An automated loading system for an inland river bulk cargo terminal

Abstract

The application discloses an automatic loading system for an inland river bulk cargo wharf, breaks through the bottleneck of high dependence on manpower, large efficiency fluctuation and many safety risks in traditional bulk cargo operation through multi-technology fusion and system-level collaboration, and constructs an intelligent system for bulk cargo loading and unloading process. Through the application of key technologies such as three-dimensional laser scanning, GPS positioning, digital twinning, intelligent calculation, PLC control, 5G communication and material level detection, the automatic loading of bulk cargo is changed from manual experience driving to data intelligent driving, and core technical support is provided for the automation of the loading and unloading process of the inland river bulk cargo wharf.

Description

Technical Field

[0001] This invention belongs to the field of automation technology and relates to an automated loading system for inland waterway bulk cargo terminals. Background Technology

[0002] With economic development, the volume of bulk cargo at domestic and international ports has surged, leading to the rapid development of bulk cargo terminals. As global trade grows rapidly and bulk cargo transportation volume increases dramatically, the construction and expansion of bulk cargo terminals will inevitably become a crucial pathway for major ports to grow stronger and enhance their competitiveness. With the advancement of globalization and trade liberalization, the throughput of inland waterway bulk cargo terminals continues to grow, placing increasingly higher demands on loading and unloading processes. Traditional inland waterway bulk cargo terminal loading and unloading processes rely primarily on manual labor and simple mechanical equipment, resulting in low efficiency, high labor costs, and numerous safety hazards. With the continuous growth of inland waterway transportation volume and intensified market competition, traditional loading and unloading processes are no longer sufficient to meet the demands of modern logistics. Therefore, researching the automation of inland waterway bulk cargo terminal loading and unloading processes is an important way to overcome the limitations of traditional processes.

[0003] In recent years, automation technology has made significant progress and has been widely applied in various fields. With its high efficiency, accuracy, and reliability, automation technology has provided new ideas for improving the loading and unloading processes at inland waterway bulk cargo terminals. By applying automation technology to inland waterway bulk cargo terminals, the loading and unloading process can be automated, intelligent, and unmanned, improving efficiency, reducing labor costs, and ensuring operational safety. The demands of enhancing service capabilities, green development, and smart ports all point to the specialization of terminal loading and unloading. Especially for new projects, port companies often face a shortage of existing operators, and recruiting and training new operators is time-consuming. To reach full production capacity as quickly as possible and improve efficiency, projects are considering remote automated control as much as possible to reduce manual labor. Research on the automation of bulk cargo loading has become a goal for many countries. Summary of the Invention

[0004] To address the problems existing in the background art, the present invention proposes an automated loading system for inland waterway bulk cargo terminals.

[0005] To achieve the above objectives, the technical solution adopted by the present invention is as follows: an automated loading system for inland bulk cargo terminals, comprising: a three-dimensional laser scanning module, a positioning and navigation module, a digital twin module, an intelligent computing module, and a level detection module;

[0006] The three-dimensional laser scanning module is used to collect three-dimensional point cloud data of the surface of the ship and the materials inside the cabin;

[0007] The positioning and navigation module is used to provide coordinated positioning and navigation for the ship loader and auxiliary equipment;

[0008] The digital twin module is used to construct a virtual model of the system and simulate the work process;

[0009] The intelligent computing module is used to optimize the loading and fabric placement strategies through machine learning algorithms.

[0010] The level detection module is used to monitor the status of materials inside the chamber in real time.

[0011] Specifically, the three-dimensional laser scanning module includes a lidar mounted on the boom of the ship loader. The lidar scans the ship's outline and hatch position during the ship-finding phase and scans the surface of the material pile during the loading phase to generate a dynamic three-dimensional model.

[0012] Specifically, the positioning and navigation module integrates the BeiDou satellite navigation system and an inertial measurement unit, and uses RTK differential technology to achieve centimeter-level positioning of the ship loader and multi-device collaborative path planning.

[0013] Specifically, the digital twin module integrates a physics engine and receives real-time data from the 3D laser scanning module and the level detection module for operation process simulation and abnormal state deduction.

[0014] Specifically, the intelligent computing module includes a lightweight algorithm model deployed on edge computing nodes and a training model deployed in the cloud. The algorithm model includes a dynamic obstacle avoidance algorithm, a PID control algorithm, and a machine learning-based material placement strategy optimization model.

[0015] Specifically, the loading system also includes a control execution module;

[0016] The control execution module includes a PLC controller, which receives instructions from the intelligent computing module and controls the actions of each mechanism of the ship loader. The control execution module supports safety interlocking and emergency stop protection.

[0017] Specifically, the loading system also includes a communication module;

[0018] The communication module uses 5G private network technology to provide a low-latency transmission channel for the point cloud data of the 3D laser scanning module, the positioning data of the positioning and navigation module, and control commands.

[0019] Specifically, the level detection module integrates lidar and visual recognition sensors, and uploads the monitored material status data to the digital twin module via the MQTT protocol.

[0020] Specifically, the loading system also includes a human-machine interface;

[0021] The human-machine interface integrates and displays the position and posture of the ship loader, the ship status and the operation progress information, and supports one-click start of the automated operation process.

[0022] Specifically, the loading system can also perform remote centralized control mode;

[0023] The remote centralized control mode automatically executes the entire process of ship search, cabin relocation, and cabin loading based on the received ship loading plan.

