Mechanical seabed garbage intelligent cleaning robot and seabed garbage cleaning method

Abstract

The application discloses a mechanical seabed garbage intelligent cleaning robot and a seabed garbage cleaning method.The robot adopts a modular integrated design and comprises a main frame, a buoyancy material, a track chassis, a material cabin, a mechanical arm, an electric control cabin, a conveying belt, an underwater sensing system, a water quality monitoring module and an optoelectrical composite cable.A mechanical grabbing-transporting-storing integrated operation chain is used to realize the garbage cleaning concept of accurate grabbing and minimum disturbance.The mechanical arm has multiple degrees of freedom and a grabbing and shearing integrated function.The track chassis is adapted to complex seabed topography.The double-layer material cabin realizes efficient garbage storage and rapid unloading, and the multi-modal sensing system supports intelligent autonomous operation.The application greatly improves the seabed garbage cleaning efficiency, significantly reduces the disturbance to the seabed ecology and is suitable for large-scale garbage cleaning operation in different water depths and various bottom environments.

Description

Technical Field

[0001] This invention belongs to the field of marine environmental protection equipment technology, specifically relating to a mechanical intelligent underwater debris cleaning robot and an underwater debris cleaning method. Background Technology

[0002] Marine debris pollution has become one of the most serious global environmental challenges of the 21st century. Every year, a large amount of garbage enters the ocean through rivers, coastlines, and maritime activities. This garbage eventually sinks to the seabed and accumulates in nearshore continental shelves, submarine canyons, and areas where ocean currents converge. The composition of seabed debris is extremely complex, mainly including plastic products, metal products, glass products, fishery waste, wood products, and rubber products. Among them, plastic products mainly consist of packaging bags, beverage bottles, food containers, and plastic tableware; fishery waste includes discarded fishing nets, ropes, buoys, and fishing cages. This debris is difficult to degrade and continues to harm marine ecosystems.

[0003] The harm of marine debris to marine ecology is comprehensive and long-term: microplastics and nanoplastics produced by the decomposition of plastic waste accumulate through the food chain and eventually threaten human health. Large debris blocks the breathing and feeding channels of benthic organisms, leading to a decline in biodiversity; discarded fishing nets and ropes entangle important ecosystems such as corals and seagrass beds, causing physical damage. Currently, the mainstream marine debris cleaning technologies are divided into three categories: human diving cleaning technology, hydraulic collection equipment technology, and remotely operated underwater vehicle (ROV) assisted cleaning technology, but all of them have significant drawbacks: (1) Human diving cleaning technology is extremely inefficient. A single professional diver can only clean a small area per day, and the operating depth is limited (within 30 meters), resulting in high safety risks, high economic costs, and difficulty in large-scale operation. (2) Hydraulic collection equipment uses high-pressure water flow to flush up the debris and then collect it. Although it is suitable for large areas of flat seabed, it will stir up a large amount of seabed sediment, forming a high-concentration suspended cloud, covering habitats, blocking the respiratory organs of organisms, causing secondary pollution, resulting in a significant decline in benthic biodiversity in the operating area, and seriously interfering with underwater optical sensing systems. (3) The existing remotely operated underwater vehicle (ROV) robotic arms are not optimized for garbage cleaning, have insufficient degrees of freedom and dexterity, and have low cleaning efficiency; they lack efficient garbage transportation and storage mechanisms, have limited storage capacity, are prone to secondary leakage, and lack an effective internal monitoring system, which affects the efficiency of operation.

[0004] Developing efficient, environmentally friendly, and safe seabed debris cleanup technologies faces multiple challenges: the high pressure, low temperature, low visibility, and strong corrosiveness of the seabed environment require equipment with extremely high reliability; the diverse forms of debris require equipment with strong adaptability; and there is an inherent contradiction between environmental protection and cleanup efficiency. Therefore, the development of a purely mechanical, intelligent seabed debris cleaning robot that achieves "precise grasping and minimal disturbance" is an urgent need to solve the above problems. Summary of the Invention

[0005] The purpose of this invention is to provide a mechanical intelligent underwater debris cleaning robot to solve the problems of low efficiency, large environmental disturbance and high safety risks of existing technologies, and to achieve efficient, environmentally friendly and safe large-scale underwater debris cleaning.

[0006] To achieve the above objectives, the present invention adopts the following technical solution:

[0007] A mechanical intelligent underwater garbage cleaning robot includes a main frame (2), buoyancy material (1), tracked chassis (5), material bin (3), robotic arm (6), first electrical control compartment (4), second electrical control compartment (10), conveyor belt (8), underwater sensing system, water quality monitoring module and optical fiber composite cable (11).

[0008] The buoyancy material (1) is fixed to the top of the main frame (2) to provide underwater buoyancy balance for the robot; the tracked chassis (5) is installed at the bottom of the main frame (2) to enable the robot to move on the seabed; the hopper (3) is fixed to the middle of the main frame (2) to store collected seabed debris; the robotic arm (6) is installed at the front end of the main frame (2) to grab seabed debris; the conveyor belt (8) extends partly outside the hopper (3) and partly inside the hopper (3) to transport the robotic arm (6). The captured garbage is transported to the material bin (3); the first electrical control cabin (4) and the second electrical control cabin (10) are respectively fixed on the left and right sides of the main frame (2) for power supply, signal transmission and motion control; the underwater sensing system is distributed in different positions of the robot for environmental perception and garbage identification; the water quality monitoring module is integrated into the main frame (2) for monitoring water quality parameters in the working area; the photoelectric composite cable (11) is connected to the second electrical control cabin (10) for communication and power transmission between the robot and the host computer.

