A control method, device and equipment for a bridge-type grab ship unloader

Abstract

The application relates to the technical field of ship unloader control, and discloses a bridge type grab ship unloader control method, device and equipment, which comprises the following steps: acquiring multi-source sensing data of a ship unloader; performing time reference alignment and weighted fusion processing on the multi-source sensing data to obtain fusion state data; performing filtering processing on the fusion state data to generate smooth state data; analyzing the smooth state data to extract operation characteristic parameters of a current working condition; inputting the operation characteristic parameters into a preset characteristic and parameter mapping model to match and generate dynamic adjustment parameters of a basic controller; inputting the dynamic adjustment parameters into the basic controller to perform online updating on gain parameters of the basic controller; generating control instructions through the basic controller after updating the gain parameters, and controlling a movement mechanism and a grab bucket of the ship unloader based on the control instructions. The application effectively reduces the positioning overshoot of the grab bucket, accelerates the convergence speed of anti-shaking and vibration suppression, and improves the positioning accuracy and stability of the ship unloader.

Description

Technical Field

[0001] This invention relates to the field of ship unloader control technology, and more specifically, to a control method, device, and equipment for a bridge grab ship unloader. Background Technology

[0002] With the rapid growth in demand for bulk cargo transportation at ports, the bridge grab unloader, as the core equipment for unloading bulk materials at terminals, directly affects the overall throughput capacity of ports. In recent years, port loading and unloading equipment has been developing towards larger scale and intelligence. However, due to the harsh unloading environment, the variable distribution of materials in the hold, and the obvious unstructured characteristics, achieving high-precision and high-reliability fully automated operation still faces enormous challenges.

[0003] Currently, conventional automated retrofits of bridge grab unloaders mostly rely on single sensors for status monitoring and fixed-parameter PID control algorithms at the underlying execution level. This technology often fails to accurately construct three-dimensional features of the materials inside the hold when facing complex operating conditions such as dust obstruction and mechanical vibration. Furthermore, due to the lack of a dynamic adaptive mechanism, the system struggles to adjust flexibly like a skilled operator when faced with nonlinear grab swaying and varying dynamic loads. Therefore, existing technologies suffer from a lack of multi-source environmental fusion sensing capabilities under complex operating conditions, and the fixed control strategy struggles to overcome nonlinear disturbances, resulting in large grab positioning overshoot and slow anti-sway and vibration-damping convergence. Summary of the Invention

[0004] To overcome the shortcomings of existing technologies, such as the lack of multi-source environmental fusion perception capabilities under complex working conditions and the difficulty of fixed control strategies in overcoming nonlinear disturbances, resulting in large overshoot in grab bucket positioning and slow convergence in anti-sway and vibration suppression, the present invention proposes the following technical solution: In a first aspect, the present invention proposes a control method for a bridge-type grab unloader, comprising: Acquire multi-source sensor data from the ship unloader; The multi-source sensor data is time-referenced and weighted to obtain fused state data. The fused state data is filtered to generate smooth state data; The smoothed state data is parsed to extract the operational characteristic parameters of the current working condition; The operation feature parameters are input into a preset feature-parameter mapping model to match and generate dynamic adjustment parameters for the basic controller; The dynamic adjustment parameters are input into the base controller to update the gain parameters of the base controller online. The base controller generates control commands by updating the gain parameters, and controls the motion mechanism and grab bucket operation of the unloader based on the control commands.

[0005] As a preferred technical solution, multi-source sensor data of the ship unloader is acquired, including: The position data of the motion mechanism is obtained through the encoder; Three-dimensional point cloud data of the ship's cabin and materials in the work area are obtained using a laser scanner; Image data of the grab bucket is acquired using a visual sensor; The location data, the 3D point cloud data, and the image data are combined to construct the multi-source sensing data.

[0006] As a preferred technical solution, the multi-source sensor data is subjected to time-base alignment and weighted fusion processing to obtain fused state data, including: Align the location data, the 3D point cloud data, and the image data to the same timestamp; According to the following formula, the aligned position data, the 3D point cloud data, and the image data are weighted and fused to obtain the fused state data:

[0007] In the formula, For the fusion state data, For the first The weighting coefficients of each sensor For the first The aligned position data, the 3D point cloud data, or the image data collected by each sensor. The number of sensors participating in the fusion.

[0008] As a preferred technical solution, the fused state data is filtered to generate smooth state data, including: The fused state data is processed using the Kalman filter algorithm; Construct the state prediction equation:

[0009] Construct the measurement update equation:

[0010] The result is obtained by calculating the state prediction equation and the measurement update equation. State estimate at time 1 The state estimate As the smoothed state data; in, for The predicted state value at time 10:00. Here is the state transition matrix. for State estimate at time 10:00 To control the input matrix, for Time-based control input, The state prediction covariance matrix, The process noise covariance matrix is... For Kalman gain, For the observation matrix, To observe the noise covariance matrix, for The observed value at time, It is the identity matrix. for The state covariance matrix at time t.

