A motion trajectory control method and system for container reach crane

Abstract

The present application provides a motion trajectory control method and system for a container front crane, which relates to the field of intelligent control technology. The method includes: executing data collection, constructing the lifting area as a three-dimensional fitting scene, then executing three-dimensional scene identification, establishing object sensitive factors, performing sensitive monitoring through a static sensor group, updating the three-dimensional fitting scene in real time, and then performing path fitting optimization to establish an optimal path, establishing M levels of early warning limits, controlling the crane to perform trajectory control fitting, and then performing real-time trajectory tracking, generating tracking compensation and performing motion trajectory control. This application can solve the technical problems of inaccurate motion trajectory control in the prior art, lack of real-time data information interaction, resulting in the inability to perform safety limits according to the trajectory position in a timely and accurate manner, and further affecting the control accuracy and efficiency, thereby achieving precise control and optimization of the crane motion trajectory and improving operating efficiency, safety and reliability.

Description

Technical Field

[0001] The present application relates to the field of intelligent control technology, and in particular to a motion trajectory control method and system for a container reach crane. Background Art

[0002] Container reach cranes are important loading and unloading equipment widely used in ports, docks, cargo yards and other places. Their main function is to lift, transport and stack containers. With the rapid development of the logistics industry and the continuous growth of container transportation volume, higher requirements are placed on the motion trajectory control of container reach cranes. In ports, docks and other places, the working environment is often complex and changeable. Factors such as wind force and uneven ground may affect the motion trajectory of the crane. Therefore, a motion trajectory control method that can adapt to environmental changes is needed to ensure the stable operation of the crane.

[0003] At present, in the actual crane operation process, the crane control method still relies more on the driver's operation, judgment and independent sensor data monitoring of the crane. There are technical problems such as inaccurate motion trajectory control and inability to perform safety limit according to the trajectory position in a timely and accurate manner.

[0004] In summary, the existing technology has inaccurate control over the motion trajectory and lacks real-time data information interaction, resulting in the inability to perform safety limits according to the trajectory position in a timely and accurate manner, further affecting the technical problem of control accuracy and efficiency. Summary of the invention

[0005] The purpose of this application is to provide a motion trajectory control method and system for a container front crane, so as to solve the technical problems that the existing technology has inaccurate motion trajectory control, lacks real-time data information interaction, and results in the inability to perform safety limiting according to the trajectory position in a timely and accurate manner, further affecting the control accuracy and efficiency.

[0006] In view of the above problems, the present application provides a motion trajectory control method and system for a container reach crane.

[0007] In the first aspect, the present application provides a motion trajectory control method for a container front crane, and the method is implemented by a motion trajectory control system of a container front crane, wherein the method includes: performing data collection through a static sensor group arranged in the lifting area, building the lifting area into a three-dimensional fitting scene, and performing three-dimensional scene identification according to the characteristics of the objects in the three-dimensional fitting scene, wherein the three-dimensional scene identification includes an active object identification and a fixed object identification; establishing an object sensitive factor with an active object identification, performing sensitive monitoring through a static sensor group, and updating the three-dimensional fitting scene in real time; performing path fitting optimization based on the three-dimensional fitting scene and the lifting task, and establishing an optimal path; establishing M levels of warning limits with the real-time updated three-dimensional fitting scene and the three-dimensional scene identification, and controlling the crane to perform trajectory control fitting of the optimal path; performing real-time trajectory tracking of the crane through the dynamic sensor group, and inputting the tracking results, the optimal path and the warning limits into the trajectory compensation network to generate path following compensation, and completing the motion trajectory control of the crane through path following compensation.

[0008] In the second aspect, the present application also provides a motion trajectory control system for a container front crane, which is used to execute a motion trajectory control method for a container front crane as described in the first aspect, wherein the system includes: a data acquisition module, which is used to perform data acquisition through a static sensor group arranged in the lifting area, build the lifting area into a three-dimensional fitting scene, and perform three-dimensional scene identification according to the characteristics of the objects in the three-dimensional fitting scene, wherein the three-dimensional scene identification includes an active object identification and a fixed object identification; a sensitive monitoring module, which is used to establish a sensitive factor of an object with an active object identification, perform sensitive monitoring through a static sensor group, and update the information in real time. A new three-dimensional fitting scene; an optimal path establishment module, the optimal path establishment module is used to perform path fitting optimization based on the three-dimensional fitting scene and the lifting task, and establish an optimal path; a trajectory control fitting module, the trajectory control fitting module is used to establish M levels of early warning limits with the real-time updated three-dimensional fitting scene and the three-dimensional scene identifier, and control the crane to perform the trajectory control fitting of the optimal path; a path following compensation module, the path following compensation module is used to perform real-time trajectory tracking of the crane through the dynamic sensor group, and input the tracking results, the optimal path and the early warning limit into the trajectory compensation network, generate path following compensation, and complete the motion trajectory control of the crane through path following compensation.

