A kind of floating multi-legged ore collecting robot with autonomous perception and positioning grabbing function and method

Abstract

This invention belongs to the field of deep-sea resource development and intelligent robot technology, specifically relating to a planktonic multi-legged ore-collecting robot and method with autonomous sensing, positioning, and grasping functions. This invention is used for low-disturbance, high-efficiency collection of discontinuous and unevenly distributed mineral resources such as polymetallic nodules and cobalt-rich crusts in complex deep-sea terrain. It includes a biomimetic multi-legged pressure-resistant shell module, a multimodal autonomous sensing and positioning system, a multi-legged cooperative flexible grasping system, an intelligent buoyancy and cooperative propulsion system, an autonomous control system, an energy management module, and a data processing and storage unit. This invention employs a composite movement mode combining a biomimetic multi-legged structure and planktonic drive, enabling intelligent identification and compliant grasping of ore, avoiding the collection of waste rock, and improving resource recovery grade and economic benefits.

Description

Technical Field

[0001] This invention belongs to the field of deep-sea resource development and intelligent robot technology, specifically relating to a planktonic multi-legged ore-gathering robot and method with autonomous perception, positioning and grasping functions. Background Technology

[0002] The deep sea contains abundant strategic mineral resources such as polymetallic nodules and cobalt-rich crusts, and their extraction is of great significance to sustainable development. However, the deep-sea environment is characterized by high pressure, low temperature, darkness, complex terrain, and fragile ecology, which places extremely high demands on mining technology. Current mainstream deep-sea mining technologies, such as trawlers and tracked ore collection machines, mainly rely on direct contact and shoveling or hydraulic pumping of seabed sediments. Although widely used in the industry, these technologies have significant drawbacks: First, these devices agitate seabed sediments over large areas during operation, forming "sediment clouds" that cause severe and long-term damage to benthic communities, resulting in high environmental costs. Second, their locomotives are poorly adapted to complex terrains (such as seamounts, steep slopes, and soft mud areas), making them prone to getting stuck or overturning, and compromising operational continuity. Third, their automation level is limited, often requiring pre-programmed paths or real-time remote control, making it difficult to achieve intelligent identification and selective collection when facing sparsely distributed and diverse mineral targets, resulting in low efficiency.

[0003] Although autonomous underwater vehicles (AUVs) and remotely operated underwater vehicles (ROVs) have been used for deep-sea exploration and sampling in recent years, the former typically lacks precise grasping and manipulation capabilities, while the latter heavily relies on mother ship cables and real-time manual control, resulting in high operating costs, low efficiency, and limited operational windows. Therefore, there is an urgent need for a new type of deep-sea mineral collection equipment that can combine highly autonomous mobility with precise control capabilities while minimizing environmental disturbance. Summary of the Invention

[0004] The purpose of this invention is to provide a robotic system capable of autonomously identifying, locating, grasping, and temporarily storing mineral resources in extreme and complex deep-sea environments. This robot integrates planar movement, precision operations, and intelligent sensing, aiming to achieve deep-sea mining operations with low environmental disturbance, high terrain adaptability, and strong operational autonomy. It effectively solves problems existing in current technologies such as high ecological damage risks, poor terrain mobility, high reliance on manual labor, and low collection efficiency.

[0005] A floating multi-legged ore-gathering robot with autonomous sensing, positioning, and grasping functions includes: a biomimetic multi-legged pressure-resistant shell module, a multimodal autonomous sensing and positioning system, a multi-legged cooperative flexible grasping system, an intelligent buoyancy and cooperative propulsion system, an autonomous control system, an energy management module, and a data processing and storage unit.

[0006] The biomimetic multi-legged pressure-resistant shell module constitutes the main support and sealing structure of the robot;

[0007] The multimodal autonomous sensing and positioning system is integrated into a biomimetic multi-legged pressure-resistant shell module, used for scanning, mapping, obstacle detection and water environment monitoring of the deep seabed environment, as well as identifying and three-dimensionally locating target minerals.

[0008] The multi-legged cooperative flexible grasping system is located on both sides of the biomimetic multi-legged pressure-resistant shell module. It is used to perform adaptive grasping and release of target ore, and at the same time, it is used for autonomous walking, turning and attitude balance control in complex terrain.

[0009] The intelligent buoyancy and cooperative propulsion system is used to adjust the overall buoyancy of the robot and provide multi-degree-of-freedom propulsion power, enabling the robot to float, hover, and adjust its attitude.

[0010] The autonomous control system is housed within the biomimetic multi-legged pressure-resistant shell module and is communicatively connected to the multimodal autonomous sensing and positioning system, the multi-legged cooperative flexible grasping system, and the intelligent buoyancy and cooperative propulsion system, respectively, for executing fully autonomous operation processes.

[0011] The energy management module is housed within the biomimetic multi-legged pressure-resistant shell module and is used to power the multimodal autonomous sensing and positioning system, the multi-legged cooperative flexible grasping system, and the intelligent buoyancy and cooperative propulsion system.

