Unmanned ship grabbing and positioning system based on multi-body configuration design

Abstract

The invention provides an unmanned ship grabbing and positioning system based on multi-body configuration design, and belongs to the technical field of unmanned ships, the unmanned ship grabbing and positioning system comprises a multi-body ship body, a machine vision module, a positioning calculation unit, a stable platform, a grabbing device and a control module, the multi-body ship body is composed of a double-hull structure or a three-hull structure; the machine vision module is used for identifying a target object and acquiring visual information of the target object; the positioning calculation unit is connected with the machine vision module and is used for calculating spatial position coordinates of the target object according to the acquired vision information; the stable platform is arranged on the multi-hull ship body; the grabbing device is installed on the stable platform and used for grabbing the target object according to the coordinates calculated by the positioning calculation unit. The robot has the comprehensive advantages of being high in environmental adaptability, high in positioning precision, flexible in operation, good in controllability, high in maintainability and the like, and can efficiently cope with target object grabbing operation in the complex water environment.

Description

Technical Field

[0001] The present invention relates to the technical field of unmanned ships, and in particular to an unmanned ship grasping and positioning system based on a multi-body configuration design. Background Art

[0002] In traditional water target grabbing operations, a single-body ship equipped with simple mechanical devices is often used, which has many limitations: the hull stability of the single-body ship is poor and it is easily affected by waves and water currents, resulting in shaking, low positioning accuracy and insufficient grasping stability; target recognition mostly relies on manual observation or simple image recognition, which has difficulty in coping with changes in target characteristics in complex water environments and low recognition accuracy; positioning calculations mostly use traditional algorithms, without combining dynamic adjustment of the hull posture, making it difficult to achieve accurate spatial positioning of the target; there is a lack of an effective stable platform, and grasping devices such as robotic arms are easily affected by hull shaking, resulting in a low grasping success rate; the degree of automation is low, and manual remote control is often required, resulting in low operation efficiency and a large impact on human factors.

[0003] Therefore, an unmanned vessel grasping and positioning system based on multi-body configuration design is proposed. Summary of the Invention

[0004] The purpose of the present invention is to provide an unmanned boat grasping and positioning system based on a multi-body configuration design, which integrates multi-hull stability design, machine vision recognition, deep learning positioning calculation, dynamic leveling control and other technologies. Through the synergistic effect of components such as the multi-hull body, machine vision module, and stable platform, it makes up for the shortcomings of traditional systems in stability, positioning accuracy, and degree of automation, meets the needs of automated grasping operations in complex water environments, and solves the defects of current unmanned boats.

[0005] The specific technical solutions are as follows:

[0006] An unmanned vessel grasping and positioning system based on a multi-body configuration design includes:

[0007] Multi-hull ship body, consisting of a double hull structure or a triple hull structure;

[0008] Machine vision module, used to identify the target and obtain visual information of the target;

[0009] A positioning calculation unit, connected to the machine vision module, for calculating the spatial position coordinates of the target object based on the acquired visual information;

[0010] A stabilizing platform is provided on the multihull body;

[0011] a grasping device, mounted on the stable platform, for grasping the target object according to the coordinates calculated by the positioning calculation unit;

[0012] The control module is electrically connected to the machine vision module, the stabilizing platform and the grasping device respectively, and is used to control the automatic operation of the machine vision module, the stabilizing platform and the grasping device.

[0013] As a preferred solution of the present invention, the stabilizing platform is installed between the double hull structure or the triple hull structure through a plurality of connecting seats, the connecting seats are fixedly connected to the double hull structure or the triple hull structure, and the connecting seats are fixedly connected to the stabilizing platform.

[0014] As a preferred solution of the present invention, the machine vision module includes a high-definition pan-tilt camera and an image processor. The high-definition pan-tilt camera is electrically connected to the control module, and the high-definition pan-tilt camera is used to capture images of target objects. The image processor is electrically connected to the high-definition pan-tilt camera, and the image processor is used to process and extract features from the captured images.

[0015] As a preferred solution of the present invention, the positioning calculation unit adopts an algorithm model based on deep learning to calculate the precise spatial position coordinates of the target object relative to the hull based on the target object visual information obtained by the machine vision module and the hull posture information.