[0024] Compared with existing technologies, this invention has the following advantages: An automated ship loading system for inland bulk cargo terminals utilizes the synergistic effects of a 3D laser scanning module, a positioning and navigation module, a digital twin module, an intelligent computing module, a level detection module, a control and execution module, a communication module, a human-machine interface, and a remote centralized control mode. The 3D laser scanning module collects 3D point cloud data of the ship and the surface of materials inside the hold. The positioning and navigation module achieves centimeter-level positioning of the ship loader and collaborative path planning for multiple devices. The digital twin module constructs a virtual model and simulates the operation process. The intelligent computing module optimizes the loading and material placement strategies. The level detection module monitors the status of materials inside the hold in real time. The control and execution module precisely controls the actions of each mechanism of the ship loader and supports safety interlocks and emergency stop protection. The communication module provides a low-latency transmission channel. The human-machine interface integrates and displays key information and supports one-click start of the automated operation process. The remote centralized control mode automatically executes the entire process of ship finding, hold shifting, and loading. It effectively breaks through the bottlenecks of traditional bulk cargo operations, which are characterized by high reliance on manual labor, large efficiency fluctuations, and numerous safety risks. It promotes the transformation of bulk cargo loading from manual experience-driven to data-driven intelligence, significantly improves the automation and intelligence level of loading and unloading processes at inland waterway bulk cargo terminals, reduces labor costs, ensures operational safety and efficiency, and meets the modern logistics demand for professional, unmanned, and remote control of terminal loading and unloading. Attached Figure Description

[0025] Figure 1 This is a pre-scanned image of the ship type for the automated loading system of the present invention;

[0026] Figure 2 This is a diagram of the main interface of the automated ship loading system of the present invention;

[0027] Figure 3 This is a system operation screen diagram of the automated ship loading system of the present invention;

[0028] Figure 4 This is a diagram of the walking mechanism of the automated ship loading system of the present invention;

[0029] Figure 5 This is a diagram of the boom mechanism of the automated ship loading system of the present invention;

[0030] Figure 6 This is a diagram of the belt conveyor mechanism of the automated ship loading system of the present invention;

[0031] Figure 7 This is a diagram of the chute mechanism of the automated ship loading system of the present invention;

[0032] Figure 8 This is a diagram of the pitching mechanism of the automated ship loading system of the present invention. Detailed Implementation

[0033] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0034] like Figures 1-8 As shown, the technical solution adopted by the present invention is as follows: an automated loading system for inland bulk cargo terminals, comprising: a three-dimensional laser scanning module, a positioning and navigation module, a digital twin module, an intelligent computing module, and a level detection module.

[0035] The three-dimensional laser scanning module is used to collect three-dimensional point cloud data of the surface of the ship and the materials inside the cabin.

[0036] The positioning and navigation module is used to provide coordinated positioning and navigation for the ship loader and auxiliary equipment.

[0037] The digital twin module is used to build a virtual model of the system and simulate the work process.

[0038] The intelligent computing module is used to optimize loading and fabric placement strategies through machine learning algorithms.

[0039] The level detection module is used to monitor the status of materials inside the chamber in real time.

[0040] Specifically, the three-dimensional laser scanning module includes a lidar mounted on the boom of the ship loader. The lidar scans the ship's outline and hatch position during the ship-finding phase and scans the surface of the material pile during the loading phase to generate a dynamic three-dimensional model.

[0041] The lidar mounted on the ship loader boom is the core foundation for the module's functionality. By mounting the lidar on the boom, its mobility allows it to cover the critical spaces required for operations within the ship and its hold. This ensures a comprehensive scanning view of the ship's overall outline while also enabling precise detection of the material pile surface deep within the hold. This provides stable and flexible hardware support for subsequent scanning tasks at different stages, and is a prerequisite for the 3D laser scanning module to effectively acquire data.

[0042] The lidar scans the ship's outline and hatch positions during the ship-finding phase, a crucial preparatory step before loading operations begin. During this phase, the loader needs to pinpoint the ship's exact location and key hatch information. The lidar, by emitting laser beams and receiving reflected signals, accurately captures the ship's overall external shape, including its length, width, and hull outline features. It also accurately identifies the specific spatial location, size, and relative distances between each hatch. This acquired ship outline and hatch position data is transmitted in real-time to other modules of the system, serving as a critical basis for the loader's positioning adjustments and determining its initial operational position. This ensures the loader can accurately align with the target hatch, preventing operational efficiency issues or safety problems caused by positioning deviations.

[0043] The lidar system scans the surface of the material pile during the loading phase to generate a dynamic 3D model. This function is crucial for ensuring the accuracy and safety of the loading operation. During loading, as material continuously falls into the bin, the height, shape, and distribution of the material pile inside change constantly. The lidar continuously scans the surface of the material pile at high frequency, collecting spatial coordinate data of various points on the surface in real time. Based on this real-time data, the system dynamically constructs and updates the 3D model of the material pile. This model clearly and accurately reflects the actual state of the material pile at different times, such as the maximum height of the pile, densely packed areas of material, and areas with low-lying areas or gaps. This dynamic 3D model not only provides real-time data support for the intelligent computing module to optimize material placement strategies, helping to adjust the drop point position and drop volume to avoid localized collapses or uneven material distribution due to excessively high pile heights, but also provides accurate data references for operators or the system to judge the loading progress and calculate the loading volume, ensuring efficient and safe loading operations.

