Intelligent forklift management system and method based on AI driving behavior analysis

Abstract

The invention provides an intelligent forklift management system and method based on AI driving behavior analysis, and the method comprises the steps: building a three-dimensional model of a forklift, and building a digital twin model of the forklift according to the three-dimensional model; generating a cargo loading and unloading basic model according to the digital twin model; generating a driver basic operation model in combination with the cargo loading and unloading basic model and the digital twin model; acquiring current driver data, current driving behavior data, current working environment data and current cargo data; generating a first loading and unloading model according to the current working environment data, the current cargo data and the cargo loading and unloading basic model; generating a first operation model in combination with the first loading and unloading model, the current driver data and the driver basic operation model; generating an operation adjustment scheme according to the current driving behavior data and the first operation model; and working parameters of the forklift are adjusted according to the operation adjusting scheme. By comprehensively utilizing various data sources and models, intelligent management of forklift operation can be realized, and the working efficiency and safety are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of artificial intelligence, and particularly to an intelligent forklift management system and method based on AI driving behavior analysis. Background Art

[0002] In recent years, with the rapid progress of mobile robot technology, its applications in fields such as logistics warehousing and industrial production have become increasingly widespread. Among them, forklifts, with their unique multi-layer cargo transfer capabilities, have become indispensable key equipment in these scenarios. However, in practical applications, forklift operations face many challenges: limited storage space, complex cargo picking and placing environments, narrow operating spaces, etc. To ensure operation safety and avoid problems such as collision accidents, failed cargo grasping, and positioning deviations, the forklift system must have high-precision operation control capabilities. However, the existing forklift management solutions have low intelligence levels, low control precision, and low management efficiency. Summary of the Invention

[0003] Based on the above problems, the present invention proposes an intelligent forklift management system and method based on AI driving behavior analysis. By comprehensively utilizing various data sources and models, it can achieve intelligent management of forklift operations, improving work efficiency and safety.

[0004] In view of this, one aspect of the present invention proposes an intelligent forklift management method based on AI driving behavior analysis, including: Obtaining the three-dimensional data of the forklift, the forklift attribute data, the forklift historical work data, the forklift historical work environment data, the forklift driver data, and the forklift driver historical operation data; Establishing a three-dimensional model of the forklift according to the three-dimensional data of the forklift and the forklift attribute data; Combining the three-dimensional model of the forklift, the forklift historical work data, the forklift historical work environment data, the forklift driver data, and the forklift driver historical operation data to establish a digital twin model of the forklift; Generating a basic cargo loading and unloading model according to the digital twin model of the forklift; Generating a basic operation model for the driver according to the basic cargo loading and unloading model and the digital twin model of the forklift; Obtaining the current driver data, the current driving behavior data, the current work environment data, and the current cargo data of the forklift; Generating a first loading and unloading model for the current cargo according to the current work environment data, the current cargo data, and the basic cargo loading and unloading model; Generating a first operation model for the current driver according to the first loading and unloading model, the current driver data, and the basic operation model for the driver; Generate an operation adjustment plan based on the current driving behavior data and the first operation model; Adjust the working parameters of the forklift according to the operation adjustment plan.

[0005] Optionally, the step of establishing a digital twin model of the forklift by combining the three-dimensional model of the forklift, the historical working data of the forklift, the historical working environment data of the forklift, the data of the forklift driver, and the historical operation data of the forklift driver includes: Establish a physical characteristic model of the forklift based on the three-dimensional model of the forklift, including: extracting the structural parameter data in the three-dimensional model of the forklift and establishing a basic forklift structure model; mapping the power system parameters, braking system parameters, and steering system parameters in the forklift attribute data into the basic forklift structure model to generate a physical characteristic model of the forklift; Establish a dynamic behavior model of the forklift based on the historical working data of the forklift, including: performing time series analysis on the historical working data of the forklift to extract the motion characteristic data of the forklift under different working conditions; using machine learning algorithms to train the motion characteristic data to establish a dynamic behavior prediction model of the forklift; associating and mapping the dynamic behavior prediction model of the forklift with the physical characteristic model of the forklift to form a dynamic behavior model of the forklift; Establish an environment interaction model based on the historical working environment data of the forklift, including: analyzing the historical working environment data of the forklift to identify the environmental characteristics and their influencing factors on the operation of the forklift; establishing an association model between the environmental characteristics and the dynamic behavior of the forklift; integrating the association model with the dynamic behavior model of the forklift to obtain an environment interaction model; Establish a driver operation model based on the data of the forklift driver and the historical operation data of the forklift driver, including: extracting features from the historical operation data of the forklift driver to identify the driver operation mode; combining the data of the forklift driver to establish a personalized operation characteristic model of the driver; associating the personalized operation characteristic model of the driver with the environment interaction model to obtain a driver operation model; Integrate the above models to establish a digital twin model of the forklift, including: constructing a multi-dimensional data association matrix to realize data interaction between sub-models; establishing a unified data update mechanism to ensure the real-time nature of the model; generating a digital twin model of the forklift.

[0006] Optionally, the step of generating a basic model for loading and unloading goods according to the digital twin model of the forklift includes: Build a fork-lift pallet picking sub-model, including: extracting fork-lift action parameters and power parameters from the forklift digital twin model; analyzing historical successful pallet picking case data based on machine learning algorithms to extract the optimal pallet picking trajectory features; constructing a fork-lift movement trajectory prediction model according to the fork-lift action parameters, power parameters and optimal pallet picking trajectory features; establishing a self-adaptive alignment mechanism between the fork-lift and the pallet based on the fork-lift movement trajectory prediction model and combining with environmental perception data to obtain the fork-lift pallet picking sub-model; Build a pallet balance detection sub-model, including: extracting fork-lift load-bearing data and pressure distribution data from the forklift digital twin model; establishing a pallet force analysis model according to the fork-lift load-bearing data and pressure distribution data to calculate the force state of the pallet at different positions; constructing a pallet stability evaluation index system in combination with the pallet force analysis model; designing a real-time monitoring algorithm for the pallet balance state according to the pallet force analysis model and the pallet stability evaluation index system to obtain the pallet balance detection sub-model; Generate a center-of-gravity change prediction model, including: extracting the forklift load state data from the forklift digital twin model; analyzing the influence law of different cargo types on the forklift center of gravity according to the forklift load state data and the forklift three-dimensional model, and establishing a correlation model between the cargo weight and the forklift center-of-gravity position; constructing a real-time prediction algorithm for the center of gravity of the forklift-cargo system to generate the center-of-gravity change prediction model; Build a forklift driving sub-model, including: extracting forklift movement feature data from the forklift digital twin model; calculating the optimal driving parameters under different load conditions in combination with the forklift movement feature data and the center-of-gravity change prediction model, establishing a speed-steering-load safety constraint model, and generating an adaptive driving path planning algorithm to obtain the forklift driving sub-model.

