A control method and system of an amphibious operation and maintenance device

Abstract

The application discloses a kind of amphibious operation and maintenance device's multi-scene adaptive control method and system, device includes four independent driving propulsion unit.The method real-time acquisition environment perception data, simultaneously acquisition short-range route data, according to this identification current operation scene and predict future operation scene, from multiple preset drive mode match target mode, respectively control the rotating speed, torque and rotating direction of four propulsion units.When predicting that scene will switch, execute transition process control-control parameter pre-adjustment before switching occurs, according to switching mode determine time window and smooth transition to target value when switching occurs.The application also has propulsion unit state monitoring, fault-tolerant reconstruction and emergency escape function.The application solves the problem of scene switching instability in multi-medium environment in the prior art, and is suitable for fish-light complementary photovoltaic power station, aquaculture, environmental monitoring and other fields of amphibious operation and maintenance operation.

Description

Technical Field

[0001] This invention relates to the field of control technology for waterborne operation equipment, specifically to a control method and system for an amphibious operation and maintenance device capable of adaptive and stable movement and operation in multi-media environments such as deep water, shallow water, tidal flats, silt, and hard ground. This invention is particularly applicable to construction, installation, operation, and maintenance operations under the "waterborne power generation, underwater aquaculture" mode in solar-fishery complementary photovoltaic power station projects, as well as in fields such as aquaculture, environmental monitoring, and shallow sea exploration. Background Technology

[0002] Currently, floating devices used in "land-sea interface" scenarios for solar-fishery complementary power plant projects, as well as in aquaculture, environmental monitoring, and shallow-sea exploration, mostly rely on traditional marine propellers or paddle wheels for movement. These typically only allow for forward and backward movement and limited turning, proving cumbersome and lacking in positioning accuracy when lateral, diagonal, or stationary rotation is required. In challenging environments such as mudflats with high frictional resistance and complex terrain, the maneuverability and controllability issues of existing amphibious operation and maintenance devices are particularly pronounced.

[0003] In recent years, technical solutions have emerged that employ the coordinated control of multiple propulsion units to achieve omnidirectional movement. For example, one solution uses a sensing module to collect real-time position, attitude, and environmental disturbance data of the unmanned surface vessel (USV), while the control module calculates the target thrust and direction combination of each propulsion unit based on preset hovering and omnidirectional movement algorithms, enabling movement in any direction while maintaining a constant heading. Another solution achieves underwater multi-degree-of-freedom motion by symmetrically arranging multiple propulsion units and dynamically configuring their rotational speed and steering.

[0004] Furthermore, in the field of amphibious vehicles, some solutions are based on multimodal signal recognition to build an automatic switching model for full-state machine transition modes, enabling automatic switching between different operating modes. Other solutions are based on structural feature models of amphibious vehicles operating in different environments such as land, water, and underwater, to build adaptive mode switching methods. In the field of solar-aquaculture hybrid photovoltaic operation and maintenance, existing solutions have proposed amphibious photovoltaic operation and maintenance platforms that include components such as a working platform and an electric propeller.

[0005] Despite the progress made in their respective directions, the aforementioned existing technologies still have the following shortcomings:

[0006] First, there is a lack of scene adaptation capability in multi-media environments. Existing multi-propulsion unit omnidirectional movement control technology is mainly designed for single-media environments (pure water or pure land), and does not fully consider the drastic changes in the load characteristics of the propulsion unit when switching between scenarios such as entering mudflats from the water surface or entering silt from the mudflats. The technical difficulties faced by amphibious operation and maintenance devices when switching between scenarios stem from the fundamental differences in the physical properties of the media. When the device is navigating in deep water, the propulsion unit is completely submerged in the water, and the load is mainly manifested as viscous fluid resistance, which is relatively stable and changes gradually. However, when the device moves from the water surface into the mudflats, the load of the propulsion unit suddenly changes from fluid resistance to solid-liquid mixed friction or even solid friction. This transformation is not simply an increase in resistance value, but a fundamental change in load characteristics. Fluid resistance has a velocity square characteristic, which increases gradually with the increase of rotational speed; while solid-liquid mixed friction has the abrupt change characteristics of static friction and dynamic friction. At the instant the propulsion unit contacts the tidal flat surface, the load can suddenly increase from the fluid resistance level (about 50 N·m to 100 N·m) to the solid-liquid mixed friction level (about 180 N·m to 350 N·m), increasing by several times to tens of times, and this change occurs in a very short time.

[0007] If constant control parameters are still used, it will lead to motor overload, sudden drop in speed, or even stall and burnout. At the same time, the hull attitude may become uncontrollable due to sudden changes in thrust. This technical difficulty does not exist in unmanned surface vessels in pure water, and it is a new problem that existing technologies have not solved. Existing multi-propulsion unit control schemes are all based on prior assumptions about the load characteristics of the propulsion units (i.e., load changes are continuous and predictable). Their control algorithms are essentially optimization problems in a continuous domain and cannot handle discontinuous events such as sudden changes in load characteristics.

[0008] Second, the scene switching process lacks fine-grained transition control. Although existing technologies involve mode switching, the switching process is mostly a step-like jump, directly switching from one control mode to another, without considering the smooth transition of control parameters when switching from one medium environment to another. Step-like switching is acceptable for mode switching in a single medium environment, but in water-land interface scenarios, the superposition of sudden load changes and parameter jumps can easily lead to instantaneous overload of the propulsion unit or a sudden drop in thrust, causing attitude instability.

[0009] Third, existing scene recognition technologies struggle to meet the predictive requirements of scene transitions. Current scene recognition largely relies on water depth or image information to determine the current scene, which is "post-event recognition"—meaning the scene change can only be identified after the device has entered the new scene, making it impossible to predict scene transitions in advance. Those skilled in the art know that if predictions cannot be made and control parameters adjusted before a scene transition occurs, the sudden load change at the moment of transition will inevitably lead to control failure. Water depth sensors can only measure the water depth at the current location and cannot detect terrain changes in areas yet to be reached; while image recognition can observe the water-land boundary ahead, it cannot quantify the load characteristics that the propulsion unit will soon bear. Existing technologies fail to utilize the propulsion unit's own current / torque feedback signals for early identification and prediction of the medium type, resulting in recognition accuracy and response speed that cannot meet the requirements of scene transition control. Summary of the Invention

[0010] This invention provides a control method and system for an amphibious operation and maintenance device, aiming to solve the technical problems of poor control adaptability, unstable transition process, and lack of fault tolerance in existing amphibious operation and maintenance devices when switching between multiple media environments.