[0009] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0010] The static sensor group arranged in the lifting area performs data collection, builds the lifting area into a three-dimensional fitting scene, and performs three-dimensional scene identification according to the characteristics of the objects in the three-dimensional fitting scene, wherein the three-dimensional scene identification includes an active object identification and a fixed object identification; establishes an object sensitive factor with an active object identification, performs sensitive monitoring through the static sensor group, and updates the three-dimensional fitting scene in real time; performs path fitting optimization based on the three-dimensional fitting scene and the lifting task, and establishes an optimal path; establishes M levels of early warning limits with the real-time updated three-dimensional fitting scene and the three-dimensional scene identification, and controls the crane to perform trajectory control fitting of the optimal path; performs real-time trajectory tracking of the crane through the dynamic sensor group, and inputs the tracking results, the optimal path and the early warning limit into the trajectory compensation network to generate path following compensation, and completes the motion trajectory control of the crane through the path following compensation, which effectively solves the technical problems that the existing technology is inaccurate in motion trajectory control and lacks real-time data information interaction, resulting in the inability to timely and accurately perform safety limits according to the trajectory position, further affecting the control accuracy and efficiency, and realizes precise control and optimization of the crane motion trajectory, improving the operation efficiency, safety and reliability.

[0011] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented according to the contents of the specification, and in order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are specifically cited below. It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present application, nor is it intended to limit the scope of the present application. Other features of the present application will become easy to understand through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] In order to more clearly illustrate the technical solutions in the present application or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only exemplary, and for ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0013] Figure 1 A schematic flow chart of a motion trajectory control method of a container reach crane in this application;

[0014] Figure 2 This is a structural schematic diagram of a motion trajectory control system of a container reach crane in the present application.

[0015] Description of reference numerals:

[0016] Data acquisition module 11, sensitive monitoring module 12, optimal path establishment module 13, trajectory control fitting 14, path following compensation module 15. DETAILED DESCRIPTION

[0017] The present application provides a motion trajectory control method and system for a container front crane, thereby solving the technical problems in the prior art of inaccurate motion trajectory control, lack of real-time data information interaction, resulting in failure to timely and accurately perform safety limiting according to trajectory position, and further affecting control accuracy and efficiency. The application achieves precise control and optimization of the crane's motion trajectory, and improves operational efficiency, safety, and reliability.

[0018] Below, the technical solutions in the present application will be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments of the present application. It should be understood that the present application is not limited to the example embodiments described herein. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present application. It should also be noted that, for the convenience of description, only the parts related to the present application are shown in the accompanying drawings, rather than all of them.

[0019] Embodiment 1

[0020] Please see attached Figure 1 The present application provides a motion trajectory control method for a container reach crane, wherein the method is applied to a motion trajectory control system of a container reach crane, the system is integrated with an intelligent processor, the intelligent processor digitally communicates with a static sensor group and a dynamic sensor group, the dynamic sensor group is arranged on the crane, and the method specifically comprises the following steps:

[0021] Step 1: Perform data collection by a static sensor group arranged in the lifting area, build the lifting area into a three-dimensional fitting scene, and perform three-dimensional scene identification according to the characteristics of the objects in the three-dimensional fitting scene, wherein the three-dimensional scene identification includes a movable object identification and a fixed object identification;

[0022] Specifically, a static sensor group is arranged in the lifting area to sense and collect data of the lifting area environment in real time. These sensors may include laser scanners, cameras, infrared sensors, etc., to capture information such as the shape, size, and position of various objects in the lifting area. By arranging these sensors in key positions, it is possible to ensure full coverage of the entire lifting area, thereby obtaining complete environmental data. Based on the collected environmental data, a three-dimensional fitting scene of the lifting area can be constructed. The collected data is preprocessed and feature extracted. The preprocessed sensor data is converted into a point cloud format, and each point in the point cloud represents a position in a three-dimensional space. Through algorithm processing such as Poisson surface reconstruction and greedy projection triangulation, the point cloud data is converted into a continuous three-dimensional surface model, making the spatial structure of the lifting area clearer and more accurate. The key features of each object are extracted from the three-dimensional fitting scene, including the shape, size, position, color, texture, etc. of the object. According to the extracted key features, the objects are divided into active objects and fixed objects, where active objects are objects that need to be operated or moved during the operation, such as containers, and fixed objects can be obstacles in the scene, such as buildings and road signs. For active objects, the motion status of active objects, such as position change, moving speed, direction, etc., is monitored in real time through sensor data or video streams. A unique identification code is assigned to each active object so that it can be accurately identified and managed in subsequent trajectory control and safety limit. This identification code can be a number, QR code, or other form of identifier. By tracking the position and status of these objects in real time, it can be ensured that the crane can accurately identify and grab the target container during operation. At the same time, the future position of the active objects can be predicted based on information such as the moving trajectory and speed of the active objects, so as to plan the movement path of the crane in advance. For fixed objects, their static features, such as position, shape, size, etc., are monitored. These features are usually relatively stable in the three-dimensional fitting scene, do not change significantly over time, and do not need to be updated frequently. By identifying these fixed objects in the three-dimensional scene, this identification code can be a number, QR code, or other form of identifier. It can avoid collision or friction between the crane and them during operation.