[0007] Build a pallet unloading sub-model, including: extracting fork-lift lowering control parameters from the forklift digital twin model, analyzing the spatial constraint conditions of the target position, establishing a precise pallet placement trajectory model, and constructing a dynamic adjustment mechanism for the unloading process to obtain the pallet unloading sub-model; Integrate each sub-model to generate a complete basic model for cargo loading and unloading, including: establishing a data interaction interface between sub-models; designing a collaborative operation mechanism for sub-models; constructing an overall model optimization and adjustment algorithm.

[0008] Optionally, the step of generating a driver's basic operation model according to the basic model for cargo loading and unloading and the forklift digital twin model includes: Construct a standard operation sequence model based on the basic model for cargo loading and unloading, including: extracting the key operation nodes of each sub-model from the basic model for cargo loading and unloading; performing a time sequence sorting on the key operation nodes to form a basic operation chain; setting safety parameter thresholds and operation tolerance ranges for each operation node; establishing logical associations and dependencies between operation nodes; Build an environmental adaptation model based on the forklift digital twin model, including: extracting environmental characteristic parameters from the forklift digital twin model; analyzing the influence degree of different environmental characteristics on operations based on the environmental characteristic parameters; establishing a mapping relationship between environmental characteristics and operation parameters; constructing an environmental adaptive adjustment mechanism for operation parameters; Build a driver ability evaluation model, including: extracting operation characteristics from historical driving data; establishing a driver skill level evaluation system; constructing a driver operation habit model; generating driver personalized operation preference characteristics; Establish a cargo characteristic recognition model, including: analyzing the loading and unloading characteristics of different cargo types; establishing an association model between cargo attributes and operation requirements; constructing a real-time monitoring mechanism for cargo status; generating operation constraint conditions related to the cargo; Generate an operation paradigm template, including: adjusting the standard operation sequence in combination with the environmental adaptation model; setting personalized operation parameters according to the driver ability evaluation results; integrating cargo characteristic constraints to form specific operation guidance; establishing a dynamic optimization mechanism for the operation paradigm; Build a driver basic operation model, including: integrating the operation paradigm template to establish a multi-scenario operation rule library; designing a real-time adjustment algorithm for the operation model; constructing an operation feedback evaluation mechanism; realizing the self-optimization function of the operation model.

[0009] Optionally, the step of generating the first loading and unloading model of the current cargo according to the current working environment data, the current cargo data, and the basic cargo loading and unloading model includes: Analyze the current working environment data to generate environmental constraint conditions, including: identifying the spatial limitation parameters of the current working area; extracting environmental characteristic data such as ground conditions and lighting conditions; detecting the distribution of surrounding obstacles; generating an environmental constraint parameter set; Analyze the current cargo data to extract cargo characteristic parameters, including: obtaining basic parameters such as the weight, size, and shape of the cargo; identifying the center of gravity position and force characteristics of the cargo; determining the loading and unloading requirements and precautions of the cargo; generating a cargo characteristic parameter set; Select an adapted sub-model from the basic cargo loading and unloading model, including: matching a corresponding fork pallet picking sub-model according to the cargo characteristic parameters; selecting an appropriate pallet balance detection sub-model based on the cargo weight; configuring a center of gravity change prediction model in combination with the cargo characteristics; selecting a suitable forklift driving sub-model and pallet unloading sub-model according to the environmental constraints; Optimize the parameters of the selected sub-models, including: adjusting the operation parameters of each sub-model according to the environmental constraint conditions; modifying the threshold settings of the model based on the cargo characteristics; optimizing the collaborative configuration between sub-models; generating an optimized model parameter set; Construct the first loading and unloading model for the current goods, including: each sub-model after integration and optimization; establish a data interaction mechanism between models; set the real-time adjustment strategy of the model; generate a complete loading and unloading model.

[0010] Optionally, the step of generating the first operation model of the current driver according to the first loading and unloading model, the current driver data, and the driver's basic operation model includes: Analyze the current driver data and establish a personalized feature model, including: extract the driver's historical operation data and behavior characteristics; analyze the driver's skill level and area of expertise; identify the driver's operation habits and preferences; establish an evaluation index for the driver's current state; Perform model matching based on the driver's basic operation model, including: select an operation paradigm that matches the current driver's characteristics from the driver's basic operation model; adjust the operation parameter threshold according to the driver's skill level; correct the operation sequence in combination with the driver's operation habits; generate a preliminary personalized operation plan; Perform operation optimization in combination with the first loading and unloading model, including: compare the personalized operation plan with the requirements of the first loading and unloading model; identify possible operation risk points and difficulties; adjust the operation parameters according to the loading and unloading requirements; establish an operation safety protection mechanism; Construct a real-time adaptation mechanism, including: design a dynamic adjustment algorithm for operation parameters; establish an operation feedback evaluation system; construct an emergency response plan; generate an operation deviation correction strategy.

[0011] Generate the first operation model of the current driver, including: the operation plan after integration and optimization; establish a real-time monitoring and feedback mechanism; set personalized operation prompts and warnings; form a complete operation guidance model.

[0012] Optionally, the step of generating an operation adjustment plan according to the current driving behavior data and the first operation model includes: Real-time analyze the current driving behavior data, including: collect the driver's real-time operation parameters, including steering angle, acceleration, and braking force; extract the current driving state characteristics, including speed control, path selection, and forklift operation; identify abnormal driving behavior points, including sharp turns, sudden braking, and forklift shaking; generate a set of current driving behavior characteristics; Perform a difference analysis with the first operation model, including: compare the current driving behavior characteristics with the first operation model in real time; calculate the deviation values of each operation parameter; evaluate the impact degree of the deviation on operation safety and efficiency; generate a deviation feature report; Perform a risk assessment based on the deviation characteristics, including: quantify the safety risks of the deviation characteristics; predict the possible consequences of the deviation behavior; determine the operation items that need to be adjusted first; generate a risk level assessment report; Formulate targeted adjustment strategies, including: designing adjustment priorities according to risk levels; generating specific improvement suggestions for each item to be adjusted; designing progressive adjustment steps; establishing evaluation criteria for adjustment effects; Generate an operation adjustment plan, including: integrating adjustment strategies to form a complete adjustment plan; designing the timing arrangement for plan execution; establishing a feedback mechanism for plan execution; constructing a dynamic optimization mechanism for the plan.