[0011] To achieve the above objectives, the present invention first provides a control method for an amphibious operation and maintenance device. In this method, the amphibious operation and maintenance device is equipped with four independently driven propulsion units. The method first acquires environmental perception data in real time, which characterizes the state of the external environment in which the amphibious operation and maintenance device is currently located. Simultaneously, the method acquires short-range flight path data, which is either a preset flight path within a predetermined distance from the current position, or a theoretical flight path within a predetermined distance from the current position, constructed in real time based on the current heading and speed.

[0012] Based on the aforementioned data, this method identifies the current operational scenario using environmental perception data and predicts future operational scenarios based on the changing trends of environmental perception data along short-range flight path data. These operational scenarios include single-medium steady-state scenarios and cross-medium transitional scenarios. Based on the identified current operational scenario and the predicted future operational scenario, a corresponding drive mode is matched from multiple preset drive modes, each drive mode corresponding to a different combination of control parameters. After determining the drive mode, the rotational speed, torque, and rotation direction of the four propulsion units are controlled according to the control parameter combinations corresponding to the matched drive mode, thereby driving the amphibious maintenance device to execute a predetermined operational state.

[0013] Specifically, when a change in the operating scenario is predicted based on the changing trend of environmental perception data along the short-distance flight path, the method executes transition process control. This transition process control includes a pre-adjustment phase and a switch execution phase executed sequentially. In the pre-adjustment phase, before the switch occurs, the control parameters are gradually adjusted from their current values ​​to the target values ​​corresponding to the target driving mode, with the adjustment range being 30% to 70% of the difference between the current and target values. In the switch execution phase, when the switch occurs, a preset time window is determined based on the identified operating scenario switch mode. Within this preset time window, the control parameters are smoothly transitioned from their current values ​​at the end of the pre-adjustment phase to the target values, using either a linear or exponential transition method. Different operating scenario switch modes correspond to preset time windows of different sizes.

[0014] The system has several preset driving modes, including deep-water navigation mode, hard-surface driving mode, tidal flat propulsion mode, and silt evacuation mode. Each mode has different control objectives and constraints: In deep-water navigation mode, the control objectives are heading accuracy and speed accuracy, with balanced torque distribution among the propulsion units; in hard-surface driving mode, the control objectives are heading accuracy and speed accuracy, with a slip ratio threshold not exceeding 5%; in tidal flat propulsion mode, the control objective is to maximize the output torque of each propulsion unit, with speed following control, and a slip ratio threshold not exceeding 15%; in silt evacuation mode, the control objective is short-term peak torque output, with overload duration not exceeding 30 seconds, and coordinated activation of thermal protection.

[0015] The above-mentioned combination of control parameters is determined as follows: Based on the target heading and target speed corresponding to the predetermined operating state, the required resultant thrust and resultant torque are calculated; the optimization objective is determined based on the matched drive mode; under this optimization objective, the optimal speed and torque distribution scheme for each propulsion unit is solved using the rated speed and rated torque of each propulsion unit as constraints. The optimization objective varies depending on the drive mode: in deep-water navigation mode or hard-surface driving mode, the optimization objective is to minimize total energy consumption; in tidal flat propulsion mode, the optimization objective is to maximize the output torque of each propulsion unit, allowing for speed deviations.

[0016] The method also monitors the operating current, speed, and temperature of each propulsion unit in real time. When a failure is detected in any propulsion unit, the control strategy is dynamically reconfigured based on the number of remaining available propulsion units—if a single propulsion unit fails, the thrust of the remaining three propulsion units is redistributed; if two diagonally opposite propulsion units fail, the system switches to a dual-propulsion differential mode. When a propulsion unit is detected to be entangled or stranded, an emergency escape control sequence is executed, which includes alternating forward and reverse rotation and pulsed high-torque output.

[0017] To implement the above method, the present invention also provides a control system for an amphibious operation and maintenance device. This system includes a scene recognition module, a pattern matching module, a parameter optimization module, a drive execution module, and a transition control module.

[0018] The scene recognition module acquires environmental perception data in real time, identifies the current operating scene based on this data, and predicts future operating scenes based on the changing trends of the environmental perception data along short-distance route data. The environmental perception data characterizes the current external environment of the amphibious operation and maintenance device, including at least one of the following: water depth data, motor current / torque feedback data of each propulsion unit, hull attitude data, and image data of the waters ahead. The short-distance route data is either a preset route data within a predetermined distance from the current position, or theoretical route data constructed in real time based on the current heading and speed, within a predetermined distance from the current position, with the predetermined distance ranging from 10m to 50m. The scene recognition module identifies the current operating scene by: projecting the water depth data forward along the heading direction along the short-distance route data to predict the water depth change trend of the area the device will pass through within the predetermined distance; and determining the current operating scene and predicting future operating scenes based on the water depth change trend and the changing trend of the motor current / torque feedback data, combined with the hull attitude data and the image data of the waters ahead.

[0019] The pattern matching module has multiple preset drive modes, each corresponding to different combinations of control parameters, including deep-water navigation mode, hard-surface driving mode, tidal flat propulsion mode, and silt escaping mode. The parameter optimization module matches the target drive mode based on the identified current operating scenario and the predicted future operating scenario, and generates four independent drive control signals. The drive execution module controls the speed, torque, and rotation direction of the four propulsion units respectively according to the drive control signals.

[0020] The transition control module is used to execute transition process control when a change in the operating scenario is predicted based on the changing trend of environmental perception data along the short-distance flight path. This transition process control includes a pre-adjustment phase and a switch execution phase executed sequentially. In the pre-adjustment phase, before the switch occurs, the control parameters are gradually adjusted from their current values ​​to the target values ​​corresponding to the target driving mode. In the switch execution phase, when the switch occurs, a preset time window is determined based on the identified operating scenario switch mode, and within this preset time window, the control parameters are smoothly transitioned from their current values ​​at the end of the pre-adjustment phase to the target values.