[0009] In a preferred embodiment of the present invention, the robotic arm (6) includes a base, a multi-joint arm body, and an end gripper connected in sequence. The base integrates an electric lifting mechanism, which is linked with a depth sensor and a posture sensor to adaptively adjust the working height of the robotic arm (6). The multi-joint arm body includes at least four joints, the first two joints for large-range positioning, and the last two joints for fine posture adjustment. Each joint is equipped with a high-precision encoder. The end gripper integrates a high-strength shearing mechanism at its bottom, forming a gripping and shearing integrated structure. The inner side of the end gripper is provided with biomimetic anti-slip texture and is made of composite material.

[0010] In a preferred embodiment of the present invention, the tracked chassis (5) is a double track structure, with an aluminum alloy frame and anodized. The tracks of the tracked chassis (5) are designed in an inverted trapezoidal shape and are driven by a front drive. The track consists of four load wheels, one drive wheel, one guide wheel and one support wheel. The drive wheel is connected to a deep-sea waterproof integrated motor reducer. The tracked chassis (5) has a waterproof sealed chamber inside. The waterproof sealed chamber contains a chassis motor control driver and is connected to the first electrical control chamber (4) or the second electrical control chamber (10).

[0011] In a preferred embodiment of the present invention, the material hopper (3) has a double-layer structure, including an outer material hopper (3-1), an inner material hopper (3-4), an internal monitoring system (3-2), a lifting door (3-3), and a door structure (3-5); the outer material hopper (3-1) is fixedly connected to the main frame (2) and the tracked chassis (5), with an opening at its front end and the conveyor belt (8) fixed thereon, and an openable door structure (3-5) at its rear end.

[0012] The inner material compartment (3-4) is hinged to the outer material compartment (3-1) via a horizontal slide rail. It has a handle at its rear and a fixed door and a lifting door (3-3) at its front. The fixed door and the lifting door (3-3) are at the same height and lower than the end of the conveyor belt (8) during normal operation. When the inner material compartment (3-4) is pulled out, the lifting door (3-3) automatically rises. The internal monitoring system (3-2) of the material compartment includes a bracket, a searchlight and a camera. The bracket is fixed to the outer material compartment (3-1), and the searchlight and camera are installed on the bracket. The inner material compartment (3-4) is also equipped with a quality sensor.

[0013] In a preferred embodiment of the present invention, the conveyor belt (8) is designed to be inclined inward, with a grid structure set at intervals on the surface and a spring tensioning device inside; the spring tensioning device is used to realize the buffering of the conveyor belt (8) and the adjustment of the fit with the garbage.

[0014] In a preferred embodiment of the present invention, the underwater sensing system includes three sets of observation and communication units, namely a first sensing system (7), a second sensing system (9), and a cargo bin internal monitoring system (3-2); the first sensing system (7) is installed on the bottom front of the robot, and the second sensing system (9) is installed on the top front of the robot. Both the first sensing system (7) and the second sensing system (9) include a camera, a searchlight, and a sonar; the camera and searchlight of the cargo bin internal monitoring system (3-2) are used for both cargo bin internal monitoring and close-range environmental observation; the underwater sensing system realizes three-dimensional reconstruction of the seabed environment based on the Kalman filter method and realizes intelligent garbage identification based on the deep learning method.

[0015] In a preferred embodiment of the present invention, the first electrical control compartment (4) is used to control the low-voltage actuator and receive the low-voltage and host computer communication signals converted by the second electrical control compartment (10); the second electrical control compartment (10) is connected to the optoelectronic composite cable (11) and is used for high-voltage transformation, receiving host computer communication signals and controlling the high-voltage actuator; both the first electrical control compartment (4) and the second electrical control compartment (10) are watertight structures with high pressure resistance and corrosion resistance.

[0016] In a preferred embodiment of the present invention, the water quality monitoring module includes a sensor probe and a data acquisition device. The data acquisition device is installed in the electrical control cabin, and the sensor probe is fixed to the main frame (2) for real-time monitoring of water quality parameters such as temperature, pH, turbidity, and dissolved oxygen.

[0017] In a preferred embodiment of the present invention, the robotic arm (6) is equipped with a force-position hybrid control algorithm module for real-time monitoring of gripping force feedback and automatic adjustment of gripping force; the tracked chassis (5) is equipped with a path planning algorithm and an adaptive power adjustment system for achieving stable movement along the planned path.

[0018] This invention provides a method for cleaning up seabed debris based on the aforementioned robot, comprising the following steps:

[0019] S1: Precisely deploy the robot to the target sea area via a high-strength optical-electric composite cable (11) to complete underwater attitude calibration;

[0020] S2: Intelligent environmental perception. The underwater perception system is activated, and the environmental reconstruction of the work area is completed through sound and light fusion perception. It identifies the density, type and terrain features of garbage distribution, and the water quality monitoring module is activated simultaneously.

[0021] S3: Undersea automatic driving, based on environmental perception data, the optimal cleaning path is generated by the optimization algorithm, the tracked chassis (5) moves according to the planned trajectory, and the power input of the drive motor is adaptively adjusted to avoid obstacles;

[0022] S4: Target precise grabbing, underwater sensing system identifies the location and type of garbage, robotic arm (6) selects grabbing strategy according to garbage type, flexible garbage starts shearing-grabbing sequence, semi-buried garbage adjusts grabbing force through force-position mixing control to complete precise grabbing;

[0023] S5: Efficient transport and storage. The conveyor belt (8) transports the grabbed garbage to the inner hopper (3-4). The internal monitoring system (3-2) monitors the storage status in real time and judges the hopper filling rate by combining the data from the quality sensor.