[0013] Optionally, the step of adjusting the working parameters of the forklift according to the operation adjustment plan includes: Analyze the operation adjustment plan to determine the parameter adjustment range, including: extracting specific parameter adjustment items in the operation adjustment plan; obtaining the target adjustment values and adjustment tolerance ranges of each parameter; determining the priority order of parameter adjustment; establishing the correlation constraint relationship between parameters; Perform a safety pre-inspection, including: evaluating the impact of parameter adjustment on the stability of the forklift; verifying whether the adjusted parameters are within the safety threshold range; analyzing potential risks during the parameter adjustment process; generating a safety assessment report; Formulate a parameter adjustment execution strategy, including: designing adjustment steps according to the safety assessment results; formulating a progressive adjustment curve for each parameter; establishing a buffer mechanism for parameter adjustment; setting emergency termination conditions for the adjustment process; Execute parameter adjustment, including: gradually adjusting the parameters of the power system, including maximum speed, acceleration, etc.; optimizing the parameters of the steering system, including steering sensitivity, maximum steering angle; adjusting the parameters of the hydraulic system, including the lifting speed and tilt angle of the fork; updating the parameters of the braking system, including braking force and response time; Establish a real-time monitoring and feedback mechanism, including: real-time monitoring of the parameter adjustment effect; collecting driver operation feedback; detecting the operating status of the equipment; generating an adjustment effect evaluation report.

[0014] Optionally, the step of extracting the key operation nodes of each sub-model from the basic model of goods loading and unloading includes: Analyze the key operation nodes of the fork taking pallet sub-model, including: extracting the key parameter points in the fork leveling stage, including the initial height and horizontal angle; identifying the control points in the fork insertion stage, including insertion speed, depth, and angle; extracting the state points in the fork lifting stage, including lifting speed and target height; establishing a node sequence model for the pallet taking process; Extract the key nodes of the pallet balance detection sub-model, including: identifying the weight distribution detection points, including the left and right balance points, front and back balance points; extracting the key pressure sensing points, including the pressure values of each support point; determining the stability judgment nodes, including tilt angle and sway amplitude; generating a node state matrix for balance detection; Analyze the key nodes of the center-of-gravity change prediction model, including: identifying static center-of-gravity measurement points, including the no-load center of gravity and the full-load center of gravity; extracting dynamic center-of-gravity change points, including the center-of-gravity offset during acceleration and turning; determining critical state points, including the maximum allowable offset; establishing a monitoring node network for center-of-gravity changes; Extract the key nodes of the forklift driving sub-model, including: identifying speed control nodes, including start, cruise, and deceleration points; extracting steering control points, including steering start, maximum steering angle, and return-to-zero points; determining path planning points, including obstacle avoidance points and turning points; generating a control node chain for the driving process; Analyze the key nodes of the pallet unloading sub-model, including: identifying positioning alignment points, including horizontal position and vertical height; extracting descent control points, including descent speed change points and buffer points; determining release judgment points, including pallet contact points and separation points; establishing a node sequence table for the unloading process.

[0015] Another aspect of the present invention provides an intelligent forklift management system based on AI driving behavior analysis for implementing an intelligent forklift management method based on AI driving behavior analysis, including: a cloud server and an Internet of Things server; The cloud server is configured to: Obtain the three-dimensional data of the forklift, the forklift attribute data, the forklift historical work data, the forklift historical work environment data, the forklift driver data, and the forklift driver historical operation data; Establish a three-dimensional model of the forklift based on the three-dimensional data of the forklift and the forklift attribute data; Combine the three-dimensional model of the forklift, the forklift historical work data, the forklift historical work environment data, the forklift driver data, and the forklift driver historical operation data to establish a digital twin model of the forklift; Generate a basic model for cargo loading and unloading based on the digital twin model of the forklift; Generate a basic operation model for the driver based on the basic model for cargo loading and unloading and the digital twin model of the forklift; The Internet of Things server is configured to: Obtain the current driver data, the current driving behavior data, the current work environment data, and the current cargo data of the forklift; Generate a first loading and unloading model for the current cargo based on the current work environment data, the current cargo data, and the basic model for cargo loading and unloading; Generate a first operation model for the current driver based on the first loading and unloading model, the current driver data, and the basic operation model for the driver; Generate an operation adjustment plan based on the current driving behavior data and the first operation model; Adjust the working parameters of the forklift according to the operation adjustment plan.

[0016] Adopting the technical solution of the present invention, a smart forklift management method based on AI driving behavior analysis includes: obtaining the three-dimensional data of the forklift, the forklift attribute data, the historical working data of the forklift, the historical working environment data of the forklift, the forklift driver data, and the historical operation data of the forklift driver; establishing a three-dimensional model of the forklift according to the three-dimensional data of the forklift and the forklift attribute data; combining the three-dimensional model of the forklift, the historical working data of the forklift, the historical working environment data of the forklift, the forklift driver data, and the historical operation data of the forklift driver to establish a digital twin model of the forklift; generating a basic model for cargo loading and unloading according to the digital twin model of the forklift; generating a basic operation model for the driver according to the basic model for cargo loading and unloading and the digital twin model of the forklift; obtaining the current driver data, the current driving behavior data, the current working environment data, and the current cargo data of the forklift; generating a first loading and unloading model for the current cargo according to the current working environment data, the current cargo data, and the basic model for cargo loading and unloading; generating a first operation model for the current driver according to the first loading and unloading model, the current driver data, and the basic operation model for the driver; generating an operation adjustment plan according to the current driving behavior data and the first operation model; and adjusting the working parameters of the forklift according to the operation adjustment plan. Through the solution of this embodiment, a full-range modeling of the physical characteristics, dynamic behavior, environmental interaction, and driver operation of the forklift is achieved, and clear association relationships are established among the sub-models to ensure the consistency and integrity of the data; personalized modeling can be carried out according to the operation characteristics of different drivers, and the operation requirements under different working environments can be adapted; accurate prediction of the dynamic behavior of the forklift is realized through machine learning algorithms, and the impact of environmental changes on forklift operation can be predicted; a unified data update mechanism is established to ensure that the model can reflect the working state of the forklift in real time and support real-time operation optimization and adjustment; the model architecture supports the integration of new data sources and features, which is convenient for subsequent function expansion and optimization. This constitutes an intelligent, dynamic, and reliable forklift digital twin system, improving the timeliness, intelligence, safety, and accuracy of forklift management. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 is a flowchart of a smart forklift management method based on AI driving behavior analysis provided by an embodiment of the present invention; Figure 2 is a schematic block diagram of a smart forklift management system based on AI driving behavior analysis provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0018] In order to more clearly understand the above-mentioned objects, features, and advantages of the present invention, the present invention will be further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that, without conflict, the embodiments of the present application and the features in the embodiments can be combined with each other.

[0019] In the following description, many specific details are set forth in order to fully understand the present invention. However, the present invention may also be implemented in other ways different from those described herein. Therefore, the protection scope of the present invention is not limited by the specific embodiments disclosed below.

[0020] The terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or units is not limited to the listed steps or units, but optionally further includes steps or units not listed, or optionally further includes other steps or units inherent to these processes, methods, products, or devices.

[0021] Referring to "embodiment" herein means that a specific feature, structure, or characteristic described in connection with the embodiment can be included in at least one embodiment of the present application. The phrase appears in various places in the specification and does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment mutually exclusive with other embodiments. Those skilled in the art will explicitly and implicitly understand that the embodiments described herein can be combined with other embodiments.

