Intelligent bucket throwing operation method and system for remotely controlling grab ship unloader

Abstract

The invention relates to the field of remote control, and discloses an intelligent bucket throwing operation method and system for remotely controlling a grab ship unloader, which are used for realizing intelligent bucket throwing operation for remotely controlling the grab ship unloader. The intelligent bucket throwing operation method for remotely controlling the grab ship unloader comprises the steps of generating a full-state expected trajectory containing position, speed and acceleration information, ensuring that the trajectory is smooth and accurate, constructing a nonlinear model prediction controller, integrating a nonlinear disturbance observer, effectively estimating and compensating wind disturbance and unmodeled dynamics, and enhancing the anti-jamming capability of the system. A discrete element method is utilized to simulate interaction of material particles, and in combination with bilge topographic data, spatial distribution after material throwing is accurately predicted, system stability is monitored in real time, and a control instruction is dynamically adjusted. The precision, the stability and the efficiency of bucket throwing operation of the grab ship unloader are remarkably improved, the operation difficulty and the labor cost are reduced, and the grab ship unloader can well adapt to complex working conditions.

Description

Technical Field

[0001] This invention relates to the field of remote control, and in particular to a method and system for remotely controlling a grab unloader with intelligent bucket-swing operation. Background Technology

[0002] In the field of port logistics and bulk cargo handling, grab unloaders, as key equipment, undertake the important task of efficiently and accurately unloading bulk cargo from ships. With the continuous growth of global trade volume and the ever-increasing cargo throughput at ports, higher demands are being placed on the efficiency, precision, and automation level of unloading operations. Traditional grab unloader operation modes mainly rely on operator experience, manually controlling the lifting, moving, and releasing of the grab to achieve cargo unloading.

[0003] However, this operating method has many limitations: on the one hand, manual operation is easily affected by factors such as the operator's skill level and fatigue, resulting in unstable operating efficiency, low accuracy of the grab bucket trajectory, difficulty in achieving precise bucket throwing, and easy to cause cargo spillage and equipment damage; on the other hand, under complex working conditions, such as strong wind interference and variable cargo characteristics, traditional control methods are difficult to deal with effectively, cannot guarantee the stable operation of the system, and increase the risk and cost of operation.

[0004] Therefore, we propose a remote-controlled intelligent bucket-swing operation method and system for grab unloaders to solve the above problems. Summary of the Invention

[0005] This invention provides a remote-controlled intelligent bucket-swing operation method and system for a grab unloader, which enables remote-controlled intelligent bucket-swing operation of the grab unloader.

[0006] The first aspect of this invention provides a remote-controlled intelligent bucket-swing operation method for a grab unloader. The method includes: receiving operator instructions and performing trajectory planning to generate a desired trajectory in all states; inputting the desired trajectory and real-time collected system state data into a nonlinear model predictive controller to generate preliminary control instructions through rolling optimization calculations; predicting and outputting spatial distribution data of the material after it lands by simulating the mechanical interactions between material particles based on the preliminary control instructions; feeding back the material spatial distribution data and real-time stability indicators to the nonlinear model predictive controller to dynamically adjust its optimization objective function and control constraints, generating anti-interference control instructions; and driving an actuator after conversion of the optimized anti-interference control instructions to complete a precise bucket-swing operation.

[0007] Optionally, in a first implementation of the first aspect of the present invention, the method includes: receiving a target release point spatial coordinate instruction from an operator input interface; combining the current grab rope length and trolley position data obtained from the system parameter storage unit to determine the initial state and target state of trajectory planning; determining the spatial position coordinates of the grab system as a differential flat output; characterizing all state variables and control inputs of the system based on the flat output and its derivatives; connecting the initial state and target state using a quintic spline curve algorithm to generate a smooth trajectory in the flat output space, which ensures that the velocity and acceleration at the start and end points are both zero; performing time differentiation on the smooth trajectory in the flat output space to obtain the first and second derivative sequences of the flat output, which correspond to the intermediate trajectories of the system's velocity and acceleration, respectively; and converting the trajectory in the flat output space and its derivative sequence into the expected full-state trajectory in the original system state space through the inverse mapping relationship defined by the differential flatness theory.

