Intelligent clean-up and floating debris mother ship based on cooperative control

Abstract

The application relates to an intelligent clean-up mother ship based on cooperative control, which comprises a separable mother ship body and at least one sub ship body, the mother ship body is provided with at least a first data processing module and a first visual auxiliary module, the first visual auxiliary module sends attribute information of a target garbage collected at least two moments to the first data processing module, the first data processing module is configured to extract at least position offset information and environment information of the target garbage based on the attribute information of the target garbage, predict at least one target position where a distance between the sub ship body and the target garbage is less than a salvage distance threshold based on the position offset information, the environment information and first position information of the sub ship body, and set at least one moving path and / or moving speed from a first position to the target position. Through the cooperative control method of the mother ship body and the sub ship body, the efficiency of garbage salvage is improved under the condition of reducing the data processing amount and operation power of the mother ship.

Description

Technical Field

[0001] This invention relates to the field of intelligent technology, and in particular to an intelligent floating debris removal mother and daughter vessel based on collaborative control. Background Technology

[0002] Currently, the main methods for cleaning up floating garbage on the surface of reservoirs, lakes, or seas are manual and mechanical dredging. However, both have drawbacks: manual dredging is labor-intensive, creates a harsh environment, harms the health of workers, and plastic products cannot be compressed, taking up space on the vessel; mechanical dredging is convenient and efficient, but existing vessels are mostly large and cannot adapt to complex waters, limiting their functionality and resulting in high economic costs. Therefore, existing technology proposes a mother-daughter vessel solution. The mother vessel is used for cleaning in open terrain or when garbage is piled up in large quantities, while the daughter vessel is used for cleaning in complex terrain, dangerous terrain, or when garbage is scattered. This collaborative approach between the mother and daughter vessels avoids situations where the mother vessel is unable to clean garbage in complex terrain or consumes excessive power to handle scattered garbage.

[0003] However, when the subsidiary vessel is cleaning up trash in complex environments or cleaning up scattered trash, it needs to obtain the specific location of the scattered trash. Common methods include using image recognition devices mounted on both the mother and subsidiary vessels for location identification, or relying on equipment such as drones to collaboratively identify the trash's location. Regardless of the method, the current approach is to first obtain the trash's location through image recognition, and then move the subsidiary vessel to that location based on the trash's coordinates for collection. However, this is only an ideal technical solution. In practice, it takes time for the subsidiary vessel to move from its current position to the trash's coordinates obtained from the image. During this time, the trash may shift due to water flow and wind direction, causing its new coordinates to deviate from the coordinates obtained during image recognition. When the subsidiary vessel moves to the specific location according to the initially identified coordinates, it may fail to find the trash, or the above steps need to be repeated until the trash collection is complete. Therefore, the existing method of determining the trash location by the mother and subsidiary vessels has shortcomings, including: the time, energy, and computing power spent in the process of collecting trash far exceed the theoretical consumption, severely impacting the cleaning efficiency of the mother and subsidiary vessels.

[0004] For example, patent CN114802618A discloses a system and method for collaboratively handling surface garbage using unmanned aerial vehicles (UAVs) and unmanned vessels. The system includes a mother ship and a garbage collection device, a garbage compactor, an unmanned aerial vehicle (UAV) cabin, and two smaller vessels located on either side of the mother ship. The mother ship includes a garbage storage compartment, a power supply compartment, an engine room, and a bridge. The power supply compartment contains a power source, and the engine room contains a propulsion system and a control system. The garbage collection device, garbage compactor, garbage storage compartment, and engine room are arranged sequentially on the hull from front to back. The garbage collection compartment includes a garbage inlet guide plate. The smaller vessels include a propulsion device and a garbage retrieval device. The propulsion device drives the smaller vessels to move on the water surface, and a detachable garbage compartment is located at the stern of each smaller vessel. The salvage method includes the following steps: 1) When the mother ship sails to the salvage area, the garbage-searching drone on the mother ship is first launched to search for the location of garbage on the water surface; 2) When the garbage-searching drone finds garbage on the water surface, the controller determines the location and amount of garbage to determine whether the mother ship or the sub-ship should go to the designated location for salvage; 3) When there is little garbage on the water surface, a command is sent to the control device of the mother ship to release the sub-ship. The first locking device on the mother ship receives the command to release the locking status between the mother ship and the sub-ship, and then the second locking device receives the command to release it as well. 4) After the child ship separates from the mother ship, it goes to the designated location to retrieve the trash according to the location instructions given by the trash-searching drone. After retrieval, the child ship returns to the mother ship, and the child ship is locked to the mother ship through the first locking device and the second locking device; 5) When there is a lot of trash on the water surface, the mother ship travels to the location given by the trash-searching drone, and the trash-searching drone lands in the drone charging compartment on the mother ship to complete fixation and charging. Then the mother ship carries out large-area retrieval; 6) After retrieval is completed, the mother ship returns to the shore to complete the collection of trash located in the ship's hull. Obviously, in this device, the drone only determines the location of the trash once, ignoring the influence of water flow on the location of the trash. Therefore, the time, energy, and computing power spent in the collection of trash are relatively large, reducing the efficiency of retrieval.