[0012] The data processing and storage unit is housed within the biomimetic multi-legged pressure-resistant shell module and is used to store data transmitted by the modal autonomous sensing and positioning system, the multi-legged cooperative flexible grasping system, and the intelligent buoyancy and cooperative propulsion system.

[0013] The biomimetic multi-legged pressure-resistant shell module includes a main pressure-resistant shell and an inner cavity. The main pressure-resistant shell adopts a flattened, six- or eight-legged symmetrical biomimetic configuration and is made of carbon fiber reinforced composite material or titanium alloy, with its outer surface covered by an anti-bioadhesion coating. The inner cavity is divided into an independent atmospheric pressure sealed chamber and a non-pressure-resistant functional chamber. The atmospheric pressure sealed chamber is equipped with an autonomous control system, an energy management module, a data processing and storage unit, and a multimodal autonomous sensing and positioning system. The non-pressure-resistant functional chamber is equipped with a buoyancy adjustment unit for an intelligent buoyancy and cooperative propulsion system, a hydraulic drive unit for a multi-legged cooperative flexible grasping system, and an ore storage unit for storing ore.

[0014] The multimodal autonomous sensing and positioning system includes an optical sensing unit, a 3D scanning unit, an acoustic sensing and navigation unit, and an environmental sensing unit. The optical sensing unit includes a deep-sea high-definition camera and a multispectral imager, used to acquire seabed optical images and perform preliminary identification and classification of target ores. The 3D scanning unit includes a laser structured light scanner or a stereo vision system, used to collect surface information of target ores and obtain high-precision 3D point cloud data of the target ores. The acoustic sensing and navigation unit includes a side-scan sonar, a forward-looking multibeam imaging sonar, a combined inertial navigation system, a Doppler log, and an ultra-short baseline acoustic positioning system, used for large-scale terrain mapping, obstacle detection, and global and relative positioning. The environmental sensing unit includes a depth sensor, a turbidity meter, and a current meter, used to monitor seawater depth, turbidity, and flow conditions.

[0015] The multi-legged cooperative flexible grasping system includes multiple robotic legs, each of which includes a robotic arm unit, a hydraulic drive unit, and an end effector. Each robotic arm unit has at least three degrees of freedom of motion. The hydraulic drive unit provides power to the robotic arm unit. The end effector is hinged to the end of the robotic arm unit, and a force / torque sensor is integrated on the end effector. The end effector is an adaptive multi-finger gripper or an adsorption device. The hydraulic drive unit, the force / torque sensor, and the autonomous control system are electrically connected.

[0016] The intelligent buoyancy and cooperative propulsion system includes a buoyancy adjustment unit and a multi-legged vector propulsion unit. The buoyancy adjustment unit includes a ballast water tank and a variable volume buoyancy adjustment mechanism. The ballast water tank is used to adjust the overall weight by adding and removing water, thereby achieving a wide range of coarse buoyancy adjustments and load balance. The variable volume buoyancy adjustment mechanism is used to change the drainage volume by expanding and contracting the volume, thereby achieving precise fine-tuning of buoyancy and dynamic compensation for underwater environmental disturbances. The multi-legged vector propulsion unit consists of multiple miniature vector thrusters distributed at the joints of the robotic arm unit or on the main pressure-resistant shell. The buoyancy adjustment unit and the multi-legged vector propulsion unit are communicatively connected to the autonomous control system.

[0017] The ore storage unit is located in the middle of the main pressure-resistant shell. The ore storage unit forms a modular ore storage bin. The modular ore storage bin has a slidingly connected layered drawer structure or modular compartment structure. The layered drawers can slide against each other, and each drawer has a buffer layer. The modular compartment structure consists of multiple compartments connected to each other in a gradient arrangement. Each compartment has a buffer layer. Each drawer or compartment is equipped with a bin status monitoring sensor to monitor whether the ore storage level in the bin has reached full capacity. The bin status monitoring sensor is communicatively connected to the autonomous control system.

[0018] The autonomous control system software layer integrates deep learning target detection and recognition algorithms, 3D point cloud processing algorithms, motion planning algorithms, and multi-task scheduling and fault diagnosis modules.

[0019] The floating multi-legged ore-gathering robot consists of multiple robots, which interact and assign tasks through a collaborative communication module in the autonomous control system to perform multi-robot cluster collaborative operations.

[0020] A method for collecting ore using a floating multi-legged ore-collecting robot with autonomous sensing, positioning, and grasping capabilities includes the following steps:

[0021] Step 1: The autonomous control system controls the intelligent buoyancy and cooperative propulsion system to enable the robot to dive to the predetermined cruising altitude and cruise along the path in floating mode. At the same time, it controls the multimodal autonomous perception and positioning system to scan the seabed topography and conduct initial mineral exploration, build a map of the operating environment and locate potential mineral-rich areas.

[0022] Step 2: In areas with potential mineral enrichment, the autonomous control system controls the intelligent buoyancy and cooperative propulsion system to switch to hovering or low-speed floating state, and controls the multimodal autonomous sensing and positioning system to perform fine optical imaging, three-dimensional scanning, obstacle detection, global and relative positioning on the seabed, and collect information on seawater depth, turbidity, and flow state, identify target ore and calculate its precise three-dimensional spatial coordinates.