[0024] S6: Recycling and unloading. When the filling rate of the material bin reaches the preset threshold or the cumulative cleaning area reaches the standard, the robot starts the return program, recovers the waste to the water surface vessel through the photoelectric composite cable (11), opens the door structure (3-5) of the outer material bin (3-1), and pulls out the inner material bin (3-4) to complete the unloading of the waste.

[0025] Compared with the prior art, the present invention has the following beneficial effects:

[0026] (1) High efficiency and low disturbance: The pure mechanical grab-transport-storage integrated operation chain is adopted, abandoning the traditional hydraulic suction method. The precise grabbing of the robotic arm only affects a very small area of ​​the bottom sediment around the garbage, reducing the amount of suspended sediment by more than 95%, reducing the area of ​​bottom sediment disturbance by 90%, and minimizing the impact on benthic biodiversity. At the same time, the multi-degree-of-freedom design of the robotic arm, the efficient transport of the conveyor belt and the large-capacity hopper enable a single operation to cover a large area of ​​sea, greatly improving the garbage cleaning efficiency compared with traditional technology, and the underwater transport efficiency of the conveyor belt can be increased by 60%.

[0027] (2) Strong terrain adaptability: The dual-track chassis adopts an inverted trapezoidal track and a multi-wheel system design, which increases the ground contact area, improves the load capacity and terrain adaptability, and can move stably in muddy and sandy bottoms and complex terrains (such as around rock crevices and gentle slopes), realizing a variety of movement actions, solving the problem of poor mobility of traditional equipment in complex seabed environments.

[0028] (3) Wide adaptability to waste: The robotic arm adopts an integrated gripping and shearing design, combined with a multi-joint high-flexibility structure and force-position hybrid control algorithm, which can adapt to seabed waste of different sizes, shapes and materials (including plastic products, metal products, glass products, fishery waste, etc.). Whether it is large-volume flexible waste, semi-buried waste or small hard waste, it can achieve efficient gripping with a high success rate.

[0029] (4) High degree of intelligence and autonomy: Based on the multimodal perception system of three sets of observation and communication units, combined with deep learning and multi-sensor fusion technology, it realizes functions such as environmental reconstruction, garbage identification, path planning and autonomous obstacle avoidance. It can operate normally in low light and low visibility environments. The operation process is highly automated. From deployment, perception, driving, grabbing, transportation to recycling and unloading, each link is seamlessly connected, reducing human intervention and reducing operation risks.

[0030] (5) Integrated environmental monitoring: The integrated water quality monitoring module can monitor water quality parameters in the work area in real time, evaluate the effect of cleaning operations on water quality improvement, and monitor whether secondary pollution is caused. This enables garbage cleaning and environmental monitoring to be carried out simultaneously, avoiding the paradox of pollution control leading to greater pollution.

[0031] (6) Strong continuity of operation: The double-layer material hopper design greatly increases the storage capacity, and the storage capacity of a single operation is 2-3 times that of traditional equipment. The inner material hopper can be quickly pulled out and unloaded, significantly shortening the operation interval time. The internal monitoring system of the material hopper monitors the storage status in real time, avoiding overload or premature return, and improving operation efficiency and continuity.

[0032] In summary, this invention, through modular integrated design, achieves efficient, environmentally friendly, and safe seabed debris cleanup, overcomes many shortcomings of existing technologies, and provides an economically feasible technical solution for large-scale seabed debris cleanup. It has significant environmental protection value, technological innovation value, and economic benefits. Attached Figure Description

[0033] To more clearly illustrate the technical solutions in the embodiments or prior art, the drawings used in the description of the embodiments or prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0034] Figure 1 This is a structural schematic diagram of a mechanical intelligent underwater debris cleaning robot provided in an embodiment of the present invention.

[0035] Figure 2 This is a front view of a mechanical intelligent underwater debris cleaning robot provided in an embodiment of the present invention.

[0036] Figure 3 This invention provides a schematic diagram of a double-layered cargo compartment structure for a mechanical intelligent underwater debris cleaning robot.

[0037] Figure 4 This invention provides a schematic diagram of a double-layer hopper (with the inner hopper pulled out) structure as an embodiment of the present invention. Detailed Implementation

[0038] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application. The terms "upper," "lower," "front," "rear," "left," and "right," etc., used when describing the installation position or direction of the structure or components in this embodiment are based on the orientation shown in the accompanying drawings. They are merely for convenience of description, used to distinguish the relative positions of various components or directions, and do not represent the orientation of the system or functional components in this embodiment during use.

[0039] This invention provides a mechanical intelligent underwater debris-collecting robot and a method for underwater debris collection. It employs a purely mechanical recycling solution specifically optimized for underwater debris cleanup. The core advantage of this purely mechanical solution lies in its ability to achieve precise grasping and minimal disturbance. By directly grasping debris with a robotic arm, it avoids the large-scale disturbance to seabed sediment caused by hydraulic equipment, fundamentally solving the problem of ecological damage. Compared to manual cleanup, the mechanical solution can operate continuously at greater depths and in more complex underwater environments, significantly improving cleanup efficiency and safety. Furthermore, the specially optimized mechanical structure and intelligent control system can be customized to meet the specific needs of debris collection, achieving a perfect balance between efficiency and environmental protection. The purely mechanical underwater debris-collecting robot solution has significant technical implications. From an environmental protection perspective, it can protect the underwater ecosystem to the greatest extent possible while cleaning up debris, avoiding the paradox of pollution control leading to greater pollution. From a technological development perspective, it promotes the integrated development of underwater robot technology, intelligent sensing technology, and precision mechanical design technology, providing new ideas for the specialized and intelligent design of marine engineering equipment. From an economic perspective, it can significantly reduce the cost of a single operation, improve cleaning efficiency, and provide an economically feasible technical solution for large-scale seabed debris cleanup.