[0044] Specifically, the positioning and navigation module integrates the BeiDou satellite navigation system and an inertial measurement unit, and uses RTK differential technology to achieve centimeter-level positioning of the ship loader and multi-device collaborative path planning.

[0045] The positioning and navigation module integrates the BeiDou Navigation Satellite System and an inertial measurement unit (IMU). This integrated design is the core foundation for ensuring the continuity and stability of positioning. The BeiDou Navigation Satellite System provides all-weather, wide-coverage basic positioning signals for ship loaders and auxiliary equipment, while the IMU supplements positioning data by sensing the device's own motion state (such as acceleration and angular velocity) when obstructions such as port buildings and cranes interfere with or interrupt the BeiDou satellite signal, thus avoiding positioning interruption. The integration of the two forms a dual positioning guarantee of "satellite positioning + inertial blind spot compensation," ensuring that the positioning and navigation module can always provide reliable position data support for the system in complex port environments, laying a data foundation for subsequent accurate positioning and collaborative operations.

[0046] The positioning and navigation module employs RTK differential technology to achieve centimeter-level positioning of the ship loader, a key technological means to meet the high-precision requirements of automated ship loading operations. RTK differential technology involves setting up fixed ground-based augmentation stations (base stations) in the port area. These base stations receive BeiDou satellite signals in real time and calculate positioning errors, then transmit the error correction information to the positioning and navigation module (rover) on the ship loader via a communication link. The rover, combining its received BeiDou satellite signals with the correction information from the base station, uses differential calculations to eliminate interference factors such as satellite orbit errors and atmospheric refraction errors, improving the ship loader's positioning accuracy from meter-level to centimeter-level using conventional satellite positioning. This high-precision positioning ensures that the ship loader accurately aligns with the hatch position during ship search, hatch relocation, and loading processes, avoiding problems such as material spillage and equipment collisions caused by positioning deviations. It is the core precision guarantee for achieving automated and unmanned ship loading operations.

[0047] The positioning and navigation module employs RTK differential technology to achieve collaborative path planning for multiple devices, which is a crucial support for improving the operational efficiency and safety of the entire ship loading system. In automated ship loading operations, the equipment requiring collaborative work includes not only the ship loader but also auxiliary equipment such as loaders. All equipment must operate in an orderly manner within the same work area to avoid overlapping or collisions. The positioning and navigation module provides high-precision positioning data in a unified coordinate system for all collaborative equipment through RTK differential technology. Based on this real-time positioning data, the system can clearly understand the current position, direction of movement, and operational status of each device. Furthermore, the positioning and navigation module, in conjunction with the intelligent computing module, can generate collaborative path planning schemes for multiple devices, clearly defining the work area, route, and sequence of actions for each device. This ensures that the ship loader and auxiliary equipment do not interfere with each other during operation, achieving seamless integration of loading and material supply processes, effectively improving overall operational efficiency while mitigating safety risks in multi-device collaborative operations.

[0048] Specifically, the digital twin module integrates a physics engine and receives real-time data from the 3D laser scanning module and the level detection module for operation process simulation and abnormal state deduction.

[0049] The digital twin module integrates a physics engine, which is the core technological support for ensuring that the digital twin can accurately map the physical operation scenario. The physics engine has the ability to simulate real physical laws and can accurately calculate and restore physical phenomena such as material accumulation characteristics (e.g., angle of repose and flowability of bulk cargo) and equipment motion mechanics (e.g., gravity balance of the boom pitch of the ship loader and inertial effects of chute extension and retraction) during loading operations. This ensures that the constructed system virtual model is not only structurally consistent with the physical equipment, but also closely matches the actual operation situation in terms of motion and interaction logic, providing a virtual environment foundation that conforms to real physical rules for subsequent operation process simulation and abnormal state inference.

[0050] The digital twin module receives real-time data from the 3D laser scanning module and the level detection module. This data interaction mechanism is key to achieving the "real-time synchronization" characteristic of the digital twin. The real-time data provided by the 3D laser scanning module includes 3D point cloud data of the ship's outline, hatch positions, and material pile surface, dynamically reflecting the ship's berthing status and changes in the spatial morphology of the material pile within the hold. The real-time data provided by the level detection module includes the real-time pose and stacking height of the materials within the hold, accurately reflecting the material loading process. By continuously receiving and parsing these two types of real-time data, the digital twin module can update the status of elements such as the ship, equipment, and materials in the system's virtual model in real time, ensuring dynamic consistency between the virtual model and the physical work site and preventing a disconnect between virtual simulation and actual operation due to data lag.

[0051] The digital twin module is used for work process simulation, a function that plays a crucial role before the actual loading operation begins or during the work plan optimization phase. Based on an integrated physics engine and a real-time synchronized virtual model, the digital twin module can completely simulate the entire work process from ship location, cargo handling, cargo entry, to loading and cargo transfer, clearly presenting the action sequence of each piece of equipment, the flow path of materials, and the progress of the operation. Through work process simulation, unreasonable aspects in the process can be identified in advance, such as delays in equipment action connections and overlapping material placement paths. Based on this, work parameters and steps can be optimized, reducing redundancy in the actual operation and improving the overall efficiency and consistency of the loading operation.