[0023] Step 3: Based on the precise three-dimensional spatial coordinates of the target ore and the working environment map, the autonomous control system plans a collision-free approach path, controls the intelligent buoyancy and cooperative propulsion system to drive the robot to move precisely to the grasping preparation position, and then controls the multi-legged cooperative flexible grasping system to complete the adaptive and compliant grasping of the target ore under the visual servo guidance of the optical sensing unit and the real-time force feedback of the force / torque sensor.

[0024] Step 4: The autonomous control system controls the multi-legged cooperative flexible grasping system to transfer the successfully grasped ore to the modular ore storage bin, and then returns to step 2 to perform the next operation cycle until the preset operation termination condition is met.

[0025] Step 5: When the operation termination conditions are met, the autonomous control system controls the intelligent buoyancy and cooperative propulsion system to make the robot float to the recovery position and communicate and automatically dock with the surface support platform to unload the collected ore.

[0026] The preset operation termination conditions include at least one of the following: the full load rate of the modular ore storage bin exceeds a predetermined threshold or the remaining power of the robot energy management module is lower than a predetermined threshold.

[0027] Compared with existing technologies, the floating multi-legged ore-gathering robot provided by this invention has the following significant advantages:

[0028] 1. Minimal environmental disturbance: The selective grabbing method completely avoids the large-scale disturbance of seabed sediments caused by traditional scraping and suction operations, thus protecting the fragile deep-sea ecosystem to the greatest extent.

[0029] 2. Extremely adaptable to terrain: The combination of biomimetic multi-legged structure and floating drive enables the robot to operate stably on rugged and soft seabeds in walking mode, and to easily cross obstacles such as ditches and steep slopes in floating mode, so that the operating range is not limited by terrain.

[0030] 3. High degree of operational autonomy: It integrates a complete autonomous intelligent chain from perception and decision-making to execution, enabling long-term and large-scale unattended operations, which greatly reduces the dependence on surface support systems and manpower, and improves operational efficiency and reliability.

[0031] 4. High collection accuracy and selectivity: The vision and force-based grasping system can intelligently identify and accurately grasp target minerals, avoiding the collection of waste rock and improving the quality and economic benefits of resource recovery.

[0032] 5. High degree of system scalability and modularity: The robot adopts a modular design, and the sensing payload, grasping tools, storage units, etc. can be quickly replaced or upgraded according to specific task requirements, making the platform highly versatile.

[0033] 6. This robot is specially designed for low-disturbance and high-efficiency collection of discontinuous and unevenly distributed mineral resources such as polymetallic nodules and cobalt-rich crusts in complex deep-sea terrain. Attached Figure Description

[0034] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0035] Figure 1 This is a schematic diagram of the overall structure of the floating multi-legged mineral-gathering robot with autonomous sensing, positioning and grasping functions of the present invention.

[0036] Figure 2 This is a schematic diagram of the bottom structure of the floating multi-legged ore-gathering robot with autonomous sensing, positioning and grasping functions of the present invention;

[0037] Figure 3This is a schematic diagram of the mechanical legs of the floating multi-legged mineral-gathering robot with autonomous sensing, positioning and grasping functions of the present invention.

[0038] Figure 4 This is a flowchart illustrating the operation of the floating multi-legged ore-gathering robot with autonomous sensing, positioning, and grasping functions of the present invention.

[0039] In the attached diagram: 1. Main pressure-resistant shell; 2. Modular ore storage bin; 3. Machine foot; 301. Robotic arm unit; 302. End effector; 303. Force / torque sensor; 304. Hydraulic drive unit; 4. Vertical thruster; 5. Vector thruster; 6. Optical sensing unit. Detailed Implementation

[0040] To make the objectives, technical solutions, and advantages of this invention clearer, the following will provide a more detailed description of a floating deep-sea multi-legged ore-gathering robot with autonomous sensing, positioning, and grasping functions, in conjunction with the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are for illustrative purposes only and are not intended to limit the invention. All other implementations obtained by those skilled in the art based on the embodiments of this invention without inventive effort are within the scope of protection of this invention.

[0041] It should be noted that the relative arrangement, shape, size ratio, and specific values, materials, and control parameters of the components in the following embodiments and accompanying drawings are merely examples provided to clearly and concisely illustrate the technical principles and core concepts of the present invention. These specific details should not be construed as limitations on the present invention. Any equivalent substitutions, parameter adjustments, or structural modifications guided by the concept of the present invention, such as increasing or decreasing the number of feet, changing the propeller layout, selecting the type of sensing sensor, or replacing the control algorithm model, should be covered within the scope of protection of the present invention, as long as they do not depart from the overall technical solution and innovative spirit disclosed in the present invention.

[0042] Furthermore, for ease of description, the accompanying drawings may not be drawn strictly to scale, and some structures may be depicted using schematic or emphatic methods. While the specification may not elaborate on well-known general technologies, component manufacturing processes, or fundamental control methods in the field of deep-sea robotics (such as the basic operation of the ROS architecture, the basic principles of sonar, and the conventional sealing design of pressure hulls), these well-known technologies should be considered, where appropriate, as integral parts of this specification supporting the described technical solutions.