[0016] As a preferred solution of the present invention, the stable platform includes a hollow support platform, a support plate, a level sensor and a leveling mechanism, the hollow support platform is hinged to the connecting seat, and the hollow support platform is fixedly connected to the double hull structure or the triple hull structure, the support plate is fixedly installed at the upper opening of the hollow support platform through a sealing elastic rubber frame, and under the action of the sealing elastic rubber frame, the support plate can float up and down relative to the hollow support platform, the level sensor is fixedly installed at the bottom of the support plate, and the leveling mechanism is provided with four groups, and the four groups of leveling mechanisms are evenly distributed between the bottom of the support plate and the inner bottom wall of the hollow support platform, wherein:

[0017] The level sensor is used to monitor the horizontal state of the support plate in real time, and the level sensor is electrically connected to the control module. The control module can control the leveling mechanism to automatically adjust the level of the support plate according to the signal data of the level sensor.

[0018] As a preferred solution of the present invention, the leveling mechanism is provided with four groups, and the four groups of the leveling mechanisms are evenly distributed between the support plate and the upper part of the inner bottom wall of the hollow support platform. Each group of the leveling mechanism includes an electric push rod and a connecting rod. The electric push rod is vertically and fixedly installed on the upper part of the inner bottom wall of the hollow support platform, and the connecting rod is vertically and fixedly installed on the bottom of the support plate, and the bottom end of the connecting rod is hinged to the piston rod end of the electric push rod through a ball hinge, and the electric push rod is electrically connected to the control module.

[0019] As a preferred solution of the present invention, the grasping device includes a robotic arm and an end effector, the robotic arm has multiple degrees of freedom, the end effector is a clamping claw or suction cup structure for grasping the target object, and the end effector and the robotic arm are both electrically connected to the control module.

[0020] As a preferred solution of the present invention, the motion control of the robotic arm adopts an adaptive control algorithm of a built-in control module to adjust the motion trajectory of the robotic arm in real time according to the hull shaking and the position of the target object.

[0021] As a preferred solution of the present invention, it further includes a communication module electrically connected to the control module, and the communication module is used to transmit the position information and grasping status information of the unmanned boat to the remote control center.

[0022] As a preferred solution of the present invention, it also includes a data storage module electrically connected to the control module, and the data storage module is used to store image data collected by the machine vision module, calculation data of the positioning calculation unit and operation data of the grasping device.

[0023] The present invention has the following beneficial effects:

[0024] The unmanned vessel grasping and positioning system, based on a multi-hull design, provided by the present invention, significantly optimizes automated grasping operations for aquatic targets through the synergistic effects of its various components. The multi-hull double or triple hull structure of the multi-hull vessel provides surface stability far exceeding that of a monohull, laying a solid foundation for the entire system. The machine vision module, using a high-definition pan-tilt camera and image processor, combined with the deep learning algorithm of the positioning calculation unit, achieves precise identification and spatial positioning of targets. The stabilization platform, utilizing a hollow support platform, support plates, level sensors, and four sets of leveling mechanisms, counteracts hull sway in real time, ensuring horizontal stability in the grasping device's working environment. The grasping device's multi-degree-of-freedom robotic arm, coupled with an adaptive control algorithm and a gripper or suction cup end effector, flexibly adapts to different targets and achieves stable grasping. The control module's automated control of various components, supplemented by remote monitoring by the communication module and information retention by the data storage module, gives the system comprehensive advantages, including strong environmental adaptability, high positioning accuracy, flexible operation, good controllability, and strong maintainability, enabling efficient grasping of targets in complex aquatic environments. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 A schematic structural diagram of an unmanned vessel grasping and positioning system based on a multi-body configuration design provided by an embodiment of the present invention;

[0026] Figure 2A schematic cross-sectional view of a stabilizing platform in an unmanned vessel grasping and positioning system based on a multi-body configuration design provided by an embodiment of the present invention;

[0027] Figure 3 This is a workflow diagram of the distributed collaborative control equations in the unmanned ship grasping and positioning system based on multi-body configuration design provided by an embodiment of the present invention.

[0028] In the attached figure:

[0029] 1. Multihull body;

[0030] 2. Stable platform; 201. Hollow support platform; 202. Support plate; 203. Level sensor; 204. Sealing elastic rubber frame; 205. Leveling mechanism; 2051. Electric push rod; 2052. Connecting rod; 2053. Ball hinge;

[0031] 3. Robotic arm;

[0032] 4. End effector;

[0033] 5. Machine vision module;

[0034] 6. Connecting seat. DETAILED DESCRIPTION

[0035] The technical solution of the present invention will be further described below with reference to the accompanying drawings and through specific implementation methods.