[0008] Optionally, in a second implementation of the first aspect of the present invention, the method includes: real-time acquisition of data on the trolley position, lifting height, and grab bucket swing angle to form real-time system state data; construction of a system energy model, which includes a Hamiltonian function characterizing energy storage characteristics, an interconnection matrix describing energy flow paths, and a damping matrix characterizing energy dissipation characteristics; inputting the real-time system state data into a nonlinear disturbance observer to generate a total disturbance estimate using a dynamic estimation method; performing rolling optimization calculations in a finite time domain using the full-state expected trajectory as the tracking target and the port-Hamilton energy model as the prediction model to generate a nominal control command; and fusing the nominal control command with the total disturbance estimate in a feedforward channel to generate a preliminary control command after disturbance compensation.

[0009] Optionally, in a third implementation of the first aspect of the present invention, the method includes: establishing an initial state model of the material particle group based on material physical parameters from a material characteristic database and real-time grab bucket opening data from a grab bucket status monitoring system; combining the grab bucket motion trajectory in the preliminary control command with the initial state model, and calculating the contact force, friction force, and collision behavior between material particles within each time step using the discrete element method to generate particle motion trajectory data; constructing a material landing point collision detection model based on the hull topography data provided by the hull three-dimensional scanning device, correcting the landing point position of the particle motion trajectory data to obtain corrected particle distribution data; performing spatial gridding statistical processing on the corrected particle distribution data to generate a spatial distribution density map of the material at the hull bottom; and extracting key feature parameters from the spatial distribution density map to form material spatial distribution data.

[0010] Optionally, in a fourth implementation of the first aspect of the present invention, the method includes: extracting a distribution uniformity index and a stacking height index from the material spatial distribution data; collecting system state data in real time and inputting it into a Lyapunov exponent analyzer to generate a real-time stability index by calculating the eigenvalues ​​of the system state matrix; converting the distribution uniformity index and the stacking height index into material distribution optimization weight coefficients; comparing the real-time stability index with a preset safety threshold to generate dynamic constraint adjustment parameters; real-time correcting the optimization objective function weight matrix and control quantity constraint range of the nonlinear model predictive controller based on the material distribution optimization weight coefficients and dynamic constraint adjustment parameters; and re-optimizing and calculating the initial control command using the corrected optimization objective function and constraint conditions to generate an anti-interference control command that simultaneously ensures stability and operational effectiveness.

[0011] Optionally, in the fifth implementation of the first aspect of the present invention, the method includes: receiving an anti-interference control command, which is a dimensionless control command after material distribution optimization and stability constraint adjustment; reading the physical parameter conversion coefficients of the current equipment from the system configuration database, and converting the dimensionless control command into a motor control command through inverse dimensionless calculation; distributing the motor control command to the corresponding actuator drive units, including a hoisting mechanism drive, an opening and closing mechanism drive, and a trolley traveling mechanism drive; collecting the operating status data of each actuator and the current and speed feedback data of the motor in real time to form drive status monitoring data; inputting the drive status monitoring data and the grab space attitude data from the sensor system into a Lyapunov stability monitor to generate a real-time stability evaluation index; continuously comparing the real-time stability evaluation index with a preset stability safety threshold, and generating a dynamic constraint adjustment command when the index approaches the safety threshold; and feeding back the dynamic constraint adjustment command to the nonlinear model predictive controller in real time to dynamically limit its optimization space and ensure that the system always operates within a stable region.

[0012] Optionally, in the sixth implementation of the first aspect of the present invention, the method further includes: during the system initialization phase, identifying the dynamic parameters of the grab bucket-material system to generate a system characteristic parameter set; based on the system characteristic parameter set, constructing an energy balance controller using a port-Hamilton energy shaping method to generate a reference energy control command; weighting and fusing the reference energy control command with the anti-interference control command output by the nonlinear model predictive controller to generate a composite control command; acquiring multi-dimensional operating data during the grab bucket operation through a real-time data acquisition module and inputting it to a dynamic performance evaluator; the dynamic performance evaluator generating a real-time performance evaluation result based on energy consumption indicators and operating efficiency indicators; and adaptively adjusting the fusion weight coefficient of the reference energy control command and the anti-interference control command according to the real-time performance evaluation result.