[0052] The digital twin module is used for anomaly simulation, a crucial means of ensuring safe loading operations and reducing accident risks. Based on a virtual model built from real-time data, the digital twin module can simulate various potential abnormal scenarios, such as the risk of collision between the loader boom and the ship's hatch, the potential collapse hazard caused by localized overloading of the cargo hold, and motion deviations caused by equipment sensor malfunctions. By simulating these abnormal states, the impact range and chain reactions when an anomaly occurs can be clearly identified in advance, helping technicians to pre-plan response strategies (such as equipment emergency stop trigger conditions and work path adjustment plans). If similar abnormal signs appear during actual operations, the system can quickly respond and execute the pre-planned solutions, minimizing the impact of abnormal states on operational safety and efficiency.

[0053] Specifically, the intelligent computing module includes a lightweight algorithm model deployed on edge computing nodes and a training model deployed in the cloud. The algorithm model includes a dynamic obstacle avoidance algorithm, a PID control algorithm, and a machine learning-based material placement strategy optimization model.

[0054] The intelligent computing module includes a lightweight algorithm model deployed on edge computing nodes and a training model deployed in the cloud. The edge computing nodes are located close to physical equipment such as the ship loader. The deployed lightweight algorithm model can quickly process real-time data collected by the equipment, including pose, flow rate, and obstacle information. It can complete local real-time decision-making without transmitting large amounts of data to the cloud, meeting the low-latency requirements for mechanism motion control and emergency obstacle avoidance in ship loading operations.

[0055] The cloud has stronger computing power support, and the deployed training model can be deeply trained and iteratively optimized based on massive historical operation data (such as different material loading records, ship cabin type adaptation data, anomaly handling cases, etc.), continuously improving the decision accuracy of the algorithm model. The optimized model parameters are then sent to edge computing nodes to ensure that the lightweight model always maintains efficient decision-making capabilities, forming a closed loop of "local real-time response + cloud continuous optimization".

[0056] The algorithm model includes a dynamic obstacle avoidance algorithm, which is crucial for ensuring the safe operation of ship loaders in complex working environments. During ship loading operations, the ship loader's boom, chute, and other mechanisms need to move within or around the ship's hold, and may face unexpected situations such as temporary obstacles within the hold (e.g., leftover hatch cover parts, maintenance tools) and overlapping work areas with other cooperating equipment (e.g., loaders).

[0057] The dynamic obstacle avoidance algorithm can receive equipment location data provided by the positioning and navigation module and environmental obstacle data captured by the 3D laser scanning module in real time, quickly calculate a safe avoidance path, and transmit adjustment instructions to the control execution module to drive the ship loader mechanism to adjust its movement trajectory in a timely manner to avoid collisions with obstacles or other equipment, thus ensuring operational safety and equipment integrity.

[0058] The algorithm model includes a PID control algorithm, which is mainly used to achieve precise control of the feeder flow rate. During ship loading operations, the feeder's material flow rate directly affects loading efficiency and stockpile stability. Excessive flow rate can lead to rapid material accumulation and overflow in the hold, while insufficient flow rate will prolong loading time and reduce operational efficiency. The PID control algorithm collects real-time actual feeder flow data (e.g., detected by a belt scale), compares it with the target flow rate preset by the intelligent calculation module, calculates the flow deviation, and then automatically adjusts the feeder's operating parameters (e.g., vibration frequency, conveyor belt speed) according to proportional (P), integral (I), and derivative (D) control laws. This continuously corrects the flow deviation, ensuring the actual feed flow rate remains stably within the target range, thus guaranteeing the continuity and stability of material supply during the loading process.

[0059] The algorithm model includes a machine learning-based material placement strategy optimization model, which is the core support for improving loading quality and meeting ship stowage requirements. Based on historical loading operation data, this model covers the stacking characteristics of different materials (such as coal and ore), the structural parameters of various ship compartments, and the material distribution effects corresponding to different loading strategies. It is trained using machine learning algorithms (such as neural networks and decision trees) to form a material placement strategy model adaptable to various scenarios. In actual loading operations, this model receives real-time dynamic 3D models of the material pile inside the compartment generated by a 3D laser scanning module and real-time material pose data provided by a level detection module. Combined with information such as the ship's compartment capacity and material characteristics, it automatically optimizes the material drop point location, drop sequence, and placement path to ensure uniform material distribution within the compartment, avoiding material segregation (such as stratification of materials with different particle sizes) or localized overloading. This ensures that the loading quality meets the ship's representative's process requirements while reducing safety risks during subsequent ship navigation.

[0060] Specifically, an automated loading system for inland bulk cargo terminals also includes a control execution module.

[0061] The control execution module includes a PLC controller, which receives instructions from the intelligent computing module and controls the actions of each mechanism of the ship loader. The control execution module supports safety interlocking and emergency stop protection.

[0062] The control execution module includes a PLC controller. As the hardware core of the control execution module, the PLC controller possesses high reliability, strong anti-interference capabilities, and real-time response characteristics. It can adapt to the harsh operating environment of inland waterway bulk cargo terminals, such as dust, humidity, and vibration, and operate stably. Its hardware characteristics and adaptability to the terminal operating environment provide a fundamental guarantee for the continuous and reliable functioning of the control execution module, serving as the hardware foundation for its operation.