[0040] Specifically, such as Figures 1-4 As shown, the mechanical underwater garbage intelligent cleaning robot includes a main frame 2, buoyancy material 1, tracked chassis 5, material bin 3, robotic arm 6, first electrical control compartment 4, second electrical control compartment 10, conveyor belt 8, underwater sensing system, water quality monitoring module and optical fiber composite cable 11.

[0041] The main frame 2 is the core load-bearing structure of the robot, made of high-strength, corrosion-resistant alloy material to ensure structural stability in the high-pressure environment of the deep sea. All functional modules are fixed to the main frame 2 by bolts or welding, achieving modular integration. The buoyancy material 1 is made of high-performance polyurethane foam material and is fixed to the top of the main frame 2. Its volume is designed according to the overall weight of the robot and the operating water depth to ensure that the robot achieves buoyancy balance underwater, ensuring stability during bottom landing and reducing energy consumption during recovery.

[0042] The tracked chassis 5 is installed at the bottom of the main frame 2 for enabling the robot to move underwater; the feed bin 3 is fixed in the middle of the main frame 2 for storing collected seabed debris; the robotic arm 6 is installed at the front end of the main frame 2 for grabbing seabed debris; the conveyor belt 8 extends partly outside the feed bin 3 and partly inside the feed bin 3 for transporting the debris grabbed by the robotic arm 6 into the feed bin 3; the first electrical control compartment 4 and the second electrical control compartment 10 are fixed on the left and right sides of the main frame 2 respectively for power supply, signal transmission and motion control; the underwater sensing system is distributed at different positions of the robot for environmental perception and debris identification; the water quality monitoring module is integrated into the main frame 2 for monitoring water quality parameters in the operating area; the optical fiber composite cable 11 is connected to the second electrical control compartment 10 for communication and power transmission between the robot and the host computer.

[0043] The cleaning robot uses its robotic arm 6 to grasp debris from the seabed surface, achieving efficient and low-disturbance collection of seabed litter. The robotic arm 6 comprises a base, a multi-jointed arm, and an end effector gripper connected in sequence. The base integrates an electric lifting mechanism, which is linked to depth and attitude sensors to adaptively adjust the working height of the robotic arm 6. It automatically adjusts the working height based on seabed topography and the height of debris distribution. In complex terrain, it maintains the optimal working distance between the robotic arm's end effector and the seabed surface, ensuring both grasping accuracy and efficiency.

[0044] The multi-joint arm comprises at least four joints. The first two joints are used for wide-range positioning, while the latter two are used for fine-tuning posture. Each joint is equipped with a high-precision encoder, resulting in excellent repeatability. Each joint has a wide range of deflection and pitch freedom, expanding the workspace and gripping coverage. A high-strength shearing mechanism is integrated at the bottom of the end effector gripper, forming a unified gripping and shearing structure. The inner side of the end effector gripper features biomimetic anti-slip textures and is made of composite materials. The robotic arm 6 is equipped with a force-position hybrid control algorithm module for real-time monitoring of gripping force feedback and automatic adjustment of gripping force.

[0045] The tracked chassis 5 features a dual-track structure with an aluminum alloy frame treated with an anodizing process. The tracks are designed in an inverted trapezoidal shape, with front-drive capability, and consist of four road wheels, one drive wheel, one guide wheel, and one support wheel. The drive wheel is connected to a deep-sea waterproof integrated motor reducer. The tracked chassis 5 has an internal waterproof sealed chamber housing a chassis motor control driver, which is connected to either the first electrical control compartment 4 or the second electrical control compartment 10. This tracked chassis 5 increases the ground contact area, adapting to seabed sediment, improving load-bearing capacity and terrain adaptability, enabling actions such as straight-line movement, reversing, turning, U-turns, and climbing, ensuring stable movement of the robot in complex seabed environments. The tracked chassis 5 is equipped with a path planning algorithm and an adaptive power adjustment system to achieve stable movement along the planned path.

[0046] The material bin 3 is a structure used by the cleaning robot to store the collected seabed debris. The material bin 3 is fixed in the middle of the main frame 2. The material bin 3 has a double-layer structure, including an outer material bin 3-1, an inner material bin 3-4, an internal monitoring system 3-2, a lifting door 3-3, and a door structure 3-5. The outer material bin 3-1 is fixedly connected to the main frame 2 and the tracked chassis 5. It has an opening at its front end and a fixed conveyor belt 8. Its rear end has an openable door structure 3-5.

[0047] The inner storage compartment 3-4 is hinged to the outer storage compartment 3-1 via a horizontal slide rail. It has a handle at its rear and a fixed door and a lifting door 3-3 at its front. During normal operation, the fixed door and lifting door 3-3 are at the same height and lower than the end of the conveyor belt 8, ensuring that waste enters smoothly and does not fall out. When the inner storage compartment 3-4 is pulled out, the lifting door 3-3 automatically rises to seal and prevent waste from falling out. The internal monitoring system 3-2 includes a bracket, a searchlight, and a camera. The bracket is fixed to the outer storage compartment 3-1, and the searchlight and camera are mounted on the bracket. A quality sensor is also installed inside the inner storage compartment 3-4 to monitor the amount and quality of waste stored in real time. This double-layer storage compartment design significantly increases storage capacity, achieving 2-3 times the storage capacity of traditional equipment in a single operation, and enables rapid waste unloading, avoiding secondary pollution.