[0036] Among them, the drawings are only used for illustrative purposes and represent only schematic diagrams rather than actual pictures, and should not be understood as limiting this patent; in order to better illustrate the embodiments of the present invention, some parts of the drawings may be omitted, enlarged or reduced, and do not represent the size of the actual product; for those skilled in the art, it is understandable that some well-known structures and their descriptions in the drawings may be omitted.

[0037] The same or similar numbers in the drawings of the embodiments of the present invention correspond to the same or similar parts; in the description of the present invention, it should be understood that if the terms "upper", "lower", "left", "right", "inside", "outside" and the like indicate an orientation or position relationship based on the orientation or position relationship shown in the drawings, it is only for the convenience of describing the present invention and simplifying the description, and does not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operate in a specific orientation. Therefore, the terms describing the position relationship in the drawings are only used for illustrative purposes and cannot be understood as limiting this patent. For ordinary technicians in this field, the specific meanings of the above terms can be understood according to specific circumstances.

[0038] In the description of the present invention, unless otherwise expressly specified or limited, when the term "connection" or the like appears to indicate a connection relationship between components, such term should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium; it can be internal communication between two components or an interaction between two components. For those skilled in the art, the specific meanings of the above terms in the present invention can be understood in specific circumstances.

[0039] Example

[0040] The unmanned ship grasping and positioning system based on multi-body configuration design provided in this embodiment is as follows: Figure 1-Figure 3 As shown, it includes: a multi-hull body 1, a machine vision module 5, a positioning calculation unit, a stabilizing platform 2, a grasping device and a control module, wherein:

[0041] The multi-hull body 1 is composed of a double hull structure or a triple hull structure;

[0042] The machine vision module 5 is used to identify the target object and obtain the visual information of the target object;

[0043] The positioning calculation unit is connected to the machine vision module and is used to calculate the spatial position coordinates of the target object based on the acquired visual information;

[0044] The stabilizing platform 2 is provided on the multihull 1;

[0045] The grasping device is installed on the stable platform 2 and is used to grasp the target object according to the coordinates calculated by the positioning calculation unit;

[0046] The control module is electrically connected to the machine vision module 5, the stabilizing platform 2 and the grasping device respectively, and is used to control the automatic operation of the machine vision module 5, the stabilizing platform 2 and the grasping device.

[0047] The unmanned boat grasping and positioning system based on the multi-body configuration design adopts the above-mentioned technical solution, and the multi-hull body 1 is set with a double-hull structure or a triple-hull structure, which can significantly improve the stability of the hull on the water surface compared with the single-hull structure, and provide a stable installation basis for the machine vision module 5, the stabilization platform 2, the grasping device, etc.; the machine vision module 5 can effectively identify the target object and obtain visual information, providing data support for the positioning calculation unit; the positioning calculation unit calculates the spatial position coordinates of the target object based on the visual information, and can realize the precise positioning of the target object; the stabilization platform 2 can reduce the impact of the hull shaking on the grasping device; the grasping device grasps according to the positioning coordinates, and combined with the automatic control of each component by the control module, the entire system can realize the automatic, stable grasping and precise positioning of the target object in a complex environment, thereby improving the environmental adaptability and operational reliability of the system in water operations.

[0048] Specifically, in this embodiment, the stabilizing platform 2 is installed between the double hull structure or the triple hull structure through a plurality of connecting seats 6, the connecting seats 6 are fixedly connected to the double hull structure or the triple hull structure, and the connecting seats 6 are fixedly connected to the stabilizing platform 2.

[0049] By adopting the above technical solution, the stabilizing platform 2 is installed between the hulls through multiple connecting seats 6, and the connecting seats 6 are fixedly connected to the double-hull structure or triple-hull structure and the stabilizing platform 2 respectively. This fixed connection method enhances the connection firmness between the stabilizing platform 2 and the multi-hull body 1, reduces the loosening or displacement of the stabilizing platform 2 when the hull shakes, thereby improving the installation stability of the stabilizing platform 2 and providing a guarantee for the stable operation of the grabbing device.

[0050] Specifically, in this embodiment, the machine vision module 5 includes a high-definition pan-tilt camera and an image processor. The high-definition pan-tilt camera is electrically connected to the control module, and the high-definition pan-tilt camera is used to capture images of the target object. The image processor is electrically connected to the high-definition pan-tilt camera, and the image processor is used to process and extract features from the captured images.

