Collaborative operation control method for float cleaning boat cluster facing complex water area

Abstract

The invention relates to a complex water area-oriented float-garbage boat cluster collaborative operation control method. The invention discloses a cluster fluid-solid coupling collaborative cleaning method and system based on drift flux field prediction. The method integrates environmental perception, fluid mechanics and multi-agent control technologies, and comprises the following steps: firstly, constructing a water area drift flux density field at an Euler perspective, and identifying a main transport channel by calculating a coupling relationship between floating object surface density and a flow velocity vector; and based on a maximum flux potential energy integral principle, planning an optimal interception section curve and topological distribution of a float cleaning ship cluster. In the intercepting operation stage, a fluid-solid coupling differential control strategy is adopted as the core, and an induced flow field with a specific pressure gradient is constructed around a ship body by setting the induced slippage speed between the floating debris cleaning ship and local water flow. The float cleaning boat can be changed into an active flow field regulator from passive blocking, so that the operation resistance and energy consumption are remarkably reduced, and the problems that the float cleaning efficiency is low and clusters are difficult to cooperatively maintain in a complex dynamic water area are solved.

Description

Technical Field

[0001] This invention relates to the field of aquatic environment management and unmanned system control technology, specifically a collaborative operation control method for a cluster of clean-up vessels in complex waters. Background Technology

[0002] With the increasing severity of aquatic pollution, the use of unmanned surface vehicle (USV) swarms for automated surface cleanup has become an important technological means to improve operational efficiency and reduce labor costs. Existing surface cleanup technologies mainly rely on patrol operations by single vessels or simple multi-vehicle convoys for coverage, typically equipped with physical filters, conveyor belts, or robotic arms to intercept and collect floating debris.

[0003] However, existing debris removal operations have significant technical limitations when facing complex and dynamic water flow environments. Traditional path planning methods are mostly based on static geometric maps, employing a comprehensive "lawnmower" path or random search strategy, ignoring the drifting characteristics of floating debris as it moves with the water flow. This "traditional approach" fails to consider the transport effect of water flow on floating debris, causing debris removal vessels to frequently idle in areas without debris, or to fail to intercept high-density accumulation areas before the debris spreads, resulting in poor targeting and low efficiency.

[0004] More importantly, existing interception and capture technologies generally employ the principle of "physical filtration," which relies on the thrust generated by a propeller to forcefully push a filter or collection bucket forward against water resistance, thereby trapping floating debris. This method has serious hydrodynamic defects: on the one hand, the dense physical filter generates enormous pressure drag when moving in water, and as the operation progresses, the filter gradually becomes clogged, causing the drag coefficient to increase exponentially, resulting in a sharp increase in system energy consumption and severely shortening the endurance of the unmanned vessel; on the other hand, in high-velocity or turbulent environments, forcibly operating against the current can easily lead to hull instability, and rigid interception devices are difficult to adapt to the instantaneous impact of the water flow, often causing floating debris to escape from the edge of the interception device.

[0005] Furthermore, in multi-vessel collaborative operations, existing formation control systems mostly employ rigid holding strategies, requiring each vessel to maintain a strictly fixed geometric spacing at all times. When water flow velocity is unevenly distributed in time and space, this "hard-wired" control logic forces each vessel's propellers to frequently adjust acceleration and deceleration to counteract local water flow disturbances. This not only wastes propulsion system energy but also easily induces internal oscillations, making it difficult to maintain steady-state operation of the swarm while ensuring interception effectiveness. Therefore, how to utilize water flow energy rather than counteracting it to achieve low-energy consumption and high-capture-rate swarm collaborative cleanup is a pressing technical challenge in this field. Summary of the Invention

[0006] The present invention aims to solve the technical problems of high energy consumption in path planning, low capture efficiency, and difficulty in dealing with the dynamic drift characteristics of floating objects caused by the excessive reliance on geometric position tracking when existing cleanup vessel clusters operate in complex and dynamic waters.