[0048] During normal robot operation, the fixed door and the lifting door are at the same height (3-3), and the height of both doors is slightly lower than the end of the conveyor belt (8). This allows waste to smoothly enter the material bin while preventing stored waste from falling out during robot movement and recycling. When the rear door of the outer material bin is opened and the inner material bin is pulled out, the lifting door automatically rises during the pulling process, sealing the inner material bin together with the fixed door to prevent waste from falling out.

[0049] The first electrical control compartment 4 and the second electrical control compartment 10 are fixed to the left and right sides of the main frame 2, respectively. Both are watertight structures with high pressure resistance and corrosion resistance. The first electrical control compartment 4 is used to control the low-voltage actuators and receive low-voltage and host computer communication signals converted by the second electrical control compartment 10. The second electrical control compartment 10 is connected to the optoelectronic composite cable 11, which is used for high-voltage transformation, receiving host computer communication signals, and controlling high-voltage actuators, realizing signal transmission between the robot and the host computer, power supply to each actuator, and motion control. In this embodiment, the optoelectronic composite cable 11 is connected to the second electrical control compartment 10, which has both power supply and communication functions, ensuring stable data transmission and continuous power supply between the robot and the host computer on the water surface when the robot is operating underwater. It also has sufficient strength for the deployment and retrieval of the robot.

[0050] The conveyor belt 8 transports the seabed debris grabbed by the robotic arm to the storage bin. It is the downstream link of the mechanical grabbing device and an important support for the debris cleaning operation. The conveyor belt 8 adopts an inward inclined design with a grid structure at intervals on its surface and a spring tensioning device inside. The spring tensioning device is used to buffer the conveyor belt 8 and adjust the fit with the debris, thereby improving the reliability of transportation.

[0051] The underwater sensing system comprises three observation and communication units: a first sensing system 7, a second sensing system 9, and a cargo hold internal monitoring system 3-2. The first sensing system 7 is installed at the bottom front of the robot, and the second sensing system 9 is installed at the top front of the robot. Both the first sensing system 7 and the second sensing system 9 include cameras, searchlights, and sonar. The cameras and searchlights of the cargo hold internal monitoring system 3-2 are used for both cargo hold internal monitoring and close-range environmental observation. The underwater sensing system uses the Kalman filter method to achieve three-dimensional reconstruction of the seabed environment and deep learning methods to achieve intelligent identification of debris. It can optimize image quality through adaptive imaging algorithms in low-light and low-visibility environments to ensure accurate identification of terrain features, obstacles, and debris information.

[0052] By installing an underwater sensing system, the cleaning robot can perform 3D reconstruction of the seabed environment based on methods such as Kalman filtering, thereby identifying the terrain features and obstacle distribution of the surrounding seabed. Using deep learning methods, it can intelligently identify debris targets in the seabed environment, determining the location and type of debris, providing a data foundation for the robotic arm's automatic grasping. Simultaneously, based on the sensing and identification results of the surrounding seabed environment, the robot can also pinpoint the target area for cleaning operations and determine a crawling path through path planning, helping it to move safely and efficiently to areas rich in debris.

[0053] The water quality monitoring module includes sensor probes and a data acquisition unit. The data acquisition unit is installed inside the electrical control compartment, while the sensor probes are fixed to the main frame 2. It is used to monitor water quality parameters such as temperature, pH, turbidity, and dissolved oxygen in real time. It can compare the water quality differences between waste-rich areas and waste-free areas, monitor whether the operation causes secondary pollution, and evaluate the cleaning effect.

[0054] Through its water quality monitoring module, the cleaning robot can not only perform garbage collection operations but also: observe the differences in water quality data between garbage-rich areas and garbage-free areas; monitor in real time whether secondary pollution such as suspended sediment will occur during the garbage collection process; and monitor the improvement of water quality caused by garbage collection operations in the same area, demonstrating the intuitive effect of garbage collection.

[0055] This invention provides a method for cleaning up seabed debris based on the aforementioned robot, comprising the following steps:

[0056] S1: Precisely deploy the robot to the target sea area via a high-strength optical fiber composite cable 11, and complete the underwater attitude calibration; in this embodiment, the host computer adjusts the robot's attitude through attitude sensor data to ensure a stable landing and complete the underwater initialization calibration.

[0057] S2: Intelligent environmental perception. The underwater perception system is activated, and the environmental reconstruction of the work area is completed through sound and light fusion perception, identifying the density, type, and terrain features of garbage distribution. The water quality monitoring module is activated simultaneously. Specifically, the intelligent environmental perception system is activated, and the cameras, searchlights, and sonar of the second perception system 9 work together to acquire large-scale environmental information. Combined with the Kalman filtering method, the three-dimensional environmental reconstruction of the work area is completed, identifying the density, type, and obstacle distribution of garbage. The first perception system 7 assists in identifying close-range terrain details. The water quality monitoring module is activated simultaneously, collecting parameters such as temperature, pH, turbidity, and dissolved oxygen in real time and transmitting them to the host computer.

[0058] S3: Undersea autonomous driving. Based on environmental perception data, an optimal cleaning path is generated through an optimization algorithm. The tracked chassis 5 moves along the planned trajectory, adaptively adjusting the power input of the drive motor to avoid obstacles. Specifically, in underwater autonomous driving, the host computer generates the optimal cleaning path based on environmental perception data through an optimization algorithm. The tracked chassis 5 moves along the planned trajectory, and the adaptive power adjustment system adjusts the power input of the drive motor in real time according to terrain changes and water flow conditions, ensuring stable movement under complex terrain and flow field conditions, with excellent path tracking accuracy, while avoiding obstacles.