[0051] By adopting the above technical solution, the high-definition pan-tilt camera in the machine vision module 5 can capture clear images of the target object, and the pan-tilt structure is convenient for adjusting the shooting angle, which can adapt to the acquisition of target images at different positions and angles; the image processor processes and extracts features of the collected image, can remove interference information in the image, and accurately extract the features of the target object, thereby improving the accuracy and reliability of target object recognition, and providing high-quality visual data support for the positioning calculation unit.

[0052] Specifically, in this embodiment, the positioning calculation unit adopts an algorithm model based on deep learning to calculate the precise spatial position coordinates of the target object relative to the hull based on the target object visual information obtained by the machine vision module and the hull posture information.

[0053] Using the above technical solution, the positioning calculation unit set up adopts an algorithm model based on deep learning. This algorithm model can effectively process the complex visual information obtained by the machine vision module 5, and perform calculations in combination with the hull posture information. It can fully tap the effective features in the visual information and reduce the impact of environmental interference on positioning, thereby improving the calculation accuracy of the spatial position coordinates of the target object relative to the hull and realizing accurate spatial positioning of the target object.

[0054] Specifically, in this embodiment, the stabilizing platform 2 includes a hollow support platform 201, a support plate 202, a level sensor 203 and a leveling mechanism 205. The hollow support platform 201 is hinged to the connecting seat 6, and the hollow support platform 201 is fixedly connected to the double hull structure or the triple hull structure. The support plate 202 is fixedly installed at the upper opening of the hollow support platform 201 through a sealing elastic rubber frame 204. Under the action of the sealing elastic rubber frame 204, the support plate 202 can float up and down relative to the hollow support platform 201. The level sensor 203 is fixedly installed at the bottom of the support plate 202. The leveling mechanism 205 is provided with four groups, and the four groups of leveling mechanisms 205 are evenly distributed between the bottom of the support plate 202 and the inner bottom wall of the hollow support platform 201, wherein:

[0055] The level sensor 203 is used to monitor the horizontal state of the support plate 202 in real time, and the level sensor 203 is electrically connected to the control module. The control module can control the leveling mechanism 205 to automatically adjust the level of the support plate 202 according to the signal data of the level sensor 203.

[0056] By adopting the above technical solution, in the stable platform 2, the hollow support platform 201 is hinged to the connecting seat 6, providing a movable basis for the leveling of the platform; the support plate 202 is installed on the hollow support platform 201 through the sealing elastic rubber frame 204, and can float relatively under the action of the sealing elastic rubber frame 204, which can buffer part of the vibration of the hull; the horizontal sensor 203 monitors the horizontal state of the support plate 202 in real time, and transmits the signal to the control module, and the control module controls the four sets of leveling mechanisms 205 for horizontal adjustment, which can correct the inclination of the support plate 202 in time, ensure that the support plate 202 always remains in a horizontal state, reduce the impact of the hull shaking on the grasping device installed on the support plate 202, and improve the working stability and positioning accuracy of the grasping device.

[0057] Specifically, in this embodiment, four groups of leveling mechanisms 205 are provided, and the four groups of leveling mechanisms 205 are evenly distributed between the support plate 202 and the upper part of the inner bottom wall of the hollow support platform 201. Each group of leveling mechanisms 205 includes an electric push rod 2051 and a connecting rod 2052. The electric push rod 2051 is vertically and fixedly installed on the upper part of the inner bottom wall of the hollow support platform 201, and the connecting rod 2052 is vertically and fixedly installed on the bottom of the support plate 202. The bottom end of the connecting rod 2052 is hinged to the piston rod end of the electric push rod 2051 through a ball hinge 2053, and the electric push rod 2051 is electrically connected to the control module.

[0058] By adopting the above-mentioned technical solution, the four groups of electric push rods 2051 and connecting rods 2052 of the leveling mechanism 205 are evenly distributed between the support plate 202 and the hollow support platform 201. The electric push rod 2051 can accurately adjust the length by telescoping. The bottom end of the connecting rod 2052 is hinged to the electric push rod 2051 through a ball hinge 2053, allowing a certain angle of rotation to adapt to the angle change during the leveling process; the four groups of evenly distributed structures can support and adjust the support plate 202 from multiple directions. Combined with the precise telescopic control of the electric push rod 2051, it can quickly and accurately respond to the signal of the horizontal sensor 203, efficiently realize the horizontal adjustment of the support plate 202, and enhance the leveling effect and response speed of the stable platform 2.