[0007] The first aspect of this invention provides a collaborative operation control method for a swarm of clean-up vessels in complex waters. This method transforms the control objective of the clean-up vessel swarm from traditional geometric position tracking to flow flux interception, and optimizes the micro-operational behavior by incorporating hydrodynamic characteristics. The method mainly includes the following steps:

[0008] Step 1: Construct the drift flux density field of the target water area: Obtain flow field environmental data and floating object distribution data of the target water area, calculate the transport flux distribution of floating objects moving with the water flow, and construct the drift flux density field. Step 2: Planning the optimal three moss oil and cluster interception: Based on the above drift flux density field, search for the geometric section with the largest cumulative flux benefit within the detection time window as the optimal interception section, and generate the operation anchor point of the cleanup vessel cluster according to the characteristics of the section. Step 3: Execute fluid-structure interaction differential control: Control the debris removal vessel cluster to move to the operation anchor point, and based on the inertial difference between the target floating object and the water flow, establish a fluid-structure interaction differential capture mechanism by applying an induced slip velocity to carry out cooperative interception operations.

[0009] Furthermore, when constructing the drift flux density field of the target water area, a two-dimensional grid map of the water area is first established, and the position x of each grid cell at time [time value missing] is obtained. Real-time flow velocity vector and the areal density of floating objects Subsequently, the drift flux density vector, representing the amount of floating matter passing through a unit cross-sectional width per unit time, is obtained through dot product calculation. Its calculation expression is:

[0010] ; Based on the convection-diffusion evolution mechanism, the system uses the drift flux density vector at the current moment to predict the spatiotemporal evolution distribution of floating debris flux within a future time window, thereby identifying high-flux transport paths. Furthermore, when planning the optimal interception section and cluster topology, the system searches for a curved section orthogonal to the velocity direction within the drift flux density field. This ensures that the curve cross section is within the prediction time window. The integral value of flux potential energy is maximized within the area. Optimal interception section. The determination follows the following optimization principles:

[0011] ; in The normal vector of the cross section is defined. Based on the geometric width of the optimal interception cross section and the dispersion of the floating debris distribution, the number of cleanup vessels participating in the operation and the target anchor point position of each cleanup vessel on the cross section are determined, realizing a strategy shift from "chasing garbage" to "intercepting flux".

[0012] Furthermore, in co-operating fluid-structure interaction differential control, this invention utilizes the differences in inertial response of different objects in a fluid. First, a range of characteristic Stokes numbers for the target floating object is set, which characterizes the degree of hysteresis in the object's response to changes in the flow field. When the cleanup vessel is in a high-velocity region, it no longer maintains complete synchronization with the water flow (i.e., zero slip), but instead sets a non-zero induced slip velocity between the cleanup vessel and the local water flow. Induced slip velocity is defined as the desired velocity of the cleanup vessel. With local flow velocity The difference:

[0013] ; By controlling the motion of the cleanup vessel to maintain the induced slip velocity, an induced flow field with a specific pressure gradient is created around the cleanup vessel.

[0014] Optionally, the induced slip velocity is set by setting a specific angle of attack between the induced slip velocity vector and the local flow velocity vector according to the opening direction of the collecting device. This angle of attack causes the fluid streamlines flowing around the cleanup vessel to bend. Because the target floating object has a specific characteristic Stokes number range, its inertial forces prevent it from following the sharply curved streamlines, resulting in a deviation in motion relative to the fluid streamlines (centrifugal detachment), and it is guided into the cleanup vessel's capture area. This mechanism achieves preliminary water-material separation without the need for physical filters.

[0015] Furthermore, in the process of cooperative fluid-structure interaction differential control, an anisotropic virtual spring model is used to maintain the formation of the cleanup vessel cluster. The connection relationship between adjacent cleanup vessels is modeled as a virtual spring-damped system, and the downstream direction and the cross-flow direction perpendicular to the water flow direction are defined. To adapt to the flow field characteristics, different virtual spring stiffness coefficients are set for the downstream direction and the cross-flow direction, respectively. Specifically, a higher stiffness coefficient is set for the cross-flow direction. To maintain the continuity of the interception section and prevent floating debris from leaking out of the gaps in the cleanup vessel; a lower stiffness coefficient is set in the downstream direction. ( This design allows the cleaning vessels to passively fluctuate their position in the downstream direction. This anisotropic design enables the swarm to adapt to water flow pulsations like a flexible screen, reducing energy consumption in combating water flow impacts.

[0016] Optionally, maintaining the formation of the cleanup vessel cluster also includes an energy-optimized obstacle avoidance strategy. When an obstacle is detected, the flow field distortion and streamline separation point caused by the obstacle are calculated, and the cleanup vessels are controlled to maneuver around the obstacle along the natural bifurcation path of the streamline, using water flow thrust to assist in obstacle avoidance, and restoring the formation through the action of the virtual spring damping system after the obstacle avoidance.