[0063] The PLC controller receives instructions from the intelligent computing module and controls the actions of various mechanisms of the ship loader. The specific process is as follows: The intelligent computing module generates ship loading strategies (such as shifting paths, material drop point adjustments, and feed rate control) and mechanism action instructions based on multi-source data (such as point cloud data from the 3D laser scanning module and material status data from the level detection module). These instructions are transmitted to the PLC controller via a system communication link (such as a 5G private network). The PLC controller parses and processes the received instructions, generates corresponding control signals, and sends them to various actuators of the ship loader (such as the traveling mechanism, boom mechanism, chute mechanism, and belt mechanism). This drives each mechanism to complete specified actions, such as controlling the traveling mechanism to move along the track to the target hatch position, controlling the boom to adjust the material drop height, controlling the chute to extend and retract to align with the target material drop point inside the hatch, and controlling the belt mechanism to adjust the material conveying speed. This achieves precise conversion between "decision instructions" and "mechanism actions," ensuring that the ship loading operation is strictly executed according to the optimized strategy.

[0064] The control execution module supports safety interlock protection. This safety interlock protection function is achieved through real-time acquisition of status signals (such as limit signals, load signals, and equipment operating temperature signals) and environmental signals (such as collision detection signals and wind speed signals) from various mechanisms of the ship loader via the PLC controller. When a signal triggers a safety threshold (such as the boom extension exceeding its limit, the distance between the chute and the bulkhead approaching the collision threshold, the equipment load exceeding its rated value, or the wind speed exceeding the safe operating range), the safety interlock mechanism will immediately activate. The PLC controller will automatically prevent the relevant mechanisms from continuing to perform dangerous actions and simultaneously feed back the abnormal signal to the human-machine interface or remote control center, alerting the operator to the anomaly and preventing equipment damage, material spillage, or safety accidents caused by malfunctions or over-limit operation, thus ensuring the safety of the equipment and the environment during operation.

[0065] The control execution module supports emergency stop protection, a crucial safety mechanism for handling sudden emergencies such as equipment failure, personnel accidentally entering hazardous work areas, or the risk of material collapse within the cargo hold. When operators at the remote control center or on-site inspectors detect an emergency, they can send an emergency stop signal to the control execution module via the emergency stop button on the human-machine interface, the local emergency stop switch on the ship loader, or the emergency stop command channel of the remote control system. Upon receiving the emergency stop signal, the control execution module immediately cuts off the power supply to all actuators of the ship loader, forcibly stopping all mechanism actions and locking the current equipment state to prevent continued operation and potential escalation of the accident during an emergency. After the emergency is resolved and the fault is diagnosed and repaired, the operator can unlock the equipment state via a "reset" operation, allowing the control execution module to resume receiving commands and driving the mechanisms to operate normally, maximizing the safety of personnel, equipment, and cargo.

[0066] Specifically, an automated loading system for inland bulk cargo terminals also includes a communication module.

[0067] The communication module uses 5G private network technology to provide a low-latency transmission channel for the point cloud data of the 3D laser scanning module, the positioning data of the positioning and navigation module, and control commands.

[0068] The communication module employs 5G private network technology, which features ultra-low latency, high reliability, and flexible deployment, making it suitable for the complex environments of inland waterway bulk cargo terminals, characterized by high dust levels, dense equipment, and challenging operational scenarios. Through network slicing technology, the communication module can allocate dedicated network resources for different types of data transmission within the system, avoiding interference between different data flows, ensuring the priority of critical data and command transmissions, and guaranteeing the stability of the communication link even under scenarios with multiple devices operating concurrently and large amounts of data interaction. This lays the technical foundation for low-latency transmission of subsequent data and commands.

[0069] The communication module provides a low-latency transmission channel for the point cloud data of the three-dimensional laser scanning module. The three-dimensional laser scanning module generates ship three-dimensional point cloud data during the ship search phase and dynamic three-dimensional point cloud data of the material pile during the loading phase. The data volume is huge and the real-time requirements are extremely high.

[0070] During the ship-finding phase, point cloud data needs to be transmitted in real time to quickly determine the hatch coordinates. During the loading phase, data needs to be transmitted in real time to dynamically update the stockpile model and adjust the material drop points. The communication module utilizes the low latency characteristics of the 5G private network to quickly and completely transmit this point cloud data to the digital twin module and the intelligent computing module, avoiding ship positioning deviations and stockpile model lags caused by data transmission delays, which would affect the accuracy and efficiency of the loading operation.

[0071] The communication module provides a low-latency transmission channel for the positioning data of the positioning and navigation module. The centimeter-level positioning data of the ship loader and the collaborative positioning data of multiple devices generated by the positioning and navigation module are the core basis for the ship loader's path planning and the avoidance of overlapping areas by multiple devices. The communication module transmits this positioning data in real time to the intelligent computing module and the control execution module through a 5G private network. This allows the intelligent computing module to optimize path planning based on the latest positioning data, and the control execution module to adjust the actions of the ship loader mechanism based on real-time positioning data. This avoids positioning deviations caused by positioning data transmission delays, which could lead to equipment collisions, overlapping work areas, and other problems, thus ensuring the safety of precise navigation of mobile devices and collaborative operation of multiple devices.