[0043] Inspired by the remarkable adaptability and dexterity exhibited by marine organisms such as octopuses and crabs in complex seabed environments, this invention provides a planktonic multi-legged mineral-gathering robot with autonomous sensing, positioning, and grasping capabilities. This robot is specifically designed for low-disturbance, high-efficiency collection of discontinuous and unevenly distributed mineral resources such as polymetallic nodules and cobalt-rich crusts in complex deep-sea terrain. It achieves deep-sea mineral collection tasks through a highly integrated, closed-loop, autonomous intelligent workflow. This robot integrates biomimetic design, intelligent sensing, collaborative control, and planktonic actuation technology, aiming to achieve fully automated operations from autonomous target search and identification, precise 3D positioning, adaptive compliant grasping, to safe temporary storage. It fundamentally overcomes the pain points of traditional technologies, such as large environmental disturbances, poor terrain adaptability, and insufficient autonomy, providing an innovative solution for environmentally friendly, precise, and efficient deep-sea mineral mining.

[0044] Reference Figures 1-3 The present invention provides a further description of a floating multi-legged ore-gathering robot with autonomous sensing, positioning, and grasping functions.

[0045] A floating multi-legged ore-gathering robot with autonomous sensing, positioning, and grasping functions includes: a biomimetic multi-legged pressure-resistant shell module, a multimodal autonomous sensing and positioning system, a multi-legged cooperative flexible grasping system, an intelligent buoyancy and cooperative propulsion system, an autonomous control system, an energy management module, and a data processing and storage unit;

[0046] The biomimetic multi-legged pressure-resistant shell module constitutes the main support and sealing structure of the robot;

[0047] The multimodal autonomous sensing and positioning system is integrated into a biomimetic multi-legged pressure-resistant shell module, used for scanning, mapping, obstacle detection and water environment monitoring of the deep seabed environment, as well as identifying and three-dimensionally locating target minerals.

[0048] The multi-legged cooperative flexible grasping system is located on both sides of the biomimetic multi-legged pressure-resistant shell module. It is used to perform adaptive grasping and release of target ore, and at the same time, it is used for autonomous walking, turning and attitude balance control in complex terrain.

[0049] The intelligent buoyancy and cooperative propulsion system is used to adjust the overall buoyancy of the robot and provide multi-degree-of-freedom propulsion power, enabling the robot to float, hover, and adjust its attitude.

[0050] The autonomous control system is housed within the biomimetic multi-legged pressure-resistant shell module and is communicatively connected to the multimodal autonomous sensing and positioning system, the multi-legged cooperative flexible grasping system, and the intelligent buoyancy and cooperative propulsion system, respectively, for executing fully autonomous operation processes.

[0051] The energy management module is housed within the biomimetic multi-legged pressure-resistant shell module and is used to power the multimodal autonomous sensing and positioning system, the multi-legged cooperative flexible grasping system, and the intelligent buoyancy and cooperative propulsion system.

[0052] The data processing and storage unit is housed within the biomimetic multi-legged pressure-resistant shell module and is used to store data transmitted by the modal autonomous sensing and positioning system, the multi-legged cooperative flexible grasping system, and the intelligent buoyancy and cooperative propulsion system.

[0053] This invention integrates a biomimetic multi-legged pressure-resistant shell module, a multimodal autonomous sensing and positioning system, a multi-legged cooperative flexible grasping system, an intelligent buoyancy and attitude adjustment system, and an integrated autonomous control and energy management system. These systems work together to form a complete autonomous operating unit.

[0054] The biomimetic multi-legged pressure-resistant shell module includes a main pressure-resistant shell 1 and an inner cavity. The main pressure-resistant shell 1 adopts a flattened six-legged or eight-legged symmetrical biomimetic configuration and is made of carbon fiber reinforced composite material or titanium alloy. Its outer surface is covered with an anti-bioadhesion coating. The inner cavity is divided into an independent atmospheric pressure sealed chamber and a non-pressure-resistant functional chamber. The atmospheric pressure sealed chamber is equipped with an autonomous control system, an energy management module, a data processing and storage unit, and a multimodal autonomous sensing and positioning system. The non-pressure-resistant functional chamber is equipped with a buoyancy adjustment unit of an intelligent buoyancy and cooperative propulsion system, a hydraulic drive unit of a multi-legged cooperative flexible grasping system, and an ore storage unit for storing ore.

[0055] In this embodiment, the robot adopts a lightweight biomimetic multi-legged structure as its main frame, preferably with a six- or eight-legged symmetrical layout to provide excellent static stability and motion redundancy. The main pressure-resistant shell 1 is made of carbon fiber composite material or titanium alloy and adopts a streamlined flat design to reduce water resistance; the interior of the shell is divided into an independent atmospheric pressure sealed chamber and a non-pressure-resistant functional chamber.