[0013] A second aspect of the present invention provides a remotely controlled intelligent bucket-swinging operation system for a grab unloader. The system includes: an acquisition module for receiving operator instructions and performing trajectory planning to generate a desired trajectory in all states; a processing module for inputting the desired trajectory in all states and real-time acquired system state data into a nonlinear model predictive controller, generating preliminary control instructions through rolling optimization calculations; a prediction module for predicting and outputting spatial distribution data of the material after it lands by simulating the mechanical interactions between material particles based on the preliminary control instructions; a setting module for feeding back the material spatial distribution data and real-time stability indicators to the nonlinear model predictive controller, dynamically adjusting its optimization objective function and control constraints, and generating anti-interference control instructions; and an allocation module for converting the optimized anti-interference control instructions to drive the actuator to complete precise bucket-swinging operations.

[0014] The mechanism of this invention is as follows: Based on the cross-integration of multiple theories of first principles, it fundamentally solves the problems of poor adaptability of traditional methods under nonlinear working conditions and weak interpretability of machine learning methods, and realizes the true unity of human-like intelligent decision-making and object-like precise execution. Beneficial effects: The accurate full-state expected trajectory generated by the trajectory planning method that integrates multiple theories, and the accurate prediction of material spillage pattern by the discrete element method, enable the grab bucket to release materials precisely according to the predetermined trajectory, reduce cargo spillage, improve the uniformity of cargo distribution at the bottom of the hold and the control accuracy of stacking height, thereby improving the overall operational accuracy. The combination of nonlinear model predictive control based on the port-Hamilton framework and a nonlinear disturbance observer effectively compensates for the effects of wind disturbance and unmodeled dynamics on the system. Meanwhile, real-time monitoring and dynamic constraint adjustment of the Lyapunov stability index ensures that the system always operates in the stable region under various operating conditions, greatly enhancing the system's stability and reliability. The adaptive control strategy can automatically adjust control parameters based on real-time operational data, ensuring optimal system performance under various operating conditions. Precise trajectory planning and material spillage prediction reduce unnecessary operation and adjustment time, improving the operational efficiency of the grab unloader and lowering operating costs. Remote control allows operators to send commands through an interface in the control room, avoiding operation in harsh on-site environments, reducing operational difficulty and labor intensity. The intelligent control method reduces reliance on operator experience, lowers labor costs, and improves the level of automation in port operations. Attached Figure Description

[0015] Figure 1 This is a schematic diagram of an embodiment of the intelligent bucket-swinging operation method for remotely controlled grab unloaders according to the present invention; Figure 2 This is a schematic diagram of another embodiment of the intelligent bucket-swinging operation method for remotely controlled grab unloaders according to the present invention; Figure 3 This is a schematic diagram of an embodiment of the intelligent bucket-swinging operation system for remotely controlled grab unloaders according to the present invention. Detailed Implementation

[0016] This invention provides a remote-controlled intelligent bucket-swing operation method and system for a grab unloader, enabling intelligent bucket-swing operations of the grab unloader via remote control. The terms "first," "second," "third," "fourth," etc. (if present) in the specification, claims, and accompanying drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" or "having" and any variations thereof are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or device that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or devices.

[0017] For ease of understanding, the specific process of the embodiments of the present invention is described below. Please refer to [link / reference]. Figure 1 One embodiment of the intelligent bucket-swinging operation method for remotely controlled grab unloaders in this invention includes: 101. Receive operator instructions and perform trajectory planning based on differential flatness theory to generate a full-state desired trajectory containing position, velocity, and acceleration information; It is understood that the executing entity of this invention can be a remotely controlled intelligent bucket unloader system, or it can be a terminal or a server; the specific implementation is not limited here. This embodiment of the invention will be described using a server as an example.

[0018] Specifically, it receives the target release point spatial coordinate command from the operator input interface, and combines it with the current grab rope length and trolley position data obtained from the system parameter storage unit to determine the starting state and target state of trajectory planning; The spatial position coordinates of the grab system are determined as a differential flat output, and all state variables and control inputs of the system are characterized based on this flat output and its derivatives. A quintic spline curve algorithm is used to connect the initial state and the target state to generate a smooth trajectory in a flat output space. This trajectory ensures that the velocity and acceleration at the start and end points are both zero. By taking the time derivative of the smooth trajectory in the flat output space, we obtain the first and second derivative sequences of the flat output, which correspond to the intermediate trajectories of the system's velocity and acceleration, respectively. Using the inverse mapping relationship defined by the differential flatness theory, the trajectory and its derivative sequence in the flat output space are transformed into the full-state expected trajectory in the original system state space. This trajectory includes the trolley position, lifting height, grab bucket swing angle and its time derivatives of each order. The expected trajectory in the full state is used as the tracking target input for the subsequent nonlinear model predictive controller to generate initial control commands.