[0023] Step 2: Establish item sensitive factors with active item identification, perform sensitive monitoring through static sensor groups, and update the three-dimensional fitting scene in real time;

[0024] Specifically, according to the characteristics of the moving objects, determine the sensitive factors that affect their motion trajectory. Sensitive factors may include dynamic factors such as the moving speed, acceleration, and direction change of the object, as well as static factors such as the relative position and distance between the object and the surrounding environment. Select sensors suitable for monitoring moving objects, such as laser radar, high-definition camera, infrared sensor, etc., and arrange them reasonably according to the actual environment of the lifting area. The motion data of the moving objects, including position, speed, direction, etc., are collected in real time through the static sensor group. The collected data is preprocessed, such as filtering and denoising, to improve the accuracy and reliability of the data. The motion state of the moving objects is monitored in real time according to the sensitive factors of the objects. The sensor data is processed through algorithm analysis, for example, the position and type of the object are identified using support vector machine or decision tree algorithm; the speed and acceleration changes of the object are predicted using regression algorithm. Thus, feature information related to sensitive factors is extracted, including the position, speed, acceleration, direction and other features of the object. According to the data collected in real time by the static sensor group, the position and state information of the moving objects in the three-dimensional fitting scene are continuously updated. Ensure that the scene model is synchronized with the actual environment to reflect the latest status of the moving objects.

[0025] Step 3: Perform path fitting optimization based on the three-dimensional fitting scene and the lifting task to establish an optimal path;

[0026] Specifically, first, it is necessary to conduct a detailed analysis of the lifting task to clarify the task's objectives, constraints, and optimization indicators. For example, the task objective may be to move an object from the starting point to the end point, and the constraints may include avoiding collisions, keeping the speed within a certain range, etc. The optimization indicators are the shortest path, the shortest time, or the lowest energy consumption. According to the results of the task analysis, set the objectives and constraints of the path planning. The objectives are used to quantitatively evaluate the pros and cons of different paths, and the constraints are used to limit the scope of path planning to ensure that the planned path meets the actual needs. The three-dimensional fitting scene is used as the basis for path planning. This scene contains information such as the location, shape, and size of all fixed and movable objects in the lifting area. By using this information, various potential obstacles and constraints can be taken into account when planning the path. According to the objectives and constraints of path planning, select a suitable path fitting optimization algorithm. These algorithms can be graph-based search algorithms such as A* algorithm, Dijkstra algorithm, etc., or sampling-based planning algorithms such as RRT algorithm, PRM algorithm, etc., or optimization-based planning algorithms such as gradient descent method, genetic algorithm, etc. The selected algorithm is used to perform path fitting optimization in the three-dimensional fitting scene. The algorithm searches for possible paths in the scene based on the goals and constraints, and evaluates the pros and cons of each path. Through iterative optimization, the algorithm gradually converges to an optimal path that meets the requirements. After the algorithm converges, an optimal path that meets the task requirements is obtained.

[0027] Step 4: establishing M levels of warning limits with the real-time updated three-dimensional fitting scene and the three-dimensional scene identifier, and controlling the crane to execute the trajectory control fitting of the optimal path;

[0028] Specifically, according to the real-time updated three-dimensional fitting scene and three-dimensional scene identification, the position, state and relative relationship of the active objects and fixed objects in the lifting area are analyzed. According to the results of the scene analysis, combined with the crane's movement ability and operation requirements, M different warning levels are determined. Among them, M is a positive integer. These levels can be divided according to factors such as the distance between the object and the crane, the moving speed of the object, and the operation priority of the crane. At each warning level, the corresponding warning limit is set. The warning limit can be a spatial boundary, a distance threshold or a time limit, such as a safe operation area, a dangerous approach area, a prohibited entry area, etc., which are used to trigger different warning responses and trajectory adjustment strategies. Based on the real-time updated three-dimensional fitting scene and three-dimensional scene identification, a path planning algorithm is used to generate an optimal path for the crane from the current position to the target position. This path should try to avoid collisions with active objects and fixed objects and shorten the operation time as much as possible. According to the planned optimal path, the trajectory of the crane is fitted by the trajectory control algorithm. This includes controlling the crane's motion parameters such as speed, acceleration, and direction to ensure that it can move accurately according to the planned path. During the movement of the crane, its distance and relative position with surrounding objects are monitored in real time. When approaching or reaching a certain warning limit, the corresponding warning response is triggered, such as deceleration, avoidance or re-planning the path, etc. By continuously adjusting the movement trajectory of the crane, it is ensured that it can complete the task safely and efficiently.