[0005] For example, patent application CN113658250A discloses a method and apparatus for predicting the position of floating objects. The method includes: acquiring multiple frames of images of the floating object at multiple historical moments; determining the center coordinates of the smallest circumcircle and the largest incircle in each frame, where the smallest circumcircle is the circumcircle containing all pixels of the floating object and has the smallest radius, and the largest incircle is the incircle containing all pixels of the floating object and has the largest radius; determining an equivalent pixel coordinate of the floating object based on the center coordinates of the smallest circumcircle and the largest incircle in each frame; identifying the equivalent moving speed and positional offset of the floating object based on the multiple equivalent pixel coordinates at multiple historical moments; and predicting the position of the floating object at future moments based on the equivalent moving speed and positional offset. This patent improves upon existing technology by identifying the equivalent moving speed and positional offset of floating objects from historical images of debris and predicting their future positions. However, this patent application also has shortcomings, including: it requires a large amount of image processing and ignores the influence of natural factors such as wind and water flow on the further displacement of the debris. Therefore, how to reduce the frequency of debris image acquisition, reduce energy consumption and computing power requirements, and accurately determine the future location of the debris for accurate retrieval is the computational problem that this invention aims to solve.

[0006] As mentioned above, all existing technologies have drawbacks. Reducing the number of location identifications decreases the accuracy of garbage location determination and reduces garbage retrieval efficiency; increasing the number of location identifications requires increasing the number of garbage images and the extraction frequency, as well as increasing the hardware requirements for computing power. This invention aims to provide a novel intelligent cleaning mother-daughter vessel model that simultaneously solves multiple problems of existing technologies, achieving the goals of low computing power requirements, small data processing volume, and high retrieval efficiency.

[0007] Furthermore, on the one hand, there are differences in understanding among those skilled in the art; on the other hand, the applicant studied a large number of documents and patents when making this invention, but due to space limitations, not all details and contents were listed in detail. However, this does not mean that the present invention does not possess the features of these prior art. On the contrary, the present invention already possesses all the features of the prior art, and the applicant reserves the right to add relevant prior art to the background art. Summary of the Invention

[0008] Existing mother-daughter salvage vessels have limited computing power, limiting their ability to process large amounts of dynamic or static images. They only perform a single location determination of the target debris, leading to salvage failures when the debris moves with the water flow. Frequent image acquisition and position prediction for the target debris require significant computing power, increasing hardware demands and substantially raising the cost of the mother-daughter vessels. Poor communication between the mother and daughter vessels also hinders continuous operation; if communication is interrupted, the daughter vessels cannot operate continuously. Furthermore, if the daughter vessels encounter obstacles, the mother vessel cannot coordinate their operations in an orderly manner, dividing them into areas or time slots, resulting in a limited range of salvage types. Even with multiple salvage types, the mother vessel cannot scientifically control and allocate these daughter vessels.

[0009] To address the shortcomings of existing technologies, this invention provides an intelligent debris-clearing mother-daughter vessel based on collaborative control, comprising at least a detachable mother vessel body and at least one daughter vessel body. The mother vessel body is equipped with at least a first data processing module and a first vision assistance module. The first vision assistance module sends attribute information of the collected target debris at at least two time points to the first data processing module. The first data processing module is configured to: extract at least the target debris's position offset information and environmental information based on the target debris's attribute information; predict at least one target position where the distance between the daughter vessel body and the target debris is less than a retrieval distance threshold based on the position offset information, environmental information, and the first position information of the daughter vessel body; and set at least one movement path and / or movement speed from the first position to the target position.

[0010] The intelligent debris-collecting mother and daughter vessels of this invention can plan reasonable movement paths and speeds for either the mother vessel or the daughter vessel, enabling them to quickly reach the target location for salvage operations. The rational setting of the movement path and speed is crucial. A longer movement path or a slower speed will delay the vessel's arrival at the target area, thus prolonging its travel time. The longer the travel time, the greater the deviation of the target debris's location, and the higher the likelihood of salvage failure for either the mother vessel or the daughter vessel. Therefore, rationally planning the movement path or speed to shorten the vessel's travel time is an important way to improve salvage efficiency.

[0011] Preferably, the first data processing module is further configured to update the movement path and / or the movement rate based on changes in the second location information and / or environmental information of the target waste.

[0012] When the location of the target waste changes significantly, the ship's path and speed should be updated to avoid situations where the ship and the target waste cannot meet.

[0013] Preferably, the sub-ship body is provided with at least a second data processing module. The second data processing module establishes a communication connection with the first data processing module and performs information exchange. The first data processing module is configured to: in response to the first position information and movement speed of the sub-ship body sent by the second data processing module, calculate the predicted time to reach the target position, and correct the movement path and / or adjust the movement speed of the sub-ship body if the error between the predicted time and the expected time exceeds the error range.

[0014] Existing technologies generally only predict the movement location of the target debris, ignoring the prediction time of the vessel. Therefore, once the vessel reaches the target location, debris retrieval is often affected by tides or natural light, reducing the vessel's retrieval efficiency. This invention reduces the impact of adverse factors by predicting the movement time and adjusting the movement path or speed.

[0015] Preferably, the sub-boat body is further provided with a second vision assistance module connected to the second data processing module. The second vision assistance module identifies the target garbage in a fan-shaped range and sends the attribute information of the target garbage within the salvage range to the second data processing module. The second data processing module selects a salvage angle based on the attribute information, and the sub-boat body adjusts its hull based on the salvage angle to salvage the target garbage.

[0016] The second vision-assisted module can reconfirm the attributes of the target waste and select the appropriate angle for retrieval to improve retrieval efficiency.