[0056] The ore storage unit is located in the middle of the main pressure-resistant shell. The ore storage unit forms a modular ore storage bin 2. The modular ore storage bin 2 is equipped with guide grooves. Layered drawer structures or modular compartment structures are slidably connected to the guide grooves. The layered drawer structures can slide against each other. Each drawer has a buffer layer to achieve orderly and low-damage storage of the ore. The modular compartment structure consists of multiple interconnected compartments arranged in a gradient. Each compartment has a buffer layer to achieve orderly and low-damage storage of the ore. Each drawer or compartment is equipped with a bin status monitoring sensor to monitor whether the ore storage level in the bin has reached full capacity. The bin status monitoring sensor is communicatively connected to the autonomous control system.

[0057] The multimodal autonomous perception and positioning system is the "eyes" and "brain" of the robot to achieve autonomous operation. Its hardware layer includes an optical sensing unit 6, a 3D scanning unit, an acoustic sensing and navigation unit, and an environmental sensing unit. The optical sensing unit 6 integrates a high-sensitivity deep-sea wide-angle high-definition camera and a multispectral imager, and is equipped with a high-power LED illumination array to provide light source assistance, acquire seabed optical images, identify the surface morphology, texture, and spectral characteristics of minerals, and achieve preliminary classification and two-dimensional positioning of targets. The 3D scanning unit uses a laser structured light scanner or a binocular stereo vision system to reconstruct 3D point clouds of targets identified by optical scanning, accurately acquiring their size, shape, and spatial coordinates relative to the robot. The acoustic sensing and navigation unit is equipped with a forward-looking multibeam imaging sonar for obstacle detection and avoidance, a side-scan sonar for large-scale seabed topography mapping, and combines an inertial navigation system, a Doppler log, and an ultra-short baseline acoustic positioning system to achieve centimeter-level precise positioning and track tracking of the robot in global and relative coordinate systems. The environmental sensing unit integrates a depth sensor, a turbidity meter, and a current meter to monitor seawater depth, turbidity, and flow status, and monitor operational environment parameters in real time to provide a basis for control decisions.

[0058] The multi-legged cooperative flexible grasping system includes multiple robotic legs 3, each of which not only serves as a moving mechanism but also integrates a working terminal at its end. Each robotic leg 3 includes a robotic arm unit 301, a hydraulic drive unit 304, and an end effector 302. One end of the robotic arm unit 301 is hinged to the main pressure-resistant housing 1, and the end of the robotic arm unit 301 is hinged to the end effector 302. Each robotic arm unit 301 has at least three degrees of freedom of motion. The hydraulic drive unit 304 provides power to the robotic arm unit 301. The end effector 302 integrates a force / torque sensor 303. The end effector 302 is an adaptive multi-finger gripper or adsorption device with tactile sensing capabilities. The autonomous control system is based on the principles of visual servoing and impedance control. It integrates visual information from the optical sensing unit 6 and force feedback information from the robotic arm unit 301 to dynamically plan the grasping trajectory and force, so as to achieve stable, adaptive, and compliant grasping of ores of different sizes, shapes, and surface conditions, avoiding slippage or crushing. At the same time, when walking on the seabed, each robotic leg serves as a bionic support limb. It relies on the coordinated action of the robotic arm unit 301 and the hydraulic drive unit 304 to realize the bionic robot's autonomous walking, turning, and posture balance control in complex terrain.

[0059] The intelligent buoyancy and collaborative propulsion system endows the robot with unique floating movement and hovering capabilities. It includes a buoyancy adjustment unit and a multi-legged vector propulsion unit. The buoyancy adjustment unit comprises a ballast water tank and a variable-volume buoyancy adjustment mechanism, which is a flexible oil bladder (or air bladder). The ballast water tank is used to adjust the overall weight by adding and removing water, achieving a wide range of coarse buoyancy adjustments and load balance. The variable-volume buoyancy adjustment mechanism is used to change the drainage volume by expanding and contracting, achieving precise fine-tuning of buoyancy and dynamic compensation for underwater environmental disturbances. Through precision water pumps and valve control, the robot's overall buoyancy can be steplessly and rapidly adjusted, thereby achieving energy-saving vertical ascent / descent and precise depth maintenance (hovering). The multi-legged vector propulsion unit consists of multiple miniature vector thrusters (such as duct thrusters) arranged at the joints of each robot leg or on the main pressure-resistant shell 1. When multiple micro-vector thrusters are arranged on the main pressure hull 1, multiple vector thrusters 5 are located at its tail and sides, and vertical thrusters 4 are also located on the sides. Through the coordinated distribution of the thrust magnitude and direction of each thruster by the central controller, the robot can achieve flexible movement with six degrees of freedom, including forward and backward, left and right translation and rotation in the horizontal plane, as well as pitch and roll attitude adjustment. This hybrid drive mode of "buoyancy adjustment as the main method and vector thrust as the auxiliary method" enables the robot to perform precise close-to-the-seabed operations in "walking" mode during collection, and to efficiently traverse complex terrain in "floating" mode during transfer.

[0060] The autonomous control system is the "brain" of the robot. It is a distributed autonomous control system based on the robot operating system. Its software layer integrates deep learning target detection and recognition algorithms, 3D point cloud processing algorithms, motion planning algorithms, multi-task scheduling and fault diagnosis modules.