[0059] Specifically, in this embodiment, the grasping device includes a robotic arm 3 and an end effector 4. The robotic arm 3 has multiple degrees of freedom. The end effector 4 is a clamping claw or suction cup structure for grasping the target object, and the end effector 4 and the robotic arm 3 are both electrically connected to the control module.

[0060] By adopting the above technical solution, the robotic arm 3 in the grasping device has multiple degrees of freedom, and can flexibly adjust the position and posture of the end effector 4 to adapt to the grasping needs of targets at different positions and angles; the end effector 4 adopts a clamping claw or suction cup structure, and can select a suitable grasping method according to the shape, material and other characteristics of the target object, thereby improving the adaptability and grasping success rate of the grasping device to different types of targets and enhancing the operational flexibility of the system.

[0061] Specifically, in this embodiment, the motion control of the robotic arm 3 adopts an adaptive control algorithm of a built-in control module to adjust the motion trajectory of the robotic arm in real time according to the hull shaking and the position of the target object.

[0062] Using the above technical solution, the motion control of the robotic arm 3 adopts the adaptive control algorithm of the built-in control module. This algorithm can sense the shaking of the hull in real time and dynamically adjust the motion trajectory of the robotic arm 3 according to the position of the target object. It can offset the interference of the hull shaking on the movement of the robotic arm 3, ensure that the robotic arm 3 can accurately reach the target object position, and improve the stability and accuracy of the grasping process.

[0063] Specifically, in this embodiment, a communication module electrically connected to the control module is further included, and the communication module is used to transmit the position information and the grasping status information of the unmanned boat to the remote control center.

[0064] By adopting the above technical solution, the communication module is electrically connected to the control module, which can transmit the location information and grasping status information of the unmanned ship to the remote control center, so that the remote control center can grasp the operating status of the system in real time, facilitate remote monitoring and scheduling of the operating process of the unmanned ship, and improve the system's remote collaboration capability and operation controllability.

[0065] Specifically, in this embodiment, a data storage module electrically connected to the control module is also included, and the data storage module is used to store image data collected by the machine vision module, calculation data of the positioning calculation unit, and operation data of the grasping device.

[0066] By adopting the above technical solution, the data storage module is electrically connected to the control module, and can store the image data collected by the machine vision module 5, the calculation data of the positioning calculation unit, and the operation data of the grasping device. These data provide a basis for subsequent analysis, troubleshooting, algorithm optimization, etc. of the system, which helps to continuously improve the system performance and enhance the system's maintainability and long-term reliability.

[0067] Specifically, in this embodiment, in order to better cooperate with the grasping device to perform precise grasping work, the machine vision module 5 is installed near the connection between the robotic arm 3 and the end effector 4, and a battery for powering the unmanned boat grasping and positioning system can be installed in the hollow support platform 201.

[0068] The positioning calculation unit uses a deep learning-based algorithm model that combines the target visual information obtained by the machine vision module 5 with the ship's posture information to accurately calculate the target's spatial position coordinates relative to the ship. The specific solution is as follows:

[0069] (1) Model architecture design

[0070] Feature Extraction Subnetwork: Utilizing a modified YOLO (You Only Look Once) network as its foundational framework, this subnetwork receives target image data (including features such as shape, color, and outline) processed by the image processor in Machine Vision Module 5. Through multi-layer convolution operations, it gradually extracts both shallow features (such as edges and textures) and deep semantic features (such as the overall shape and key areas of the target) from the image. Furthermore, an attention mechanism is introduced to enhance feature capture in key target areas while minimizing interference from water reflections and background debris.

[0071] The attitude fusion subnetwork takes in ship attitude information (such as roll, pitch, and heading) and preprocesses it through a fully connected layer, converting it into a vector that matches the feature dimensions of the feature extraction subnetwork output. The attitude vector is then fused with the target feature vector output by the feature extraction subnetwork using a concatenation method, allowing the model to simultaneously focus on the correlation between the target's visual features and the changes in the ship's attitude.

[0072] Coordinate regression subnetwork: Taking the fused feature vector as input, it performs nonlinear mapping through multiple fully connected layers and activation functions (such as ReLU), ultimately outputting the 3D spatial coordinates (X, Y, and Z coordinates) of the target object relative to the hull. To improve regression accuracy, a smooth L1 loss function is used in the output layer to reduce the impact of outliers on model training.