[0019] 102. The full-state expected trajectory and the real-time acquired system state data are input together into a nonlinear model predictive controller based on the port-Hamilton framework. This controller generates preliminary control commands through rolling optimization calculations, and at the same time integrates a nonlinear disturbance observer to estimate and compensate for wind disturbances and unmodeled dynamics. Specifically, the system sensors collect data on the trolley's position, lifting height, and grab bucket swing angle in real time to form real-time system status data; A system energy model is constructed based on port-Hamilton theory. This model includes a Hamiltonian function characterizing energy storage characteristics, an interconnection matrix describing energy flow paths, and a damping matrix characterizing energy dissipation characteristics. The system's real-time state data is input into the nonlinear disturbance observer, and a total disturbance estimate is generated through a dynamic estimation method. This estimate includes the wind disturbance effect and the system's unmodeled dynamic characteristics. Using the full-state expected trajectory as the tracking target and the port-Hamilton energy model as the prediction model, rolling optimization calculations are performed in the finite time domain to generate nominal control commands. The nominal control command and the total disturbance estimate are fused in the feedforward channel to generate a preliminary control command after disturbance compensation. The initial control command is output to the subsequent material spill pattern prediction module, and the total disturbance estimate is fed back to the port-Hamilton energy model for real-time correction of the prediction model accuracy.

[0020] 103. Input the initial control command into the material spill pattern prediction module based on the discrete element method. By simulating the mechanical interaction between material particles, predict and output the spatial distribution data of the material after it lands. Specifically, an initial state model of the material particle group is established based on the material physical parameters from the material property database and the real-time grab opening data from the grab state monitoring system. By combining the grab's motion trajectory in the initial control command with the initial state model, the contact force, friction force and collision behavior between material particles within each time step are calculated using the discrete element method to generate particle motion trajectory data. Based on the hull top terrain data provided by the in-cabin 3D scanning device, a material impact point collision detection model is constructed, and the impact point position of the particle motion trajectory data is corrected to obtain the corrected particle distribution data. The corrected particle distribution data is subjected to spatial gridding statistical processing to generate a spatial distribution density map of the material at the bottom of the tank. Key feature parameters, including distribution uniformity index and stacking height index, are extracted from the spatial distribution density map to form material spatial distribution data; Material spatial distribution data is fed back to the nonlinear model predictive controller and used as an adjustment term in the optimization objective function to participate in the optimization calculation of control commands.

[0021] 104. The material spatial distribution data and the real-time stability index based on Lyapunov index analysis are fed back to the nonlinear model predictive controller to dynamically adjust its optimization objective function and control constraints, and generate optimized anti-interference control commands. Specifically, it receives material spatial distribution data from the material spill pattern prediction module and extracts distribution uniformity index and accumulation height index from it. Real-time system status data is collected and input into the Lyapunov index analyzer, and real-time stability indexes are generated by calculating the eigenvalues ​​of the system status matrix. The distribution uniformity index and stacking height index are converted into material distribution optimization weighting coefficients; The real-time stability index is compared with the preset safety threshold to generate dynamic constraint adjustment parameters. Based on the material distribution optimization weight coefficients and dynamic constraint adjustment parameters, the optimization objective function weight matrix and control quantity constraint range of the nonlinear model predictive controller are corrected in real time. Using the revised objective function and constraints, the initial control commands are recalculated to generate anti-interference control commands that simultaneously ensure stability and operational effectiveness. The anti-interference control command is output to the actuator drive unit to complete the precise bucket-throwing operation.

[0022] 105. The optimized anti-interference control commands are converted to drive the actuator to complete the precise bucket-throwing operation. The Lyapunov stability index is used for online monitoring to ensure that the system always operates in a stable range.

[0023] Specifically, it receives anti-interference control commands from a nonlinear model predictive controller, which are dimensionless control quantities adjusted by material distribution optimization and stability constraints. Read the physical parameter conversion coefficients of the current device from the system configuration database, and convert the dimensionless control commands into motor control commands with actual physical units through inverse dimensionless calculation; The motor control commands are distributed to the corresponding actuator drive units, including the hoisting mechanism driver, the opening and closing mechanism driver, and the trolley traveling mechanism driver; Real-time collection of operating status data of each actuator and current and speed feedback data of the motor to form drive status monitoring data; The drive status monitoring data and the grab space attitude data from the sensor system are input together into the Lyapunov stability monitor to generate real-time stability evaluation indicators. The real-time stability assessment index is continuously compared with the preset stability safety threshold. When the index approaches the safety threshold, a dynamic constraint adjustment instruction is generated. The dynamic constraint adjustment command is fed back to the nonlinear model predictive controller in real time, dynamically limiting its optimization space and ensuring that the system always operates within a stable region to complete safe and accurate bucket-throwing operations.