[0029] Step 5: Perform real-time trajectory tracking of the crane through the dynamic sensor group, and input the tracking results, optimal path and warning limit into the trajectory compensation network to generate path following compensation, and complete the motion trajectory control of the crane through path following compensation.

[0030] Specifically, a dynamic sensor group is installed on the crane, including but not limited to gyroscopes, accelerometers, displacement sensors, etc., for real-time monitoring of the motion state of the crane. The dynamic sensor group collects the position, speed, acceleration and other motion data of the crane in real time, and filters, denoises and calibrates them through data processing algorithms to ensure the accuracy and reliability of the data. Based on the collected motion data, the actual motion trajectory of the crane is calculated in real time through a trajectory tracking algorithm, and compared with the planned optimal path. A trajectory compensation network is established, which can be a deep learning model or other type of machine learning model, which is used to generate path following compensation based on real-time trajectory tracking results, optimal paths and warning limits. The real-time trajectory tracking results, optimal paths and warning limits are used as input features and preprocessed to meet the input requirements of the trajectory compensation network. The trajectory compensation network learns and generates path following compensation based on the input features, which can be an adjustment amount for the crane motion parameters, such as speed, acceleration or direction adjustment. The generated path following compensation is applied to the motion control system of the crane, and the motion parameters of the crane are adjusted so that it can better follow the planned optimal path and avoid conflicts with the warning limits.

[0031] Furthermore, the present application also includes:

[0032] Execute network initialization of trajectory compensation network, including:

[0033] A1: Acquire real-time weather data and lifting data, wherein the lifting data includes the mass data and volume data of the lifting object, and the real-time weather data is collected;

[0034] A2: Reset the constraint layer data using the real-time weather data and the lifting data, wherein the constraint layer is a hidden layer of the trajectory compensation network, and complete network initialization according to the reset of the constraint layer data.

[0035] Specifically, real-time weather information such as wind speed, wind direction, temperature, humidity, etc. is obtained through meteorological sensors. These weather factors will directly affect the working environment of the crane, and then affect its motion trajectory and stability. Lifting data, including mass data and volume data of the lifting object, are obtained through weight sensors. These data describe the load conditions that the crane needs to handle at present, which is crucial for determining the motion parameters and trajectory compensation strategy of the crane. The constraint layer data of the trajectory compensation network is reset using the collected real-time weather data and lifting data. The constraint layer here refers to the hidden layer in the network, which is responsible for extracting features from the input data and passing them to the output layer. The real-time weather data and lifting data are preprocessed, such as normalization, standardization or encoding, to adapt them to the input requirements of the network. Features that have a significant impact on trajectory compensation, such as temperature, humidity, wind speed, lifting mass, volume, etc., are selected from the preprocessed data. These features are spliced ​​into a comprehensive feature vector in a certain order. The length of the vector should be equal to the total number of selected features. This vector will be used as the input of the constraint layer. According to the dimension of the fused feature vector, the weights of each neuron in the constraint layer are randomly initialized. Including uniform distribution initialization, normal distribution initialization or He / Glorot-based initialization methods. Randomly initialize the bias value for each neuron in the constraint layer. The initialization of the bias is usually a small constant. Select an activation function that can introduce nonlinearity, such as ReLU, Sigmoid, etc., and fine-tune the weight of the constraint layer according to the fused feature vector and specific constraints such as safety thresholds, job requirements, etc.

[0036] Furthermore, the present application also includes:

[0037] The tracking fitting layer is used to perform deviation evaluation on the optimal path and the tracking result, and N levels of preview distances are configured according to the deviation evaluation results, and corresponding N preview points are established according to the N levels of preview distances, and trajectory compensation fitting is performed according to the N preview points to generate N path compensation results, wherein the tracking fitting layer is a hidden layer coupled to the trajectory compensation network for performing trajectory fitting processing;

[0038] The N path compensation results are shared to the position limit constraint layer, the position limit constraint layer is updated by the early warning position limit, and the position limit evaluation of the N path compensation results is performed based on the early warning position limit to complete a round of screening;

[0039] The first round of screening is continued through the constraint layer to obtain the second round of screening;

[0040] A path following compensation is generated based on the two rounds of screening.