[0017] Preferably, the first data processing module is further configured to: select a sub-boat body that matches the size, shape, relative orientation, and / or material of the target waste based on the attribute information of the target waste. There are many types of waste, and therefore many corresponding salvage sub-boat bodies. Therefore, selecting the appropriate type of sub-boat body can improve salvage efficiency.

[0018] Preferably, the first data processing module is further configured to: determine the predicted arrival time of each of the sub-vessels at the target location in response to the first position information and / or movement speed information of each of the sub-vessels sent by the second data processing module; sort the predicted times; and select at least one sub-vessel with the earliest predicted time to retrieve the target debris. Selecting the sub-vessel that arrives at the target location earlier to perform the task can minimize the positional offset distance of the target debris.

[0019] Preferably, the first data processing module is further configured to: in response to abnormal attribute information of the target waste sent by the second data processing module, add and / or reselect the sub-boat body to salvage the target waste. When the sub-boat body detects abnormal attributes, timely replacement or addition of the sub-boat body can help salvage the waste and prevent it from being abandoned in lakes or seas due to salvage failure.

[0020] Preferably, when the first data processing module locks the garbage distribution area, it divides the garbage distribution area into several sub-distribution areas and allocates the sub-vessel bodies according to the location information of the sub-distribution areas. In response to the attribute anomaly information of the target garbage sent by the second data processing module, the first data processing module adjusts the salvage sequence, movement path, and / or movement speed of the sub-vessel bodies based on the attribute anomaly information. Currently, the multiple sub-vessel bodies are operating out of order, which means that when the types of garbage to be salvaged are different, one sub-vessel body needs to finish salvaging that area before another sub-vessel body continues its work, extending the operation time. This invention, through a reasonable arrangement of the salvage sequence and types of sub-vessel bodies, allows two types of sub-vessel bodies to operate in the same area in a staggered manner, reducing the overall salvage operation time.

[0021] Preferably, the mother ship body is further provided with at least one lifting mechanism for transporting the daughter ship body to the top of the mother ship body. The lifting mechanism includes at least a transmission plate driven by a synchronous belt. When the toothed groove at the bottom of the daughter ship body engages with the synchronous belt, the lifting mechanism lifts the daughter ship body. The lifting mechanism of the present invention can further ensure the stable entry or stable movement of the daughter ship body into the water.

[0022] Preferably, when the number of failed salvage attempts on the sub-ship exceeds a preset threshold, the second data processing module determines that the attributes are abnormal and sends the attribute abnormality information of the target waste to the first data processing module. The attribute abnormality information includes at least one or more of the following: abnormal number of failures, abnormal size, abnormal shape, abnormal material, and abnormal structure. This invention's setting of attribute abnormality information enables the first data processing module to promptly detect abnormal salvage phenomena. Attached Figure Description

[0023] Figure 1 This is a simplified schematic diagram of the modular connection relationship of an intelligent floating debris cleaning mother and daughter vessel according to a preferred embodiment of the present invention;

[0024] Figure 2 This is a structural schematic diagram of an intelligent cleanup mother and daughter vessel according to a preferred embodiment of the present invention from one angle.

[0025] Figure 3 This is a structural schematic diagram of the intelligent clean-up mother and daughter vessel according to a preferred embodiment of the present invention from another angle;

[0026] Figure 4 This is a simplified schematic diagram of the lifting mechanism for the mother ship hull of the present invention;

[0027] Figure 5 This is a simplified structural diagram of the compression component of the present invention.

[0028] List of reference numerals

[0029] 100: Mother ship body; 200: Daughter ship body; 300: First data processing module; 110: First vision aid module; 120: Waste storage bin; 130: Lifting mechanism; 210: Second vision aid module; 310: Second data processing module; 400: Compression assembly; 410: Compression plate; 420: First link; 430: Second link; 440: Steering gear. Detailed Implementation

[0030] The following is a detailed explanation with reference to the accompanying drawings.

[0031] Existing mother-daughter salvage vessels have limited computing power, limiting their ability to process large amounts of dynamic or static images. They only perform a single location determination of the target debris, leading to salvage failures when the debris moves with the water flow. Frequent image acquisition and position prediction for the target debris require significant computing power, increasing hardware demands and substantially raising the cost of the mother-daughter vessels. Poor communication between the mother and daughter vessels also hinders continuous operation; if communication is interrupted, the daughter vessels cannot operate continuously. Furthermore, if the daughter vessels encounter obstacles, the mother vessel cannot coordinate their operations in an orderly manner, dividing them into areas or time slots, resulting in a limited range of salvage types. Even with multiple salvage types, the mother vessel cannot scientifically control and allocate these daughter vessels.

[0032] The attribute information of this invention includes at least the size, shape, relative orientation, and / or material of the target waste. Existing technologies only predict the location of the target waste, ignoring the influence of size, shape, relative orientation to the mother ship, and material, resulting in the allocation of unsuitable sub-ships for salvage and subsequent salvage failure. Therefore, this invention uses attribute information as basic data to allocate sub-ships, thereby improving salvage efficiency.

[0033] The first location information refers to the location information of the sub-ship itself, where the coordinate data represents the first location. The second location information refers to the location information of the target debris, where the coordinate data represents the second location. The target location refers to the location that is advantageous for salvaging the target debris, calculated and determined by the first data processing module of the mother ship.