[0073] (2) Training process

[0074] Dataset Construction: We collected image data of objects in various aquatic environments (e.g., calm lakes, wavy waters, and strong / low light conditions). The high-definition pan-tilt camera in Machine Vision Module 5 captured the images from multiple angles and distances, while simultaneously recording the corresponding ship's posture information (obtained via the ship's onboard inertial measurement unit). We annotated the objects in the images, determined their actual spatial coordinates within the ship's coordinate system, and constructed a training dataset consisting of "image-posture-coordinate" triplets.

[0075] Preprocessing: Perform data enhancement operations such as scaling, cropping, rotation, and brightness adjustment on the image data to expand the dataset size and improve the generalization ability of the model; normalize the hull posture information and coordinate data and map them to the [0,1] interval to facilitate rapid convergence of the model.

[0076] Training Strategy: The Adam optimizer was used, with an initial learning rate of 0.001. The learning rate was adjusted using a learning rate decay strategy (e.g., decaying to 0.1 times the original value every 10 epochs). During training, data was input in batches, and model parameters were updated via backpropagation until the loss function converged to a stable value. Early stopping was also used to prevent overfitting; training was terminated when the validation set loss did not decrease for five consecutive epochs.

[0077] (3) Reasoning process

[0078] Real-time input: The machine vision module 5 collects the target object image in real time and processes it through the image processor, and then inputs the image data and the synchronously acquired hull posture information into the trained model.

[0079] Feature fusion and calculation: The feature extraction subnetwork extracts features from the image, the posture fusion subnetwork preprocesses the hull posture information and fuses it with the image features, and the coordinate regression subnetwork calculates the spatial position coordinates of the target object relative to the hull based on the fused features. The entire process is efficiently executed in the positioning calculation unit to meet real-time requirements.

[0080] Result output: The calculated coordinate data is transmitted to the control module to provide accurate positioning basis for the motion control of the robotic arm 3 of the grasping device.

[0081] (4) Model optimization strategy

[0082] Incremental learning: Regularly use newly collected operation data (stored by the data storage module) to fine-tune the model to adapt it to new target types or environmental changes, and continuously improve positioning accuracy.

[0083] Lightweight design: Through methods such as model pruning (removing redundant convolution kernels) and knowledge distillation (using complex models to guide simple model training), the number of model parameters is reduced, the inference speed is improved, and efficient operation is ensured within the hardware resource constraints of the positioning computing unit.

[0084] Dynamic calibration: Combined with the horizontal state information of the support plate 202 fed back by the horizontal sensor 203 of the stable platform 2, the output coordinates are dynamically calibrated during the inference process to offset the slight position deviation that may occur during the leveling process of the stable platform.

[0085] The thrust distribution of the leveling mechanism 205 satisfies the following distributed cooperative control equation:

[0086]

[0087] in:

[0088] F i is the output thrust of the electric push rod 2051 of the i-th group of leveling mechanisms 205 (i=1, 2, 3, 4);

[0089] θ err The tilt angle error (radian) of the support plate 202 detected in real time by the horizontal sensor 203;

[0090] θ j is the local tilt angle corresponding to the jth group of leveling mechanisms (converted by the displacement of the electric push rod);

[0091] k is the leveling gain coefficient (range 500–2000 N / rad), which is negatively correlated with the hull rolling frequency;

[0092] β is the cooperative coupling factor (range 0.2–0.5) used to balance global error and local consistency.

[0093] Example:

[0094] Input: The horizontal sensor detects the tilt angle error of the support plate θ err =0.02rad (about 1.15°).

[0095] Parameter setting: set k = 800 N / rad, β = 0.3 according to the current wave frequency (such as 2 Hz).

[0096] calculate:

[0097] If the local inclination angle θ4 at the No. 4 push rod is 0.018 rad, then its thrust is:

[0098] F4=800×(0.02+0.3×(0.02-0.018))=16.48N.

[0099] Execution: The control module drives electric push rod No. 4 to output a thrust of 16.48N, and the other push rods are calculated and executed synchronously.

[0100] The technical effect of this equation is:

[0101] Multi-executor coordination: through ∑(θ err -θ j ) item realizes dynamic load balancing of four groups of push rods to avoid single point overload.

[0102] Improved response speed: The tilt error convergence time is significantly shortened (compared to traditional PID control), especially suitable for high-frequency wave interference.