[0041] Specifically, the tracking and fitting layer receives the real-time trajectory tracking results and the planned optimal path as input. This layer is responsible for performing trajectory fitting processing and quantitatively evaluating the deviation between the two. Based on the processing results of the tracking and fitting layer, the deviation value between the real-time trajectory and the optimal path is calculated, and the deviation value can be calculated by deviation metrics such as Euclidean distance and Manhattan distance. These deviation values ​​reflect the difference between the actual movement of the crane and the ideal path. According to the calculated deviation value, the performance of the current trajectory tracking is evaluated to provide a basis for the subsequent preview distance configuration. Based on the deviation evaluation results, N preview distances of different levels are dynamically configured. Among them, N is a positive integer. The preview distance can be adjusted according to factors such as the size of the deviation, the movement state of the crane, and the working environment. A corresponding preview point is established on the optimal path with each preview distance. These preview points will be used as reference points for trajectory compensation. The preview points are fitted using the tracking and fitting layer, which can include interpolation and curve fitting to generate a smooth and constrained compensation trajectory. For each preview point, a corresponding path compensation result is generated. These results will be used as candidate compensation paths for subsequent screening and selection. The limit constraint layer dynamically updates the warning limit according to factors such as the current state of the crane, the operating environment, and safety requirements. The N path compensation results are input into the limit constraint layer, and the limit evaluation is performed based on the warning limit. This includes checking whether each compensation path exceeds the physical limit, safety threshold, or other constraints. According to the results of the limit evaluation, the path compensation results that do not meet the limit requirements are screened out. The path compensation results after being screened by the limit constraint layer are input into the constraint layer for further processing and screening. The constraint layer may contain other types of constraints, such as smoothness constraints, speed constraints, etc. The constraint layer performs a second round of screening on the path compensation results according to its own constraints to further screen out better candidate paths. Based on the final candidate path after two rounds of screening, the final path following compensation is generated. This compensation will be used to adjust the motion trajectory of the crane to achieve more accurate and safe trajectory control.

[0042] Furthermore, the present application also includes:

[0043] Configure the evaluation allocation ratio of the evaluation layer, wherein the evaluation allocation ratio is the normalized weight ratio of the limit warning and the approach speed;

[0044] The path optimization of the second round of screening is performed through the path approach speed, path limit warning and the evaluation allocation ratio of the second round of screening, and the path following compensation is constructed based on the path optimization result.

[0045] Specifically, first, identify the main evaluation factors that affect path optimization. The main evaluation factors include displacement limit warning and approach speed. Each evaluation factor is normalized to determine their weight ratio. The normalized weight ratio should reflect the relative importance of each factor in the path optimization process. It can be obtained through machine learning and other methods. Based on the normalized weight ratio, set the evaluation allocation ratio of the evaluation layer. Extract the approach speed information of each path from the results of the second round of screening. The approach speed reflects the speed of the crane moving to the target position. At the same time, obtain the limit warning information of each path. The limit warning indicates the degree of proximity between the path and the warning limit, which is an important indicator to ensure safe operation. Combine the path approach speed, path limit warning and the evaluation allocation ratio of the evaluation layer to perform path optimization calculation. The specific calculation method can be weighted summation, weighted product, etc. Based on the results of path optimization, determine the optimal path. The optimal path should comprehensively consider factors such as approach speed and limit warning to ensure that the crane meets safety requirements while operating efficiently. According to the optimal path, construct path following compensation. Path following compensation should be able to guide the crane to move along the optimal path and achieve accurate tracking of the target position.

[0046] Furthermore, step five of this application also includes:

[0047] Parsing the three-dimensional scene identifier to establish an item level, wherein the item level is an importance level evaluated according to the importance of the item, and the item level includes an active item level and a fixed item level;

[0048] The M levels of warning limit construction are completed based on the item level and the real-time updated three-dimensional fitting scene.

[0049] Specifically, obtain three-dimensional scene data containing the crane operating environment. This data may come from lidar, cameras or other sensors, which together build a three-dimensional, real-time operating environment model. In the three-dimensional scene data, identify and extract the identification information of various items. This identification information may include the location, size, shape, material, etc. of the item, as well as any tags related to the importance or function of the item. According to the extracted identification information, the importance of the items in the scene is evaluated. The basis for the evaluation may include factors such as the value of the item, the impact on the operating process, and safety. Based on the results of the importance evaluation, the items are divided into different levels. Including active item level and fixed item level. Active items may be goods being operated by the crane, while fixed items may be fixed obstacles or structures in the operating environment. According to the item level, determine the type and number of warning limits required for items of different levels. For example, for items with high value or high safety risks, set stricter and more intensive warning limits. Since the operating environment may change at any time, such as the addition of new items, the removal of old items, changes in the environment, etc., the three-dimensional fitting scene is updated in real time to ensure the accuracy of the warning limit. Based on the real-time updated 3D fitting scene and item level information, M different levels of warning limits are constructed. M is a positive integer. These warning limits may include distance limits, height limits, speed limits, etc. They will form a multi-level safety protection network to prevent the crane from colliding with or approaching dangerously with items of different levels.

[0050] Furthermore, the present application also includes:

[0051] Real-time data monitoring is performed through the dynamic sensor group, and path obstacle triggering verification is performed based on the real-time monitoring results and path following compensation;

[0052] If the preset trigger threshold is met, an operation abnormality is reported and a shutdown instruction is generated;

[0053] The crane is controlled to stop and maintain according to the stop instruction, and an abnormal warning is issued based on the reported operating abnormality.