[0061] The floating multi-legged ore-gathering robot consists of multiple robots, which interact and assign tasks through a collaborative communication module in the autonomous control system to perform multi-robot cluster collaborative operations.

[0062] like Figure 4 As shown, the method for collecting ore using a floating multi-legged ore-collecting robot with autonomous sensing and positioning grasping functions is as follows:

[0063] (1) First stage: initialization, diving and wide-area exploration

[0064] The operation began with the safe deployment of the robot from the water surface, supporting the mother ship. Upon entry into the water, the robot immediately initiated its autonomous initialization program, and the ballast tanks began controlled water injection to provide adequate negative buoyancy, allowing it to descend to the designated operational area at a stable speed of approximately 0.5 meters per second. During the descent, the acoustic sensing and navigation unit activated, performing real-time depth monitoring, course correction, and underwater obstacle detection to ensure controllable course and depth. After reaching the preset cruising altitude of 5-10 meters above the seabed, the robot entered wide-area exploration mode. In this mode, the robot primarily relies on its multi-legged vector propulsion unit for efficient horizontal movement. Simultaneously, it activates its environmental perception unit to collect data on seawater temperature, salinity, depth, turbidity, and current velocity and direction in the work area. This data provides sound velocity compensation and environmental error correction for sonar detection and topographic mapping. Side-scan sonar and forward-looking multibeam imaging sonar are also activated concurrently. The robot systematically cruises along a pre-planned grid path. The acoustic perception and navigation unit collects data in real time and transmits it to the autonomous control system to create a high-resolution seabed topographic map. Preliminary analysis is then conducted using acoustic reflection characteristics and the optical perception unit to identify potential target areas rich in polymetallic nodules or cobalt-rich crusts. The local environmental map constructed at this stage provides a global spatial reference and guidance for key areas in subsequent refined operations.

[0065] (2) Second stage: Target fine identification and three-dimensional localization

[0066] Once the robot determines it has entered a potential mineral-rich area, it immediately switches to a precision operation mode. First, by adjusting its buoyancy control unit and multi-legged vector propulsion unit, the robot achieves stable hovering or low-speed floating above the target area. The acoustic perception and navigation unit continues to operate online, while the environmental perception unit continuously monitors changes in seawater depth, turbidity, and flow patterns. Then, the high-power LED lighting array on the bottom, along with a deep-sea high-definition camera and multispectral imager, is activated to perform high-brightness, multi-spectral imaging of the seabed below. The autonomous control system processes the acquired images in real time, accurately identifying individual ore targets that meet set characteristics (e.g., diameter greater than 4 cm) and completing preliminary classification. To further obtain the precise spatial information required for grasping, the robot activates a laser structured light scanner or stereo vision system to scan the identified target area, generating a 3D point cloud with millimeter-level precision. By fusing the visual recognition results with the 3D point cloud data, the autonomous control system can calculate the precise 3D coordinates (X, Y, Z) and orientation of each target ore in the robot's coordinate system, laying the foundation for subsequent grasping actions.

[0067] (3) Third stage: Intelligent approach and adaptive compliant grasping

[0068] After obtaining the target's precise pose, the autonomous control system calculates an optimal collision-free approach path by integrating the robot's current state, the target's position, the storage compartment's position, and the real-time environmental map (including obstacle information). The robot then, following instructions, fine-tunes its buoyancy adjustment unit (entering "buoyancy-driven mode") and coordinates its multi-legged vector propulsion unit to move smoothly and precisely along the planned path to the optimal grasping position directly above the target (typically placing the target at the center of the robotic arm's workspace and maintaining a certain height, such as Z+0.3 meters). Immediately afterward, the multi-legged cooperative flexible grasping system begins operation: the designated robotic arm extends towards the target ore under visual servo control, ensuring the end effector is aligned with the target. The moment the gripper contacts the ore, the built-in force / torque sensor generates a real-time feedback signal. The autonomous control system employs an impedance control algorithm to dynamically adjust the gripping force and trajectory of the gripper based on preset compliance parameters and force feedback information. This allows the gripper to adaptively envelop and hold the ore with a constant and gentle force (e.g., 50N), ensuring a firm and reliable grip while effectively preventing damage to structurally fragile ore or target slippage due to excessive force.

[0069] (4) Fourth stage: Ore recovery, storage and operation cycle

[0070] After successfully grasping the ore, the robotic arm lifts it from the seabed and plans an efficient path to transfer it to the modular storage bin entrance above the main pressure-resistant shell. The bin's buffer design prevents collision damage between ore fragments. The autonomous control system continuously updates the collection count and bin status. After completing one "identification-grabbing-storage" unit operation, the robot does not need to ascend to its cruising altitude but returns directly to the second stage near the current work point. Utilizing its multimodal autonomous perception and positioning system, it finds and locates the next nearest or optimal target, initiating a new work cycle. The autonomous control system autonomously reports battery status, while bin status monitoring sensors continuously operate throughout the process, assessing the storage bin's full load rate and remaining battery power in real time. Once the preset work termination conditions are met (e.g., storage volume reaches 85%, battery power is below 20%), the system automatically terminates the collection cycle and seamlessly transitions to the final stage.