[0017] Optionally, the method further includes a decision-making switching step based on the energy flux benefit ratio. The energy flux benefit ratio, associated with the amount of floating debris intercepted per unit of energy consumption, is calculated in real time. When this ratio is higher than a preset threshold, it indicates a significant water flow transport effect, and the fluid-structure interaction differential control is executed for passive interception; when the ratio is lower than the preset threshold, it indicates a weak water flow transport effect, and the method switches to an active search and capture mode.

[0018] A second aspect of the present invention provides an electronic device including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to perform the steps of the method described in the first aspect.

[0019] This invention provides a control method for collaborative operations of cleanup vessels in complex waters. It offers the following advantages:

[0020] 1. This invention abandons the traditional geometric full-coverage search logic and creatively integrates water flow transport characteristics with floating debris distribution data by constructing a drift flux density field from an Eulerian perspective. The optimal interception section calculated by the system effectively locks onto the "transport highway" where floating debris naturally converges in the water. This strategy allows a cluster of debris-collecting vessels to be pre-positioned in high-flux areas with minimal maneuvering costs, continuously delivering dispersed floating debris into the capture zone using water flow, greatly increasing the amount of debris intercepted per unit of energy consumption, and completely solving the problem of ineffective patrolling in low-density waters using traditional methods.

[0021] 2. Addressing the shortcomings of traditional physical filters, such as easy clogging and high drag, this invention utilizes fluid-structure interaction differential control to construct an induced flow field, cleverly leveraging the significant physical difference in Stokes number between the target floating object and the water body to achieve separation. By precisely controlling the induced slip velocity, the high-inertia floating object is forced to detach centrifugally instead of following the curved streamline, and is directly "thrown" into the collection device. This non-contact inertial screening mechanism not only eliminates the exponential drag increase caused by filter clogging but also avoids excessive wear and tear on the propulsion system caused by forced operation against the current, significantly extending the single-operation endurance of the unmanned vessel.

[0022] 3. The anisotropic virtual spring formation model proposed in this invention endows rigid clusters with rheological characteristics of "lateral rigidity and longitudinal flexibility." High lateral stiffness forces adjacent vessels to maintain strict alignment in the direction perpendicular to the water flow, constructing a seamless interception line to prevent missed catches; while low downstream stiffness allows vessels to undergo moderate elastic drift with water flow pulsations in the flow direction. This design is similar to the adaptive deformation of a flexible fishing net in a surge, effectively buffering and absorbing the instantaneous impact energy of the water flow, avoiding frequent high-power adjustments by the propellers to counteract minor flow velocity fluctuations, thereby minimizing system energy consumption while maintaining formation steady state. Attached Figure Description

[0023] Figure 1 This is a flowchart illustrating the overall process of an embodiment of the present invention. Figure 2 This is a schematic diagram of the fluid-structure interaction differential screening principle based on inertial differences of the present invention. Figure 3 This is a schematic diagram of the anisotropic virtual spring formation model of the present invention. Detailed Implementation

[0024] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0025] See attached document Figures 1 to 3 , Figure 1 This is a flowchart of a collaborative operation control method for a group of clean-up vessels in complex waters, according to an embodiment of the present invention. The present invention provides a collaborative operation control method for a group of clean-up vessels based on flow field flux potential energy coupling and inertial differential screening. This method constructs a dynamic environment model through shore-based monitoring, UAV inspection, and USV-integrated sensors, transforming the control objective of the clean-up vessel group from traditional geometric position tracking to flow field flux interception. Furthermore, it optimizes micro-operational behavior by combining fluid dynamic characteristics to solve the problems of low clean-up efficiency and high energy consumption in dynamic waters.

[0026] The method first performs step S100: constructing the drift flux density field of the target water area. This step aims to establish an environmental dynamics model from an Eulerian perspective to quantify the spatiotemporal distribution characteristics of floating objects transported by water currents.

[0027] In practice, the system first delineates the target water area for the operation. The data is then discretized into a two-dimensional Cartesian grid map. The system acquires real-time environmental status data through multi-source information fusion technology. This includes utilizing current meters (such as ADCP) or clear water meters deployed in the water body.