[0054] Specifically, the dynamic sensor group includes weight sensors, height sensors, tilt sensors, etc. to collect various operating data of the crane in real time, such as weight, height, speed, angle, etc. The collected data is transmitted to the data processing unit of the trajectory compensation network for necessary preprocessing and conversion for subsequent analysis and judgment. The real-time monitoring results are compared with the path following compensation to analyze the difference between the current actual operating trajectory of the crane and the expected trajectory. According to the preset obstacle trigger conditions such as deviation threshold, speed threshold, angle threshold, etc., it is judged whether there is a path obstacle trigger at present. These thresholds can be set according to factors such as the type of crane, the operating environment, and safety requirements. If the real-time monitoring results meet the preset trigger threshold, it is judged that there is an operation abnormality or path obstacle. Immediately report the operation abnormality and generate the corresponding shutdown command. According to the generated shutdown command, the control system immediately executes the shutdown operation to stop the crane at the current position and keep it stationary. At the same time, the reported operation abnormality information is sent to relevant personnel or equipment for abnormal warning. This can be achieved through sound and light alarms, SMS notifications, email reminders, etc., so that relevant personnel can understand and handle abnormal situations in time.

[0055] Furthermore, the present application also includes:

[0056] Collecting user response feedback, wherein the response feedback is a feedback result of an actual control result of the crane;

[0057] sending the response feedback to the trajectory compensation network and executing the trajectory compensation network additional constraints;

[0058] The subsequent trajectory control adjustment is performed according to the trajectory compensation network after the additional constraints.

[0059] Specifically, after the user completes the task using the crane, the user's response feedback is collected. These feedback data may include the user's evaluation of the control accuracy, satisfaction with the crane's motion trajectory, whether a collision occurs or exceeds the safety limit, etc. After the feedback data is collected, it is sorted and analyzed to extract useful information for the trajectory compensation network. This includes classification, statistics and quantitative processing of the feedback data. Before the user's response feedback is sent to the trajectory compensation network, some preprocessing operations are performed on the feedback data, such as data cleaning, standardization or encoding conversion, to ensure that the data format matches the input requirements of the network. The preprocessed feedback data will be integrated into the input of the trajectory compensation network. After the feedback data is integrated into the trajectory compensation network, additional constraint operations are performed. Constraints include restrictions or optimization requirements in terms of operating speed, accuracy, stability, etc. Based on the user's response feedback and additional constraints, the parameters of the trajectory compensation network are updated. This includes adjusting the weights, biases or other hyperparameters of the network through optimization algorithms such as gradient descent and back propagation to improve the performance and accuracy of the network. The updated network parameters are reapplied to the crane operation, new feedback data is collected, and the above steps are repeated for iterative optimization. Through continuous iteration and optimization, the trajectory compensation network can gradually adapt to different operating environments and user needs, achieving more precise and safer trajectory control.

[0060] In summary, the motion trajectory control method of a container reach crane provided in the present application has the following technical effects:

[0061] The static sensor group arranged in the lifting area performs data collection, builds the lifting area into a three-dimensional fitting scene, and performs three-dimensional scene identification according to the characteristics of the objects in the three-dimensional fitting scene, wherein the three-dimensional scene identification includes an active object identification and a fixed object identification; establishes an object sensitive factor with an active object identification, performs sensitive monitoring through the static sensor group, and updates the three-dimensional fitting scene in real time; performs path fitting optimization based on the three-dimensional fitting scene and the lifting task, and establishes an optimal path; establishes M levels of early warning limits with the real-time updated three-dimensional fitting scene and the three-dimensional scene identification, and controls the crane to perform trajectory control fitting of the optimal path; performs real-time trajectory tracking of the crane through the dynamic sensor group, and inputs the tracking results, the optimal path and the early warning limit into the trajectory compensation network to generate path following compensation, and completes the motion trajectory control of the crane through the path following compensation, which effectively solves the technical problems that the existing technology is inaccurate in motion trajectory control and lacks real-time data information interaction, resulting in the inability to timely and accurately perform safety limits according to the trajectory position, further affecting the control accuracy and efficiency, and realizes precise control and optimization of the crane motion trajectory, improving the operation efficiency, safety and reliability.