[0022] The following refers to Figures 1 to 2 to describe a smart forklift management system and method based on AI driving behavior analysis provided according to some embodiments of the present invention.

[0023] As Figure 1 shown, an embodiment of the present invention provides a smart forklift management method based on AI driving behavior analysis, including: Obtaining the three-dimensional data of the forklift, the forklift attribute data, the forklift historical work data, the forklift historical work environment data, the forklift driver data, and the forklift driver historical operation data; It is understandable that a LiDAR device can be used to perform three-dimensional scanning of the forklift from multiple angles to obtain high-precision three-dimensional data; sensors (such as temperature sensors, pressure sensors, etc.) can be installed on the forklift to monitor the performance parameters of the forklift in real time; through the API or database provided by the forklift manufacturer, the technical specifications and attribute information of the forklift can be obtained; a data recorder can be installed on the forklift to record historical data such as the working hours, load conditions, and driving distances of the forklift; the historical working data can be uploaded to the cloud platform for subsequent data analysis and query; environmental monitoring sensors can be installed in the working environment to record environmental parameters such as temperature, humidity, and light; by installing cameras, the operation of the forklift in different working environments can be recorded to assist in analyzing the impact of the environment on the forklift's work; biometric technologies (such as fingerprint recognition, face recognition) can be used to record the identity information of the driver; through the enterprise's internal management system, the training and certification records of the driver can be obtained; monitoring devices can be installed on the forklift to record the operation behaviors of the driver (such as acceleration, braking, steering, etc.); the operation data of the driver can be uploaded to the data analysis platform for behavior pattern analysis and historical record storage. Through this step, various types of data of the forklift can be comprehensively obtained, providing a basis for subsequent intelligent management and optimization.

[0024] Establish a three-dimensional model of the forklift based on the three-dimensional data of the forklift and the forklift attribute data; In this step, select a suitable three-dimensional modeling software (such as SolidWorks, AutoCAD, CATIA, etc.). These software can handle complex geometric shapes and structural features and support the import of various formats of data; import the collected three-dimensional data of the forklift into the selected modeling software; according to the imported data, use modeling tools to create a three-dimensional model of the forklift. This includes drawing each component of the forklift (such as the frame, forks, wheels, etc.) and ensuring the accurate size and position relationship between components; assign the forklift attribute data (such as material type, weight, etc.) to the corresponding model components for subsequent analysis and simulation; on the premise of ensuring model accuracy, the model can be simplified to reduce the calculation amount and improve the efficiency of subsequent analysis. For example, ignore the detail features with less impact on structural strength, such as small holes and welds, etc.; if finite element analysis (FEA) is required subsequently, the three-dimensional model needs to be meshed and discretized into finite elements for mechanical analysis; verify the accuracy of the three-dimensional model by comparing it with the actual forklift. If differences are found, the model needs to be adjusted; after the model is completed, functional tests can be carried out to ensure that the performance of the model under different working conditions meets expectations. Through this step, a three-dimensional model of the forklift can be successfully established based on the three-dimensional data and attribute data of the forklift, providing a basis for subsequent digital twin model construction and other analyses.

[0025] Establish a forklift digital twin model by combining the three-dimensional model of the forklift, the historical working data of the forklift, the historical working environment data of the forklift, the forklift driver data, and the historical operation data of the forklift driver; Generate a basic model for cargo handling according to the forklift digital twin model (including a sub-model for the forklift forks to pick up the pallet, a sub-model for pallet balance detection, a sub-model for predicting the change of the center of gravity, a sub-model for forklift driving, a sub-model for pallet unloading, etc.); Generate a basic operation model for the driver according to the basic model for cargo handling and the forklift digital twin model (the basic operation paradigm for the driver can be generated by combining different working environments, different drivers, different cargos, etc.); Obtain the current driver data of the forklift (including historical driving behavior data, historical driving vehicle type data, historical working environment data, historical transported cargo data, etc.), current driving behavior data, current working environment data, and current cargo data; It can be understood that by installing a variety of sensors on the forklift, including GPS, accelerometers, gyroscopes, and load sensors, to monitor the operating status of the forklift and the operation behavior of the driver in real time; configure in-vehicle terminal devices that can record and store the operation data of the driver, the working environment data, and the cargo information; establish a data recording system that can store the real-time collected data in a local or cloud database. The system should support the query and management of historical data; design the database structure to ensure that historical driving behavior data, historical driving vehicle type data, historical working environment data, and historical transported cargo data can be effectively stored and retrieved; obtain the current driver data in real time through the in-vehicle terminal and sensors, including current driving behavior data, current working environment data, and current cargo data; obtain the historical data related to the current driver through the database query interface, including historical driving behavior, historical driving vehicle type, historical working environment, and historical transported cargo data; integrate the real-time data with the historical data to form a complete driver data profile. This can be achieved through data analysis tools to help identify the operation patterns and behavior characteristics of the driver; use data analysis techniques (such as machine learning algorithms) to analyze the integrated data to identify the behavior trends and potential problems of the driver; display the analysis results in the form of charts or dashboards through visualization tools for easy understanding and use by managers and drivers; provide real-time feedback to the driver according to the analysis results to help optimize their operation behavior and improve safety and efficiency. Through this step, the current driver data and related historical data of the forklift can be effectively obtained, providing a basis for the intelligent management and optimization of the forklift.

[0026] Generate a first loading and unloading model for the current cargo according to the current working environment data, the current cargo data, and the basic model for cargo handling; Generate the first operation model of the current driver based on the first loading and unloading model, the current driver data, and the driver's basic operation model; Generate an operation adjustment plan based on the current driving behavior data and the first operation model; Adjust the working parameters of the forklift according to the operation adjustment plan.

[0027] In this embodiment, by comprehensively utilizing a variety of data sources and models, intelligent management of forklift operations can be achieved, improving work efficiency and safety. This AI-based management method has broad application prospects in modern logistics and warehousing management.

[0028] In some possible implementation manners of the present invention, the step of establishing a forklift digital twin model of the forklift in combination with the forklift three-dimensional model, the forklift historical working data, the forklift historical working environment data, the forklift driver data, and the forklift driver historical operation data includes: Establish a forklift physical feature model based on the forklift three-dimensional model, including: Extract the structural parameter data in the forklift three-dimensional model to establish a forklift basic structure model; Map the power system parameters, braking system parameters, and steering system parameters in the forklift attribute data to the forklift basic structure model to generate a forklift physical feature model; Establish a forklift dynamic behavior model based on the forklift historical working data, including: Perform time series analysis on the forklift historical working data to extract the motion feature data of the forklift under different working conditions; Use machine learning algorithms to train the motion feature data to establish a forklift dynamic behavior prediction model; Perform correlation mapping between the forklift dynamic behavior prediction model and the forklift physical feature model to form a forklift dynamic behavior model; Establish an environment interaction model based on the forklift historical working environment data, including: Analyze the forklift historical working environment data to identify environmental features and their influencing factors on forklift operation; Establish a correlation model between environmental features and forklift dynamic behavior; Integrate the correlation model with the forklift dynamic behavior model to obtain an environment interaction model; Establish a driver operation model based on the forklift driver data and the forklift driver historical operation data, including: Extract features from the forklift driver historical operation data to identify the driver's operation mode; Combine the forklift driver data to establish a driver personalized operation feature model; Associate the driver personalized operation feature model with the environment interaction model to obtain a driver operation model; Integrate the above models to establish a forklift digital twin model, including: Construct a multi-dimensional data association matrix to realize data interaction between sub-models; Establish a unified data update mechanism to ensure the real-time nature of the model; Generate a forklift digital twin model.