[0024] In this embodiment of the invention, the trajectory planning method based on differential flatness theory can generate a full-state expected trajectory containing position, velocity, and acceleration information, ensuring that the velocity and acceleration at the beginning and end of the trajectory are both zero, making the grab bucket movement more stable and precise, effectively reducing the impact and shaking during the material throwing process, and improving the accuracy and stability of the operation.

[0025] By integrating a nonlinear disturbance observer, wind disturbances and unmodeled dynamics are estimated and compensated in real time during nonlinear model predictive control. This enables the system to accurately capture and respond to various complex disturbance factors, allowing it to maintain good control performance even in harsh environments and enhancing its robustness and adaptability.

[0026] This method utilizes the discrete element method to simulate the mechanical interactions between material particles, combines various real-time data to predict the spatial distribution of the material after it lands, and feeds the distribution data back to the controller for optimization calculations. This closed-loop control approach can dynamically adjust control commands based on the actual material distribution, effectively optimizing the material spillage pattern and improving operational efficiency and material accumulation quality.

[0027] A real-time stability index based on Lyapunov exponent analysis is introduced to monitor system stability online. By comparing the stability index with a preset safety threshold, the optimization objective function and control constraints of the controller are dynamically adjusted to ensure that the system always operates in a stable region, providing a strong guarantee for safe operation.

[0028] This invention forms a complete closed-loop control system from trajectory planning, control command generation, material spillage prediction to stability monitoring. It comprehensively considers factors such as the accuracy, stability, anti-interference ability and material spillage effect of the operation, and realizes the comprehensive performance improvement of intelligent bucket unloader operation.

[0029] Please see Figure 2 Another embodiment of the remote control grab unloader intelligent bucket-swinging operation method in this invention includes: 106. In the system initialization stage, the grab bucket-material system is identified by a dedicated calibration module to generate a set of system characteristic parameters; Based on the system characteristic parameter set, an energy balance controller is constructed using the port-Hamilton energy shaping method to generate a reference energy control command. The reference energy control command and the anti-interference control command output by the nonlinear model predictive controller are weighted and fused to generate a composite control command; The real-time data acquisition module acquires multi-dimensional operational data during the grab bucket operation and inputs it into the dynamic performance evaluator. The dynamic performance evaluator generates real-time performance evaluation results based on energy consumption and operational efficiency metrics. The fusion weighting coefficients of the baseline energy control command and the anti-interference control command are adaptively adjusted based on the real-time performance evaluation results. The adjusted fusion weight coefficients are fed back to the control command fusion module to achieve online self-optimization of the control strategy, ensuring that the system can maintain optimal performance under different operating conditions.

[0030] The above describes the intelligent bucket-swing operation method of the remote-controlled grab bucket unloader in an embodiment of the present invention. The following describes the intelligent bucket-swing operation system of the remote-controlled grab bucket unloader in an embodiment of the present invention. Please refer to [link to relevant documentation]. Figure 3 An embodiment of the intelligent bucket-slinging operation system for remotely controlled grab unloaders in this invention includes: an acquisition module 201, used to receive operator instructions and perform trajectory planning to generate a full-state expected trajectory; a processing module 202, used to input the full-state expected trajectory and real-time collected system state data into a nonlinear model predictive controller, and generate preliminary control instructions through rolling optimization calculations; a prediction module 203, used to predict and output the spatial distribution data of the material after landing by simulating the mechanical interaction between material particles based on the preliminary control instructions; a setting module 204, used to feed back the material spatial distribution data and real-time stability indicators to the nonlinear model predictive controller, dynamically adjust its optimization objective function and control constraints, and generate anti-interference control instructions; and an allocation module 205, used to drive the actuator after conversion of the optimized anti-interference control instructions to complete the precise bucket-slinging operation.