[0011] As a preferred technical solution, the smoothed state data is parsed to extract the operational characteristic parameters of the current working condition, including: Extract single grab cycle data from the smoothed state data; The single grab cycle data is analyzed to extract grab motion rhythm parameters, acceleration and deceleration control parameters, attitude adjustment law parameters, and material height distribution parameters; The operation characteristic parameters are constructed by combining the grab movement rhythm parameters, the acceleration and deceleration control parameters, the attitude adjustment law parameters, and the material height distribution parameters.

[0012] As a preferred technical solution, before inputting the job feature parameters into a preset feature-parameter mapping model, the method further includes: Retrieve historical job data; The historical operation data is parsed into historical operation feature parameters and historical operation control parameters; The historical operation feature parameters are used as input samples, and the historical operation control parameters are used as corresponding labels to train a deep learning model. The deep learning model after training is extracted as the preset feature and parameter mapping model.

[0013] As a preferred technical solution, the dynamic adjustment parameters are input into the base controller, and the gain parameters of the base controller are updated online, including: The basic controller is a proportional-integral-derivative controller; The dynamic adjustment parameters include the proportional gain adjustment amount; The proportional gain of the proportional-integral-derivative controller is updated according to the following formula:

[0014] In the formula, For the updated proportional gain, This is the initial proportional gain. The proportional gain adjustment amount, For error-based and error change rate The adaptive adjustment function.

[0015] As a preferred technical solution, control commands are generated by updating the gain parameters of the base controller, and the motion mechanism and grab bucket of the unloader are controlled based on the control commands, including: The basic controller, after updating the gain parameters, generates initial control commands. Based on the smoothing state data, the relative position data between the grab bucket and the working area is extracted, and the real-time distance is calculated; The real-time distance is compared with a preset safety threshold; When the real-time distance is greater than or equal to the preset safety threshold, the preliminary control command is used as the control command; When the real-time distance is less than the preset safety threshold, a deceleration or shutdown anti-collision control command is generated, and the anti-collision control command is used as the control command; The control commands are sent to the actuator to control the movement mechanism and grab bucket of the unloader.

[0016] Secondly, the present invention proposes a control device for a bridge grab unloader, applied to the control method for a bridge grab unloader as described in any of the embodiments of the first aspect, comprising: The acquisition module is used to acquire multi-source sensor data from the ship unloader; The fusion module is used to perform time-base alignment and weighted fusion processing on the multi-source sensor data to obtain fused state data; The filtering module is used to filter the fused state data to generate smooth state data; The extraction module is used to parse the smoothed state data and extract the operational characteristic parameters of the current working condition; The generation module is used to input the operation feature parameters into a preset feature-parameter mapping model and match and generate dynamic adjustment parameters for the basic controller. The update module is used to input the dynamic adjustment parameters into the base controller and update the gain parameters of the base controller online. The control module is used to generate control commands by updating the gain parameters of the base controller, and to control the motion mechanism and grab bucket operation of the unloader based on the control commands.

[0017] Thirdly, the present invention also proposes an electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to perform the operations performed by the bridge grab unloader control method as described in any of the embodiments of the first aspect, or to perform the operations performed by the bridge grab unloader control method as described in any of the embodiments of the second aspect.

[0018] The beneficial effects of the present invention include at least the following: This invention acquires multi-source sensor data from a ship unloader and sequentially performs time-base alignment, weighted fusion processing, and filtering to generate accurate and reliable smooth state data. This overcomes the shortcomings of existing technologies in lacking multi-source environmental fusion perception capabilities under complex working conditions, effectively eliminating measurement noise caused by multi-sensor asynchrony and on-site environmental interference, and providing highly stable state input for control. Furthermore, this invention extracts operational feature parameters of the current working condition by parsing the smooth state data and inputs them into a preset feature-parameter mapping model to match and generate dynamic adjustment parameters for the basic controller. The gain parameters of the basic controller are then updated online. Finally, control commands are generated based on the updated gain parameters to drive the mechanism. This process achieves dynamic adaptive adjustment of the control strategy according to real-time working conditions, successfully overcoming the shortcomings of existing fixed control strategies in overcoming nonlinear disturbances. This effectively reduces the positioning overshoot of the grab bucket during equipment operation, significantly accelerates the convergence speed of anti-sway and vibration suppression, and significantly improves the positioning accuracy and dynamic stability of the ship unloader in complex working environments. Attached Figure Description

[0019] Figure 1 This is a flowchart illustrating the control method for a bridge grab unloader provided in an embodiment of the present invention.