[0029] The solution of this embodiment realizes all-round modeling of the physical characteristics, dynamic behavior, environment interaction and driver operation of the forklift. Clear association relationships are established between sub-models to ensure data consistency and integrity; it can perform personalized modeling according to the operation characteristics of different drivers and adapt to operation requirements in different working environments; through machine learning algorithms, accurate prediction of the dynamic behavior of the forklift can be achieved, and the impact of environmental changes on forklift operation can be predicted; a unified data update mechanism is established to ensure that the model can reflect the working state of the forklift in real time and support real-time operation optimization and adjustment; the model architecture supports the incorporation of new data sources and features, facilitating subsequent function expansion and optimization. This constitutes an intelligent, dynamic and reliable forklift digital twin system, improving the timeliness, intelligence, safety and accuracy of forklift management.

[0030] In some possible implementation manners of the present invention, the step of generating a basic model for cargo loading and unloading according to the forklift digital twin model includes: Construct a sub-model for the forklift forks to pick up the pallet, including: Extract the fork action parameters and power parameters from the forklift digital twin model; Based on machine learning algorithms, analyze the historical successful pallet picking case data and extract the optimal pallet picking trajectory features; According to the fork action parameters, power parameters and optimal pallet picking trajectory features, construct a fork movement trajectory prediction model; According to the fork movement trajectory prediction model, combined with environment perception data, establish a self-adaptive alignment mechanism for the position of the fork and the pallet to obtain a sub-model for the forklift forks to pick up the pallet; Establish a sub-model for detecting the balance of the pallet, including: Extract the fork load-bearing data and pressure distribution data from the forklift digital twin model; According to the fork load-bearing data and pressure distribution data, establish a pallet force analysis model to calculate the force state of the pallet at different positions; Combined with the pallet force analysis model, construct a pallet stability evaluation index system; According to the pallet force analysis model and the pallet stability evaluation index system, design a real-time monitoring algorithm for the balance state of the pallet to obtain a sub-model for detecting the balance of the pallet; Generate a center of gravity change prediction model, including: Extract the forklift load status data from the forklift digital twin model; According to the forklift load status data and the 3D model of the forklift, analyze the influence law of different cargo types on the center of gravity of the forklift, and establish a correlation model between the cargo weight and the center of gravity position of the forklift; Construct a real-time prediction algorithm for the center of gravity of the forklift-cargo system to generate a center of gravity change prediction model; Construct a forklift driving sub-model, including: Extract the forklift motion feature data from the forklift digital twin model; Combine the forklift motion feature data and the center of gravity change prediction model, calculate the optimal driving parameters under different load conditions, establish a speed-steering-load safety constraint model, and generate an adaptive driving path planning algorithm to obtain the forklift driving sub-model.

[0031] Establish a pallet unloading sub-model, including: extract the fork lowering control parameters from the forklift digital twin model, analyze the spatial constraint conditions of the target position, establish a pallet precise placement trajectory model, and construct a dynamic adjustment mechanism for the unloading process to obtain the pallet unloading sub-model; Integrate each sub-model to generate a complete basic model for cargo loading and unloading, including: establish a data interaction interface between sub-models; design a collaborative operation mechanism for sub-models; construct an overall model optimization and adjustment algorithm.

[0032] The solution of this embodiment significantly reduces the risk of cargo tipping through accurate center of gravity prediction and balance detection; real-time monitoring and adjustment ensure the stability of the loading and unloading process; adaptive path planning reduces ineffective movements, and the precise alignment mechanism improves the loading and unloading speed, and the collaborative optimization algorithm realizes the optimal loading and unloading strategy; multi-dimensional monitoring and warning mechanisms improve operation reliability, and the dynamic adjustment ability adapts to different working conditions; realize intelligent decision-making in the loading and unloading process, and the adaptive adjustment ability improves the intelligence level of the system; the model architecture supports the loading and unloading requirements of different types of cargo, and has good scene adaptability and scalability; enable the system to achieve safe, efficient, and intelligent cargo loading and unloading operations, and at the same time have strong adaptability and scalability.

[0033] In some possible implementation manners of the present invention, the step of generating a driver's basic operation model according to the basic model for cargo loading and unloading and the forklift digital twin model includes: Construct a standard operation sequence model based on the basic model for cargo loading and unloading, including: Extract the key operation nodes of each sub-model from the basic model for cargo loading and unloading; Perform chronological sorting on the key operation nodes to form a basic operation chain; In this step, first, establish an operation node dependency matrix, including: analyzing the pre-and post-dependency relationships between key operation nodes; identifying the mandatory sequential constraints between nodes; determining the groups of nodes that can be executed in parallel; generating a node dependency table; second, construct a node timing weight model, including: assigning an execution duration weight to each operation node; setting the time interval requirements between nodes; calculating the buffer time for node switching; establishing a timing weight matrix; then, perform node grouping optimization, including: grouping nodes according to functional relevance; identifying the critical path nodes within each group; determining the connecting nodes between groups; generating an optimized node group structure; then, execute the timing sorting algorithm, including: performing topological sorting based on the dependency matrix; applying the timing weights for sequential optimization; handling the timing arrangement of parallel nodes; generating a preliminary timing sequence; furthermore, establish a basic operation chain, including: integrating the sorted node sequence; adding transitional operations between nodes; setting checkpoints for the operation chain; forming a complete basic operation chain structure; finally, verify and optimize the operation chain, including: checking the logical integrity of the operation chain; verifying the rationality of the timing arrangement; optimizing the execution efficiency of the operation chain; generating a final basic operation chain model. The solution of this step ensures the rationality of the operation sequence, maintains the integrity of the operation chain, optimizes the sorting of operation nodes, improves the execution efficiency, ensures the continuity of operations, reduces the risk of operation interruption, supports dynamic node adjustment, adapts to different job requirements, accurately controls the node timing, and ensures the operation quality.

