Unmanned loading model construction and application method based on digital twinning

Abstract

The invention provides an unmanned loading model construction and application method based on digital twinning, and aims to improve the efficiency and accuracy of unmanned loading. According to the method, an accurate three-dimensional digital twin model of the loader, goods and the environment is constructed, and data synchronization between the model and the actual loader is realized, so that the model state is updated in real time. Functionally, according to the technology, the unmanned loading process can be simulated through simulation software, the loading strategy is optimized, meanwhile, potential faults are predicted, and maintenance is carried out in advance. The unmanned loading system has the advantages that compared with a traditional unmanned loading system, more accurate loading process control can be achieved, cargo damage is reduced, the downtime is shortened, and the overall operation efficiency is improved. The innovation is reflected in that the digital twin technology is applied to the field of unmanned loading, and a new thought is provided for intelligence and automation of the loading process.

Description

Technical Field

[0001] This invention relates to the field of intelligent logistics and automated control technology, and in particular to a method for constructing and applying an unmanned loading model based on digital twins, which realizes efficient modeling and intelligent application of unmanned loading operations through digital twin technology. Background Technology

[0002] In the modern logistics industry, loading operations are a crucial link in cargo transportation. Traditional loading operations mainly rely on manual labor, which is not only inefficient but also susceptible to human error, leading to operational errors and safety hazards. With the development of automation technology, unmanned loading equipment has gradually been applied, but current unmanned loading systems still have many problems.

[0003] Existing unmanned loading models are insufficient in adapting to complex operating environments and diverse cargo, making it difficult to accurately simulate various situations during actual loading processes. Furthermore, due to a lack of effective data processing and model optimization methods, the accuracy and reliability of these models need improvement. In addition, in practical applications of unmanned loading, there is a lack of intuitive and effective methods for evaluating and adjusting operational plans, and real-time monitoring and intelligent control of equipment also have limitations, failing to meet the demands of modern logistics for efficiency, safety, and intelligence. Summary of the Invention

[0004] Purpose of the invention The purpose of this invention is to provide a method for constructing and applying unmanned loading models based on digital twins, so as to solve the above-mentioned problems in existing unmanned loading technologies and realize accurate modeling, efficient application and intelligent control of unmanned loading operations. Technical solution

[0005] Data Acquisition and Processing: Comprehensive data from the loading operation site is collected in real time using multiple sensors, including attributes such as cargo size, weight, and shape; status information such as the location, attitude, and operating parameters of the loading equipment; and environmental information such as terrain and obstacle distribution. The collected raw data is preprocessed, employing filtering algorithms to remove noise interference, data interpolation to fill in missing values, and converting the data into a standard format suitable for subsequent model processing to ensure accuracy and completeness.

[0006] Digital twin model construction: 3D geometric model construction: Based on the processed data, a precise 3D geometric model of the unmanned loading equipment is constructed using parametric modeling methods. This model can accurately simulate the mechanical structure and appearance of the equipment and can be quickly adjusted and modified according to different equipment models and specifications to adapt to diverse equipment needs.

[0007] Dynamic model construction: Based on multibody dynamics theory, dynamic models of each component of the equipment are established, taking into full account nonlinear factors such as friction and collision between components, accurately describing the mechanical characteristics and motion laws of the components during the motion process, and providing an accurate basis for equipment motion simulation.

[0008] Workflow Model Construction: Construct a loading operation flow model to simulate the loading sequence of goods, route planning, and collaborative operation logic between multiple devices, thereby achieving a digital simulation of the entire loading operation flow.

[0009] Model fusion: Through a data-driven approach, the above-mentioned three-dimensional geometric model, dynamic model and operation process model are deeply integrated to construct an unmanned loading model based on digital twin, which can reflect the status and behavior of the actual loading operation in real time and accurately.

[0010] Model Validation and Optimization: Actual loading operation data is input into the constructed digital twin model. By comparing the model's output with the actual operation results, the accuracy and reliability of the model are evaluated using metrics such as root mean square error (RMSE) and mean absolute error (MAE). Optimization algorithms are used to adjust and optimize the model's parameters, continuously improving its accuracy and performance. Simultaneously, based on feedback from actual operations, the model is continuously updated and improved to adapt to different loading tasks and complex, ever-changing operating environments.

[0011] Applications of unmanned loading models: Virtual simulation: Based on a digital twin model, virtual reality (VR) or augmented reality (AR) technology is used to simulate loading operations. Before actual operations, operators can intuitively evaluate and optimize the operation plan through an immersive virtual environment, identify potential problems in advance, and develop solutions, effectively reducing the risks of actual operations.

[0012] Real-time monitoring: Utilizing the real-time monitoring function of the digital twin model, the actual loading operation process is monitored in all aspects, and information such as the operating status of the equipment and the progress of the operation is obtained in real time. Abnormal situations are detected and warned in a timely manner to ensure the safe operation.

[0013] Predictive Function: Based on historical and real-time data, the model predicts the future development trend of loading operations, providing a scientific basis for job scheduling and decision-making, and improving operational efficiency and resource utilization.

[0014] Intelligent control: Intelligent control for unmanned loading is achieved based on a digital twin model. Adaptive control algorithms are used to automatically adjust the operating parameters and operation process of the loading equipment according to real-time data and model prediction results during the operation. This enables the equipment to adapt to different operating conditions and working conditions, achieving efficient and safe unmanned loading operations.