[0020] Figure 2 This is a diagram of the task perception interface provided in an embodiment of the present invention.

[0021] Figure 3 This is a schematic diagram of grab bucket image recognition provided in an embodiment of the present invention.

[0022] Figure 4 This is a comparison chart of positioning errors between single sensor and multi-source fusion provided in an embodiment of the present invention.

[0023] Figure 5 This is a comparison diagram of velocity signals before and after Kalman filtering provided in an embodiment of the present invention.

[0024] Figure 6 This is a comparison diagram of the positioning response of three control strategies provided in the embodiments of the present invention.

[0025] Figure 7 This is a comparison diagram of the grab bucket swing angles for the three control strategies provided in the embodiments of the present invention.

[0026] Figure 8 This is a comparison chart of single-cycle time and relative energy consumption provided in an embodiment of the present invention.

[0027] Figure 9 This is a diagram of the automated ship unloader system architecture provided in an embodiment of the present invention.

[0028] Figure 10 This is a schematic diagram of the control device for a bridge grab unloader provided in an embodiment of the present invention.

[0029] Figure 11 This is a schematic diagram of the structure of the electronic device provided in an embodiment of the present invention. Detailed Implementation

[0030] The embodiments of the present invention will be described below with reference to the accompanying drawings and preferred technical solutions. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be understood that the preferred technical solutions are only for illustrating the present invention and not for limiting the scope of protection of the present invention.

[0031] It should be noted that the illustrations provided in the following embodiments are only schematic representations of the basic concept of the present invention. Therefore, the drawings only show the components related to the present invention and are not drawn according to the actual number, shape and size of the components in the actual implementation. In the actual implementation, the form, quantity and proportion of each component can be arbitrarily changed, and the layout of the components may also be more complex.

[0032] In the following description, numerous details are explored to provide a more thorough explanation of embodiments of the invention. However, it will be apparent to those skilled in the art that embodiments of the invention may be practiced without these specific details. In other embodiments, well-known structures and devices are shown in block diagram form rather than in detail to avoid obscuring embodiments of the invention.

[0033] Example 1 This embodiment proposes a control method for a bridge-type grab unloader, such as... Figure 1 As shown, Figure 1 This is a flowchart illustrating a control method for a bridge grab unloader provided in this embodiment. The method includes the following steps: S1: Acquire multi-source sensor data from the ship unloader; S2: Perform time-base alignment and weighted fusion processing on the multi-source sensor data to obtain fused state data; S3: Filter the fused state data to generate smooth state data; S4: Analyze the smoothed state data and extract the operational characteristic parameters of the current working condition; S5: Input the operation feature parameters into a preset feature and parameter mapping model to match and generate dynamic adjustment parameters for the basic controller; S6: Input the dynamic adjustment parameters into the base controller and update the gain parameters of the base controller online; S7: Generate control commands by updating the gain parameters of the base controller, and control the motion mechanism and grab bucket operation of the unloader based on the control commands.

[0034] Understandably, by acquiring multi-source sensor data from the ship unloader and sequentially performing time-base alignment, weighted fusion processing, and filtering, accurate and reliable smooth state data is generated. This overcomes the shortcomings of existing technologies in lacking multi-source environmental fusion perception capabilities under complex working conditions, effectively eliminating measurement noise caused by multi-sensor asynchrony and on-site environmental interference, and providing highly stable state input for control. Furthermore, by parsing the smooth state data to extract the operational feature parameters of the current working condition, and inputting them into a preset feature-parameter mapping model to match and generate dynamic adjustment parameters for the basic controller, the gain parameters of the basic controller are updated online. Finally, based on the updated gain parameters, control commands are generated to drive the mechanism to operate. This process realizes the dynamic adaptive adjustment of the control strategy according to real-time working conditions, successfully overcoming the shortcomings of existing fixed control strategies in overcoming nonlinear disturbances. Thus, during equipment operation, the positioning overshoot of the grab bucket is effectively reduced, the convergence speed of anti-sway and vibration suppression is greatly accelerated, and the positioning accuracy and dynamic stability of the ship unloader in complex working environments are significantly improved.

[0035] Example 2 This embodiment improves upon the bridge grab unloader control method proposed in Embodiment 1.

[0036] In this embodiment, acquiring multi-source sensor data from the ship unloader includes: The position data of the motion mechanism is obtained through the encoder; Three-dimensional point cloud data of the ship's cabin and materials in the work area are obtained using a laser scanner; Image data of the grab bucket is acquired using a visual sensor; The location data, the 3D point cloud data, and the image data are combined to construct the multi-source sensing data.