[0072] The communication module provides a low-latency transmission channel for control commands, including mechanism action commands (such as shifting, pitching, and telescopic commands) issued by the intelligent computing module to the control execution module (PLC controller), emergency stop commands, and mode switching commands issued by the remote control center. Leveraging the low-latency characteristics of the 5G private network, the communication module ensures that these control commands are transmitted to the execution end in real time. Low-latency transmission of mechanism action commands ensures synchronized operation of all mechanisms of the ship loader, improving material unloading accuracy. Low-latency transmission of emergency stop commands allows the equipment to respond quickly and stop in the event of sudden anomalies (such as anti-collision alarms or overloads), minimizing safety risks and ensuring operational safety and reliable equipment operation.

[0073] Specifically, the level detection module integrates lidar and visual recognition sensors, and uploads the monitored material status data to the digital twin module via the MQTT protocol.

[0074] The level detection module integrates lidar and visual recognition sensors. This fusion design is key to ensuring the accuracy and comprehensiveness of material status monitoring within the hold. LiDAR, with its non-contact measurement characteristics, can accurately capture data on the height, volume, and spatial distribution of materials within the hold, even in harsh operating environments such as dusty and humid conditions at the dock, unaffected by environmental interference.

[0075] Visual recognition sensors can assist in identifying the surface condition of materials, such as the presence of agglomerates and surface unevenness caused by differences in particle size. The integration of these two technologies forms a dual monitoring capability of precise quantification and detailed identification, compensating for the limitations of single sensors in complex environments. This ensures that the material state data acquired by the level detection module is both accurate and comprehensive, providing a reliable data foundation for subsequent system decisions.

[0076] The level detection module uploads the monitored material status data to the digital twin module via the MQTT protocol. As a lightweight IoT communication protocol, MQTT features low bandwidth consumption, high reliability, and strong adaptability, making it suitable for network fluctuations that may occur at inland waterway bulk cargo terminals. Using this protocol ensures data transmission stability while avoiding excessive network resource consumption and preventing conflicts with critical data transmissions such as point cloud data from the 3D laser scanning module and positioning data from the positioning and navigation module. This ensures efficient and uninterrupted transmission of material status data to the digital twin module, preventing data transmission delays or loss from causing a disconnect between the virtual model of the digital twin module and the actual material status.

[0077] The level detection module uploads the monitored material status data to the digital twin module, serving as the core data source for the digital twin module to dynamically update the virtual model. The digital twin module relies on real-time data to construct a virtual model synchronized with the physical operation scenario. The material status data uploaded by the level detection module (such as real-time material level height and changes in material accumulation patterns) allows the digital twin module to adjust the virtual representation of materials within the cargo hold in a timely manner, ensuring that the virtual model accurately reflects the actual material status. This data interaction mechanism provides the digital twin module with real-time and accurate material status data for conducting operational process simulations, abnormal scenario deductions, and for the intelligent computing module to optimize loading and material placement strategies, ensuring that the decisions and execution of the entire automated loading system always align with actual operational conditions.

[0078] Specifically, an automated loading system for inland bulk cargo terminals also includes a human-machine interface.

[0079] The human-machine interface integrates and displays the position and posture of the ship loader, the ship status and the operation progress information, and supports one-click start of the automated operation process.

[0080] The human-machine interface integrates and displays the ship loader's position and posture information, including key data such as the position of the ship loader's trolley, pitch angle, extension length, and rotation angle. This data directly reflects the real-time position and movement status of each mechanism of the ship loader, and is the core basis for operators to judge whether the ship loader is aligned with the target hatch and whether it is within the safe operating range.

[0081] The human-machine interface integrates and presents this information in a graphical way (such as simulating the three-dimensional posture of the ship loader and displaying specific values ​​digitally), avoiding operators from switching between multiple independent interfaces to query, ensuring that they can quickly and comprehensively grasp the current status of the ship loader, and providing data support for operation monitoring and anomaly judgment.

[0082] The human-machine interface integrates and displays ship status information, including hatch coordinates, material level in the hold, and ship berthing position verification results. Hatch coordinates serve as the target reference for the loader's shifting and loading operations; material level reflects the progress of material loading within the hold; and ship berthing position verification results determine whether the ship is in the correct working position before operation. The human-machine interface centrally displays this key information related to the work object, allowing operators to intuitively understand the ship's current status, promptly identify issues such as ship position deviations and abnormal material accumulation in the hold (e.g., localized overloading), provide a basis for intervention and adjustment, and ensure operational accuracy.

[0083] The human-machine interface integrates and displays operation progress information, including data such as the completed loading volume of a single hold, the total loading volume, and the percentage of completed volume to total volume. This data intuitively reflects the progress of the loading operation, helping operators grasp the work rhythm and determine whether the loading task can be completed as planned. Through integrated display, operators can understand the operation progress in real time without manually calculating or querying other systems, facilitating timely coordination and adjustments (such as optimizing the feed rate) when the operation is lagging behind, thus ensuring operational efficiency.

[0084] The human-machine interface supports one-click startup of automated work processes, a key design feature that simplifies operation steps and reduces the complexity of manual operations. After the system receives the vessel loading plan (e.g., the control center issues the plan via TOS) and verifies the vessel's berthing position, operators no longer need to trigger actions such as vessel search, cabin transfer, and loading step by step. They only need to click the "One-Click Start" button on the human-machine interface, and the system can autonomously execute the entire automated operation from vessel search → cabin transfer → loading → cabin transfer. This function not only reduces the tedium and risk of error in manual operations but also quickly responds to operation startup requests, meeting the needs of port enterprises to reduce manpower and improve operational efficiency. Furthermore, it is compatible with the system's remote control mode, supporting remote operators to efficiently manage multiple ship loaders.