[0034] Set a safety parameter threshold and an operation tolerance range for each operation node; Establish the logical association and dependency relationship between operation nodes; Based on the forklift digital twin model, construct an environment adaptation model, including: Extract environmental characteristic parameters from the forklift digital twin model; Based on the environmental characteristic parameters, analyze the influence degree of different environmental characteristics on operations; In this step, first, construct a classification system for environmental characteristic parameters, including: classifying environmental characteristic parameters according to physical attributes, including ground conditions, lighting conditions, space limitations, etc.; classifying environmental characteristic parameters according to time-varying characteristics, including fixed parameters, periodically changing parameters, randomly changing parameters, etc.; classifying environmental characteristic parameters according to the scope of influence, including global influence parameters, local influence parameters, etc.; generating a classification matrix for environmental characteristic parameters; second, establish an environmental-operation impact assessment model, including: constructing an impact model of ground conditions on the driving stability of forklifts; establishing an impact model of lighting conditions on operation accuracy; establishing an impact model of space limitations on operation paths; establishing a coupling impact model of multiple environmental factors; third, conduct a quantitative analysis of environmental characteristic impacts, including: setting the reference values and change ranges of each environmental characteristic parameter; calculating the impact coefficients of environmental characteristic changes on operation parameters; analyzing the critical values and risk thresholds of environmental characteristic changes; generating a quantitative report on environmental characteristic impacts; fourth, establish an environmental adaptability assessment system, including: establishing an operation difficulty scoring standard for different environmental characteristics; designing response strategies for environmental changes; constructing evaluation indicators for environmental adaptability capabilities; generating environmental adaptability assessment results; fifth, construct an environmental impact early warning mechanism, including: setting monitoring thresholds for environmental characteristic changes; establishing a classification standard for environmental risk levels; formulating response strategies for different risk levels; forming a complete early warning response mechanism. The solution in this step comprehensively covers environmental impact factors, establishes a complete analysis system, accurately quantifies environmental impacts, provides reliable assessment results, predicts the impacts of environmental changes, plans response strategies in advance, quickly responds to environmental changes, dynamically adjusts operation parameters, timely identifies environmental risks, and ensures operation safety.

[0035] Establish the mapping relationship between environmental characteristics and operation parameters; Construct an environmental self-adaptive adjustment mechanism for operation parameters; Construct a driver ability assessment model, including: extracting operation characteristics from historical driving data; establishing a driver skill level assessment system; constructing a driver operation habit model; generating personalized operation preference characteristics of drivers; Establish a cargo characteristic recognition model, including: Analyze the handling characteristics of different cargo types; Establish an association model between cargo attributes and operation requirements; In this step, various attribute data related to the goods are collected, including information such as goods type, weight, volume, packaging method, and hazard level. These data can be obtained through barcode scanning, RFID tags, or manual input. The operation requirements related to the loading, unloading, transportation, and storage of the goods are collected, including the loading and unloading processes, required equipment, operation steps, safety precautions, etc. The collected data is cleaned to remove duplicate and incorrect information to ensure the accuracy and consistency of the data. The data from different sources is standardized to facilitate subsequent analysis and modeling. For example, the classification criteria for goods types and the description methods for operation requirements are unified. The basic structure of the model is determined, including the relationship between goods attributes and operation requirements. For example, an entity-relationship model (ER model) can be used to represent the association between goods attributes and operation requirements. Based on the collected data, association rules between goods attributes and operation requirements are established, which can be achieved through data mining techniques (such as association rule learning) to identify which goods attributes correspond to which operation requirements. The effectiveness of the model is verified through actual operation data to check whether the model can accurately reflect the relationship between goods attributes and operation requirements. The model is optimized according to the verification results, adjusting the association rules and model structure to improve the accuracy and applicability of the model. The established association model is applied to actual operations to guide the processes of loading, unloading, transportation, and storage of the goods. Feedback data is collected during actual application to evaluate the effect of the model, and the model is continuously improved based on the feedback. Through this step, an association model between goods attributes and operation requirements can be effectively established, thereby improving the efficiency and safety of logistics and warehousing management.

[0036] Construct a real-time monitoring mechanism for the goods status; Generate operation constraint conditions related to the goods; It can be understood that according to the characteristics of the goods (such as flammable, fragile, etc.) and the operation environment, the constraint conditions for safe operations are determined. For example, the storage and transportation of flammable items need to follow specific safety specifications. According to the equipment used (such as forklifts) and the characteristics of the goods, the operation limitations of the equipment are determined. For example, the load capacity limit and operation radius of forklifts. According to the transportation and storage requirements of the goods, the constraint conditions in terms of time are determined, such as the shelf life and delivery time of the goods. A constraint model is designed to systematically organize the collected constraint conditions, and mathematical models or logical models can be used to represent these constraint conditions. Through data analysis, association rules between goods attributes and operation constraint conditions are established to ensure that corresponding operation constraint conditions can be automatically generated under specific conditions. The effectiveness of the generated operation constraint conditions is verified through actual operation data to check whether they can be effectively implemented in actual operations. According to the verification results and actual operation feedback, the constraint conditions are adjusted and optimized to improve the safety and efficiency of operations. Through this step, operation constraint conditions related to the goods can be effectively generated, thereby improving the safety and efficiency of logistics and warehousing management.

[0037] Generate an operation paradigm template, including: Combine with the environment adaptation model to adjust the standard operation sequence; Set personalized operation parameters according to the driver ability evaluation result; Integrate the cargo feature constraints to form specific operation guidelines; Establish a dynamic optimization mechanism for the operation paradigm; Construct a basic driver operation model, including: Integrate the operation paradigm template to establish a multi-scenario operation rule library; Design a real-time adjustment algorithm for the operation model; Construct an operation feedback evaluation mechanism; Implement the self-optimization function of the operation model.

[0038] The solution of this embodiment generates customized operation guidelines according to the driver characteristics, adapts to the operation habits and skill levels of different drivers; responds to environmental changes in real time, provides operation suggestions for environmental adaptability; establishes a complete operation safety boundary, monitors and warns of dangerous operations in real time; optimizes the operation process, reduces unnecessary actions; provides the best operation suggestions; continuously optimizes the operation model, accumulates and utilizes excellent operation experience; supports multiple working scenarios, and adapts to different types of cargo requirements. This solution enables the system to provide personalized, safe, and efficient operation guidelines for different drivers, and at the same time has the ability of continuous optimization and scenario adaptation.