[0103] Energy consumption optimization: The push rod output is nonlinearly matched with the real-time error, which reduces invalid stroke and significantly reduces power consumption.

[0104] In summary, the working principle of the unmanned vessel grasping and positioning system based on the multi-body configuration design provided in this embodiment is as follows:

[0105] 1. Foundation support: The double or triple hull structure of the multihull body 1 provides stable buoyancy and support when sailing on the water, reducing hull swaying and providing a stable installation foundation for components such as the machine vision module 5, the stabilizing platform 2, and the gripping device.

[0106] 2. Target recognition and positioning: The high-definition pan-tilt camera in the machine vision module 5 captures the target image. After being processed and feature extracted by the image processor, the visual information is transmitted to the positioning calculation unit. The positioning calculation unit combines the hull posture information with a deep learning algorithm model to calculate the precise spatial position coordinates of the target relative to the hull.

[0107] 3. Stability guarantee: The horizontal sensor 203 of the stabilizing platform 2 monitors the horizontal state of the support plate 202 in real time. If tilt occurs, the control module controls the extension and retraction of the electric push rods 2051 of the four sets of leveling mechanisms 205 according to the sensor signal, and adjusts the position of the support plate 202 through the cooperation of the connecting rod 2052 and the ball hinge 2053. Under the cushioning effect of the sealing elastic rubber frame 204, the support plate 202 is ensured to be always horizontal, reducing the impact of the hull shaking on the grasping device.

[0108] 4. Grasping execution: The control module controls the movement of the multi-degree-of-freedom robotic arm 3 of the grasping device based on the coordinates output by the positioning calculation unit. The robotic arm 3 adjusts its motion trajectory in real time through an adaptive control algorithm to offset the interference of the hull shaking. The end effector 4 selects the gripper or suction cup method to complete the grasping according to the characteristics of the target object.

[0109] 5. Auxiliary functions: The communication module transmits information such as the position and grasping status of the unmanned boat to the remote control center to achieve remote monitoring; the data storage module records image data, positioning data and operation data to provide a basis for system optimization.

[0110] How to use

[0111] 1. System startup: Start the multihull 1 and the power supply of each module. The control module initializes the machine vision module 5, the stabilization platform 2, the gripping device, the communication module and the data storage module.

[0112] 2. Arriving at the operating area: Maneuver the multihull body 1 to sail to the waters where the target object is located. The stable structure of the multihull ensures the overall stability of the system during navigation.

[0113] 3. Target recognition and positioning: The machine vision module 5 starts the high-definition pan-tilt camera, adjusts the angle to capture the target image, and the image processor transmits the feature information to the positioning calculation unit after processing, calculates the spatial coordinates of the target object and feeds it back to the control module.

[0114] 4. Stability adjustment: The level sensor 203 of the stable platform 2 monitors in real time. If the support plate 202 tilts, the control module automatically controls the electric push rod 2051 of the leveling mechanism 205 to extend and retract, and quickly adjusts the support plate 202 to the horizontal.

[0115] 5. Grasping operation: The control module drives the robot arm 3 of the grasping device to move according to the positioning coordinates. The robot arm 3 adjusts its trajectory to adapt to the hull shaking through an adaptive algorithm. The end effector 4 switches between gripper and suction cup mode according to the characteristics of the target object to complete the grasping action.

[0116] 6. Operation monitoring and data recording: The communication module transmits the operation status to the remote control center in real time, and the operator can monitor remotely; the data storage module synchronously records various data during the operation. After the operation is completed, the system can be shut down or the next round of operation can be carried out.

[0117] The above are only preferred embodiments of the present invention and do not limit the implementation mode and protection scope of the present invention. For those skilled in the art, it should be aware that all solutions obtained by equivalent substitutions and obvious changes made using the description and illustrations of the present invention should be included in the protection scope of the present invention.

Claims

1. An unmanned vessel grasping and positioning system based on a multi-body configuration design, characterized in that: include: Multihull (1), consisting of a double hull structure or a triple hull structure; A machine vision module (5), used for identifying a target object and obtaining visual information of the target object; A positioning calculation unit, connected to the machine vision module, for calculating the spatial position coordinates of the target object based on the acquired visual information; A stabilizing platform (2) is provided on the multihull body (1); A grasping device, mounted on the stable platform (2), for grasping the target object according to the coordinates calculated by the positioning calculation unit; The control module is electrically connected to the machine vision module (5), the stabilizing platform (2) and the grasping device, respectively, and is used to control the automated operation of the machine vision module (5), the stabilizing platform (2) and the grasping device.