[0028] The Doppler log on the raft acquires the position of each grid cell. At the present moment Flow velocity vector Simultaneously, visual recognition algorithms using drones or high-point cameras are employed to identify floating objects on the water surface, and the recognition results are mapped onto a grid map to generate a surface density scalar field representing the distribution of these floating objects. The areal density represents the mass or quantity of floating matter per unit area.

[0029] Unlike existing technologies that only focus on the static position of floating objects, this embodiment introduces the concept of flux from fluid mechanics. The controller couples the obtained velocity vector with the surface density scalar field to calculate the drift flux density vector, which characterizes the amount of floating object transported per unit cross-sectional width per unit time. The calculation formula is as follows:

[0030] The drift flux density vector Flux is a physical quantity that combines direction and magnitude. Its direction indicates the main transport direction of floating objects, and its magnitude indicates the intensity of floating objects passing through at that location. By constructing a flux vector map of the entire field, the system can intuitively identify the "main transport channels" of floating objects in the water area rather than scattered distribution points.

[0031] Building upon this foundation, to achieve predictive control, the system predicts the flux field over time based on the convection-diffusion evolution mechanism. The motion of floating objects in the water body follows the law of conservation of mass transport, and the system uses numerical integration methods to solve the convection-diffusion equations to predict future time windows. The spatiotemporal evolution of internal floating object density. This evolution process is described as follows:

[0032] In the formula, For Hamiltonian operators, representing spatial gradient operations; Let be the turbulent diffusion coefficient tensor, which characterizes the random diffusion effect of floating objects caused by water turbulence; The source term represents the amount of floating matter continuously input from the upstream boundary.

[0033] By solving the above equations, the system obtains any future time. Predicted density distribution This leads to the updated predicted drift flux density field. This predictive model not only considers the advection transport effect of water flow but also incorporates the diffusion effect, thus accurately predicting the convergence trend of floating debris in complex flow fields (such as bends and confluence zones), providing data support for subsequent search for high-yield interception sections. The system marks the calculated high-flux areas as key operation areas, which typically correspond to the convergence zone or backflow edge of the river.

[0034] See attached document Figure 1 After constructing the drift flux density field, this embodiment further executes step S200: planning the optimal interception section and cluster topology. The core logic of this step is to utilize the transport characteristics of the flow field to find the section with the highest flux potential energy of the floating objects for a "wait-and-see" interception, rather than chasing the dispersed floating objects one by one.

[0035] In this embodiment, the system predicts the time window. Within this framework, a global search is performed on the drift flux density field to determine a spatial geometric curve as the optimal interception section. The selection principle for this cross section is: the total cumulative amount of floating debris (i.e., flux potential energy) passing through this cross section with the water flow will be maximized over a future period. This process transforms the discrete, dynamically changing problem of floating debris distribution into a continuous field energy problem.

[0036] The problem of maximizing search volume.

[0037] Specifically, define arbitrary candidate interception cross-sections A parametric curve in three-dimensional space ,in This parameter represents the arc length of the curve. (Definition) For the curve at position The unit normal vector at the location. To maximize interception efficiency, the cross-section normal vector is usually constrained to be opposite to or at a specific angle to the local velocity vector. The system solves for the optimal interception cross-section based on the following optimization objective function. :

[0038] In the above formula, the inner integral Indicates at a certain instant , passing through the entire cross section The sum of floating flux; outer integral This indicates the cumulative effect within the prediction time domain. This is the drift flux density vector predicted in the previous step. By solving this maximum value problem, the system can automatically locate the core of the river's convergence zone or the outlet of the backflow zone, which are usually the inevitable paths for floating objects to move with the water flow.

[0039] After determining the optimal interception cross section After determining the geometric position, the system generates the topology of the cleanup vessel cluster based on the physical characteristics of the cross-section. The system first calculates the effective coverage width of the optimal interception cross-section. In conjunction with the effective operating width of a single debris removal vessel Calculate the number of cleanup vessels required. Normally, The value is greater than or equal to The smallest integer to ensure seamless coverage of the intercept surface.

[0040] Subsequently, the system will perform optimal interception. Discretize along the arc length direction The system consists of several continuous sub-segments, with the geometric center of each sub-segment set as the target anchor point for the corresponding cleanup vessel. .in These anchor points together constitute the desired topology of the cluster.