[0039] In some possible implementation manners of the present invention, the step of generating the first loading and unloading model of the current cargo according to the current working environment data, the current cargo data, and the basic cargo handling model includes: Analyze the current working environment data to generate environmental constraint conditions, including: identifying the space limit parameters of the current working area; extracting environmental feature data such as ground conditions and lighting conditions; detecting the distribution of surrounding obstacles; generating an environmental constraint parameter set; Analyze the current cargo data to extract cargo feature parameters, including: obtaining basic parameters such as the weight, size, and shape of the cargo; identifying the center of gravity position and force characteristics of the cargo; determining the loading and unloading requirements and precautions of the cargo; generating a cargo feature parameter set; Select an appropriate sub-model from the basic cargo handling model, including: matching a corresponding fork pallet picking sub-model according to the cargo feature parameters; selecting an appropriate pallet balance detection sub-model based on the cargo weight; configuring a center of gravity change prediction model in combination with the cargo characteristics; selecting a suitable forklift driving sub-model and pallet unloading sub-model according to the environmental constraints; Optimize the parameters of the selected sub-models, including: adjusting the operation parameters of each sub-model according to the environmental constraint conditions; correcting the threshold settings of the model based on the characteristics of the goods; optimizing the collaborative configuration between sub-models; generating an optimized set of model parameters; Construct the first loading and unloading model for the current goods, including: integrating the optimized sub-models; establishing a data interaction mechanism between the models; setting a real-time adjustment strategy for the model; generating a complete loading and unloading model.

[0040] The solution of this embodiment can adjust the loading and unloading strategy in real time according to the current environment and the characteristics of the goods, quickly respond to changes in the environment and the state of the goods; optimize the loading and unloading parameters according to the specific characteristics of the goods to improve the accuracy of loading and unloading operations; fully consider environmental constraints and the characteristics of the goods to prevent possible safety hazards; optimize the loading and unloading process, reduce unnecessary actions, and provide the optimal loading and unloading path; support the loading and unloading requirements of different types of goods and adapt to various working environmental conditions. This solution enables the system to generate the optimal loading and unloading plan according to the actual situation, ensuring the safety and efficiency of operations.

[0041] In some possible embodiments of the present invention, the step of generating the first operation model of the current driver according to the first loading and unloading model, the current driver data, and the driver's basic operation model includes: Analyze the current driver data and establish a personalized feature model, including: extracting the driver's historical operation data and behavioral characteristics; analyzing the driver's skill level and expertise field; identifying the driver's operation habits and preferences; establishing an evaluation index for the driver's current state; Perform model matching based on the driver's basic operation model, including: selecting an operation paradigm that matches the current driver's characteristics from the driver's basic operation model; adjusting the operation parameter threshold according to the driver's skill level; correcting the operation sequence in combination with the driver's operation habits; generating a preliminary personalized operation plan; Perform operation optimization in combination with the first loading and unloading model, including: comparing the personalized operation plan with the requirements of the first loading and unloading model; identifying possible operation risk points and difficulties; adjusting the operation parameters according to the loading and unloading requirements; establishing an operation safety protection mechanism; Construct a real-time adaptation mechanism, including: designing a dynamic adjustment algorithm for operation parameters; establishing an operation feedback evaluation system; constructing an emergency response plan; generating an operation deviation correction strategy.

[0042] Generate the first operation model of the current driver, including: integrating the optimized operation plan; establishing a real-time monitoring and feedback mechanism; setting personalized operation prompts and warnings; forming a complete operation guidance model.

[0043] The solution of this embodiment fully considers the individual characteristics of the driver, provides operation suggestions that conform to the driving habits, identifies and prevents potential risks, establishes a multi-level safety protection mechanism, optimizes the personal operation process, reduces unnecessary operation actions, quickly adapts to changes in the operation state, and provides operation adjustment suggestions in a timely manner. It also supports the improvement of the driver's skills and accumulates optimized operation experience. This solution enables the system to provide the most suitable operation guidance for the current driver while ensuring the safety and efficiency of the operation.

[0044] In some possible embodiments of the present invention, the step of generating an operation adjustment plan according to the current driving behavior data and the first operation model includes: Analyze the current driving behavior data in real time, including: collecting the real-time operation parameters of the driver, including the steering angle, acceleration, and braking force; extracting the current driving state characteristics, including speed control, path selection, and forklift operation; identifying abnormal points in the driving behavior, including sharp turns, sudden braking, and forklift shaking; generating a set of current driving behavior characteristics; Perform a difference analysis with the first operation model, including: comparing the current driving behavior characteristics with the first operation model in real time; calculating the deviation values of each operation parameter; evaluating the impact degree of the deviation on the operation safety and efficiency; generating a deviation characteristic report; Conduct a risk assessment based on the deviation characteristics, including: quantifying the safety risks of the deviation characteristics; predicting the possible consequences of the deviation behavior; determining the operation items that need to be adjusted first; generating a risk level assessment report; Formulate targeted adjustment strategies, including: designing the adjustment priority according to the risk level; generating specific improvement suggestions for each item to be adjusted; designing progressive adjustment steps; establishing an evaluation criterion for the adjustment effect; Generate an operation adjustment plan, including: integrating the adjustment strategies to form a complete adjustment plan; designing the timing arrangement for the plan execution; establishing a feedback mechanism for the plan execution; constructing a dynamic optimization mechanism for the plan.

[0045] The solution of this embodiment can discover operation deviations in real time, quickly generate adjustment suggestions, accurately identify operation problems, provide targeted improvement solutions, timely warn of potential risks, prevent dangerous operation behaviors, dynamically adjust according to the actual situation, support progressive improvement, provide specific and feasible improvement steps, facilitate the driver's understanding and execution, and continuously improve the plan through feedback to promote the improvement of the driving level. This solution enables the system to effectively identify and improve problems in driving operations, ensuring the continuous improvement of the safety and efficiency of forklift operations.

[0046] In some possible embodiments of the present invention, the step of adjusting the working parameters of the forklift according to the operation adjustment plan includes: Analyze the operation adjustment plan, determine the parameter adjustment range, including: extract the specific parameter adjustment items in the operation adjustment plan; obtain the target adjustment values and adjustment tolerance ranges of each parameter; determine the priority order of parameter adjustment; establish the associated constraint relationships between parameters; Perform a safety pre-inspection, including: evaluate the impact of parameter adjustment on the forklift stability; verify whether the adjusted parameters are within the safety threshold range; analyze the potential risks during the parameter adjustment process; generate a safety assessment report; Formulate a parameter adjustment execution strategy, including: design the adjustment steps according to the safety assessment results; formulate a progressive adjustment curve for each parameter; establish a buffer mechanism for parameter adjustment; set the emergency abort conditions for the adjustment process; Execute the parameter adjustment, including: gradually adjust the power system parameters, including the maximum speed, acceleration, etc.; optimize the steering system parameters, including the steering sensitivity, maximum steering angle; adjust the hydraulic system parameters, including the fork lifting speed, tilt angle; update the braking system parameters, including the braking force, response time; Establish a real-time monitoring and feedback mechanism, including: monitor the parameter adjustment effect in real time; collect the driver operation feedback; detect the equipment operation status; generate an adjustment effect assessment report.