[0021] This invention relates to a control method and system for an amphibious operation and maintenance device, which has the following beneficial effects:

[0022] This invention integrates information from multiple sensors, including water depth data, motor current / torque feedback data, ship attitude data, and images of the water ahead, to comprehensively determine the environment in which the device operates. Based on the changing trend of water depth data along a short-distance route, it predicts future operating scenarios. This enables adaptive scene recognition and drive mode matching in various media environments, such as deep water, shallow water, tidal flats, silt, and hard ground, solving the problems of control failure and poor adaptability in existing technologies when switching media.

[0023] This invention achieves free movement in eight directions—forward, backward, left, right, left-front, right-front, left-rear, and right-rear—through the coordinated control of four independent vector propulsion units and based on a direction-motor steering mapping table. Furthermore, it enables smooth steering through left-right differential steering and zero-radius in-situ turning through left-right reverse rotation, making it particularly suitable for precision operations in places such as narrow and complex waterways and tidal flat photovoltaic power stations.

[0024] This invention employs a two-stage transition control strategy of pre-adjustment and smooth transition—the pre-adjustment stage stores torque in advance and reduces speed to cope with sudden load changes, while the switching execution stage smoothly converges to the target value within a preset time window according to a linear or exponential function—effectively solving the attitude instability and motor overload problems caused by the sudden change in propulsion unit load from fluid resistance to solid-liquid mixed friction during water surface / tidal flat switching, significantly improving operational safety.

[0025] This invention combines short-range flight path data with environmental perception data to achieve early perception and quantitative prediction of terrain change trends ahead. The short-range flight path data is used only as a spatial reference for forward projection prediction, not for navigation. The system can initiate a pre-adjustment phase several seconds to tens of seconds before the device reaches the scene switching point, reserving sufficient torque output capacity for high-load scenarios and avoiding the passive response mode of "switching only when it arrives".

[0026] This invention has the ability to self-diagnose and dynamically reconfigure propulsion unit faults—real-time monitoring of the operating current, speed and temperature of each propulsion unit; when a single unit fails, the remaining three units redistribute thrust to maintain basic navigation capability; when two diagonally opposite units fail, it switches to dual-propulsion differential mode to achieve emergency turning and docking—significantly improving the operational reliability and survivability of the device in harsh environments.

[0027] This invention pre-sets a dedicated emergency escape control sequence for common fault modes such as entanglement in aquatic plants and stranding on mudflats. By alternating forward and reverse operation and pulsed high torque output, the hull is made to sway to assist in escape. At the same time, the motor temperature and current are continuously monitored during the escape process, and the over-limit state is automatically terminated and an alarm is triggered, thus balancing escape efficiency and equipment safety.

[0028] This invention achieves stepless speed regulation from slow movement to fast navigation based on PWM motor speed control. Through scene-adaptive parameter scheduling—with the optimization goal of minimizing total energy consumption in deep water / hard ground mode and maximizing output torque in tidal flat mode—it realizes targeted control strategies for different operating conditions, resulting in smooth start-stop and controllable energy consumption.

[0029] This invention covers all operational stages of amphibious operation and maintenance devices from normal navigation and scene switching to fault handling through a complete control logic chain of "scene recognition → pattern matching → parameter optimization → transition control → fault-tolerant reconstruction". The method claims and system claims correspond to each other and can be widely applied to amphibious operation and maintenance operations in fields such as solar-fishery complementary photovoltaic power stations, aquaculture, environmental monitoring, and shallow sea exploration. Attached Figure Description

[0030] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0031] Figure 1 A flowchart illustrating a control method for an amphibious operation and maintenance device provided in this application embodiment;

[0032] Figure 2 A block diagram of a control system for an amphibious operation and maintenance device provided in an embodiment of this application. Detailed Implementation

[0033] The technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings.

[0034] In the following description, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance. The following description provides multiple embodiments of this application, which can be substituted or combined with each other. Therefore, this application can also be considered to include all possible combinations of the same and / or different embodiments described. Thus, if one embodiment includes features A, B, and C, and another embodiment includes features B and D, then this application should also be considered to include embodiments containing one or more other possible combinations of A, B, C, and D, even if such embodiments are not explicitly described in the following text.

[0035] The following description provides examples and does not limit the scope, applicability, or examples set forth in the claims. Changes may be made to the function and arrangement of the described elements without departing from the scope of this application. Various processes or components may be appropriately omitted, substituted, or added to the examples. For example, the described methods may be performed in a different order than described, and various steps may be added, omitted, or combined. Furthermore, features described with respect to some examples may be combined into other examples.

[0036] In the control process of amphibious operation and maintenance devices, the most critical and challenging technical problem lies in the transition control when the operating scenario changes. To more clearly illustrate the technical problem solved by this invention, the following analysis examines the load change mechanism during scenario switching and the resulting control difficulties.

[0037] As an example, the amphibious operation and maintenance device provided by the present invention includes a vehicle (hull), a propulsion unit installed on the vehicle, and a control system.

[0038] The vehicle is rectangular or other shapes. Taking a rectangle as an example, a propulsion unit can be installed at each of the four corners. The propulsion unit can be a cylindrical propulsion pontoon with spiral ribs to form a propeller-like propulsion effect. When the vehicle is of other shapes, the four propulsion units can be installed at the four corners.

[0039] In one embodiment of this application, when the amphibious operation and maintenance device (hereinafter referred to as the device) is navigating normally in deep water, the propellers of the four propulsion pontoons are completely submerged in the water, and the load on the propulsion pontoons is mainly manifested as viscous fluid resistance. This resistance is approximately proportional to the square of the rotational speed and has the characteristics of continuous and predictable variation. At this time, stable rotational speed and torque output can be achieved by using conventional PID speed closed-loop control, and the operating parameters (current, rotational speed, torque) of each propulsion pontoon are in a relatively stable state.

[0040] As the device gradually approaches the tidal flat area from the deep water zone, the water depth continuously decreases, and the immersion depth of the propulsion pontoon decreases accordingly. During this process, the propulsion pontoon is initially partially exposed above the water surface, or stirs up bottom sediment in the shallow water zone, and the load characteristics begin to deviate from the pure fluid resistance model. More importantly, when the device officially enters the tidal flat area, the propulsion pontoon propeller will come into direct contact with the solid-liquid mixture (wet sand, clay, gravel, etc.), and the load abruptly changes from fluid resistance to solid-liquid mixed frictional resistance. This abrupt change has three characteristics: First, the magnitude of the abrupt change is extremely large, with torque requirements rising sharply from 50 N·m to 100 N·m during conventional navigation to 180 N·m to 350 N·m; second, the abrupt change is extremely rapid, typically completed within seconds or even hundreds of milliseconds; third, the load characteristics fundamentally change after the abrupt change, transforming from a velocity-square characteristic to a complex nonlinear characteristic of solid-liquid mixed friction, including multiple mechanical components such as static friction, dynamic friction, and medium shear.