[0015] Instruction manual with accompanying drawings Figure 1 This paper describes a digital twin-based unmanned loader model and application method. The process involves multiple roles: MES (Manufacturing Execution System), mixing plant, backend, scheduling system, loader, silo, hopper, and mixer truck. The MES creates a pouring instruction and sends its ID to the mixing plant. Manual operation at the mixing plant begins and status is pushed. The backend retrieves silo information from the MES, updates hopper data in real time, and triggers an automatic task scheduling algorithm. The scheduling system distributes manually created tasks (and also supports task assignment) to the loader, which sequentially performs shoveling and unloading. The mixing plant executes a "start operation - task completion" cycle based on the task. After the task is completed, the loader sends a status signal, ending the process and establishing a collaborative mechanism for the entire unmanned loader operation process.

[0016] Figure 2: The model and application method of this invention for unmanned loaders based on digital twins, constructing a multi-level solution architecture. The perception layer deploys sensors, positioning terminals, monitoring equipment, communication equipment, and network protocol stacks to collect multi-dimensional information; the data layer aggregates diverse data such as equipment status, environmental perception, and work tasks; the support layer relies on hardware infrastructure and software platforms, combined with digital twins, AI algorithms, and network communication systems, to coordinate data management, security systems, standardized protocols, and user interaction modules; the application layer realizes functions such as scheduling, planning, and monitoring, covering multiple types of management and control; and compliance and security are ensured through standards, specifications, and network security systems, providing complete technical support for the intelligent operation management of unmanned loaders.

Claims

1. A method for constructing and applying an unmanned loading model based on digital twins, characterized in that, Includes the following steps: Data Acquisition and Processing: Real-time acquisition of various data from the loading operation site via sensors, including but not limited to cargo dimensions, weight, and shape information, loading equipment location, attitude, and operating parameter information, as well as terrain and obstacle distribution information of the operating environment; preprocessing of the acquired data to remove noise, fill in missing values, and convert the data into a format suitable for model processing.

2. Digital Twin Model Construction: Based on the processed data, a three-dimensional geometric model of the unmanned loading equipment is constructed to accurately simulate the mechanical structure and appearance of the equipment; a dynamic model of each component of the equipment is established to describe the mechanical characteristics and motion laws of the components during the movement process; a loading operation process model is constructed to simulate the loading sequence of goods, path planning, and collaborative operation logic between equipment; through a data-driven approach, the above models are integrated to construct an unmanned loading model based on a digital twin, enabling the model to reflect the status and behavior of the actual loading operation in real time.

3. Model Validation and Optimization: Input the actual loading operation data into the constructed digital twin model, compare the model output results with the actual operation results to verify the accuracy and reliability of the model; use optimization algorithms to adjust and optimize the model parameters to improve the model's accuracy and performance; continuously update and improve the model based on actual operation feedback to make it adaptable to different loading tasks and operating environments.

4. Application of Unmanned Loading Model: Based on a digital twin model, virtual simulation of loading operations is performed. Before actual operations, the operation plan is evaluated and optimized, potential problems are identified in advance, and solutions are developed. The model's real-time monitoring function monitors the actual loading process, acquiring real-time equipment operating status and operation progress, and promptly identifying and issuing warnings of abnormal situations. Through the model's prediction function, the subsequent development trend of the loading operation is predicted, providing a basis for operation scheduling and decision-making. Intelligent control of unmanned loading is achieved based on the digital twin model, automatically adjusting the operating parameters and operation process of the loading equipment according to the model's analysis results, achieving efficient and safe unmanned loading operations.

5. The method for constructing and applying an unmanned loading model based on digital twins according to claim 1, characterized in that, During the data acquisition process, various types of sensors are used for data acquisition, including lidar for acquiring three-dimensional point cloud data of the working environment, cameras for acquiring visual image information, inertial measurement units for measuring the attitude and acceleration information of the loading equipment, and force sensors for detecting forces and torques during the loading process.

6. The method for constructing and applying an unmanned loading model based on digital twins according to claim 1, characterized in that, In the construction of the digital twin model, the three-dimensional geometric model adopts a parametric modeling method, which facilitates rapid adjustment and modification according to different equipment models and specifications; the dynamic model is established based on multibody dynamics theory, taking into account nonlinear factors such as friction and collision between components.

7. The method for constructing and applying an unmanned loading model based on digital twins according to claim 1, characterized in that, In the model validation and optimization steps, indicators such as root mean square error and mean absolute error are used to evaluate the difference between the model output and the actual operation results, serving as the basis for model optimization.

8. The method for constructing and applying an unmanned loading model based on digital twins according to claim 1, characterized in that, In unmanned loading model applications, the virtual simulation uses virtual reality (VR) or augmented reality (AR) technology to provide operators with an immersive work simulation environment, facilitating intuitive evaluation and adjustment of work plans.

9. The method for constructing and applying an unmanned loading model based on digital twins according to claim 1, characterized in that, In unmanned loading model applications, the intelligent control adopts an adaptive control algorithm, which automatically adjusts control parameters based on real-time data and model prediction results during the operation, enabling the loading equipment to adapt to different operating conditions and working condition changes.

Images