[0085] Specifically, an automated loading system for inland bulk cargo terminals also has the capability to execute a remote centralized control mode.

[0086] The remote centralized control mode automatically executes the entire process of ship search, cabin relocation, and cabin loading based on the received ship loading plan.

[0087] The remote centralized control mode operates based on the received vessel loading plan. The vessel loading plan is the core basis for the remote centralized control mode's operation, containing key information such as the vessel's hatch layout (e.g., number of hatches, coordinates and dimensions of each hatch), target workload for each hatch, and material characteristics (e.g., bulk density, angle of repose). This information is typically distributed from the remote centralized control center to the automated loading system via the Terminal Operating System (TOS). After receiving the vessel loading plan, the system automatically verifies whether the vessel's current berthing position matches the planned position, ensuring consistency between the loading plan and the actual work, avoiding operational errors due to vessel berthing deviations, and laying an accurate data foundation for subsequent automated execution of the operation process.

[0088] The remote centralized control mode automatically executes the ship-finding operation, which is the initial step in the loading process. In this mode, the ship-finding process requires no manual intervention. The system drives the loader's trolley mechanism along the track while simultaneously controlling the lidar (3D laser scanning module) mounted on the loader's boom to scan the overall outline of the ship, identifying the location and size of the hatches and surrounding obstacles (such as hatch covers and pipes), and generating a 3D point cloud model of the ship. This 3D point cloud model is uploaded in real-time to the monitoring system at the remote centralized control center via a 5G private network. The system automatically parses and stores the hatch coordinates in the model, completing the ship-finding process and ensuring that subsequent operations can accurately align with the target hatch, providing a precise spatial reference for hatch relocation and loading operations.

[0089] The remote centralized control mode automatically executes the hatch relocation operation, which is typically performed after loading a single hatch is completed or when a switch of hatches is required. In remote centralized control mode, the system automatically calculates the precise coordinates of the next target hatch based on the ship's loading plan and the virtual ship model constructed by the digital twin module. It then generates a relocation path for the ship loader from its current position to the target hatch (path planning must avoid obstacles to ensure equipment safety). The generated relocation path is transmitted via a 5G private network to the PLC controller of the control execution module. The PLC controller drives the ship loader's traveling mechanism, boom mechanism, and other components to move precisely to the next target hatch position, completing the relocation operation. The entire process requires no manual operation of the mechanisms.

[0090] The remote centralized control mode automatically executes the loading operation, which is the core operational step. The loading operation in remote centralized control mode is dynamically executed based entirely on the ship's stowage plan and real-time data. Based on the target workload of the current hatch in the ship's stowage plan, combined with the dynamic 3D model of the material pile inside the hatch generated by the 3D laser scanning module and the material status data uploaded in real time by the level detection module, the intelligent calculation module automatically generates a loading strategy (such as a material distribution path of "from inside to outside, layered flattening"). The intelligent calculation module translates the loading strategy into action commands for each mechanism and sends them to the PLC controller via the 5G private network, controlling the coordinated actions of the ship loader's pitching mechanism, telescopic mechanism, chute mechanism, belt mechanism, etc., to achieve precise material placement.

[0091] Meanwhile, the system receives dynamic models of the stockpile and material status data in real time, and dynamically adjusts the drop point position and feed rate (such as adjusting the feeder flow rate through PID control algorithm) to ensure that the material distribution in the hatch is uniform and not overloaded, until the target workload of the current hatch is completed, realizing automated closed-loop control of the loading operation.

[0092] The remote centralized control mode automatically executes the entire process of ship location, cabin transfer, and loading, meaning that from the start of the operation to the completion of loading on a single ship, the entire process requires no continuous manual intervention from operators. Emergency handling is only required when the system malfunctions (such as collision avoidance alarms or overload warnings). This full-process automation not only significantly improves operational efficiency (avoiding delays and errors caused by manual operation) but also ensures the continuity of each operational stage (such as automatically triggering cabin transfer after ship location is completed, and automatically starting loading after cabin transfer is in place). At the same time, through unified remote centralized control management, it enables the coordinated scheduling of multiple ship loaders.

[0093] In a specific embodiment, an inland bulk cargo terminal uses an automated loading system to load coal onto a bulk carrier. The system first activates the remote control mode, and the remote control center issues a ship stowage plan (including the layout of the four hatches of the bulk carrier, the target workload of each hatch, and coal stacking characteristics data) through the TOS. The system automatically verifies that the ship's berthing position is consistent with the ship's position in the stowage plan, thus completing the pre-operation preparations.

[0094] The system's positioning and navigation module begins operation. This module integrates the BeiDou satellite navigation system and an inertial measurement unit, achieving centimeter-level positioning of the ship loader through RTK differential technology. It also provides collaborative path planning for the ship loader and the yard loader to avoid overlap in their operating areas. Simultaneously, the 3D laser scanning module is activated. Its lidar, mounted on the ship loader's boom, scans the overall outline of the ship and the positions of its four hatches during the ship-finding phase, generating 3D point cloud data of the ship and uploading it to the system.

[0095] The digital twin module receives the ship's 3D point cloud data from the 3D laser scanning module and integrates a physics engine to construct a system virtual model of the bulk carrier and the loader. At this time, the level detection module integrates lidar and visual recognition sensors to monitor the initial material status in the hold in real time (at this time, there is no material in the hold) and uploads the material status data to the digital twin module via the MQTT protocol. The digital twin module completes the operation process simulation based on this real-time data, and enters the formal operation stage after confirming that there are no abnormalities.