[0041] The aforementioned load surge places three contradictory demands on the control system: if the parameters are adjusted too slowly, the propulsion float will stop or stall under high load due to insufficient torque, causing the device to lose power; if the parameters are adjusted too quickly (step switching), the speed and torque commands of the propulsion float will change dramatically in an instant, which may cause electrical overload or mechanical shock, and may also lead to device malfunction; if adjustments are made solely based on feedback signals after the event (post-event response), there is an unavoidable delay between the occurrence of the load surge and the control system recognizing the change in the event and completing the parameter adjustment. During this delay, the propulsion unit is already in an overloaded or stalled state, and the adjustment is too late.

[0042] This invention addresses the aforementioned technical difficulties by employing a three-pronged mechanism: "forward projection of environmental perception data along the short-distance route to predict scene changes + advance reserve of torque / speed capacity during the pre-adjustment phase + smooth transition during the switching execution phase." Forward projection of environmental perception data along the short-distance route allows the system to predict scene changes before the device reaches the new scene, providing sufficient time lead for pre-adjustment. During the pre-adjustment phase, control parameters are partially adjusted towards the target value before the switch occurs, avoiding abrupt parameter changes and ensuring the propulsion float has the necessary torque or speed capacity to handle sudden load changes. The smooth transition during the switching execution phase completes the smooth convergence of remaining parameters based on the pre-adjustment, achieving a shock-free transition from the pre-adjustment end value to the target value.

[0043] The four propulsion units are designated V1, V2, V3, and V4. Each propulsion float is equipped with a sealed brushless DC motor to provide driving force. The four propulsion floats provide buoyancy support for floating and also generate propulsion through the rotation of the helical rib propeller. The control system can use an STM32 series microcontroller as the main controller, connected to the four DC motor drive modules, to generate four independent PWM signals to control the operation of each motor.

[0044] According to the first aspect of this application, see Figure 1 This application provides a control method for an amphibious operation and maintenance device, comprising the following steps:

[0045] S1: Real-time acquisition of environmental perception data, which is used to characterize the current external environment of the amphibious operation and maintenance device;

[0046] S2: Obtain short-distance route data. The short-distance route data is the route data within a preset distance from the current position in the preset route, or the theoretical route data within a preset distance from the current position constructed in real time based on the current heading and speed.

[0047] S3: Identify the current operating scenario based on environmental perception data, and predict future operating scenarios based on the changing trends of environmental perception data along short-distance flight path data;

[0048] S4: Based on the current operating scenario and / or future operating scenario, match the current driving mode and / or the target driving mode from multiple preset driving modes, with each driving mode corresponding to a different combination of control parameters;

[0049] S5: According to the control parameter combination corresponding to the matched drive mode, control the speed, torque and rotation direction of the four propulsion units respectively, so as to drive the amphibious operation and maintenance device to perform the predetermined operation state.

[0050] S6: When a change in the operating scenario is predicted based on the changing trend, transition process control is executed:

[0051] Transition process control includes a pre-adjustment phase and a switching execution phase, which are executed sequentially.

[0052] During the pre-adjustment phase, before the switch occurs, the control parameters are gradually adjusted from their current values ​​to the target values ​​corresponding to the target drive mode.

[0053] During the switching execution phase, when the switching occurs, a preset time window is determined based on the identified operating scenario switching mode. Within the preset time window, the control parameters are smoothly transitioned from the current value at the end of the pre-adjustment phase to the target value.

[0054] The control method first acquires environmental perception data in real time. This environmental perception data characterizes the current external environment of the amphibious operation and maintenance device, including at least one of the following: water depth data, motor current / torque feedback data from each propulsion unit, hull attitude data, and forward water area image data. Water depth data is collected by a laser depth sensor to determine the water depth of the area where the device is located, distinguishing between deep water, shallow water, and mudflats. Motor current / torque feedback data is collected by the drive feedback module of each propulsion unit's motor to identify the load characteristics borne by the propulsion unit—current is stable and fluctuates little for fluid resistance, while current fluctuates greatly and peaks high for solid friction or solid-liquid mixed friction. Hull attitude data is collected by an attitude sensor, including heel angle, pitch angle, and heading angle, to determine the hull's spatial attitude and its changing trends. Forward water area image data is collected by a visible light camera installed at the front of the device, and after image processing, is used to identify environmental features such as land-water boundaries, obstacles, and vegetation distribution.

[0055] Of course, other environmental sensing sensors, such as sonar and radar, can also be used. They can be adapted based on the actual application scenario, with the aim of providing data support for switching sensing operation scenarios.

[0056] Simultaneously, the control system acquires short-range route data. This short-range route data can be route data within a preset distance from the current position, or theoretical route data within a preset distance from the current position, constructed in real-time based on the current heading and speed. The preset distance is typically 10m to 50m, and this range can be calibrated according to the device's actual speed and braking distance—the distance should not be too long to avoid introducing interference from irrelevant environmental information at a distance; nor should it be too short to ensure the system has sufficient lead time to complete pre-adjustments for scene switching. In a preferred embodiment, when the device speed is 0.8m / s to 1.5m / s, the preset distance is preferably 20m to 30m, corresponding to a prediction lead time of approximately 13s to 37s.

[0057] It should be clarified that in this application, short-range route data is only used to provide a spatial reference direction for forward prediction—that is, "which direction to look at the environmental changes ahead"—rather than to guide the device from its starting point to its destination. When there is a preset route, the control system extracts a short-range segment from the preset complete route data; when there is no preset route, the control system constructs a theoretical segment in real time based on the current heading and speed. In either case, the short-range route data is unrelated to the actual navigation control of the device; its function is to provide a directional reference for forward projection prediction.

[0058] After acquiring environmental perception data and short-range flight path data, the control system identifies the current operating scenario based on the environmental perception data and predicts the future operating scenario based on the changing trend of the environmental perception data along the short-range flight path data.