[0037] It should be noted that, in response to the complex unloading environment and the susceptibility of single sensing methods to obstruction and interference, this invention constructs a multi-dimensional perception system with complementary advantages. When acquiring 3D point cloud data of the ship's hold and materials in the work area, multiple 3D laser scanners can be used in collaboration. For example, four 128-line 3D laser scanners can be configured, with three fixedly positioned at key locations such as cantilever arms to achieve full coverage scanning of the hatch opening and materials inside. The other scanner, equipped with a pan-tilt unit, performs a 360° rotating scan below the driver's cab to collect the overall state of the ship. After the initial scan, point cloud data filtering, noise reduction, and feature extraction algorithms are used to construct a 3D model of the ship's hold and material pile information data. The system monitors and updates the data in real time during subsequent operations, comprehensively capturing the material height distribution and cabin geometry. When acquiring image data of the grab bucket via visual sensors, a wide-angle camera can be positioned above the grab bucket, combined with a deep learning model to supplement the monitoring of grab bucket attitude, opening and closing status, and local operational details. This is especially crucial during the grab bucket approaching the material surface and during the cleaning phase, accurately determining the relative position of the grab bucket to the bottom, walls, and internal moving machinery. When acquiring position data via encoders and other devices, a high-precision position detection and redundant acquisition mechanism is introduced. This involves acquiring the same physical quantity through multiple channels and comparing and verifying the data. When a single sensor data point is abnormal, redundant information is used for correction or replacement. Figure 2 As shown, Figure 2 The diagram of the task perception interface provided in this embodiment of the invention illustrates the collaborative perception effect of the global 3D point cloud and local state; for example... Figure 3 As shown, Figure 3 The image recognition diagram of the grab provided in the embodiment of the present invention illustrates the precise positioning boundary of the grab and surrounding machinery by the AI ​​visual recognition model; Table 1 Key parameters and index settings for simulation experiments

[0038] As shown in Table 1, Table 1 is a table of key parameters and index settings for the simulation experiment provided in the embodiments of the present invention. It clarifies that the sensor configuration is an encoder combined with laser ranging, 3D scanning and visual perception. Considering the disturbance factors at the unloading site, interference parameters such as position measurement noise σ=0.12m and velocity noise σ=0.18m / s are set to simulate actual working conditions. The basic principle of its perception is to form a multi-scale, all-round perception network by deeply combining the spatial structure information of global laser scanning with local visual detail features. The beneficial effect of this is that it can effectively avoid single-point failures and perception blind spots caused by harsh on-site environment and electromagnetic interference, and significantly enhance the stability and fault tolerance under complex unstructured chemical conditions. It provides comprehensive and high-precision basic data support for accurate anti-collision safety warning, material grabbing path planning and intelligent bionic control.

[0039] In this embodiment, the multi-source sensor data undergoes time-base alignment and weighted fusion processing to obtain fused state data, including: Align the location data, the 3D point cloud data, and the image data to the same timestamp; According to the following formula, the aligned position data, the 3D point cloud data, and the image data are weighted and fused to obtain the fused state data:

[0040] In the formula, For the fusion state data, For the first The weighting coefficients of each sensor For the first The aligned position data, the 3D point cloud data, or the image data collected by each sensor. The number of sensors participating in the fusion.

[0041] It should be noted that, considering the complex environment of ship unloading operations, the susceptibility of single sensing methods to obstruction and interference, and the significant differences in time scale, accuracy, and stability of data collected by different sensors, this invention utilizes a collaborative configuration of various industrial-grade devices, such as encoders, laser rangefinders or 3D scanners, and vision sensors, to comprehensively capture the operating status of the ship unloader, the grab's attitude, and environmental information of the operating area, thereby constructing a complementary sensing network. During multi-source data processing, a unified time base and data synchronization mechanism are established to accurately align data from different physical channels to a unified timestamp, and the aforementioned weighted fusion algorithm is used to effectively integrate key parameters. For example... Figure 4 As shown, Figure 4 This is a comparison chart of single-sensor and multi-source fusion positioning errors provided in an embodiment of the present invention. From... Figure 4 As can be seen, compared with single-sensor measurement, the fluctuation range of position error is significantly reduced after multi-source fusion, the peak error is significantly reduced, and the overall change is more stable. The basic principle of achieving effective fusion of multi-source data lies in using the redundancy and complementarity of multi-dimensional information to overcome the limitations of a single sensor. It can maintain reliable and high-precision state estimation even under harsh conditions such as dust obstruction, mechanical vibration, and electromagnetic interference on site, effectively avoiding control failure due to single-point data anomalies or single sensor failures. This significantly enhances stability and fault tolerance under complex working conditions, which is completely consistent with the design goal of using a unified time base and weighted fusion mechanism. It also lays a solid and stable data input foundation for subsequent data smoothing and filtering processing, accurate extraction of working condition features, and the realization of biomimetic adaptive control.