[0034] To address the shortcomings of existing technologies, this invention provides an intelligent floating debris removal mother-daughter vessel based on collaborative control, such as... Figure 1 As shown, it includes at least a separable mother ship body 100 and at least one daughter ship body 200.

[0035] The mother ship body 100 and the daughter ship body 200 of this invention both include a hull, a power assembly, a power source storage assembly, a waste storage bin, a compression assembly, a positioning assembly, a salvage assembly, etc. Since these conventional mechanical structures are not modified, they will not be described in detail here. This invention focuses on describing the cooperative method between the mother ship and the daughter ship.

[0036] The first data processing module 300 in this invention may be a processor, server, or dedicated integrated chip capable of executing the mother ship to daughter ship cooperative control method in this invention.

[0037] The second data processing module 310 in this invention may be a processor, server, or dedicated integrated chip capable of executing the sub-ship to mother ship cooperative control method in this invention.

[0038] Both the first visual aid module 110 and the second visual aid module 210 may include a spherical camera acquisition device, an image processing device, and a display device for acquiring image information of the waste. The first visual aid module 110 identifies waste within a circumferential range. The first visual aid module 110 is at least capable of rising and rotating circumferentially. The second visual aid module 210 identifies waste within a fan-shaped range.

[0039] The first visual aid module 110 and the second visual aid module 210 are based on the YOLOv5-lite neural network model to detect debris on the water surface. The aim is to improve the detection accuracy and real-time performance. Nearly 10,000 images of debris on the water surface are collected and organized through techniques such as data augmentation, adjusting anchor boxes, model quantization, and network structure optimization, including real-world shooting, collecting publicly available datasets, and using Python web crawlers. Operations such as random cropping, rotation, flipping, and color enhancement are performed to expand the training dataset and improve the robustness and generalization ability of the model.

[0040] The mother ship hull 100 has at least one waste storage compartment 120. The daughter ship hull 200 has at least one temporary waste compartment. The daughter ship hull 200 stores the collected waste in the waste storage compartment 120. The waste storage compartment 120 is equipped with at least one compression component 400 to compress the waste within it.

[0041] like Figure 5The diagram shows a simplified structural schematic of the compression assembly 400. The rear panel of the collection compartment reciprocates to compress the waste. Compression plates 410 are connected side-by-side via a first connecting rod 420. Through the application of a crank-slider mechanism, a second connecting rod 430 connects a servo motor 440 to one of the compression plates 410. By using a transfer function, the 120° rotational motion of the servo motor 440 is converted into the reciprocating motion of the slider, reducing the volume of waste and efficiently utilizing space.

[0042] The first data processing module 300 divides the image acquired by the first vision assistance module 110 into several grid cells. The grid cells are categorized into two states: with garbage and without garbage. The size of each grid cell is limited by the size of the sub-ship body 200.

[0043] Preferably, the mother ship body 100 is further provided with at least one lifting mechanism 130 for transporting the daughter ship body 200 to the top of the mother ship body 100. The lifting mechanism 130 includes at least a transmission plate driven by a timing belt. When the toothed groove at the bottom of the daughter ship body 200 engages with the timing belt, the lifting mechanism 130 lifts the daughter ship body 200 to the top of the mother ship body 100.

[0044] like Figure 4 As shown, when the sub-ship 200 returns to port, it faces the synchronous belt and pulleys installed on the mother ship 100. A 28-stepper motor drives the synchronous pulleys clockwise, positioning the tight side of the synchronous belt at the top. With a fixed distance between the synchronous pulleys, the angle is adjusted to determine the preload, improving the feasibility of the sub-ship's return. Simultaneously, a second vision assist module 210 installed on the sub-ship 200 assists in aligning the sub-ship 200 with the conveyor belt of the mother ship 100.

[0045] After the sub-ship body 200 reaches the top of the mother ship body 100, the second data processing module 310 sends a control command to the rudder motor 440 of the sub-ship body to control the bottom plate of the sub-ship body 200 to rotate, so that the garbage falls into the collection compartment of the mother ship body 100, thereby enabling the garbage to be centrally processed.

[0046] Example 1

[0047] The mother ship body 100 is equipped with at least a first data processing module 300 and a first vision assistance module 110.

[0048] The first vision assistance module 110 sends the attribute information of the collected target waste at least at two time points to the first data processing module 300.

[0049] The first data processing module 300 is configured to: extract at least the location offset information and environmental information of the target waste based on the attribute information of the target waste; predict at least one target location where the distance between the sub-ship body 200 and the target waste is less than the salvage distance threshold based on the location offset information, environmental information and the first location information of the sub-ship body 200; and set at least one movement path and / or movement speed from the first location to the target location.

[0050] Environmental information includes at least wind direction, wind speed, water flow speed, direction of sunlight, tidal patterns, and sunrise and sunset times, and may also include other necessary environmental information.

[0051] The sub-vessel 200 has a normal moving speed. Predicted time refers to the moment when the sub-vessel 200 moves to the target position at its normal moving speed and arrives. Expected time refers to the moment or time range when the operation is most effective. Since the accuracy of debris identification by the visual assistance module is related to factors such as the light intensity, water flow speed, and direction of light, selecting the time with the best operational performance based on environmental information as the expected time, and identifying debris and determining its attributes under good lighting conditions with high image clarity, helps improve the accuracy of debris identification and the efficiency of the salvage operation.

[0052] Preferably, the first data processing module 300 uses an automatic tracking algorithm to determine the optimal movement path based on the target location.