[0059] S4: Precise target grasping. The underwater sensing system identifies the location and type of debris. The robotic arm 6 selects a grasping strategy based on the debris type. For flexible debris, a shearing-grabbing sequence is initiated. For semi-buried debris, the grasping force is adjusted through force-position hybrid control to achieve precise grasping. Specifically, for precise target grasping, the underwater sensing system uses deep learning algorithms to identify the location and type of debris and transmits the data to the electrical control cabin. The electrical control cabin then controls the robotic arm 6 to initiate the following: For large, flexible, and easily entangled debris such as discarded fishing nets and ropes, a shearing mechanism is first activated for pre-processing before grasping; for semi-buried debris, the force-position hybrid control algorithm monitors the grasping force feedback and automatically increases the grasping force; for small, hard debris, it is directly grasped precisely using biomimetic anti-slip grippers, resulting in short grasping times and a high success rate. During the grasping process, the water quality monitoring module provides real-time feedback on disturbance data to ensure minimal impact on the seabed ecosystem.

[0060] S5: Efficient transport and storage. Conveyor belt 8 transports the grabbed waste to the inner hopper 3-4. The hopper's internal monitoring system 3-2 monitors the storage status in real time and determines the hopper's filling rate based on data from mass sensors. Specifically, for efficient transport and storage, robotic arm 6 places the grabbed waste onto conveyor belt 8. Conveyor belt 8, through the coordinated action of its inclined design, baffle structure, and spring tensioning device, quickly transports the waste to the inner hopper 3-4. The camera in the hopper's internal monitoring system 3-2 captures real-time images of the interior of the inner hopper 3-4, and mass sensors monitor the total mass of the waste. The system comprehensively determines the hopper's filling rate, and the data is transmitted to the host computer in real time.

[0061] S6: Recycling and Unloading. When the material bin filling rate reaches a preset threshold or the cumulative cleaning area reaches the target, the robot initiates the return procedure, recovers the waste to the vessel on the water via the photoelectric composite cable 11, opens the door structure 3-5 of the outer material bin 3-1, and pulls out the inner material bin 3-4 to complete the waste unloading. Specifically, for recycling and unloading, when the material bin filling rate reaches a preset threshold such as 80% or the cumulative cleaning area reaches the expected target, the host computer issues a return command, and the robot returns to the deployment position along the original path, recovering the waste to the vessel on the water via the photoelectric composite cable 11; the operator opens the door structure 3-5 of the outer material bin 3-1, pulls the inner material bin 3-4 horizontally through the handle at the rear, and the lifting door 3-3 automatically rises to prevent the waste from falling, quickly completing the waste unloading; after unloading, the inner material bin 3-4 is pushed back into the outer material bin 3-1, the door structure 3-5 is closed, and the robot can be deployed again.

[0062] The present invention has the following positive effects:

[0063] 1. The tracked chassis and large-volume double-layer cargo bin design are as follows:

[0064] The robot's dual-track chassis structure increases the ground contact area, allowing it to adapt well to seabed sediment and improving its load-bearing capacity on such surfaces. This provides a reliable installation and support foundation for large-volume cargo bins. Simultaneously, this design ensures the robot's agile maneuverability in complex seabed terrain and its posture stability during cleaning operations, enhancing both cleaning efficiency and safety.

[0065] The large-volume material bin adopts a double-layer structure. The outer layer is rigidly connected to the robot body, providing reliable structural support; the inner layer can be pulled out horizontally as a whole, enabling rapid unloading of waste. This design significantly increases waste storage capacity while maintaining the robot's overall compactness, achieving a single-operation storage capacity 2-3 times that of traditional equipment. More importantly, the double-layer design completely solves the problems of difficult, time-consuming, and easily caused secondary pollution associated with traditional underwater equipment. The inner layer can be pulled out and unloaded in a very short time, significantly improving operational continuity.

[0066] This invention overcomes the traditional constraints between maneuverability and garbage collection capacity in underwater cleaning robots through an innovative integrated design of a tracked chassis and a large-volume hopper. The synergistic design of the tracked chassis and the large-volume hopper produces a significant multiplier effect: the stable support of the tracked chassis provides a reliable mounting foundation for the large-capacity hopper, avoiding stability issues caused by center-of-gravity shifts during cleaning operations, as is common with floating robots; the optimized layout of the hopper improves the overall center-of-gravity distribution, enhancing the traction efficiency and obstacle-crossing ability of the tracked chassis. This integrated design significantly increases the robot's coverage area per operation, multiplies its operational efficiency, and ensures stability and reliability in complex seabed environments, providing a solid technical foundation for large-scale seabed garbage cleanup.

[0067] 2. The purely mechanical seabed debris cleaning solution is as follows:

[0068] This invention innovatively proposes a purely mechanical seabed debris collection solution. Through a mechanical design encompassing the entire process of grabbing, transporting, and storing debris, it balances collection efficiency with environmental protection, achieving highly efficient and low-disturbance collection of seabed debris. The robot design abandons the traditional hydraulic suction method, constructing a complete mechanical debris collection and recycling operation chain through three stages: precise grabbing by a robotic arm, efficient transport by a barrier conveyor belt, and centralized storage in a large-capacity double-layer bin. This realizes the operational concept of "precise grabbing and minimal disturbance."