[0042] In this embodiment, the fused state data is filtered to generate smooth state data, including: The fused state data is processed using the Kalman filter algorithm; Construct the state prediction equation:

[0043] Construct the measurement update equation:

[0044] The result is obtained by calculating the state prediction equation and the measurement update equation. State estimate at time 1 The state estimate As the smoothed state data; in, for The predicted state value at time 10:00. Here is the state transition matrix. for State estimate at time 10:00 To control the input matrix, for Time-based control input, The state prediction covariance matrix, The process noise covariance matrix is... For Kalman gain, For the observation matrix, To observe the noise covariance matrix, for The observed value at time, It is the identity matrix. for The state covariance matrix at time t.

[0045] It should be noted that mechanical vibrations, electromagnetic interference, and environmental changes at the unloading site can easily introduce noise and outliers into the sensor data, thus affecting the effectiveness of feature analysis and the accuracy of control. Therefore, this invention introduces a Kalman filter algorithm to smooth the continuously sampled fused state data. For example... Figure 5 As shown, Figure 5 This is a comparison diagram of velocity signals before and after Kalman filtering, provided in an embodiment of the present invention. (Combined with...) Figure 5As can be seen, the original speed measurement signal exhibits significant jitter, while the curve after Kalman filtering more closely reflects the actual trend. The filtering principle lies in using recursive calculations of state prediction and measurement updates to dynamically eliminate deviations caused by measurement noise, utilizing the estimated value from the previous moment and the observed value from the current moment. This filtering process effectively suppresses high-frequency noise signals while maintaining a good dynamic response, thus providing smoother, more stable, and reliable data input for subsequent accurate analysis of work cycle characteristics, identification of changing work conditions, and adaptive adjustment of biomimetic PID control parameters.

[0046] In this embodiment, the smoothed state data is parsed to extract the operational characteristic parameters of the current working condition, including: Extract single grab cycle data from the smoothed state data; The single grab cycle data is analyzed to extract grab motion rhythm parameters, acceleration and deceleration control parameters, attitude adjustment law parameters, and material height distribution parameters; The operation characteristic parameters are constructed by combining the grab movement rhythm parameters, the acceleration and deceleration control parameters, the attitude adjustment law parameters, and the material height distribution parameters.

[0047] It should be noted that, given the obvious periodicity of unloading operations, this invention focuses on the complete process of a single grab bucket material handling, lifting, translation, and unloading. By installing high-precision, high-speed sampling sensors, it records in real-time the changes in the trolley encoder speed and acceleration, grab bucket attitude angle, speed and acceleration, and the encoder angle for grab bucket opening and closing actions—all underlying operational data—within the manual operation cycle. During feature extraction, the backend performs comprehensive analysis, calculation, and parsing of this smoothed state data, combining it with previously acquired information on ship type, material type, hull structure, and material level height to accurately extract the material height distribution and grabbing area characteristics under the current operating conditions. This invention extracts the massive and complex low-level sensor signals from the field into high-dimensional feature expressions that reflect the real operational logic in stages and under different working conditions. It can accurately capture and quantify the core control rhythm and posture adjustment patterns of skilled drivers under different specific working conditions, and transform the originally nonlinear and difficult-to-describe human experience in unloading operations into standardized feature inputs that the system can read. This lays a key data and decision foundation for the subsequent construction of standardized operational experience models, matching deep learning mapping relationships, and generating intelligent bionic control strategies that closely fit human operating habits.

[0048] In this embodiment, before inputting the job feature parameters into a preset feature-parameter mapping model, the method further includes: Retrieve historical job data; The historical operation data is parsed into historical operation feature parameters and historical operation control parameters; The historical operation feature parameters are used as input samples, and the historical operation control parameters are used as corresponding labels to train a deep learning model. The deep learning model after training is extracted as the preset feature and parameter mapping model.

[0049] It should be noted that, addressing the problem of significant nonlinear characteristics and difficulty in explicitly describing control rules in unloading operations, this invention accumulates a vast amount of historical operational data from experienced drivers and categorizes it according to working conditions such as ship type, material type, and operational stage. The basic principle of its mapping modeling lies in constructing a mapping model between operational features and control parameters using deep learning methods. Through learning from this historical experience data, the model can automatically discover and analyze potential patterns such as the trends in trolley speed and acceleration changes and the logic of grab bucket attitude adjustments during manual operation, thereby achieving intelligent mapping from on-site environmental data to the final control strategy. Figure 6 As shown, Figure 6 The figure shows a comparison of the positioning responses of the three control strategies provided in the embodiments of the present invention. The figure demonstrates that the biomimetic AI control recovers faster during disturbances and exhibits smaller steady-state fluctuations, reflecting the adaptability advantages of deep learning mapping combined with empirical strategy matching to complex working conditions; for example... Figure 7 As shown, Figure 7 This is a comparison diagram of the grab bucket swing angles for three control strategies provided in this embodiment of the invention. The diagram shows that the bionic AI control, by fitting the rhythm of a skilled driver, makes the grab bucket swing angle decay faster and the peak value lower; as shown... Figure 8 As shown, Figure 8 The comparison chart of single-cycle time and relative energy consumption provided in the embodiments of the present invention reflects that the biomimetic AI control can shorten the cycle time and reduce energy consumption fluctuations, and performs more balanced in terms of efficiency and energy consumption, making it suitable for long-term continuous unloading operations.