[0059] Specifically, the control system projects water depth data forward along the heading direction, following the direction indicated by the short-distance route data, to predict the water depth change trend of the area the device will pass through within a set distance. Then, based on the water depth change trend and the change trend of motor current / torque feedback data, combined with hull attitude data and forward water area image data, it determines the current operating scenario and predicts future operating scenarios. For example, when the water depth data sequence along the short-distance route shows that the water depth continuously decreases from 2m to below 0.5m within seconds, the system predicts that the device will soon move from deep water to shallow water or tidal flats; when the motor current feedback shows that the load torque gradually increases from a low torque stable state to a high torque fluctuating state, the system predicts that the device will soon move from the water surface to tidal flats. The mutual corroboration between the water depth change trend and the motor load change trend improves the accuracy and reliability of scenario prediction.

[0060] The control system has multiple preset drive modes, each corresponding to a different combination of control parameters. These drive modes include deep-water navigation mode, hard-surface driving mode, tidal flat propulsion mode, and silt extrication mode.

[0061] In deep-water navigation mode, with sufficient water depth and primarily controlled by fluid resistance, the control objectives are maintaining course and speed accuracy. Torque distribution among the propulsion units is balanced, and the four propulsion units coordinate their propulsion. In this mode, the power of a single motor ranges from 0.5kW to 1.0kW, the speed from 60r / min to 100r / min, and the torque from 50N·m to 100N·m. The total power of the four motors ranges from 2kW to 4kW, and the speed ranges from 0.8m / s to 2.0m / s. The control method is constant speed control, with the four motors operating at 30% to 70% power.

[0062] In hard ground driving mode, the device operates on hard surfaces such as cement / asphalt hardened surfaces, gravel roads, or compacted dirt roads. The friction coefficient is stable, and rolling resistance is low. The control objectives are heading accuracy and speed accuracy, with a slip ratio control threshold not exceeding 5%. In this mode, the single motor power is 0.3kW~0.6kW, the single motor speed is 80r / min~150r / min, and the single motor torque is 30N·m~80N·m. The control method is constant speed control, prioritizing speed and heading accuracy, with the slip ratio controlled within 5%.

[0063] In the tidal flat propulsion mode, the device operates in a solid-liquid mixed medium environment with high frictional resistance. The control objective is to maximize the output torque of each propulsion unit to overcome resistance. The rotational speed is controlled by following the direction of rotation, and the slip ratio control threshold does not exceed 15%. In this mode, the power of a single motor is 1.0kW to 1.5kW, the speed of a single motor is 40r / min to 80r / min, the torque of a single motor is 100N·m to 180N·m, the total power of the four motors is 4kW to 6kW, the control method is constant torque / limited slip control, the slip ratio is controlled within 15%, and the four motors operate at 60% to 90% power.

[0064] In the sludge escaping mode, when the device becomes trapped in high-viscosity sludge, the control objective is to output peak torque for a short period to escape the predicament. The overload duration threshold does not exceed 30 seconds, and thermal protection is activated in conjunction to prevent the motor from overheating and burning out. In this mode, the power of a single motor is 1.5kW to 3.0kW (short-term overload), the speed of a single motor is 20r / min to 50r / min, and the torque of a single motor is 180N·m to 350N·m, with a duration not exceeding 30 seconds. The control method is a combination of peak torque limiting and thermal protection control, with all four motors operating at 80% to 120% power.

[0065] Based on the identified current operating scenario and the predicted future operating scenario, the control system matches the current driving mode and the target driving mode from the above multiple driving modes.

[0066] Specifically, when the system identifies that it is currently in deep water, it matches the deep water navigation mode as the current driving mode; when the system predicts that it is about to enter the tidal flat area, it matches the tidal flat propulsion mode as the target driving mode, in order to prepare for subsequent scene switching.

[0067] As a feasible implementation, the current operating scenario can be determined based on only some environmental perception data. For example, the current operating scenario can be determined using only the operating parameters and operating environment parameters of the amphibious maintenance device and non-visual data. Specifically, this could include relevant parameters of the drive unit and water depth parameters.

[0068] Based on changes in environmental perception parameters, the system can determine the operating scenario, including the current operating scenario and the future operating scenario.

[0069] Secondly, the current operating scenario can be determined using only visualized data, such as water area image data, for image recognition.

[0070] In addition, a combination of the two methods mentioned above can be used for judgment, which is more accurate and reliable.

[0071] After determining the drive mode, the control system controls the speed, torque, and direction of rotation of the four propulsion units according to the control parameter combination corresponding to the matched drive mode, thereby driving the amphibious operation and maintenance device to execute the predetermined operating state. The control parameter combination is determined as follows: first, the required resultant thrust and resultant torque are calculated based on the target heading and target speed corresponding to the predetermined operating state; then, the optimization objective is determined according to the matched drive mode. Under this optimization objective, the optimal speed and torque distribution scheme for each propulsion unit is solved using the rated speed and rated torque of each propulsion unit as constraints. The optimization objective varies depending on the drive mode: in deep-water navigation mode or hard-surface driving mode, the propulsion unit load is relatively small and energy consumption is relatively high, so the optimization objective is to minimize total energy consumption; in tidal flat propulsion mode, the propulsion unit needs to output a large torque to overcome high resistance, so the optimization objective is to maximize the output torque of each propulsion unit, allowing for a certain degree of speed deviation.

[0072] In terms of steering control, the control system achieves multiple steering modes through differential drive of the four propulsion units. During normal steering, the left and right propulsion units steer differentially with a speed difference of 20% to 50%, with the inner unit having slightly lower torque for smooth steering. When turning on the spot, the left and right propulsion units rotate in opposite directions to achieve zero-radius turning. During lateral movement, lateral translation is achieved through vector synthesis of the four propulsion units. In case of emergency stop, the control system controls each propulsion unit to rapidly reduce torque to zero to prevent impact.

[0073] When the control system predicts a change in the operating scenario based on the changing trend of environmental perception data along the short-distance flight path, it executes transient process control. Transient process control includes a pre-adjustment phase executed sequentially and a switching execution phase.

[0074] During the pre-adjustment phase, before the switchover occurs, the control system gradually adjusts the control parameters from their current values ​​to the target values ​​corresponding to the target drive mode. For example, when it is predicted that the system will be moving from deep water to tidal flats, the system gradually reduces the speed of each motor and gradually increases the torque limit value in advance, so that the propulsion unit has sufficient torque output capacity to cope with the upcoming high-load environment. During the pre-adjustment phase, the adjustment range of the control parameters is usually 30% to 70% of the difference between the current value and the target value. This avoids both insufficient pre-adjustment effect due to too small an adjustment range and abnormal attitude or energy waste due to too large an adjustment range.