2. The unmanned ship grasping and positioning system based on multi-body configuration design according to claim 1 is characterized in that: The stabilizing platform (2) is installed between the double hull structure or the triple hull structure via a plurality of connecting seats (6), the connecting seats (6) are fixedly connected to the double hull structure or the triple hull structure, and the connecting seats (6) are fixedly connected to the stabilizing platform (2).

3. The unmanned ship grasping and positioning system based on multi-body configuration design according to claim 1 is characterized in that: The machine vision module (5) includes a high-definition pan-tilt camera and an image processor, wherein the high-definition pan-tilt camera is electrically connected to the control module and is used to collect images of the target object, and the image processor is electrically connected to the high-definition pan-tilt camera and is used to process and extract features from the collected images.

4. The unmanned vessel grasping and positioning system based on multi-body configuration design according to claim 1 is characterized in that: The positioning calculation unit adopts an algorithm model based on deep learning to calculate the precise spatial position coordinates of the target object relative to the hull based on the target object visual information obtained by the machine vision module and the hull posture information.

5. The unmanned ship grasping and positioning system based on multi-body configuration design according to claim 2 is characterized in that: The stabilizing platform (2) comprises a hollow supporting platform (201), a supporting plate (202), a level sensor (203) and a leveling mechanism (205); the hollow supporting platform (201) is hinged to the connecting seat (6), and the hollow supporting platform (201) is fixedly connected to the double hull structure or the triple hull structure; the supporting plate (202) is fixedly mounted on the upper opening of the hollow supporting platform (201) through a sealing elastic rubber frame (204); under the action of the sealing elastic rubber frame (204), the supporting plate (202) can float up and down relative to the hollow supporting platform (201); the level sensor (203) is fixedly mounted on the bottom of the supporting plate (202); and the leveling mechanism (205) is provided in four groups, and the four groups of the leveling mechanism (205) are evenly distributed between the bottom of the supporting plate (202) and the inner bottom wall of the hollow supporting platform (201), wherein: The horizontal sensor (203) is used to monitor the horizontal state of the support plate (202) in real time, and the horizontal sensor (203) is electrically connected to a control module, and the control module can control the leveling mechanism (205) to automatically adjust the horizontal state of the support plate (202) according to signal data from the horizontal sensor (203).

6. The unmanned ship grasping and positioning system based on multi-body configuration design according to claim 5 is characterized in that: The leveling mechanism (205) is provided with four groups, and the four groups of the leveling mechanism (205) are evenly distributed between the support plate (202) and the upper part of the inner bottom wall of the hollow support platform (201). Each group of the leveling mechanism (205) includes an electric push rod (2051) and a connecting rod (2052). The electric push rod (2051) is vertically and fixedly installed on the upper part of the inner bottom wall of the hollow support platform (201). The connecting rod (2052) is vertically and fixedly installed on the bottom of the support plate (202), and the bottom end of the connecting rod (2052) is hinged to the piston rod end of the electric push rod (2051) through a ball hinge (2053). The electric push rod (2051) is electrically connected to the control module.

7. The unmanned vessel grasping and positioning system based on multi-body configuration design according to claim 1 is characterized in that: The grasping device comprises a robotic arm (3) and an end effector (4), wherein the robotic arm (3) has multiple degrees of freedom, and the end effector (4) is a gripper or suction cup structure for grasping a target object, and the end effector (4) and the robotic arm (3) are both electrically connected to a control module.

8. The unmanned ship grasping and positioning system based on multi-body configuration design according to claim 7 is characterized in that: The motion control of the robotic arm (3) adopts an adaptive control algorithm of a built-in control module, and adjusts the motion trajectory of the robotic arm in real time according to the swaying of the hull and the position of the target object.

9. The unmanned ship grasping and positioning system based on multi-body configuration design according to claim 1 is characterized in that: It also includes a communication module electrically connected to the control module, and the communication module is used to transmit the position information and grasping status information of the unmanned boat to the remote control center.

10. The unmanned ship grasping and positioning system based on multi-body configuration design according to claim 1 is characterized in that: It also includes a data storage module electrically connected to the control module, and the data storage module is used to store image data collected by the machine vision module, calculation data of the positioning calculation unit, and operation data of the grasping device.

Images