[0053] The automatic pathfinding algorithm is actually the A_star algorithm, and its heuristic function is defined as:

[0054] heuristic(node)=|node.x-goal.x|+|node.y-goal.y|

[0055] Where node.x and node.y are the coordinates of the current node, and goal.x and goal.y are the coordinates of the target node.

[0056] However, this heuristic function does not consider the effects of water flow and wind direction. These factors have a significant impact on pathfinding, therefore a more accurate heuristic function is needed. This heuristic function can be optimized by considering factors such as the distance from the current node to the target node, and the angle between the direction from the current node to the target node and the water flow and wind direction. This new heuristic function is defined as follows:

[0057] heuristic(node)=(|node.x-goal.x|+|node.y-goal.y|)*factor

[0058] Here, `factor` is a factor that takes into account water flow and wind direction; its value can be greater than 1 (if the water flow and wind direction are opposite to the target direction) or less than 1 (if the water flow and wind direction are the same as the target direction). The optimized heuristic function will have higher accuracy.

[0059] Preferably, the first data processing module 300 is further configured to update the movement path and / or movement rate based on changes in the second location information and / or environmental information of the target waste.

[0060] The first visual aid module 110 on the mother ship hull 100 scans the surrounding water surface area at a first moment to obtain the first position of a certain piece of debris. The first visual aid module 110 on the mother ship hull 100 scans the surrounding water surface area at a second moment to obtain the second position of the same piece of debris. Preferably, the first data processing module 300 calculates the equivalent offset rate of the debris moving from the first position to the second position based on the time interval between the first and second moments.

[0061] The equivalent displacement rate is essentially the rate at which the direction of water flow and wind at the water surface causes the debris to float. The equivalent displacement rate of the debris can be approximated as the environmental dynamic rate primarily caused by water flow and wind direction in the current time period. Environmental dynamic rate refers to the rate at which an object without any additional propulsion will drift along a certain path due to the influence of wind and water flow on its surface; the magnitude of its speed and the direction of the wind are determined by the environmental dynamic rate. Furthermore, even when an object is placed on the water surface with some additional propulsion, its actual speed is determined by both the additional propulsion rate and the environmental dynamic rate.

[0062] Preferably, the sub-vessel body 200 is equipped with at least a second data processing module 310, which establishes a communication connection with the first data processing module 300 and exchanges information. For example, real-time communication is achieved through a Bluetooth module. A broadcast communication protocol is adopted, whereby the mother ship issues a water area cleanup task, the sub-vessel identifies and executes it, and simultaneously feeds back information to the mother ship, such as whether to return, the type of garbage to be collected, and the water area conditions, to achieve unified decision-making and efficient garbage cleanup.

[0063] Because the sub-vessel needs to operate remotely from the mother ship, its operational efficiency is affected by the environment. Insufficient light or changes in weather can decrease its salvage efficiency. Alternatively, the sub-vessel may encounter rain or strong winds after only a short time at its destination, making further operation impossible, and returning is unsafe. Therefore, to improve the sub-vessel's salvage efficiency, changes in environmental information need to be considered, and the operating time needs to be adjusted to provide an expected time. If the operating time exceeds the expected time, the second data processing module 310 directly controls the sub-vessel to stop operating and return, thus ensuring its safety and reducing its damage rate.

[0064] The first data processing module 300 obtains the current first position information of a certain sub-vessel body 200 from the second data processing module 310. Based on the straight-line distance between the first position information and the second position information of the debris, and the normal speed of the sub-vessel body 200, it calculates the predicted time required for the sub-vessel body 200 to move to the second position or target position at the normal moving speed. The predicted time is adjusted to the desired time based on environmental information, enabling the sub-vessel body 200 to operate under good ambient light conditions. For example, in the case of approaching sunset, the predicted time needs to be advanced to the desired time. In the case of approaching rainfall, the predicted time needs to be advanced to the desired time to complete the salvage operation before the rainfall.

[0065] Preferably, in response to the water flow velocity and / or wind speed change information fed back by the second data processing module 310, the first data processing module 300 updates the expected time and sends it to the second data processing module 310. The second data processing module 310 adjusts the sub-boat's operating strategy according to the updated expected time. The operating strategy includes returning to shore and continuing operation. When the continued operation time exceeds the expected time, the second data processing module 310 controls the sub-boat to return to shore. For example, if the water flow velocity increases and the increase is rapid, it may indicate rainfall or flash floods in the upstream waters. Since the sub-boat is a small boat and generally lightweight, changes in the environment can easily lead to operational failures and safety risks.

[0066] Preferably, in response to the actual distribution image of the garbage in the water area fed back by the second data processing module 310, the first data processing module 300 updates the expected time of the assigned sub-boats by increasing the number of sub-boats at the target location. For example, after the first sub-boat arrives at the target location, it is found that the amount of garbage at the target location is constantly increasing due to the influence of water flow, and the predicted operation time will exceed the expected time. In response to this situation reported by the sub-boats, the first data processing module 300 chooses to adjust its own hull to reach the target location, or chooses to add a new sub-boat and instruct it to sail to the target location. With the increase in the number of sub-boats in operation, the operation time is shortened, so the first data processing module 300 updates the expected time of the sub-boats and sends it to the corresponding second data processing module 310. This helps to improve the salvage efficiency of the sub-boats and avoids the sub-boats salvaging garbage in an unplanned manner.