[0075] During the switching execution phase, when a switch occurs, the control system determines a preset time window based on the identified operating scenario switching mode. Within this preset time window, the control parameters are smoothly transitioned from their current values ​​at the end of the pre-adjustment phase to the target values. The smooth transition is achieved by varying the control parameters according to a linear or exponential function within the preset time window, thus realizing a smooth convergence from the pre-adjustment end value to the target value.

[0076] Different operating scenario switching modes correspond to different preset time windows. These modes include water surface to mudflat switching, deep water to shallow water switching, and mudflat to hard ground switching. In the water surface to mudflat switching mode, the propulsion unit load abruptly changes from fluid resistance to solid-liquid mixed friction, resulting in the most drastic load change. Therefore, the corresponding preset time window is the shortest, preferably 2-3 seconds, to ensure rapid response of control parameters to load changes and prevent the propulsion unit from getting stuck in silt or becoming blocked due to parameter adjustment lag. In the deep water to shallow water switching mode, the load change is relatively gradual, resulting in the longest preset time window, preferably 4-5 seconds, to achieve a smooth parameter transition and avoid attitude fluctuations. In the mudflat to hard ground switching mode, the load changes from high-resistance solid-liquid mixed friction to low-resistance rolling friction, with the load change amplitude falling between the two modes mentioned above. The corresponding preset time window also falls between these two modes, preferably 3-4 seconds. The specific values ​​of these time windows can be calibrated based on the actual inertial parameters of the device and the response characteristics of the propulsion unit, and are not limited to the above preferred values.

[0077] In one feasible embodiment of this application, the operating current, rotational speed, and temperature of each propulsion unit are also monitored in real time. When any propulsion unit is detected to be overloaded, stalled, or experiencing a failure rate exceeding the limit, the control system automatically performs fault isolation and dynamically reconstructs the control strategy based on the number of remaining available propulsion units. When a single propulsion unit fails, the remaining three propulsion units redistribute thrust—by adjusting the rotational speed and torque combination of the remaining three propulsion units, the thrust lost by the failed unit is compensated, maintaining the basic navigation capability of the device. Although some maneuverability is sacrificed, forward, backward, and limited turning are still possible. When two diagonally opposite propulsion units fail, the system switches to a dual-propulsion differential mode—utilizing the differential rotation of the remaining two propulsion units to achieve an emergency turn and docking, ensuring the safe return of the device.

[0078] In one feasible embodiment of this application, an emergency escape function is also provided. When the propulsion unit is detected to be entangled in aquatic plants or stranded on the mudflat, the control system executes an emergency escape control sequence. The emergency escape control sequence includes alternating forward and reverse rotation and pulsed high-torque output—attempting to break free from the entanglement through alternating forward and reverse rotation of the propulsion unit, or attempting to break free from the stranded state through short-duration pulsed high-torque output. During the escape process, the control system continuously monitors the current and temperature of each propulsion unit. When continuous overload or temperature exceeding the safety threshold is detected, the escape sequence is automatically terminated and an alarm signal is issued, awaiting manual intervention.

[0079] Based on the amphibious operation and maintenance device, a control console can be configured to perform manual operations for product debugging or field testing, based on the aforementioned operating scenarios and drive modes. Furthermore, in practical applications, manual operation can also be used to switch operating scenarios and autonomously adjust drive modes.

[0080] According to the second aspect of this application, see Figure 2 The embodiments of this application also provide a control system for an amphibious operation and maintenance device, including a scene recognition module, a pattern matching module, a parameter optimization module, a drive execution module, and a transition control module.

[0081] The scene recognition module acquires environmental perception data in real time, identifies the current operating scene based on this data, and predicts future operating scenes based on the changing trends of the environmental perception data along short-distance route data. This module is equipped with multi-source sensor interfaces, including a laser depth sensor, a motor current / torque feedback acquisition unit, an attitude sensor, and a visible light image acquisition unit. The scene recognition module first acquires water depth data, motor current / torque feedback data from each propulsion unit, hull attitude data, and forward water area image data. Then, it compares this multi-source data with a preset scene feature library to determine the current operating scene. Simultaneously, the scene recognition module projects the water depth data forward along the direction indicated by the short-distance route data to obtain the water depth change trend. Combining this with the motor current / torque feedback data and the forward water area image data, it predicts the changing trends of the operating scene in the area the device will traverse within a future time window. The short-distance route data is extracted by the scene recognition module from a preset route or constructed in real time based on the current heading and speed, with a set distance of 10m to 50m.

[0082] The pattern matching module has multiple preset driving modes, each corresponding to different combinations of control parameters, including deep-water navigation mode, hard-surface driving mode, tidal flat propulsion mode, and silt evacuation mode. Based on the current and future operating scenarios output by the scene recognition module, the pattern matching module matches the current driving mode and the target driving mode. When the scene recognition module determines that the current area is deep water, the pattern matching module matches the deep-water navigation mode as the current driving mode; when the scene recognition module predicts that the area will enter tidal flats, the pattern matching module simultaneously matches the tidal flat propulsion mode as the target driving mode.

[0083] The parameter optimization module generates four independent drive control signals based on the matched drive mode. First, it calculates the required resultant thrust and torque based on the target heading and speed corresponding to the predetermined operating state. Then, it determines the optimization objective based on the matched drive mode—minimizing total energy consumption in deep-water or hard-surface driving modes, and maximizing the output torque of each propulsion unit in tidal flat propulsion mode. Under the determined optimization objective, and with the rated speed and rated torque of each propulsion unit as constraints, it solves for the optimal speed and torque distribution scheme for each propulsion unit. Finally, it converts the solution into four PWM drive control signals for output.

[0084] The drive execution module is used to control the speed, torque, and rotation direction of the four propulsion units respectively, based on the four drive control signals output by the parameter optimization module. The drive execution module includes four independent motor drive circuits, each connected to the corresponding brushless DC motor. After receiving the PWM control signal, it drives the motor to operate at the target speed and torque.

[0085] The transition control module is used to execute transition process control when the scene recognition module predicts that the running scene will switch.