[0050] Table 2 Comparison of Control Performance Indicators (including Disturbance Conditions)

[0051] Meanwhile, as shown in Table 2, which is a comparison table of control performance indicators (including disturbance conditions) provided by the embodiments of the present invention, the comparison results show that when using fixed PID control, the positioning RMSE is 7.081m, the overshoot is 30.9%, the settling time is 26.5s, and the maximum swing angle is 2.64°. However, after applying the bionic AI control of the present invention, the positioning RMSE is significantly reduced to 5.402m, the overshoot is greatly reduced to 4.6%, the settling time is shortened to 9.7s, and the maximum swing angle is reduced to 1.12°. The present invention can automatically match the optimal working logic and execute it accurately, flexibly adapting to changes in working conditions such as different ship structures and material distribution in the cabin, providing solid data-driven support for bionic control, and greatly improving the smoothness of operation and the accuracy of operation while ensuring system stability. In addition, the generated standardized operation execution logic can also be directly used for operation guidance and teaching of new employees, effectively shortening the training cycle of professional operators.

[0052] In this embodiment, the dynamic adjustment parameters are input into the base controller, and the gain parameters of the base controller are updated online, including: The basic controller is a proportional-integral-derivative controller; The dynamic adjustment parameters include the proportional gain adjustment amount; The proportional gain of the proportional-integral-derivative controller is updated according to the following formula:

[0053] In the formula, For the updated proportional gain, This is the initial proportional gain. The proportional gain adjustment amount, For error-based and error change rate The adaptive adjustment function.

[0054] It should be noted that this invention is based on the mature PID control algorithm, and its core formula is: ,in, For controller output, This indicates the deviation between the target position and the actual position. , , These are proportional, integral, and derivative gains, respectively. Based on this, biomimetic control logic is incorporated to adaptively adjust the PID parameters according to operational errors and their rate of change. The generated control commands are sent to the PLC execution system via the control interface. The PLC coordinates and controls the various actuators of the unloader to complete the full range of operations, including grabbing, lifting, translating, and unloading. While ensuring system stability, this makes the equipment's operating rhythm more aligned with the habits of skilled drivers, effectively integrating traditional control methods with biomimetic strategies. This significantly improves operational smoothness and precision, fundamentally solving the technical defects of traditional fixed PID controllers, such as excessive overshoot, slow convergence, and more pronounced oscillations during acceleration and deceleration.

[0055] In this embodiment, control commands are generated by updating the gain parameters of the base controller, and the motion mechanism and grab bucket of the unloader are controlled based on the control commands, including: The basic controller, after updating the gain parameters, generates initial control commands. Based on the smoothing state data, the relative position data between the grab bucket and the working area is extracted, and the real-time distance is calculated; The real-time distance is compared with a preset safety threshold; When the real-time distance is greater than or equal to the preset safety threshold, the preliminary control command is used as the control command; When the real-time distance is less than the preset safety threshold, a deceleration or shutdown anti-collision control command is generated, and the anti-collision control command is used as the control command; The control commands are sent to the actuator to control the movement mechanism and grab bucket of the unloader.

[0056] It should be noted that this invention constructs a dual hardware and software safety protection system covering all scenarios, including equipment collision avoidance, personnel safety, and operational protection, when executing control commands. At the software control level, as described in the collision avoidance logic of this embodiment, the system accurately calculates the real-time distance between the grab bucket and the ship's hold, baffle plate, and the unloader body by acquiring the coordinates of each mechanism of the unloader, the real-time updated 3D scanning model, and the grab bucket tracking data, and compares this distance with a preset safety threshold to achieve automatic collision avoidance protection. At the overall system architecture and hardware coordination level, such as... Figure 9 As shown, Figure 9This is a system architecture diagram of an automated ship unloader provided in an embodiment of the present invention. The system is data-driven at its core, decoupling the perception, analysis, decision-making, and execution functions during the unloading process. It relies on hardware such as an automated PLC and a core switch to form a highly reliable closed loop for instruction execution and monitoring. To further enhance safe execution capabilities under extreme conditions, this invention can accurately predict and promptly intervene in collision and injury risks that may be caused by ship tilting, abnormal grab swing, or unauthorized personnel intrusion. It completely compensates for the safety blind spots inherent in traditional obstacle avoidance methods that rely solely on manual visual inspection. While achieving fully automated and intelligent material handling operations, it constructs highly reliable multiple safety protection barriers for the equipment itself, berthed vessels, and on-site personnel.