[0067] Preferably, the expected time for the sub-boats is continuously updated based on the information processed by the first data processing module 300. Compared to the sub-boats freely collecting garbage at the target location, the adjustment range of the expected time update is limited. This not only avoids the sub-boats collecting garbage without a plan but also ensures their safe return. Furthermore, it allows the first data processing module 300 to systematically schedule the sub-boats based on the collection status reported by the second data processing modules 310. For example, if the water flow in the second target water area is fast, garbage is prone to drifting significantly. If the sub-boats were allowed to collect garbage freely in their assigned areas, a large amount of garbage in the second target water area would inevitably not be collected.

[0068] When there is not much garbage remaining in the first target water area, the first data processing module 300 can dispatch the first sub-boat in the first target water area to enter the second target water area to assist the second sub-boat in collecting garbage. When the garbage collection in the second target water area is completed, the first sub-boat returns to the first target water area to reposition the target garbage and clean up the remaining garbage.

[0069] The first data processing module 300 can also communicate with the third data processing module of the UAV to transmit images of the patrolled waters. Even if the visual assistance modules on the mother ship and the daughter ship can adjust the image acquisition angle 360 ​​degrees, they are limited by the ship's height and cannot fully understand the garbage disposal situation in the waters. Preferably, in response to the coordinates and movement speed of the target garbage sent by the third data processing module of the UAV, the first data processing module 300 corrects its own or the daughter ship's movement path / movement speed. For example, when the daughter ship is about to complete the salvage task and is preparing to return at the expected time, the UAV scans that garbage is about to drift towards the daughter ship from upstream, but the garbage cannot be captured by the first data visual assistance module 110 at this time. The UAV sends the position change of the garbage image to the first data processing module 300. The first data processing module 300 determines the time when the target garbage will reach the salvage range of the daughter ship and updates the expected time of the daughter ship. The first data processing module 300 sends the new expected time and a stop return command to the second data processing module 310. The second data processing module 310 receives the new expected time and drives the second vision assistance module 210 to acquire images of the target waste within the expected time range. After the second data processing module 310 locks onto the target waste based on the image sent by the second vision assistance module 210, the second data processing module 310 drives the sub-boat to continue the waste retrieval operation.

[0070] Preferably, the first data processing module 300 determines the actual speed of the sub-ship 200 based on the distance of the optimal path and the expected time. The actual speed may be a conventional transfer speed or it may not be a conventional transfer speed. This setting can minimize the risk of the sub-ship 200 performing salvage operations in unfavorable environments, avoid unclear identification of the properties of the debris, or reduce the frequency of salvage failures.

[0071] The first data processing module 300 is configured to: in response to the first position information and movement speed of the sub-ship body 200 sent by the second data processing module 310, calculate the predicted time to reach the target position, and correct the movement path and / or adjust the movement speed of the sub-ship body 200 if the error between the predicted time and the expected time exceeds the error range.

[0072] For example, when the predicted time differs too much from the expected time, and salvage operations are possible, the first data processing module 300 corrects the movement path and / or adjusts the movement speed of the sub-vessel body 200 so that the sub-vessel body 200 reaches the target position at the expected time.

[0073] The first data processing module 300 on the mother ship 100 continuously records the working status of the daughter ship 200. The working status of the daughter ship 200 is at least categorized as: busy, idle, and returning. "Busy" means that the daughter ship 200 is currently performing a garbage collection task; it may be on its way to garbage collection or currently collecting garbage. Once garbage collection is complete, the second data processing module 310 of the daughter ship 200 immediately transmits the task completion information to the first data processing module 300 of the mother ship 100. The first data processing module 300 then switches the daughter ship 200 from the busy state to the idle state. The second data processing module 310 within the daughter ship 200 can at least detect the amount of garbage in the daughter ship 200's temporary storage compartment. When the amount of garbage reaches the storage capacity of the temporary storage compartment, the second data processing module 310 requests a return trip. Upon receiving this information, the first data processing module 300 of the mother ship 100 plans the return route and changes its status to the return trip state. After returning to port, the sub-ship 200 places the waste into the waste storage bin of the mother ship 100. At this time, the sub-ship 200 changes from the returning state to the idle state.

[0074] Based on the above scheme, when the data processing module performs task assignment and path calculation, it usually assigns the task to the idle sub-ship body 200 that is closest to the garbage to be collected. That is, it selects one of the idle sub-ship bodies 200 that is closest to the garbage to be collected, calculates the straight-line distance between it and the second location of the garbage to be collected, and calculates the predicted time and expected time.

[0075] Preferably, the sub-boat body 200 is further provided with a second vision assistance module 210 connected to the second data processing module 310. The second vision assistance module 210 identifies the target waste in a fan-shaped range and sends the attribute information of the target waste within the salvage range to the second data processing module 310.

[0076] The second data processing module 310 selects the salvage angle based on the attribute information, and the sub-boat body 200 adjusts the hull based on the salvage angle to salvage the target garbage.

[0077] If the first data processing module 300 frequently guides the garbage retrieval operation, not only will the data transmission and processing volume be large, but the time delay caused by data processing may also allow the target garbage to change its position and angle again. Therefore, having the second data processing module 310 select the retrieval angle based on attribute information not only reduces the data transmission and processing volume with the mother ship 100 and reduces data delay, but also enables the daughter ship 200 to operate autonomously, avoiding retrieval failures caused by the rapid movement of the target garbage.