[0031] In this embodiment of the invention, by integrating a nonlinear disturbance observer, wind disturbance and unmodeled dynamics are estimated and compensated in real time during the nonlinear model predictive control process. This accurately captures and responds to various complex disturbance factors, enabling the system to maintain good control performance even in harsh environments, thus enhancing the system's robustness and adaptability. The discrete element method is used to simulate the mechanical interactions between material particles, and various real-time data are combined to predict the spatial distribution of the material after landing. The distribution data is then fed back to the controller for optimization calculations. This closed-loop control method can dynamically adjust control commands according to the actual material distribution, effectively optimizing the material throwing pattern and improving operational efficiency and material accumulation quality. A real-time stability index based on Lyapunov exponent analysis is introduced to monitor system stability online. By comparing the stability index with a preset safety threshold, the controller's optimization objective function and control constraints are dynamically adjusted to ensure the system always operates in a stable region, providing strong protection for safe operation. This invention forms a complete closed-loop control system from trajectory planning, control command generation, material throwing prediction to stability monitoring. It comprehensively considers factors such as operational accuracy, stability, anti-interference capability, and material throwing effect, achieving a comprehensive performance improvement in the intelligent bucket unloader operation.

[0032] The present invention also provides a remote-controlled grab unloader intelligent bucket-swinging operation device, the remote-controlled grab unloader intelligent bucket-swinging operation device including a memory and a processor, the memory storing computer-readable instructions, when the computer-readable instructions are executed by the processor, causing the processor to perform the steps of the remote-controlled grab unloader intelligent bucket-swinging operation method in the above embodiments.

[0033] The present invention also provides a computer-readable storage medium, which can be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium, wherein the computer-readable storage medium stores instructions that, when the instructions are executed on a computer, cause the computer to perform the steps of the remote-controlled grab unloader intelligent bucket-swinging operation method.

[0034] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0035] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0036] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A remote-controlled intelligent bucket unloader operation method, characterized in that, The remote-controlled grab unloader intelligent bucket-swing operation method includes: Receive operator instructions and perform trajectory planning to generate the desired trajectory in all states; The full-state expected trajectory and the real-time acquired system state data are input together into the nonlinear model predictive controller, and preliminary control commands are generated through rolling optimization calculations. Based on the preliminary control instructions, the spatial distribution data of the material after it lands is predicted and output by simulating the mechanical interaction between material particles; The material spatial distribution data and real-time stability index are fed back to the nonlinear model predictive controller, which dynamically adjusts its optimization objective function and control constraints to generate anti-interference control commands. The optimized anti-interference control command is converted and then used to drive the actuator to complete the precise bucket-throwing operation.

2. The remote-controlled grab unloader intelligent bucket-swing operation method according to claim 1, characterized in that, include: Receive the target release point spatial coordinate command from the operator input interface, and combine it with the current grab rope length and trolley position data obtained from the system parameter storage unit to determine the starting state and target state of trajectory planning; The spatial position coordinates of the grab system are determined as a differential flat output, and all state variables and control inputs of the system are characterized based on this flat output and its derivatives. A quintic spline curve algorithm is used to connect the initial state and the target state to generate a smooth trajectory in a flat output space. This trajectory ensures that the velocity and acceleration at the start and end points are both zero. By taking the time derivative of the smooth trajectory in the flat output space, the first and second derivative sequences of the flat output are obtained, which correspond to the intermediate trajectories of the system's velocity and acceleration, respectively. By using the inverse mapping relationship defined by the differential flatness theory, the trajectory and its derivative sequence in the flat output space are converted into the expected full-state trajectory in the original system state space.

3. The remote-controlled grab unloader intelligent bucket-swing operation method according to claim 2, characterized in that, include: Real-time data collection of trolley position, lifting height, and grab bucket swing angle is used to generate real-time system status data. A system energy model is constructed, which includes a Hamiltonian function characterizing energy storage characteristics, an interconnection matrix describing energy flow paths, and a damping matrix characterizing energy dissipation characteristics. The real-time state data of the system is input into a nonlinear disturbance observer, and a total disturbance estimate is generated through a dynamic estimation method. Using the full-state expected trajectory as the tracking target and the port-Hamilton energy model as the prediction model, rolling optimization calculations are performed in the finite time domain to generate nominal control commands. The nominal control command and the total disturbance estimate are fused in the feedforward channel to generate a preliminary control command after disturbance compensation.