[0057] Example 3 like Figure 10 As shown, this embodiment proposes a bridge grab unloader control device, which is applied to the bridge grab unloader control method described in the above embodiment, including: acquisition module 100, fusion module 200, filtering module 300, extraction module 400, generation module 500, update module 600 and control module 700.

[0058] The system comprises the following modules: an acquisition module 100 for acquiring multi-source sensor data from the ship unloader; a fusion module 200 for performing time-base alignment and weighted fusion processing on the multi-source sensor data to obtain fused state data; a filtering module 300 for filtering the fused state data to generate smoothed state data; an extraction module 400 for parsing the smoothed state data to extract operational feature parameters of the current working condition; a generation module 500 for inputting the operational feature parameters into a preset feature-parameter mapping model to match and generate dynamic adjustment parameters for the basic controller; an update module 600 for inputting the dynamic adjustment parameters into the basic controller to update the gain parameters of the basic controller online; and a control module 700 for generating control commands through the updated gain parameters of the basic controller and controlling the movement mechanism and grab bucket of the ship unloader based on the control commands.

[0059] It should be noted that the foregoing explanation of the control method embodiment for the bridge grab unloader also applies to the control device of the bridge grab unloader in this embodiment, and will not be repeated here.

[0060] Example 4 Figure 11 This is a schematic diagram of the structure of the electronic device 800 provided in this embodiment. The electronic device 800 includes: a memory 801, a processor 802, and a computer program stored in the memory 801 and executable on the processor 802.

[0061] When the processor 802 executes the program, it implements the bridge grab unloader control method provided in the above embodiments.

[0062] Furthermore, the electronic device 800 also includes a communication interface 803 for communication between the memory 801 and the processor 802.

[0063] The memory 801 may include high-speed RAM (Random Access Memory) memory, and may also include non-volatile memory, such as at least one disk storage.

[0064] If the memory 801, processor 802, and communication interface 803 are implemented independently, then the communication interface 803, memory 801, and processor 802 can be interconnected via a bus to complete communication between them. The bus can be an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component Interconnect) bus, or an EISA (Extended Industry Standard Architecture) bus, etc. The bus can be divided into address bus, data bus, control bus, etc. For ease of representation, Figure 11 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0065] Optionally, in a specific implementation, if the memory 801, processor 802, and communication interface 803 are integrated on a single chip, then the memory 801, processor 802, and communication interface 803 can communicate with each other through an internal interface.

[0066] The processor 802 may be a CPU (Central Processing Unit), an ASIC (Application Specific Integrated Circuit), or one or more integrated circuits configured to implement embodiments of the present invention.

[0067] This invention also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described bridge grab unloader control method.

[0068] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0069] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this invention, "N" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0070] Any process or method description in the flowchart or otherwise herein can be understood as representing a module, segment, or portion of code comprising one or more N executable instructions for implementing custom logic functions or processes, and the scope of preferred embodiments of the invention includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order depending on the functions involved, as should be understood by those skilled in the art to which embodiments of the invention pertain.

[0071] It should be understood that various parts of the present invention can be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, the N steps or methods can be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware as in another embodiment, it can be implemented using any of the following techniques known in the art, or a combination thereof: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (FPGAs), field-programmable gate arrays (FPGAs), etc.

[0072] Those skilled in the art will understand that all or part of the steps of the methods described in the above embodiments can be implemented by a program instructing related hardware, and the program can be stored in a computer-readable storage medium. When executed, the program includes one or a combination of the steps of the method embodiments.

[0073] Obviously, the above embodiments of the present invention are merely examples for clearly illustrating the present invention, and are not intended to limit the implementation of the present invention. Those skilled in the art can make other variations or modifications based on the above description. It is neither necessary nor possible to exhaustively describe all embodiments here. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the scope of protection of the claims of the present invention.

Claims

1. A control method for a bridge-type grab unloader, characterized in that, include: Acquire multi-source sensor data from the ship unloader; The multi-source sensor data is time-referenced and weighted to obtain fused state data. The fused state data is filtered to generate smooth state data; The smoothed state data is parsed to extract the operational characteristic parameters of the current working condition; The operation feature parameters are input into a preset feature-parameter mapping model to match and generate dynamic adjustment parameters for the basic controller; The dynamic adjustment parameters are input into the base controller to update the gain parameters of the base controller online. The base controller generates control commands by updating the gain parameters, and controls the motion mechanism and grab bucket operation of the unloader based on the control commands.