[0069] The robotic arm features a multi-joint, degree-of-freedom design and is equipped with an adaptive "gripper-shear" gripper, enabling it to highly maneuver and pick up seabed debris scattered across the seabed surface. Compared to traditional hydraulic collection equipment, the destructive impact of the debris collection process on the seabed ecosystem can be reduced by more than 80%. The robotic arm employs a multi-degree-of-freedom serial structure, equipped with biomimetic grippers and shearing functions, and automatically selects the optimal gripping strategy for different types of debris based on an intelligent recognition system. This precise gripping method minimizes environmental disturbance, affecting only a very small area of ​​the seabed around the debris, avoiding the problem of large-scale disturbance of seabed sediments caused by traditional hydraulic equipment.

[0070] The conveyor belt features an inclined design, fully utilizing gravity to enhance the contact between the waste and the belt surface during transport. A grid structure on the belt surface increases the conveyor belt's thrust on the waste, effectively preventing slippage during underwater transport and improving underwater transport efficiency by 60%. An elastic tensioning mechanism further enhances transport reliability by moderately deforming the belt to increase the contact area between the waste and the belt surface. The entire transport process is quiet and stable, without generating additional water flow disturbance.

[0071] The core value of the purely mechanical approach lies in its perfect balance between eco-friendliness and operational efficiency. By avoiding the use of water suction, this approach reduces suspended sediment by over 95%, minimizes bottom disturbance by 90%, and reduces the impact on benthic biodiversity to a minimum. Simultaneously, precise grasping and efficient transport significantly improve waste cleanup efficiency, allowing a single operation to cover a large area of ​​the sea. This "low-disturbance, high-efficiency" operational mode provides a sustainable technological solution for marine environmental protection, resolving the destructive impact of traditional hydraulic collection equipment on the seabed ecosystem and achieving a perfect balance between waste cleanup and ecological protection.

[0072] 3. The intelligent operation mode based on multimodal perception is as follows:

[0073] This invention constructs a comprehensive, multi-layered underwater environmental perception network through three sets of collaborative observation and communication systems and multi-sensor fusion technology, providing a solid foundation for the intelligent autonomous operation of robots. The three sets of observation and communication systems are respectively deployed at the top front, the sides front, and inside the hopper, forming a complementary perception system: the top front group is responsible for long-distance environmental perception and path planning; the sides front group is responsible for close-range observation and detail recognition of the robotic arm's operating area; and the hopper-side group is responsible for monitoring the waste storage status. This layout solves the problems of limited field of view and single function of a single observation and communication system, achieving full-coverage perception from the macroscopic environment to microscopic details.

[0074] The intelligent operation mode, based on advanced environmental perception and autonomous decision-making algorithms, achieves closed-loop control from environmental recognition to action execution. The system first acquires extensive environmental information through the frontal top observation and communication group, combining sonar and laser scanning data to generate a high-precision 3D environmental map. It automatically identifies obstacle distribution and debris-rich areas, planning the optimal operation path. During movement, the system monitors environmental conditions in real time and dynamically adjusts track power input to ensure automatic and stable movement in complex seabed environments. When a debris target is detected, the frontal side observation and communication group initiates high-precision identification, analyzes the debris type and location, and automatically selects the best grasping strategy. When the robotic arm performs the grasping action, the system adjusts the grasping force and position in real time through force sensors and visual feedback to ensure a high success rate.

[0075] The innovation of the intelligent sensing system lies in its outstanding adaptability to complex seabed environments. The system employs adaptive imaging algorithms to automatically optimize image quality under extremely low light conditions, compensating for underwater spectral loss through color correction and image enhancement techniques. Even in environments with extremely low visibility, the system can effectively identify and grasp debris, significantly expanding the robot's operational range and applicable conditions. Advanced multi-sensor fusion technology enables the system to predict environmental changes, proactively mitigating potential risks and ensuring operational safety.

[0076] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely preferred examples and are not intended to limit the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of the present invention is defined by the appended claims and their equivalents.