4. The remote-controlled grab unloader intelligent bucket-swing operation method according to claim 3, characterized in that, include: An initial state model of the material particle group is established based on the material physical parameters from the material property database and the real-time grab opening data from the grab state monitoring system. By combining the grab's motion trajectory in the initial control command with the initial state model, the contact force, friction force, and collision behavior between material particles within each time step are calculated using the discrete element method to generate particle motion trajectory data. Based on the hull top terrain data provided by the in-cabin 3D scanning device, a material impact point collision detection model is constructed, and the impact point position of the particle motion trajectory data is corrected to obtain the corrected particle distribution data. The corrected particle distribution data is subjected to spatial gridding statistical processing to generate a spatial distribution density map of the material at the bottom of the tank. Key feature parameters are extracted from the spatial distribution density map to form material spatial distribution data.

5. The intelligent bucket-swinging operation method for a remotely controlled grab unloader according to claim 4, characterized in that, include: Based on the material spatial distribution data, the distribution uniformity index and the stacking height index are extracted from it; Real-time system status data is collected and input into the Lyapunov index analyzer, and real-time stability indexes are generated by calculating the eigenvalues ​​of the system status matrix. The distribution uniformity index and stacking height index are converted into material distribution optimization weighting coefficients; The real-time stability index is compared with a preset safety threshold to generate dynamic constraint adjustment parameters; Based on the material distribution optimization weight coefficients and dynamic constraint adjustment parameters, the optimization objective function weight matrix and control quantity constraint range of the nonlinear model predictive controller are corrected in real time. By using the revised objective function and constraints, the initial control commands are recalculated to generate anti-interference control commands that simultaneously ensure stability and operational effectiveness.

6. The intelligent bucket-swinging operation method for a remotely controlled grab unloader according to claim 5, characterized in that, include: Receive anti-interference control commands, which are dimensionless control commands that have been optimized for material distribution and adjusted for stability constraints. Read the physical parameter conversion coefficients of the current device from the system configuration database, and convert the dimensionless control commands into motor control commands through inverse dimensionless calculation; The motor control commands are distributed to the corresponding actuator drive units, including the hoisting mechanism driver, the opening and closing mechanism driver, and the trolley traveling mechanism driver; Real-time collection of operating status data of each actuator and current and speed feedback data of the motor to form drive status monitoring data; The driving state monitoring data and the grab space attitude data from the sensor system are input together into the Lyapunov stability monitor to generate real-time stability evaluation indicators. The real-time stability evaluation index is continuously compared with the preset stability safety threshold, and a dynamic constraint adjustment instruction is generated when the index approaches the safety threshold. The dynamic constraint adjustment command is fed back to the nonlinear model prediction controller in real time, dynamically limiting its optimization space and ensuring that the system always operates within the stable region.

7. The intelligent bucket-swinging operation method for a remotely controlled grab unloader according to claim 6, characterized in that, Also includes: During the system initialization phase, the dynamic parameters of the grab-material system are identified, and a set of system characteristic parameters is generated. Based on the system characteristic parameter set, an energy balance controller is constructed using the port-Hamilton energy shaping method to generate a reference energy control command. The reference energy control command and the anti-interference control command output by the nonlinear model predictive controller are weighted and fused to generate a composite control command; The real-time data acquisition module acquires multi-dimensional operational data during the grab bucket operation and inputs it into the dynamic performance evaluator. The dynamic performance evaluator generates real-time performance evaluation results based on energy consumption indicators and operational efficiency indicators; The fusion weighting coefficients of the baseline energy control command and the anti-interference control command are adaptively adjusted based on the real-time performance evaluation results.

8. A remote-controlled intelligent bucket unloader system, characterized in that, The remote-controlled grab unloader intelligent bucket-swing operation system includes: The acquisition module is used to receive operator instructions, perform trajectory planning, and generate the desired trajectory in all states. The processing module is used to input the full-state expected trajectory and the real-time acquired system state data into the nonlinear model predictive controller, and generate preliminary control commands through rolling optimization calculations. The prediction module is used to predict and output the spatial distribution data of the material after it lands by simulating the mechanical interaction between material particles based on the preliminary control instructions. The setting module is used to feed the material spatial distribution data and real-time stability index back to the nonlinear model prediction controller, dynamically adjust its optimization objective function and control constraints, and generate anti-interference control commands. The distribution module is used to convert the optimized anti-interference control commands and drive the actuator to complete the precise bucket-throwing operation.

Images