[0078] The second data processing module 310 stops at a position downstream of the target waste based on its movement direction, thus enabling smoother waste retrieval and avoiding frequent relocation.

[0079] Preferably, the first data processing module 300 is further configured to: select a sub-ship body 200 that matches the size, shape, relative orientation and / or material of the target waste based on the attribute information of the target waste.

[0080] Preferably, the first data processing module 300 is further configured to: determine the predicted time for each sub-vessel body 200 to reach the target position in response to the first position information and / or movement speed information of each sub-vessel body 200 sent by the second data processing module 310; sort the predicted times; and select at least one sub-vessel body 200 with the earliest predicted time to retrieve the target waste. When the overall operating environment is good and there are multiple predictable times to choose from, selecting an earlier predicted time enables the sub-vessel body 200 to complete the waste retrieval task.

[0081] Preferably, the first data processing module 300 is further configured to: in response to abnormal attribute information of the target waste sent by the second data processing module 310, add and / or reselect the sub-boat body 200 to salvage the target waste. For waste that always fails to be salvaged, such as abnormally large or soft waste, if the current sub-boat body 200 cannot salvage it, the first data processing module 300 selects to add or replace the sub-boat body 200 to operate based on the reason for the salvage failure.

[0082] Preferably, when the first data processing module 300 locks the garbage distribution area, the first data processing module 300 divides the garbage distribution area into several sub-distribution areas and allocates the sub-ship body 200 according to the location information of the sub-distribution areas.

[0083] Preferably, it is assumed that a sub-ship 200 is responsible for all the trash in an area. Regardless of whether the sub-ship 200 is busy or idle, it will be assigned to collect the trash in that area. Therefore, some modifications are needed in route planning. Assume trash A, currently at position Y1, is collected at the first target position X1. Trash B, currently at position Y2, is collected at the second target position X2. Sub-ship 200, currently at position Y3.

[0084] Assuming the data processing module plans the time for the sub-ship 200 to travel from its current position to the first target position X1 as T1, and the time for collecting garbage A as T2, then when the sub-ship 200 travels from the first target position X1, garbage B will have reached the intermediate position X3 after a time of T1+T2. At this point, the predicted time T3 for traveling to garbage B is calculated based on the first target position X1 and the intermediate position X3. When calculating the second target position X2, the time is calculated according to T1+T2+T3, which also yields the target position of garbage B.

[0085] The first data processing module 300 sends the optimal movement path and actual speed to the second data processing module 310 of the sub-ship body 200. The second data processing module 310 controls the sub-ship body 200 to move to the target position according to the optimal movement path and actual speed. After reaching the target position, the second data processing module 310, in conjunction with the second vision assistance module 210 in the vision assistance module, collects the garbage.

[0086] After the second data processing module 310 of the sub-ship body 200 receives the path and actual speed sent by the first data processing module 300, it drives the sub-ship body 200 to move. The magnitude of its driving power is related to the actual speed and the determined environmental dynamic speed. If the environmental dynamic speed and the path are in the same direction, the driving power-generated speed plus the environmental dynamic speed equals the actual speed. If the environmental dynamic speed and the path are in opposite directions, the driving power-generated speed should equal the environmental dynamic speed plus the actual speed.

[0087] In response to the attribute anomaly information of the target waste sent by the second data processing module 310, the first data processing module 300 adjusts the salvage sequence, movement path and / or movement speed of the sub-ship body 200 based on the attribute anomaly information.

[0088] For example, when the first sub-vessel 200 is retrievaling the first batch of garbage, the second data processing module 310 feeds back the attribute information and retrieval image information of the garbage that cannot be retrieved to the first data processing module 300, and marks the second location information of these target garbage. Based on the attribute anomaly information, the first data processing module 300 selects the second sub-vessel 200 to perform the retrieval, and schedules the second sub-vessel 200 to operate in a staggered order with the first sub-vessel 200. For example, the second sub-vessel 200's operation time is half an hour later than the first sub-vessel 200's. The operation area of ​​the second sub-vessel 200 is 5 kilometers away from the operation area of ​​the first sub-vessel 200. This allows for coordinated operation of two sub-vessels 200 with different retrieval types, eliminating the need to schedule the second sub-vessel 200's operation after the first sub-vessel 200 has finished its work.

[0089] Preferably, when the number of salvage failures of the sub-ship body 200 exceeds a preset threshold, the second data processing module 310 determines that the attributes are abnormal and sends the attribute abnormality information of the target waste to the first data processing module 300. The attribute abnormality information includes at least one or more of the following: abnormal number of failures, abnormal size, abnormal shape, abnormal material, and abnormal structure.

[0090] The mother-daughter ship of this invention can collect large amounts of garbage in large, open areas quickly and efficiently, while the daughter ship can collect small amounts of garbage scattered in complex terrain. When collecting large amounts of garbage, the mother ship is more efficient and faster, significantly improving the efficiency of garbage collection. When collecting small amounts of garbage, the daughter ship has a smaller load and consumes less energy, which helps to save resources. In complex terrain, the daughter ship is more flexible and can collect garbage more comprehensively, improving the operational effect.