2. The control method for a bridge-type grab unloader according to claim 1, characterized in that, Acquire multi-source sensor data from the ship unloader, including: The position data of the motion mechanism is obtained through the encoder; Three-dimensional point cloud data of the ship's cabin and materials in the work area are obtained using a laser scanner; Image data of the grab bucket is acquired using a visual sensor; The location data, the 3D point cloud data, and the image data are combined to construct the multi-source sensing data.

3. The control method for a bridge-type grab unloader according to claim 2, characterized in that, The multi-source sensor data is time-reference aligned and weighted fused to obtain fused state data, including: Align the location data, the 3D point cloud data, and the image data to the same timestamp; According to the following formula, the aligned position data, the 3D point cloud data, and the image data are weighted and fused to obtain the fused state data: In the formula, For the fusion state data, For the first The weighting coefficients of each sensor For the first The aligned position data, the 3D point cloud data, or the image data collected by each sensor. The number of sensors participating in the fusion.

4. The control method for a bridge-type grab unloader according to claim 3, characterized in that, The fused state data is filtered to generate smooth state data, including: The fused state data is processed using the Kalman filter algorithm; Construct the state prediction equation: Construct the measurement update equation: The result is obtained by calculating the state prediction equation and the measurement update equation. State estimate at time 1 The state estimate As the smoothed state data; in, for The predicted state value at time 10:

00. Here is the state transition matrix. for State estimate at time 10:00 To control the input matrix, for Time-based control input, The state prediction covariance matrix, The process noise covariance matrix is... For Kalman gain, For the observation matrix, To observe the noise covariance matrix, for The observed value at time, It is the identity matrix. for The state covariance matrix at time t.

5. The control method for a bridge-type grab unloader according to claim 1, characterized in that, The smoothed state data is parsed to extract the operational characteristic parameters of the current working condition, including: Extract single grab cycle data from the smoothed state data; The single grab cycle data is analyzed to extract grab motion rhythm parameters, acceleration and deceleration control parameters, attitude adjustment law parameters, and material height distribution parameters; The operation characteristic parameters are constructed by combining the grab movement rhythm parameters, the acceleration and deceleration control parameters, the attitude adjustment law parameters, and the material height distribution parameters.

6. The control method for a bridge-type grab unloader according to claim 1, characterized in that, Before inputting the job feature parameters into a preset feature-parameter mapping model, the method further includes: Retrieve historical job data; The historical operation data is parsed into historical operation feature parameters and historical operation control parameters; The historical operation feature parameters are used as input samples, and the historical operation control parameters are used as corresponding labels to train a deep learning model. The deep learning model after training is extracted as the preset feature and parameter mapping model.

7. The control method for a bridge-type grab unloader according to claim 1, characterized in that, The dynamic adjustment parameters are input into the base controller, and the gain parameters of the base controller are updated online, including: The basic controller is a proportional-integral-derivative controller; The dynamic adjustment parameters include the proportional gain adjustment amount; The proportional gain of the proportional-integral-derivative controller is updated according to the following formula: In the formula, For the updated proportional gain, This is the initial proportional gain. The proportional gain adjustment amount, For error-based and error change rate The adaptive adjustment function.

8. The control method for a bridge-type grab unloader according to claim 1, characterized in that, The system generates control commands by updating the gain parameters of the base controller, and controls the motion mechanism and grab bucket operation of the unloader based on the control commands, including: The basic controller, after updating the gain parameters, generates initial control commands. Based on the smoothing state data, the relative position data between the grab bucket and the working area is extracted, and the real-time distance is calculated; The real-time distance is compared with a preset safety threshold; When the real-time distance is greater than or equal to the preset safety threshold, the preliminary control command is used as the control command; When the real-time distance is less than the preset safety threshold, a deceleration or shutdown anti-collision control command is generated, and the anti-collision control command is used as the control command; The control commands are sent to the actuator to control the movement mechanism and grab bucket of the unloader.

9. A control device for a bridge-type grab unloader, characterized in that, include: The acquisition module is used to acquire multi-source sensor data from the ship unloader; The fusion module is used to perform time-base alignment and weighted fusion processing on the multi-source sensor data to obtain fused state data; The filtering module is used to filter the fused state data to generate smooth state data; The extraction module is used to parse the smoothed state data and extract the operational characteristic parameters of the current working condition; The generation module is used to input the operation feature parameters into a preset feature-parameter mapping model and match and generate dynamic adjustment parameters for the basic controller. The update module is used to input the dynamic adjustment parameters into the base controller and update the gain parameters of the base controller online. The control module is used to generate control commands by updating the gain parameters of the base controller, and to control the motion mechanism and grab bucket operation of the unloader based on the control commands.

10. An electronic device, characterized in that, The electronic device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to perform the operations performed by the bridge grab unloader control method as described in any one of claims 1 to 8.

Images