[0062] Embodiment 2

[0063] Based on the motion trajectory control method of a container reach crane in the aforementioned embodiment, the present application also provides a motion trajectory control system for a container reach crane, see the attached Figure 2 , the system comprising:

[0064] A data acquisition module 11, the data acquisition module 11 is used to perform data acquisition through a static sensor group arranged in the lifting area, build the lifting area into a three-dimensional fitting scene, and perform three-dimensional scene identification according to the characteristics of the objects in the three-dimensional fitting scene, wherein the three-dimensional scene identification includes a movable object identification and a fixed object identification;

[0065] A sensitive monitoring module 12, which is used to establish a sensitive factor of an item with an active item identifier, perform sensitive monitoring through a static sensor group, and update the three-dimensional fitting scene in real time;

[0066] An optimal path establishment module 13, the optimal path establishment module 13 is used to perform path fitting optimization based on the three-dimensional fitting scene and the lifting task, and establish an optimal path;

[0067] A trajectory control fitting module 14, the trajectory control fitting module 14 is used to establish M levels of warning limits with the real-time updated three-dimensional fitting scene and the three-dimensional scene identifier, and control the crane to perform the trajectory control fitting of the optimal path;

[0068] The path following compensation module 15 is used to perform real-time trajectory tracking of the crane through the dynamic sensor group, and input the tracking results, optimal path and warning limit into the trajectory compensation network to generate path following compensation, and complete the motion trajectory control of the crane through path following compensation.

[0069] Furthermore, the system further includes a network initialization module, and the network initialization module is used to:

[0070] Execute network initialization of trajectory compensation network, including:

[0071] A1: Acquire real-time weather data and lifting data, wherein the lifting data includes the mass data and volume data of the lifting object, and the real-time weather data is collected;

[0072] A2: Reset the constraint layer data using the real-time weather data and the lifting data, wherein the constraint layer is a hidden layer of the trajectory compensation network, and complete network initialization according to the reset of the constraint layer data.

[0073] Furthermore, the system further comprises a path following compensation generating module, wherein the path following compensation generating module is used for:

[0074] The tracking fitting layer is used to perform deviation evaluation on the optimal path and the tracking result, and N levels of preview distances are configured according to the deviation evaluation results, and corresponding N preview points are established according to the N levels of preview distances, and trajectory compensation fitting is performed according to the N preview points to generate N path compensation results, wherein the tracking fitting layer is a hidden layer coupled to the trajectory compensation network for performing trajectory fitting processing;

[0075] The N path compensation results are shared to the position limit constraint layer, the position limit constraint layer is updated by the early warning position limit, and the position limit evaluation of the N path compensation results is performed based on the early warning position limit to complete a round of screening;

[0076] The first round of screening is continued through the constraint layer to obtain the second round of screening;

[0077] A path following compensation is generated based on the two rounds of screening.

[0078] Furthermore, the system further comprises a path optimization module, wherein the path optimization module is used to:

[0079] Configure the evaluation allocation ratio of the evaluation layer, wherein the evaluation allocation ratio is the normalized weight ratio of the limit warning and the approach speed;

[0080] The path optimization of the second round of screening is performed through the path approach speed, path limit warning and the evaluation allocation ratio of the second round of screening, and the path following compensation is constructed based on the path optimization result.

[0081] Furthermore, the system trajectory control fitting module 14 is also used for:

[0082] Parsing the three-dimensional scene identifier to establish an item level, wherein the item level is an importance level evaluated according to the importance of the item, and the item level includes an active item level and a fixed item level;

[0083] The M levels of warning limit construction are completed based on the item level and the real-time updated three-dimensional fitting scene.

[0084] Furthermore, the system further comprises a shutdown instruction generating module, wherein the shutdown instruction generating module is used for:

[0085] Real-time data monitoring is performed through the dynamic sensor group, and path obstacle triggering verification is performed based on the real-time monitoring results and path following compensation;

[0086] If the preset trigger threshold is met, an operation abnormality is reported and a shutdown instruction is generated;

[0087] The crane is controlled to stop and maintain according to the stop instruction, and an abnormal warning is issued based on the reported operating abnormality.

[0088] Furthermore, the system further comprises a trajectory control adjustment module, wherein the trajectory control adjustment module is used to:

[0089] Collecting user response feedback, wherein the response feedback is a feedback result of an actual control result of the crane;

[0090] sending the response feedback to the trajectory compensation network and executing the trajectory compensation network additional constraints;

[0091] The subsequent trajectory control adjustment is performed according to the trajectory compensation network after the additional constraints.

[0092] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. Figure 1 The motion trajectory control method and specific examples of a container front crane in the first embodiment are also applicable to a motion trajectory control system of a container front crane in this embodiment. Through the above detailed description of the motion trajectory control method of a container front crane, those skilled in the art can clearly know the motion trajectory control system of a container front crane in this embodiment, so for the sake of brevity of the specification, it will not be described in detail here. For the system disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part description.

[0093] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

[0094] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application belong to the scope of the present application and its equivalent technology, the present application is also intended to include these modifications and variations.