[0091] The invented mother-daughter vessel, when locating garbage through image recognition, fully considers the potential impact of garbage displacement caused by water and wind direction. Based on the data processing module, it analyzes and predicts the displacement distance of the garbage to determine the target position of the garbage within the expected time. Then, it calculates the optimal path from the daughter vessel to the target position, which significantly improves the efficiency of the daughter vessel moving to the target position and collecting garbage, reduces energy consumption, and avoids the need to repeatedly obtain the target position of garbage and repeatedly calculate the optimal path, saving computing resources and significantly improving the endurance of the cleaning mother-daughter vessel.

[0092] It should be noted that the specific embodiments described above are exemplary. Those skilled in the art can devise various solutions inspired by the disclosure of this invention, and these solutions all fall within the scope of this invention and its protection. Those skilled in the art should understand that this specification and its accompanying drawings are illustrative and not intended to limit the scope of the claims. The scope of protection of this invention is defined by the claims and their equivalents. This specification contains multiple inventive concepts; terms such as "preferredly," "according to a preferred embodiment," or "optionally" indicate that the corresponding paragraph discloses an independent concept. The applicant reserves the right to file divisional applications based on each inventive concept.

Claims

1. A smart debris-clearing mother-daughter vessel based on collaborative control, comprising at least a detachable mother vessel body (100) and at least one daughter vessel body (200), characterized in that, The mother ship body (100) is equipped with at least a first data processing module (300) and a first vision assistance module (110). The first visual assistance module (110) sends the attribute information of the collected target waste at at least two times to the first data processing module (300). The first data processing module (300) is configured as follows: Based on the attribute information of the target waste, at least the location offset information and environmental information of the target waste should be extracted. Based on the location offset information, environmental information and the first location information of the sub-ship body (200), predict at least one target location where the distance between the sub-ship body (200) and the target garbage is less than the salvage distance threshold, and set at least one movement path and / or movement speed from the first location to the target location.

2. The intelligent floating debris removal mother-daughter vessel based on collaborative control according to claim 1, characterized in that, The first data processing module (300) is also configured to: The movement path and / or movement rate are updated based on changes in the second location information of the target waste and / or changes in environmental information.

3. The intelligent floating debris removal mother-daughter vessel based on collaborative control according to claim 1 or 2, characterized in that, The sub-ship body (200) is equipped with at least a second data processing module (310), which establishes a communication connection with the first data processing module (300) and performs information exchange. The first data processing module (300) is configured as follows: In response to the first position information and movement rate of the sub-ship body (200) sent by the second data processing module (310), the predicted time to reach the target position is calculated, and if the error between the predicted time and the expected time exceeds the error range, the movement path is corrected and / or the movement rate of the sub-ship body (200) is adjusted.

4. The intelligent floating debris removal mother-daughter vessel based on collaborative control according to claim 3, characterized in that, The sub-ship body (200) is also provided with a second vision auxiliary module (210) connected to the second data processing module (310). The second visual assistance module (210) identifies the target waste in a fan-shaped range and sends the attribute information of the target waste within the retrieval range to the second data processing module (310). The second data processing module (310) selects the retrieval angle based on the attribute information. The sub-boat body (200) adjusts its hull based on the salvage angle and salvages the target garbage.

5. The intelligent floating debris removal mother-daughter vessel based on collaborative control according to claim 1, characterized in that, The first data processing module (300) is also configured to: Based on the attribute information of the target waste, the sub-ship body (200) is selected to match the size, shape, relative orientation and / or material of the target waste.

6. The intelligent floating debris removal mother-daughter vessel based on collaborative control according to claim 3, characterized in that, The first data processing module (300) is also configured to: In response to the first position information and / or movement speed information of each of the sub-ship bodies (200) sent by the second data processing module (310), the predicted time for each of the sub-ship bodies (200) to reach the target position is determined. The predicted times are sorted and at least one of the sub-ship bodies (200) with the earlier predicted time is selected to salvage the target garbage.

7. The intelligent floating debris removal mother-daughter vessel based on collaborative control according to claim 3, characterized in that, The first data processing module (300) is also configured to: In response to the abnormal attribute information of the target waste sent by the second data processing module (310), the sub-ship body (200) is added and / or reselected to salvage the target waste.

8. The intelligent floating debris removal mother-daughter vessel based on collaborative control according to claim 3, characterized in that, When the first data processing module (300) locks the garbage distribution area, The first data processing module (300) divides the garbage distribution area into several sub-distribution areas and allocates the sub-ship body (200) according to the location information of the sub-distribution areas. In response to the abnormal attribute information of the target waste sent by the second data processing module (310), the first data processing module (300) adjusts the salvage sequence, movement path and / or movement speed of the sub-ship body (200) based on the abnormal attribute information.

9. The intelligent floating debris removal vessel based on collaborative control according to claim 1, characterized in that, The mother ship body (100) is also provided with at least one lifting mechanism (130) for transporting the daughter ship body (200) to the top of the mother ship body (100). The lifting mechanism (130) includes at least a transmission plate driven by a synchronous belt. When the toothed groove at the bottom of the sub-boat body (200) engages with the timing belt, the lifting mechanism (130) lifts the sub-boat body (200).

10. The intelligent floating debris removal mother-daughter vessel based on collaborative control according to claim 3, characterized in that, If the number of failed salvage attempts of the sub-ship body (200) exceeds a preset threshold, the second data processing module (310) determines that the attributes are abnormal and sends the attribute abnormality information of the target waste to the first data processing module (300). The attribute anomaly information includes at least one or more of the following: anomaly in the number of failures, anomaly in size, anomaly in shape, anomaly in material, and anomaly in structure.

Images