Claims

1. A mechanical intelligent underwater debris cleaning robot, characterized in that, It includes the main frame (2), buoyancy material (1), tracked chassis (5), cargo bin (3), robotic arm (6), first electrical control compartment (4), second electrical control compartment (10), conveyor belt (8), underwater sensing system, water quality monitoring module and optical fiber composite cable (11). The buoyancy material (1) is fixed to the top of the main frame (2) to provide underwater buoyancy balance for the robot; the tracked chassis (5) is installed at the bottom of the main frame (2) to enable the robot to move on the seabed; the hopper (3) is fixed to the middle of the main frame (2) to store collected seabed debris; the robotic arm (6) is installed at the front end of the main frame (2) to grab seabed debris; the conveyor belt (8) extends partly outside the hopper (3) and partly inside the hopper (3) to transport the robotic arm (6). The captured garbage is transported to the material bin (3); the first electrical control cabin (4) and the second electrical control cabin (10) are respectively fixed on the left and right sides of the main frame (2) for power supply, signal transmission and motion control; the underwater sensing system is distributed in different positions of the robot for environmental perception and garbage identification; the water quality monitoring module is integrated into the main frame (2) for monitoring water quality parameters in the working area; the optoelectronic composite cable (11) is connected to the second electrical control cabin (10) for communication and power transmission between the robot and the host computer; The robotic arm (6) includes a base, a multi-joint arm body, and an end gripper connected in sequence. The base integrates an electric lifting mechanism, which is linked with a depth sensor and a posture sensor to adaptively adjust the working height of the robotic arm (6). The multi-joint arm body includes at least four joints. The first two joints are used for large-range positioning, and the last two joints are used for fine posture adjustment. Each joint is equipped with a high-precision encoder. The end gripper integrates a high-strength shearing mechanism at the bottom, forming a gripping and shearing integrated structure. The end gripper has a biomimetic anti-slip texture on the inner side and is made of composite material. The tracked chassis (5) is a double track structure, with an aluminum alloy frame and anodized finish. The tracks of the tracked chassis (5) are inverted trapezoidal, driven by a front drive, and consist of four load wheels, one drive wheel, one guide wheel and one support wheel. The drive wheel is connected to a deep-sea waterproof integrated motor reducer. The tracked chassis (5) has a waterproof sealed chamber inside, which contains a chassis motor control driver and is connected to the first electrical control chamber (4) or the second electrical control chamber (10). The material hopper (3) has a double-layer structure, including an outer material hopper (3-1), an inner material hopper (3-4), an internal monitoring system (3-2), a lifting door (3-3), and a door structure (3-5); the outer material hopper (3-1) is fixedly connected to the main frame (2) and the tracked chassis (5), with an opening at its front end and the conveyor belt (8) fixed thereon, and an openable door structure (3-5) at its rear end. The inner material compartment (3-4) is hinged to the outer material compartment (3-1) via a horizontal slide rail. It has a handle at its rear and a fixed door and a lifting door (3-3) at its front. The fixed door and the lifting door (3-3) are at the same height and lower than the end of the conveyor belt (8) during normal operation. When the inner material compartment (3-4) is pulled out, the lifting door (3-3) automatically rises. The internal monitoring system (3-2) of the material compartment includes a bracket, a searchlight and a camera. The bracket is fixed to the outer material compartment (3-1), and the searchlight and camera are installed on the bracket. The inner material compartment (3-4) is also equipped with a quality sensor.

2. The mechanical intelligent underwater debris cleaning robot according to claim 1, characterized in that, The conveyor belt (8) is designed to be inclined inward, with a grid structure set at intervals on the surface and a spring tensioning device inside; the spring tensioning device is used to achieve the buffering of the conveyor belt (8) and the adjustment of the fit with the garbage.

3. The mechanical intelligent underwater debris cleaning robot according to claim 2, characterized in that, The underwater sensing system includes three sets of observation and communication units, namely a first sensing system (7), a second sensing system (9), and a cargo bin internal monitoring system (3-2); the first sensing system (7) is installed on the bottom front of the robot, and the second sensing system (9) is installed on the top front of the robot. Both the first sensing system (7) and the second sensing system (9) include a camera, a searchlight, and a sonar; the camera and searchlight of the cargo bin internal monitoring system (3-2) are used for both cargo bin internal monitoring and close-range environmental observation; the underwater sensing system realizes three-dimensional reconstruction of the seabed environment based on the Kalman filter method and realizes intelligent garbage identification based on the deep learning method.

4. The mechanical intelligent underwater debris cleaning robot according to claim 3, characterized in that, The first electrical control compartment (4) is used to control the low-voltage actuator and receive the low-voltage and host computer communication signals converted by the second electrical control compartment (10); the second electrical control compartment (10) is connected to the optoelectronic composite cable (11) and is used for high-voltage transformation, receiving host computer communication signals and controlling the high-voltage actuator; both the first electrical control compartment (4) and the second electrical control compartment (10) are watertight structures with high pressure resistance and corrosion resistance.

5. The mechanical intelligent underwater debris cleaning robot according to claim 4, characterized in that, The water quality monitoring module includes a sensor probe and a data acquisition device. The data acquisition device is installed in the electrical control cabin, and the sensor probe is fixed to the main frame (2) for real-time monitoring of water quality parameters such as temperature, pH, turbidity, and dissolved oxygen.

6. The mechanical intelligent underwater debris cleaning robot according to claim 5, characterized in that, The robotic arm (6) is equipped with a force-position hybrid control algorithm module, which is used to monitor the gripping force feedback in real time and automatically adjust the gripping force; the tracked chassis (5) is equipped with a path planning algorithm and an adaptive power adjustment system, which is used to achieve stable movement according to the planned path.

7. A method for cleaning up underwater debris based on the robot described in claim 6, characterized in that, Includes the following steps: S1: Precisely deploy the robot to the target sea area via a high-strength optical-electric composite cable (11) to complete underwater attitude calibration; S2: Intelligent environmental perception. The underwater perception system is activated, and the environmental reconstruction of the work area is completed through sound and light fusion perception. It identifies the density, type and terrain features of garbage distribution, and the water quality monitoring module is activated simultaneously. S3: Undersea automatic driving, based on environmental perception data, the optimal cleaning path is generated by the optimization algorithm, the tracked chassis (5) moves according to the planned trajectory, and the power input of the drive motor is adaptively adjusted to avoid obstacles; S4: Target precise grabbing, underwater sensing system identifies the location and type of garbage, robotic arm (6) selects grabbing strategy according to garbage type, flexible garbage starts shearing-grabbing sequence, semi-buried garbage adjusts grabbing force through force-position mixing control to complete precise grabbing; S5: Efficient transport and storage. The conveyor belt (8) transports the grabbed garbage to the inner hopper (3-4). The internal monitoring system (3-2) monitors the storage status in real time and judges the hopper filling rate by combining the data from the quality sensor. S6: Recycling and unloading. When the filling rate of the material bin reaches the preset threshold or the cumulative cleaning area reaches the standard, the robot starts the return program, recovers the waste to the water surface vessel through the photoelectric composite cable (11), opens the door structure (3-5) of the outer material bin (3-1), and pulls out the inner material bin (3-4) to complete the unloading of the waste.

Images