[0086] During the pre-adjustment phase, before the switch occurs, the control parameters are gradually adjusted from their current values ​​to the target values ​​corresponding to the target drive mode.

[0087] During the switching execution phase, when the switching occurs, a preset time window is determined based on the identified operating scenario switching mode. Within the preset time window, the control parameters are smoothly transitioned from the current value at the end of the pre-adjustment phase to the target value.

[0088] Specifically, the transition control module initiates a pre-adjustment phase before the switch occurs—gradually adjusting the control parameters corresponding to the current driving mode towards the target value corresponding to the target driving mode, with the adjustment range being 30% to 70% of the difference between the current value and the target value. When the switch occurs, the transition control module initiates a switch execution phase—determining a preset time window based on the identified operating scenario switch mode, and smoothly transitioning the control parameters from the current value at the end of the pre-adjustment phase to the target value within this time window. The transition process control includes the sequentially executed pre-adjustment phase and switch execution phase.

[0089] All of the above modules can be integrated into an STM32 series microcontroller, enabling information exchange between modules via an internal data bus. The scene recognition module sends the recognition and prediction results to the pattern matching module, which in turn sends the matching results to the parameter optimization module. The parameter optimization module then sends the generated drive control signals to the drive execution module. The transition control module intervenes in the control parameter output of the parameter optimization module during scene switching, achieving smooth adjustment of the control parameters. In actual operation, these modules work together to form a complete control closed loop of "perception → decision → execution → transition".

[0090] The following is a specific application example to further illustrate the technical solution of the present invention.

[0091] In a solar-fishery hybrid photovoltaic power station project, an amphibious operation and maintenance device needs to navigate from deep water to the tidal flats to perform maintenance on the photovoltaic panels. Initially, the device is in deep water, approximately 3 meters deep. The scene recognition module acquires water depth data via a laser depth sensor and identifies typical fluid resistance characteristics through motor current feedback, indicating stable current with minimal fluctuations. Based on this, the current operation is determined to be in deep water navigation mode. The mode matching module selects this deep water navigation mode as the current drive mode. The parameter optimization module prioritizes heading and speed control, aiming to minimize total energy consumption. It solves for the optimal speed and torque distribution scheme for each propulsion unit, and the drive execution module controls the device to navigate towards the tidal flats at a speed of approximately 1.2 m / s.

[0092] During navigation, the scene recognition module continuously acquires short-distance route data, intercepts a segment within 25 meters ahead of the current position along the preset route, and projects the water depth data forward along the short-distance route direction. When the water depth data sequence shows that the water depth ahead continuously decreases from 3 meters to 0.8 meters within a few seconds, the scene recognition module predicts that the device is about to enter the mudflat area. The transition control module then initiates the pre-adjustment phase—approximately 5 seconds before the switch occurs, it gradually reduces the speed of each motor from 100 r / min to 70 r / min, while simultaneously increasing the torque limit from 100 N·m to 150 N·m to complete torque reserve. The adjustment range during the pre-adjustment phase is approximately 50% of the difference between the current value and the target value.

[0093] When the water depth further decreases to below 0.5m and the motor current feedback indicates that the load characteristics have changed from stable fluid resistance to fluctuating solid-liquid mixed friction, the scene recognition module confirms that the device has entered the tidal flat area, and the transition control module initiates the switching execution phase. The transition control module recognizes that the current switch is from water surface to tidal flat switching mode, determines the preset time window to be 2.5s, and smoothly transitions the control parameters from the pre-adjustment end value to the target value of the tidal flat propulsion mode within 2.5s—further adjusting the rotational speed to 60 r / min and increasing the torque limit to 180 N·m, switching the control mode to torque-priority output. The smooth transition uses an exponential function approach, making the parameter changes faster in the initial stage and slower in the later stage, to quickly respond to sudden load changes while avoiding parameter overshoot.

[0094] After the switchover is completed, the control system enters the post-switching stabilization phase and operates stably in the tidal flat propulsion mode. The slip ratio is controlled within 15%, and the device slowly advances in the tidal flat area at a speed of about 0.5 m / s, successfully completing the maintenance work on the photovoltaic panels.

[0095] Throughout the operation, the scene recognition module continuously monitors the operating status of each propulsion unit. If an abnormal increase in current is detected in a propulsion unit due to entanglement with seaweed, the transition control module automatically executes an emergency escape control sequence—first, it performs alternating forward and reverse operations to attempt to break free from the entanglement; if this fails, it executes a short-duration pulse high-torque output while simultaneously monitoring current and temperature. If escape is successful, the device resumes normal operation; if escape fails and the temperature exceeds a safety threshold, the system automatically stops the propulsion unit and issues an alarm signal. Simultaneously, it reconstructs the control strategy based on the remaining available propulsion units to maintain the device's basic navigation capability for a safe return.

[0096] In another application scenario, the device needs to be moved from the tidal flats to the hardened shoreline for equipment maintenance. At this time, the laser depth sensor detects that the water depth is zero, the motor current feedback shows that the load torque rapidly decreases from the high torque (100 N·m to 180 N·m) in tidal flat mode to the low torque (30 N·m to 80 N·m), the current fluctuation rate decreases from the higher value in tidal flat mode to below 5%, and the attitude sensor shows that the hull attitude tends to stabilize. The scene recognition module determines that the current scene is a hard ground driving mode, and the mode matching module matches the hard ground driving mode as the target driving mode. The transition control module automatically executes scene switching transition control: after predicting that it will enter the hard ground, it starts the pre-adjustment phase, gradually reducing the torque limit of each motor from 180 N·m to 80 N·m within 3 seconds, while increasing the target speed from 60 r / min to 120 r / min; after entering the hard ground, it executes the switching execution phase, smoothly transitioning the control parameters to the target values ​​of the hard ground driving mode within 2 seconds; after the switching is completed, the parameter optimization module prioritizes heading maintenance and speed control, and allocates the power output of each propulsion unit with the goal of minimizing total energy consumption, and the drive execution module controls the slip ratio to be within 5%.

[0097] The above examples are merely one specific application of the present invention. The present invention is also applicable to amphibious operations in other land-sea interface scenarios such as aquaculture, environmental monitoring, and shallow sea exploration. Depending on the application scenario and device parameters, the above control parameters can be adaptively adjusted within the scope of the present invention, and all such adjustments fall within the protection scope of the present invention.