[0041] It is worth noting that this topology generation process is adaptive. When the flux field shows that the floating debris distribution is characterized by "high density and narrow bandwidth," the anchor points generated by the system will overlap or be distributed in multiple layers in space, directing the swarm to form a "pocket-shaped" array to enhance local interception capabilities. When the floating debris distribution is characterized by "low density and wide bandwidth," the anchor points will spread out along the cross-section, directing the swarm to form a "line-shaped" or "arc-shaped" array to expand the search area. In this way, the formation of the cleanup vessel swarm can be dynamically reconstructed according to the flux distribution pattern of the floating debris in the flow field.

[0042] See attached document Figure 1 After the cleanup vessel cluster reaches the planned operational anchor point, this embodiment executes step S300: performing fluid-structure interaction differential control. This step constitutes the core microscopic control strategy of this invention, aiming to solve the problems of high resistance and easy clogging of traditional net-type interception under high-speed water flow, and to achieve efficient inertial screening by actively manipulating the local flow field.

[0043] In this embodiment, the control strategy utilizes the differences in inertial response of different objects moving in a fluid. In physics, the "Stokes number (St)" is used to characterize the degree of hysteresis in the response of suspended particles to changes in the flow field. For the water body itself or tiny particles (such as algae), its... The value is extremely small, almost completely following the streamline motion; however, for target floating objects with a certain mass and volume (such as plastic bottles, branches), its When the streamlines are sharply curved, these objects cannot change their direction of motion in time due to inertial forces, resulting in a "centrifugal detachment" phenomenon. This invention utilizes this physical mechanism to artificially create high-curvature streamlines by controlling the movement of the cleanup vessel.

[0044] Specifically, when the cleanup vessel is in a high-velocity operating area, the controller no longer aims for the vessel to remain stationary relative to the water flow (i.e., drifting with the current or completely stationary against it), but instead sets the desired speed for the cleanup vessel. With local flow velocity There exists a non-zero vector difference, defined as the induced slip velocity. In this embodiment, the induced slip velocity The direction and size are carefully designed to create an induced flow field with a specific pressure gradient around the hull, particularly in front of the collection device opening. The system is based on the geometric orientation of the collection device. Set the induced slip speed With local flow velocity A specific angle of attack is formed between them. Driven by the controller, the cleanup vessel maintains this angle of attack, causing the streamlines flowing through the hull to bend in a controlled manner.

[0045] Under the influence of this induced flow field, the equations of motion show that it has a target Stokes number range. The drag force of the fluid acting on the floating object becomes unbalanced with its own inertial force. This causes the floating object's trajectory to change. With the fluid streamlines that are bent This produces a significant separation bias. The direction of this bias is designed to point towards the capture area of ​​the debris removal vessel (such as the collection port or flank guide net). In this way, the target floating debris is "thrown" into the collection device, while most of the water flow flows around the hull along the streamline, thus achieving preliminary water-debris separation without the need for physical filters, significantly reducing the inlet resistance and energy consumption of the collection device.

[0046] Furthermore, to enhance this screening effect, the cleanup vessel in this embodiment can also perform periodic dynamic maneuvers. The controller sends commands to the propulsion system, causing the cleanup vessel to perform slight lateral oscillations (Heave) or speed surges (Surge). This periodic motion induces a Kármán vortex street or similar shear vortex structure on the side and rear of the hull. Taking advantage of the fact that large-mass floating objects are easily captured by the low-pressure core of the vortex or ejected by the shear layer, floating objects outside the boundary layer are further drawn into the capture trajectory.

[0047] Ultimately, the bottom motion controller of the debris removal vessel adopts a closed-loop feedback mechanism to measure the actual sliding speed and the desired induced sliding speed in real time. The deviation is addressed by adjusting the thrust of the propeller and the rudder angle to eliminate it, ensuring the fluid-structure interaction differential screening mechanism remains effective in dynamically changing water flow environments. This proactive control method, utilizing hydrodynamic characteristics, essentially transforms the cleanup vessel from a passive interceptor into an active flow field regulator.

[0048] Furthermore, while performing the fluid-structure interaction differential control described in step S300, in order to maintain the cooperative stability of the cluster in dynamic water flow and optimize operational energy consumption, this embodiment also simultaneously executes anisotropic flexible cooperative formation and energy optimization decision-making strategies. This strategy aims to solve the problem that rigid formations are difficult to maintain and have excessive energy consumption in complex flow fields, while giving the system adaptive decision-making capabilities under different flow conditions.