[0071] (5) Fifth stage: mission termination, ascent and return to base and automatic unloading

[0072] Upon receiving the termination command, the robot first performs status verification and data storage. Then, it initiates the return procedure: the ballast tanks are completely emptied, and the variable-volume buoyancy adjustment mechanism expands to provide sufficient positive buoyancy, allowing the robot to begin vertical ascent in energy-saving mode. During the ascent, when it reaches a safe depth close to the surface, the robot establishes contact with the mother ship via underwater acoustic communication or, after surfacing, via radio and reports its position. The mother ship navigates to the rendezvous point, and the robot, utilizing its precise positioning and propulsion capabilities, automatically docks with the receiving device on the bottom of the mother ship through a standardized docking interface. After successful docking, the bottom hatch of the robot's modular ore storage compartment opens under command control, and the collected ore is unloaded in batches and efficiently into the mother ship's collection system under gravity or auxiliary mechanisms, thus completing a full mission cycle from diving and operation to recovery.

Claims

1. A floating multi-legged ore-gathering robot with autonomous sensing, positioning, and grasping functions, characterized in that, include: Bionic multi-legged pressure-resistant shell module, multimodal autonomous sensing and positioning system, multi-legged cooperative flexible grasping system, intelligent buoyancy and cooperative propulsion system, autonomous control system, energy management module, and data processing and storage unit; The biomimetic multi-legged pressure-resistant shell module constitutes the main support and sealing structure of the robot; The multimodal autonomous sensing and positioning system is integrated into a biomimetic multi-legged pressure-resistant shell module, used for scanning, mapping, obstacle detection and water environment monitoring of the deep seabed environment, as well as identifying and three-dimensionally locating target minerals. The multi-legged cooperative flexible grasping system is located on both sides of the biomimetic multi-legged pressure-resistant shell module. It is used to perform adaptive grasping and release of target ore, and at the same time, it is used for autonomous walking, turning and attitude balance control in complex terrain. The intelligent buoyancy and cooperative propulsion system is used to adjust the overall buoyancy of the robot and provide multi-degree-of-freedom propulsion power, enabling the robot to float, hover, and adjust its attitude. The autonomous control system is housed within the biomimetic multi-legged pressure-resistant shell module and is communicatively connected to the multimodal autonomous sensing and positioning system, the multi-legged cooperative flexible grasping system, and the intelligent buoyancy and cooperative propulsion system, respectively, for executing a fully autonomous operation process; The energy management module is housed within the biomimetic multi-legged pressure-resistant shell module and is used to power the multimodal autonomous sensing and positioning system, the multi-legged cooperative flexible grasping system, and the intelligent buoyancy and cooperative propulsion system. The data processing and storage unit is housed within the biomimetic multi-legged pressure-resistant shell module and is used to store data transmitted by the modal autonomous sensing and positioning system, the multi-legged cooperative flexible grasping system, and the intelligent buoyancy and cooperative propulsion system.

2. The floating multi-legged ore-gathering robot with autonomous sensing and positioning grasping functions according to claim 1, characterized in that, The biomimetic multi-legged pressure-resistant shell module includes a main pressure-resistant shell and an inner cavity. The main pressure-resistant shell adopts a flattened six-legged or eight-legged symmetrical biomimetic configuration and is made of carbon fiber reinforced composite material or titanium alloy. Its outer surface is covered with an anti-bioadhesion coating. The internal cavity is divided into an independent atmospheric pressure sealed chamber and a non-pressure-resistant functional chamber. The atmospheric pressure sealed chamber is equipped with an autonomous control system, an energy management module, a data processing and storage unit, and a multi-modal autonomous sensing and positioning system. The non-pressure-resistant functional chamber is equipped with a buoyancy adjustment unit for an intelligent buoyancy and cooperative propulsion system, a hydraulic drive unit for a multi-legged cooperative flexible gripping system, and an ore storage unit for storing ore.

3. The floating multi-legged ore-gathering robot with autonomous sensing and positioning grasping functions according to claim 1, characterized in that, The multimodal autonomous sensing and positioning system includes an optical sensing unit, a 3D scanning unit, an acoustic sensing and navigation unit, and an environmental sensing unit. The optical sensing unit includes a deep-sea high-definition camera and a multispectral imager, used to acquire seabed optical images and perform preliminary identification and classification of target ores. The 3D scanning unit includes a laser structured light scanner or a stereo vision system, used to collect surface information of target ores and obtain high-precision 3D point cloud data of the target ores. The acoustic sensing and navigation unit includes a side-scan sonar, a forward-looking multibeam imaging sonar, a combined inertial navigation system, a Doppler log, and an ultra-short baseline acoustic positioning system, used for large-scale terrain mapping, obstacle detection, and global and relative positioning. The environmental sensing unit includes a depth sensor, a turbidity meter, and a current meter, used to monitor seawater depth, turbidity, and flow conditions.

