the ship's real-time speed and draft. Under high-speed or light-load conditions, the control proportional gain is automatically reduced and a rudder angle rate limit is introduced to prevent the ship from rolling and overshooting oscillations. Under low-speed or heavy-load conditions, the control proportional gain is automatically increased and the integral limit is relaxed.

4. The rapid retrofit device for an unmanned vessel navigation control system according to claim 3, characterized in that, The aforementioned rudder effect gain scheduling model, based on the ship's real-time speed and draft, specifically includes: Based on the hydrodynamic derivatives of the original ship or the actual ship maneuverability test data, the ship turning index under different speed and draft combinations is extracted, and a two-dimensional benchmark gain lookup table with speed and draft as two independent variables is constructed. During navigation, the surface speed and draft are collected in real time. A reference proportional gain is matched in the two-dimensional reference gain lookup table using a bilinear interpolation algorithm. Considering the hydrodynamic characteristics that the rudder force is proportional to the square of the speed, a speed compensation coefficient is introduced to dynamically correct the reference proportional gain. The speed compensation coefficient is the ratio of the square of the design cruise speed to the square of the current real-time surface speed. At the same time, a low-speed stall threshold and a high-speed saturation threshold are set to limit the amplitude of the corrected proportional gain. The ship's turning inertia correction factor is calculated based on the current draft. When an increase in draft or a shallow water channel is detected that causes an increase in the ship's turning resistance, the integral limit threshold is increased to allow a larger steady-state rudder angle to be output. The corrected proportional gain and integral limiting threshold are input into the adaptive control method. When the external unmanned servo actuator is detected to have reached the physical limit rudder angle or the maximum push stick speed, the anti-integral saturation mechanism is triggered to freeze or reverse the decay of the integral term in the PID control algorithm, so as to prevent the heading overshoot caused by the continuous accumulation of control error when the system goes out of the limit state.

5. The rapid retrofit device for an unmanned vessel navigation control system according to claim 1, characterized in that, The underlying software of the navigation control system module adopts a C++ / QT multi-threaded modular architecture, and its functional modules include at least: electronic chart module, integrated situation module, route management module, autonomous navigation management module and base station remote control navigation module; the navigation control system module is used to complete the full life cycle management of control tasks, including editing, storing, issuing, executing, pausing and resetting.

6. The rapid retrofit device for an unmanned vessel navigation control system according to claim 1, characterized in that, The environmental perception system module includes a combined navigation device, a navigation radar, an ultrasonic weather station, and optoelectronic equipment; wherein, the combined navigation device is used to provide centimeter-level position, velocity, and attitude information; the navigation radar is used for water surface target identification and tracking; the ultrasonic weather station is used to collect wind speed, wind direction, temperature, humidity, and air pressure data; and the optoelectronic equipment is used for day and night video observation.

7. The rapid retrofit device for an unmanned vessel navigation control system according to claim 1, characterized in that, The communication module uses an RS485 bus link. When issuing control tasks, it uses structured data frames for transmission. The data frame includes a header structure, a tail structure, and a coordinate point structure containing longitude, latitude, and elevation information. It also achieves batch interaction of task data by traversing the waypoint list in the task object.

8. The rapid retrofit device for an unmanned vessel navigation control system according to claim 1, characterized in that, The construction process of the nonlinear dead zone and hysteresis model of the original ship machinery specifically includes: During system initialization or when the ship is safely moored, a low-frequency multi-step triangular wave test excitation signal covering the entire stroke is injected into the external actuator, and the command input sequence of the external actuator and the actual physical response sequence of the original ship machinery are acquired at high frequency simultaneously. The collected physical response sequence is filtered by a sliding window to eliminate hydrodynamic and mechanical vibration interference noise. The zero-crossing point and steady-state dwell point of the velocity during forward and reverse motion are extracted. The forward and reverse motion deviation curves of the original ship's machinery at different stroke positions are calculated, and the static dead zone boundary and dynamic hysteresis loop width are identified accordingly. A piecewise linear function is used to construct a dead zone compensation sub-model, and a discrete Play operator with memory effect is used to construct a hysteresis compensation sub-model. The static dead zone boundary and the dynamic hysteresis loop width are used as initial parameter inputs to combine and generate the nonlinear dead zone and hysteresis model to adapt to the limited computing power of the shipboard controller. During normal navigation of the ship, steady-state error data of daily control commands and actual responses are extracted. The parameters of the nonlinear dead zone and hysteresis model are dynamically corrected online using the recursive least squares method with forgetting factor to compensate for the hysteresis drift caused by long-term wear and tear of the ship's machinery and changes in hydraulic oil temperature.

9. The rapid retrofit device for an unmanned vessel navigation control system according to claim 1, characterized in that, In autonomous navigation or base station remote control mode, the flight control system module inputs the generated basic control commands into the hysteresis model for feedforward compensation, specifically including: Based on the nonlinear dead zone and hysteresis model, a series-parallel inverse compensator is constructed. The expected original ship mechanical physical response, i.e. the basic control command, is used as the input of the inverse compensator to calculate the theoretical driving command. When the change trajectory of the basic control command is detected to cross the static dead zone boundary, a step compensation amount proportional to the width of the original ship's mechanical dead zone is superimposed on the theoretical drive command to overcome the static friction and mechanical clearance of the original ship's machinery and eliminate the response dead zone. Calculate the first-order rate of change of the basic control command, and based on the width of the dynamic hysteresis loop, generate a hysteresis advance compensation amount that is proportional to the rate of change of the command and superimpose it onto the theoretical drive command to counteract the phase lag caused by mechanical transmission and hydraulic response. Apply speed change rate limiting and low-pass filtering to the corrected control command after superimposed compensation to suppress instantaneous destructive torque output by the external actuator caused by dead zone boundary jumps or high-frequency control commands, and prevent rigid impact between the modified mechanism and the original ship machinery. The smoothed and corrected control command is output to the external actuator, and the actual physical feedback of the original ship machinery is collected. The closed-loop residual between the actual physical feedback and the basic control command is calculated. The residual is then corrected a second time by a proportional-integral controller to generate the final control signal.

10. A control method for a rapid retrofit device for an unmanned vessel navigation control system, applied to a rapid retrofit device for an unmanned vessel navigation control system as claimed in any one of claims 1-8, characterized in that, Includes the following steps: S1. Inject test excitation signals into the modified actuator under safe conditions, synchronously collect the command input and the actual physical response of the original ship machinery, extract motion deviations to identify the static dead zone boundary and dynamic hysteresis loop width, and construct the nonlinear dead zone and hysteresis model of the original ship machinery. S2. Switch between manual driving, base station remote control, and autonomous navigation modes according to mission requirements; In unmanned control mode, basic control commands for the original ship's main engine and rudder angle are calculated and generated based on the ship's perception information. S3. Input the basic control command into the hysteresis model for feedforward compensation, superimpose the dead zone crossing compensation and hysteresis advance, and apply the speed change rate limit to generate correction commands for execution; during the execution process, collect the original ship hydraulic pipeline pressure to perform torque coordination of the dual power sources, and calculate the closed-loop residual based on the actual physical feedback for secondary correction and online model update; S4. Real-time monitoring of the stress status of the control channel, communication link, and control lever; dual redundancy seamless switching in case of channel failure; When the manual intervention torque exceeds the limit, the electromagnetic clutch is disengaged to transfer control to the operator at the machine; when a serious fault is triggered, the original machine is forced to stop through an independent hardware circuit.

Images