[0049] In this embodiment, a flexible formation control method based on an anisotropic virtual spring model is adopted to control the cluster formation. The system will control adjacent cleanup vessels. and The connection between them is modeled as a virtual spring-damped system, but it is given completely different mechanical properties in different directions.

[0050] First, define a local flow field coordinate system, which includes a unit vector along the direction of water flow. (In the direction of flow) and unit vector perpendicular to the direction of water flow (Crossflow direction).

[0051] In this coordinate system, the system calculates the position error vector between adjacent vessels. ,in The actual position of the ship. The desired spacing is then defined. Subsequently, a cooperative potential function incorporating anisotropic stiffness is constructed.

[0052] In the formula, and These are the virtual spring stiffness coefficients for the downstream and cross-flow directions, respectively. The key technical feature of this embodiment lies in setting... This parameter configuration endows the cluster with rheological characteristics of "lateral stiffness and longitudinal flexibility": high lateral stiffness. Forced debris removal vessels maintain strict formation alignment perpendicular to the water flow to ensure a seamless interception section and prevent floating debris from escaping; while low downstream stiffness It allows vessels to drift moderately forward and backward in the direction of the current. This flexible design is similar to a fishing net spread out in the water, which can absorb the impact energy of the water flow through deformation, thereby avoiding the need for the propeller to make frequent high-power adjustments to counteract instantaneous fluctuations in flow velocity, and significantly reducing the steady-state energy consumption of the swarm to maintain formation.

[0053] While maintaining formation, if a channel obstacle is detected, the system executes an energy-optimized obstacle avoidance strategy. Unlike traditional geometric obstacle avoidance, this system calculates the flow field distortion around the obstacle and identifies the streamline separation point. The controller guides the cleanup vessels to maneuver along the natural bifurcation path of the streamline, utilizing the lateral component of the water flow to bypass the obstacle, rather than relying entirely on propeller power for forced turning. After passing the obstacle, the vessels automatically return to their original formation position under the pull of virtual spring potential energy, achieving hydrodynamic optimization of the obstacle avoidance process. Furthermore, this embodiment also introduces a dynamic decision-making mechanism based on the Energy-Flux Efficiency Ratio (EFER). The system monitors the energy consumption level of the cluster in real time. and intercepted floating object flux Calculate the current performance indicators When detected Higher than the preset threshold When the current flow transport effect is significant and the passive interception strategy is highly efficient, the system maintains the aforementioned fluid-structure interaction differential interception mode, utilizing water flow energy to transport floating objects. Conversely, when... Below the threshold (In still or stagnant water areas) this indicates that relying on water flow for transport is insufficient to maintain operational efficiency. The system automatically switches its control logic and enters an active search and fishing mode. At this point, the cluster releases its anchor point lock and uses a coverage path planning algorithm to actively search for scattered floating objects.

Claims

1. A method for collaborative operation control of a cluster of cleanup vessels in complex waters, characterized in that, Includes the following steps: Step 1: Construct the drift flux density field of the target water area: Obtain flow field environmental data and floating object distribution data of the target water area, calculate the transport flux distribution of floating objects moving with the water flow, and construct the drift flux density field. Step 2: Planning the optimal interception section and cluster topology: Based on the drift flux density field, search for the geometric section with the largest cumulative flux benefit within the prediction time window as the optimal interception section, and generate the operation anchor point of the cleanup vessel cluster according to the characteristics of the section. Step 3: Execute fluid-structure interaction differential control: Control the debris removal vessel cluster to move to the operation anchor point, and based on the inertial difference between the target floating object and the water flow, establish a fluid-structure interaction differential capture mechanism by applying an induced slip velocity to carry out cooperative interception operations.

2. The method for collaborative operation control of a cluster of cleanup vessels in complex waters according to claim 1, characterized in that, The construction of the drift flux density field of the target water area specifically includes: Establish a two-dimensional grid map of the water area and obtain the real-time flow velocity vector and the areal density of floating objects at each grid location; The areal density of the floating object distribution is multiplied by the real-time flow velocity vector to obtain the drift flux density vector representing the amount of floating object passing through a unit cross-sectional width per unit time. Based on the convection-diffusion evolution mechanism, the spatiotemporal evolution distribution of floating object flux within a future time window is predicted using the drift flux density vector at the current moment.