[0096] The intelligent computing module begins operation, with its lightweight algorithm model (including dynamic obstacle avoidance and PID control algorithms) deployed on edge computing nodes and the training model (a machine learning-based material placement strategy optimization model) deployed in the cloud working together to optimize and generate a "from the inside out, layered flattening" loading strategy and action instructions for each mechanism of the ship loader based on the ship's loading plan and the virtual model of the digital twin module. The PLC controller of the control execution module receives instructions from the intelligent computing module and controls the actions of the ship loader's traveling mechanism, boom mechanism, chute mechanism, etc. Meanwhile, the communication module uses 5G private network technology to provide a low-latency transmission channel for the point cloud data of the 3D laser scanning module, the positioning data of the positioning and navigation module, and the control instructions of the PLC controller, ensuring smooth data interaction between the modules.

[0097] The human-machine interface integrates and displays the ship loader's position and posture (traveling position, pitch angle, extension length), ship status (coordinates of the four hatches, material level in the hatches), and work progress (completed quantity / target quantity in each hatch) in real time. After the operator can "start" the automated work process with one click through the interface, no continuous intervention is required. During the operation, the level detection module continuously monitors the status of the coal pile in the hatch and uploads the data to the digital twin module in real time. The intelligent calculation module dynamically adjusts the material distribution strategy based on the data. For example, when the material pile in a certain hatch is locally too high, the position of the chute drop point is adjusted to avoid overloading of the material pile.

[0098] Once the target workload for the first hatch is completed, the remote centralized control mode automatically executes the hatch relocation operation: the intelligent calculation module calculates the coordinates of the second hatch based on the ship's loading plan, generates the relocation path, the PLC controller drives the ship loader's traveling mechanism to move to the position of the second hatch, and the 3D laser scanning module simultaneously scans the outline of the second hatch to confirm its position, and then repeats the loading process; until all four hatches have completed the loading operation, the system automatically stops, and the human-machine interface displays that the operation is complete. The entire process realizes full automation of ship finding, hatch relocation, and hatch loading.

[0099] Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. An automated ship loading system for an inland waterway bulk cargo terminal, characterized in that, It includes: a 3D laser scanning module, a positioning and navigation module, a digital twin module, an intelligent computing module, and a level detection module; The three-dimensional laser scanning module is used to collect three-dimensional point cloud data of the surface of the ship and the materials inside the cabin; The positioning and navigation module is used to provide coordinated positioning and navigation for the ship loader and auxiliary equipment; The digital twin module is used to construct a virtual model of the system and simulate the work process; The intelligent computing module is used to optimize the loading and fabric placement strategies through machine learning algorithms. The level detection module is used to monitor the status of materials inside the chamber in real time.

2. The automated loading system for inland bulk cargo terminals according to claim 1, characterized in that, The three-dimensional laser scanning module includes a lidar mounted on the boom of the ship loader. The lidar scans the ship's outline and hatch position during the ship-finding phase and scans the surface of the material pile during the loading phase to generate a dynamic three-dimensional model.

3. The automated ship loading system for an inland waterway bulk cargo terminal according to claim 1, characterized in that, The positioning and navigation module integrates the BeiDou satellite navigation system and an inertial measurement unit, and uses RTK differential technology to achieve centimeter-level positioning of the ship loader and collaborative path planning for multiple devices.

4. The automated loading system for inland bulk cargo terminals according to claim 1, characterized in that, The digital twin module integrates a physics engine and receives real-time data from the 3D laser scanning module and the level detection module for operation process simulation and abnormal state deduction.

5. The automated loading system for inland bulk cargo terminals according to claim 1, characterized in that, The intelligent computing module includes a lightweight algorithm model deployed on edge computing nodes and a training model deployed in the cloud. The algorithm model includes a dynamic obstacle avoidance algorithm, a PID control algorithm, and a machine learning-based material placement strategy optimization model.

6. The automated ship loading system for an inland waterway bulk cargo terminal according to claim 1, characterized in that, It also includes a control execution module; The control execution module includes a PLC controller, which receives instructions from the intelligent computing module and controls the actions of each mechanism of the ship loader. The control execution module supports safety interlocking and emergency stop protection.

7. The automated loading system for inland bulk cargo terminals according to claim 1, characterized in that, It also includes a communication module; The communication module uses 5G private network technology to provide a low-latency transmission channel for the point cloud data of the 3D laser scanning module, the positioning data of the positioning and navigation module, and control commands.

8. The automated loading system for inland bulk cargo terminals according to claim 1, characterized in that, The level detection module integrates lidar and visual recognition sensors, and uploads the monitored material status data to the digital twin module via the MQTT protocol.

9. An automated ship loading system for an inland waterway bulk cargo terminal according to claim 1, characterized in that, It also includes the human-computer interaction interface; The human-machine interface integrates and displays the position and posture of the ship loader, the ship status and the operation progress information, and supports one-click start of the automated operation process.

10. An automated ship loading system for an inland waterway bulk cargo terminal according to claim 1, characterized in that, Execute remote centralized control mode; The remote centralized control mode automatically executes the entire process of ship search, cabin relocation, and cabin loading based on the received ship loading plan.

Images