4. The floating multi-legged ore-gathering robot with autonomous sensing and positioning grasping functions according to claim 1, characterized in that, The multi-legged cooperative flexible grasping system includes multiple robotic legs, each of which includes a robotic arm unit, a hydraulic drive unit, and an end effector. Each robotic arm unit has at least three degrees of freedom of motion. The hydraulic drive unit provides power to the robotic arm unit. The end effector is hinged to the end of the robotic arm unit, and a force / torque sensor is integrated on the end effector. The end effector is an adaptive multi-finger gripper or an adsorption device. The hydraulic drive unit, the force / torque sensor, and the autonomous control system are electrically connected.

5. A floating multi-legged ore-gathering robot with autonomous sensing and positioning grasping functions according to claim 1, characterized in that, The intelligent buoyancy and cooperative propulsion system includes a buoyancy adjustment unit and a multi-legged vector propulsion unit. The buoyancy adjustment unit includes a ballast water tank and a variable volume buoyancy adjustment mechanism. The ballast water tank is used to adjust the overall weight by adding or removing water, thereby achieving a wide range of coarse buoyancy adjustments and load balance. The variable volume buoyancy adjustment mechanism is used to change the drainage volume by expanding or contracting the volume, thereby achieving precise fine-tuning of buoyancy and dynamic compensation for underwater environmental disturbances. The multi-legged vector propulsion unit consists of multiple miniature vector thrusters distributed and installed at the joints of the robotic arm unit or on the main pressure-resistant shell. The buoyancy adjustment unit and the multi-legged vector propulsion unit are communicatively connected to the autonomous control system.

6. A floating multi-legged ore-gathering robot with autonomous sensing and positioning grasping functions according to claim 1, characterized in that, The ore storage unit is located in the middle of the main pressure-resistant shell. The ore storage unit forms a modular ore storage bin. The modular ore storage bin has a slidingly connected layered drawer structure or modular compartment structure. The layered drawers can slide against each other, and each drawer has a buffer layer. The modular compartment structure consists of multiple compartments connected to each other in a gradient arrangement. Each compartment has a buffer layer. Each drawer or compartment is equipped with a bin status monitoring sensor to monitor whether the ore storage level in the bin has reached full capacity. The bin status monitoring sensor is communicatively connected to the autonomous control system.

7. A floating multi-legged ore-gathering robot with autonomous sensing and positioning grasping functions according to claim 1, characterized in that, The autonomous control system software layer integrates deep learning target detection and recognition algorithms, 3D point cloud processing algorithms, motion planning algorithms, and multi-task scheduling and fault diagnosis modules.

8. A floating multi-legged ore-gathering robot with autonomous sensing and positioning grasping functions according to claim 1, characterized in that, The floating deep-sea multi-legged mining robot consists of multiple robots. These robots interact and assign tasks through a collaborative communication module in the autonomous control system to perform multi-robot cluster collaborative operations.

9. A method for collecting ore using a floating multi-legged ore-collecting robot with autonomous sensing and positioning grasping functions, based on the floating multi-legged ore-collecting robot with autonomous sensing and positioning grasping functions as described in claim 1, characterized in that, Includes the following steps: Step 1: The autonomous control system controls the intelligent buoyancy and cooperative propulsion system to enable the robot to dive to the predetermined cruising altitude and cruise along the path in floating mode. At the same time, it controls the multimodal autonomous perception and positioning system to scan the seabed topography and conduct initial mineral exploration, build a map of the operating environment and locate potential mineral-rich areas. Step 2: In areas with potential mineral enrichment, the autonomous control system controls the intelligent buoyancy and cooperative propulsion system to switch to hovering or low-speed floating state, and controls the multimodal autonomous sensing and positioning system to perform fine optical imaging, three-dimensional scanning, obstacle detection, global and relative positioning on the seabed, and collect information on seawater depth, turbidity, and flow state, identify target ore and calculate its precise three-dimensional spatial coordinates. Step 3: Based on the precise three-dimensional spatial coordinates of the target ore and the working environment map, the autonomous control system plans a collision-free approach path, controls the intelligent buoyancy and cooperative propulsion system to drive the robot to move precisely to the grasping preparation position, and then controls the multi-legged cooperative flexible grasping system to complete the adaptive and compliant grasping of the target ore under the visual servo guidance of the optical sensing unit and the real-time force feedback of the force / torque sensor. Step 4: The autonomous control system controls the multi-legged cooperative flexible grasping system to transfer the successfully grasped ore to the modular ore storage bin, and then returns to step 2 to perform the next operation cycle until the preset operation termination condition is met. Step 5: When the operation termination conditions are met, the autonomous control system controls the intelligent buoyancy and cooperative propulsion system to make the robot float to the recovery position and communicate and automatically dock with the surface support platform to unload the collected ore.

10. A method for collecting ore using a floating multi-legged ore-collecting robot with autonomous sensing and positioning grasping functions as described in claim 9, characterized in that, The preset operation termination conditions include at least one of the following: the full load rate of the modular ore storage bin exceeds a predetermined threshold or the remaining power of the robot energy management module is lower than a predetermined threshold.

Images