3. The method for collaborative operation control of a cluster of cleanup vessels in complex waters according to claim 1, characterized in that, The planning of the optimal interception cross-section and cluster topology specifically includes: In the drift flux density field, a curve cross section orthogonal to the flow velocity direction is searched, such that the flux potential energy integral value of the curve cross section is maximized within the prediction time window, and the curve cross section is determined as the optimal interception section. Based on the geometric width of the optimal interception section and the dispersion of the floating debris distribution, the number of cleanup vessels participating in the operation and the target anchor point position of each cleanup vessel on the section are determined.

4. The method for collaborative operation control of a cluster of cleanup vessels in complex waters according to claim 1, characterized in that, The execution of fluid-structure interaction differential control specifically includes: Define a range of characteristic Stokes numbers for the target floating object, wherein the Stokes number characterizes the degree of hysteresis in the object's response to changes in the flow field; When the cleanup vessel is in a high-velocity area, it no longer maintains complete synchronization with the water flow, but instead sets a non-zero induced slip velocity between the cleanup vessel and the local water flow. By controlling the motion of the cleanup vessel to maintain the induced slip velocity, an induced flow field with a specific pressure gradient is created around the cleanup vessel.

5. The method for collaborative operation control of a cluster of cleanup vessels in complex waters according to claim 4, characterized in that, The method for setting the induced slip velocity is as follows: Based on the opening direction of the collection device, a specific angle of attack is set between the induced slip velocity vector and the local flow velocity vector; The angle of attack is used to bend the fluid streamlines flowing around the cleanup vessel, causing target floating objects with the characteristic Stokes number range to be unable to follow the bend of the streamlines due to inertial forces, thereby generating a motion deviation relative to the fluid streamlines and entering the capture area of ​​the cleanup vessel.

6. The method for collaborative operation control of a group of clean-up vessels in complex waters according to claim 1, characterized in that, The process of performing fluid-structure interaction differential control also includes using an anisotropic virtual spring model to maintain the formation of the cleanup vessel group: The connection relationship between adjacent cleanup vessels is modeled as a virtual spring-damped system; Define the downstream direction along the direction of water flow and the cross direction perpendicular to the direction of water flow; Different virtual spring stiffness coefficients are set for the downstream direction and the crossflow direction, respectively.

7. The method for collaborative operation control of a cluster of clean-up vessels in complex waters according to claim 6, characterized in that, The specific details of setting different virtual spring stiffness coefficients are as follows: A high stiffness coefficient is set in the crossflow direction to maintain the continuity of the interception section and prevent floating debris from leaking out of the gaps in the cleanup vessel. Setting a lower stiffness coefficient in the downstream direction allows the cleanup vessel to generate passive positional fluctuations in the downstream direction to adapt to water flow pulsations and reduce energy consumption.

8. The method for collaborative operation control of a group of clean-up vessels in complex waters according to claim 6, characterized in that, Maintaining the formation of the cleanup vessel group also includes energy-optimized obstacle avoidance strategies: When an obstacle is detected, the flow field distortion and streamline separation point caused by the obstacle are calculated. The cleanup vessel is controlled to maneuver around the natural bifurcation path of the flow line, using the thrust of the water flow to assist in obstacle avoidance, and after the detour, the formation is restored through the action of the virtual spring damping system.

9. A method for collaborative operation control of a cluster of cleanup vessels in complex waters according to any one of claims 1 to 8, characterized in that, The method also includes a decision-making switching step based on the energy flux-benefit ratio: The current energy flux benefit ratio is calculated in real time, and this ratio is related to the amount of floating debris that can be intercepted per unit of energy consumption; When the energy flux benefit ratio is higher than a preset threshold, the fluid-structure interaction differential control is executed to perform passive interception. When the energy flux benefit ratio is lower than a preset threshold, the system switches to active search and catch mode.

10. The method for collaborative operation control of a cluster of cleanup vessels in complex waters according to claim 4, characterized in that, The control of the movement state of the debris removal vessel also includes: The cleanup vessel is controlled to perform periodic lateral oscillations or speed pulses, using the hull movement to create a vortex shedding zone on the side and rear, and using the vortex suction to assist in screening target floating objects.

Images