[0098] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that this application is not limited to the described order of actions, as some steps may be performed in other orders or simultaneously according to this application. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions and modules involved are not necessarily essential to this application.

[0099] In the above embodiments, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0100] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage device, which may include: a flash drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, etc.

[0101] The foregoing description is merely an exemplary embodiment of this disclosure and should not be construed as limiting the scope of this disclosure. Any equivalent changes and modifications made in accordance with the teachings of this disclosure shall still fall within the scope of this disclosure. Those skilled in the art will readily conceive of embodiments of this disclosure upon considering the specification and practicing the disclosure herein. This application is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not described herein. The specification and embodiments are to be considered exemplary only, and the scope and spirit of this disclosure are defined by the claims.

Claims

1. A control method for an amphibious operation and maintenance device, characterized in that, Includes the following steps: S1: Real-time acquisition of environmental perception data, which is used to characterize the current state of the external environment in which the amphibious operation and maintenance device is located; S2: Obtain short-distance route data, which is route data within a predetermined distance ahead from the current position in a preset route, or theoretical route data within a predetermined distance ahead from the current position, constructed in real time based on the current heading and speed. S3: Identify the current operating scenario based on the environmental perception data, and predict the future operating scenario based on the changing trend of the environmental perception data along the short-distance flight path data; S4: Based on the current operating scenario and / or the future operating scenario, match the current driving mode and / or the target driving mode from a plurality of preset driving modes, each driving mode corresponding to a different combination of control parameters; S5: According to the control parameter combination corresponding to the matched drive mode, control the speed, torque and rotation direction of the four propulsion units respectively, so as to drive the amphibious operation and maintenance device to perform the predetermined operation state. S6: When a change in the operating scenario is predicted based on the aforementioned trend, transition process control is executed: The transition process control includes a pre-adjustment phase and a switching execution phase executed sequentially; During the pre-adjustment phase, before the switching occurs, the control parameters are gradually adjusted from their current values ​​to the target values ​​corresponding to the target driving mode. During the switching execution phase, when the switching occurs, a preset time window is determined based on the identified operating scenario switching mode. Within the preset time window, the control parameters are smoothly transitioned from the current value at the end of the pre-adjustment phase to the target value.

2. The method according to claim 1, characterized in that, The environmental perception data includes at least one of the following: water depth data, motor current / torque feedback data of each propulsion unit, hull attitude data, and image data of the water ahead.

3. The method according to claim 1, characterized in that, The preset driving modes include deep-water navigation mode, hard ground driving mode, tidal flat propulsion mode, and silt extrication mode.

4. The method according to claim 2, characterized in that, In step S3, the methods for identifying the current operating scenario and predicting the future operating scenario are as follows: Along the short-distance route data, the water depth data is projected forward along the heading direction to predict the water depth change trend of the area the device will pass through within the set distance. Based on the water depth change trend and / or the change trend of the motor current / torque feedback data, combined with the hull attitude data and / or the image data of the water ahead, the current operating scenario is determined and the future operating scenario is predicted.

5. The method according to claim 1, characterized in that, In step S5, the combination of control parameters is determined in the following manner: Calculate the required resultant thrust and resultant torque based on the target heading and / or target speed corresponding to the predetermined operating state; The optimization objective is determined based on the matching drive mode. Under the optimization objective, the optimal speed and torque distribution scheme of each propulsion unit is solved with the rated speed and rated torque of each propulsion unit as constraints. The optimization objectives, determined based on the driving mode, specifically include: In deep-water navigation mode or hard-surface driving mode, the optimization objective is to minimize total energy consumption; In the tidal flat propulsion mode, the optimization objective is to maximize the output torque of each propulsion unit, while allowing for deviations in rotational speed.

6. The method according to claim 1, characterized in that, The operation scenario switching modes include water surface to mudflat switching mode, deep water to shallow water switching mode, and mudflat to hard ground switching mode; The preset time window corresponding to the water surface to mudflat switching mode is shorter than the preset time window corresponding to the deep water to shallow water switching mode. The preset time window corresponding to the mudflat to hard ground switching mode is between the preset time windows corresponding to the water surface to mudflat switching mode and the deep water to shallow water switching mode.

7. The method according to claim 1, characterized in that, In step S6, the smooth transition is achieved by changing the control parameter from its current value at the end of the pre-adjustment phase to the target value within the preset time window, following a linear or exponential function.

8. The method according to claim 1, characterized in that, Also includes: Real-time monitoring of the operating current, speed, and temperature of each propulsion unit; When a failure is detected in any propulsion unit, the control strategy is dynamically reconfigured based on the number of remaining available propulsion units. When a single propulsion unit fails, the remaining three propulsion units redistribute thrust; when two diagonally opposite propulsion units fail, the system switches to a dual-propulsion differential mode.

9. The method according to claim 1, characterized in that, Also includes: When the propulsion unit is detected to be entangled or stranded, an emergency escape control sequence is executed, which includes alternating forward and reverse operation and / or pulsed high torque output.

10. A control system for an amphibious operation and maintenance device, characterized in that, include: The scene recognition module is used to acquire environmental perception data in real time, identify the current operating scene based on the environmental perception data, and predict the future operating scene based on the changing trend of the environmental perception data along the short-distance route data. The environmental perception data is used to characterize the state of the external environment in which the amphibious operation and maintenance device is currently located. The short-distance route data is the route data within a set distance ahead from the current position in the preset route, or the theoretical route data within a set distance ahead from the current position, which is constructed in real time based on the current heading and speed. The pattern matching module has multiple preset driving modes, each of which corresponds to a different combination of control parameters. The parameter optimization module is used to match the target driving mode based on the identified current operating scenario and / or the predicted future operating scenario, and generate four independent driving control signals; The drive execution module is used to control the rotational speed, torque and rotational direction of the four propulsion units respectively according to the drive control signal; The transition control module is used to execute transition process control when the change trend predicts that the operating scenario will switch. The transition process control includes a pre-adjustment phase and a switching execution phase executed sequentially; During the pre-adjustment phase, before the switching occurs, the control parameters are gradually adjusted from their current values ​​to the target values ​​corresponding to the target driving mode. During the switching execution phase, when the switching occurs, a preset time window is determined based on the identified operating scenario switching mode. Within the preset time window, the control parameters are smoothly transitioned from the current value at the end of the pre-adjustment phase to the target value.

Images