Claims

1. A motion trajectory control method for a container reach crane, the motion trajectory control method for a container reach crane is applied to a motion trajectory control system for a container reach crane, the motion trajectory control system for a container reach crane is integrated with an intelligent processor, the intelligent processor digitally communicates with a static sensor group and a dynamic sensor group, the dynamic sensor group is arranged on the crane, and is characterized in that: The method comprises: The lifting area is constructed as a three-dimensional fitting scene by performing data collection through a static sensor group arranged in the lifting area, and three-dimensional scene identification is performed according to the characteristics of objects in the three-dimensional fitting scene, wherein the three-dimensional scene identification includes a movable object identification and a fixed object identification; Establish item sensitive factors with active item identification, perform sensitive monitoring through static sensor groups, and update the three-dimensional fitting scene in real time; Perform path fitting optimization based on the three-dimensional fitting scene and the lifting task, and establish an optimal path; Establishing M levels of early warning limits with the real-time updated three-dimensional fitting scene and the three-dimensional scene identifier, and controlling the crane to execute the trajectory control fitting of the optimal path; The real-time trajectory tracking of the crane is performed by the dynamic sensor group, and the tracking results, the optimal path and the warning limit are input into the trajectory compensation network to generate the path following compensation, and the motion trajectory control of the crane is completed by the path following compensation; Input the tracking results, optimal path and warning limit into the trajectory compensation network to generate the path following compensation, including: Execute network initialization of trajectory compensation network, including: A1: Acquire real-time weather data and lifting data, wherein the lifting data includes the mass data and volume data of the lifting object, and the real-time weather data is collected; A2: Resetting the constraint layer data using the real-time weather data and the lifting data, wherein the constraint layer is a hidden layer of the trajectory compensation network, and completing network initialization according to the reset of the constraint layer data; After the tracking results, optimal path and warning limit are input into the trajectory compensation network, it includes: The tracking fitting layer is used to perform deviation evaluation on the optimal path and the tracking result, and N levels of preview distances are configured according to the deviation evaluation results, and corresponding N preview points are established according to the N levels of preview distances, and trajectory compensation fitting is performed according to the N preview points to generate N path compensation results, wherein the tracking fitting layer is a hidden layer coupled to the trajectory compensation network for performing trajectory fitting processing; The N path compensation results are shared to the position limit constraint layer, the position limit constraint layer is updated by the early warning position limit, and the position limit evaluation of the N path compensation results is performed based on the early warning position limit to complete a round of screening; The first round of screening is continued through the constraint layer to obtain the second round of screening; A path following compensation is generated based on the two rounds of screening.

2. The method according to claim 1, characterized in that Generating the path following compensation according to the second round of screening includes: Configure the evaluation allocation ratio of the evaluation layer, wherein the evaluation allocation ratio is the normalized weight ratio of the limit warning and the approach speed; The path optimization of the second round of screening is performed through the path approach speed, path limit warning and the evaluation allocation ratio of the second round of screening, and the path following compensation is constructed based on the path optimization result.

3. The method according to claim 1, characterized in that The step of establishing M levels of warning limits by using the real-time updated three-dimensional fitting scene and the three-dimensional scene identifier includes: Parsing the three-dimensional scene identifier to establish an item level, wherein the item level is an importance level evaluated according to the importance of the item, and the item level includes an active item level and a fixed item level; The M levels of warning limit construction are completed based on the item level and the real-time updated three-dimensional fitting scene.

4. The method according to claim 1, characterized in that The method further comprises: Real-time data monitoring is performed through the dynamic sensor group, and path obstacle triggering verification is performed based on the real-time monitoring results and path following compensation; If the preset trigger threshold is met, an operation abnormality is reported and a shutdown instruction is generated; The crane is controlled to stop and maintain according to the stop instruction, and an abnormal warning is issued based on the reported operating abnormality.

5. The method according to claim 1, characterized in that The method comprises: Collecting user response feedback, wherein the response feedback is a feedback result of an actual control result of the crane; sending the response feedback to the trajectory compensation network and executing the trajectory compensation network additional constraints; The subsequent trajectory control adjustment is performed according to the trajectory compensation network after the additional constraints.

6. A motion trajectory control system for a container reach crane, characterized in that: For implementing the steps of the method according to any one of claims 1 to 5, the system comprises: A data acquisition module, the data acquisition module is used to perform data acquisition through a static sensor group arranged in the lifting area, build the lifting area into a three-dimensional fitting scene, and perform three-dimensional scene identification according to the characteristics of objects in the three-dimensional fitting scene, wherein the three-dimensional scene identification includes a movable object identification and a fixed object identification; A sensitive monitoring module, which is used to establish a sensitive factor of an item with an active item identifier, perform sensitive monitoring through a static sensor group, and update the three-dimensional fitting scene in real time; An optimal path establishment module, the optimal path establishment module is used to perform path fitting optimization based on the three-dimensional fitting scene and the lifting task, and establish an optimal path; A trajectory control fitting module, the trajectory control fitting module is used to establish M levels of warning limits with the real-time updated three-dimensional fitting scene and the three-dimensional scene identifier, and control the crane to execute the trajectory control fitting of the optimal path; A path following compensation module is used to perform real-time trajectory tracking of the crane through the dynamic sensor group, and input the tracking results, optimal path and warning limit into the trajectory compensation network to generate path following compensation, and complete the motion trajectory control of the crane through path following compensation.

Images