[0047] The solution of this embodiment can ensure the safety of the parameter adjustment process, prevent equipment failures caused by parameter adjustment; achieve the progressive adjustment of parameters, avoid the impact of mutations on operations; accurately achieve the target parameter values, maintain the coordination between parameters; quickly respond to abnormal situations, support dynamic parameter adjustment; record the parameter adjustment process, facilitate problem analysis and optimization; automate parameter adjustment, intelligent error prevention and protection. This solution enables the system to safely, reliably and efficiently complete the adjustment of forklift working parameters and improve the overall operation effect.

[0048] In some possible implementation manners of the present invention, the step of extracting the key operation nodes of each sub-model from the basic model of goods loading and unloading includes: Analyze the key operation nodes of the fork taking pallet sub-model, including: extract the key parameter points in the fork leveling stage, including the initial height, horizontal angle; identify the control points in the fork insertion stage, including the insertion speed, depth, angle; extract the state points in the fork lifting stage, including the lifting speed, target height; establish a node sequence model for the pallet taking process; Extract the key nodes of the pallet balance detection sub-model, including: identify the weight distribution detection points, including the left and right balance points, front and back balance points; extract the key pressure sensing points, including the pressure values of each support point; determine the stability judgment nodes, including the tilt angle, sway amplitude; generate a node state matrix for balance detection; Analyze the key nodes of the center of gravity change prediction model, including: identifying static center of gravity measurement points, including the no-load center of gravity and the full-load center of gravity; extracting dynamic center of gravity change points, including the center of gravity offset during acceleration and turning; determining critical state points, including the maximum allowable offset; establishing a monitoring node network for the center of gravity change; Extract the key nodes of the forklift driving sub-model, including: identifying speed control nodes, including start, cruise, and deceleration points; extracting steering control points, including the starting point of steering, the maximum steering angle, and the point of returning to the straight position; determining path planning points, including obstacle avoidance points and turning points; generating a control node chain for the driving process; Analyze the key nodes of the pallet unloading sub-model, including: identifying positioning alignment points, including horizontal position and vertical height; extracting descent control points, including the points of change in descent speed and buffer points; determining release judgment points, including pallet contact points and separation points; establishing a node sequence table for the unloading process.

[0049] The solution of this embodiment can accurately identify the key control points in each link and establish precise operation reference standards; it can completely cover the entire loading and unloading process and establish a systematic node network; it can clarify the logical relationship between nodes and ensure the continuity of operations; it can support node adjustment under different working conditions and has the ability of dynamic optimization; it can achieve precise process control and provide clear operation guidance; it can effectively identify risk control points and establish multiple safety guarantees; it can provide an accurate and complete control node basis for the subsequent operation model and ensure the accuracy and safety of the loading and unloading operations.

[0050] Please refer to Figure 2 , Another embodiment of the present invention provides an intelligent forklift management system based on AI driving behavior analysis for implementing an intelligent forklift management method based on AI driving behavior analysis, including: a cloud server and an Internet of Things server; The cloud server is configured to: Obtain the three-dimensional data of the forklift, the forklift attribute data, the forklift historical work data, the forklift historical work environment data, the forklift driver data, and the forklift driver historical operation data; Establish a three-dimensional model of the forklift according to the three-dimensional data of the forklift and the forklift attribute data; Combine the three-dimensional model of the forklift, the forklift historical work data, the forklift historical work environment data, the forklift driver data, and the forklift driver historical operation data to establish a digital twin model of the forklift; Generate a basic model for cargo loading and unloading according to the digital twin model of the forklift; Generate a basic operation model for the driver according to the basic model for cargo loading and unloading and the digital twin model of the forklift; The Internet of Things server is configured to: Obtain the current driver data, current driving behavior data, current working environment data, and current cargo data of the forklift; Generate a first loading and unloading model for the current cargo according to the current working environment data, the current cargo data, and the basic cargo loading and unloading model; Generate a first operation model for the current driver according to the first loading and unloading model, the current driver data, and the basic driver operation model; Generate an operation adjustment plan according to the current driving behavior data and the first operation model; Adjust the working parameters of the forklift according to the operation adjustment plan.

[0051] It should be known that Figure 2 The block diagram of the intelligent forklift management system based on AI driving behavior analysis shown is only for illustration, and the number of each module shown does not limit the protection scope of the present invention. The intelligent forklift management system based on AI driving behavior analysis provided in this embodiment can be used to execute the corresponding implementation solutions of the intelligent forklift management method based on AI driving behavior analysis. For the specific implementation process, please refer to the descriptions of each method embodiment, which will not be elaborated here.

[0052] It should be noted that for the foregoing method embodiments, for the sake of simple description, they are all expressed as a series of action combinations. However, those skilled in the art should know that this application is not limited by the described action sequence, because according to this application, certain steps can be in other sequences or carried out simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily essential to this application.

[0053] In the above embodiments, the descriptions of each embodiment have their own emphases. For the parts not detailed in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0054] In several embodiments provided by this application, it should be understood that the disclosed device can be implemented in other ways. For example, the device embodiments described above are only illustrative. For example, the above-mentioned unit division is only a logical function division. In actual implementation, there can be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point, the displayed or discussed coupling or direct coupling or communication connection to each other can be through some interfaces. The indirect coupling or communication connection of the device or unit can be in an electrical or other form.

[0055] The units described above as separate components may or may not be physically separated. The components shown as units may or may not be physical units, that is, they may be located in one place or distributed over multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0056] In addition, each functional unit in various embodiments of the present application may be integrated into a processing unit, or each unit may exist physically alone, or two or more units may be integrated into one unit. The above-mentioned integrated units can be implemented in the form of hardware or in the form of software functional units.

[0057] If the above-mentioned integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable memory. Based on this understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art, or all or part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a memory and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the above methods in various embodiments of the present application. The aforementioned memory includes: USB flash drives, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), mobile hard disks, magnetic disks, or optical discs and other various media that can store program codes.

[0058] Those of ordinary skill in the art can understand that all or part of the steps in the various methods of the above embodiments can be completed by instructing relevant hardware through a program. This program can be stored in a computer-readable memory. The memory can include: flash drives, read-only memories (abbreviation: ROM), random access memories (abbreviation: RAM), magnetic disks, or optical discs, etc.

[0059] The above has introduced the embodiments of the present application in detail. Specific examples are used in this article to elaborate on the principle and implementation manner of the present application. The description of the above embodiments is only used to help understand the method and its core idea of the present application; at the same time, for those of ordinary skill in the art, according to the idea of the present application, there will be changes in the specific implementation manner and application scope. In summary, the content of this specification should not be construed as a limitation to the present application.

[0060] Although the present invention is disclosed as above, the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions without departing from the spirit and scope of the present invention, and can make various changes and modifications, including combinations of the above different functions and implementation steps, including software and hardware implementation manners, all within the protection scope of the present invention.

Claims

1. A smart forklift management method based on AI driving behavior analysis, characterized in that: include: Obtaining forklift three-dimensional data, forklift attribute data, forklift historical working data, forklift historical working environment data, forklift driver data, and forklift driver historical operation data of the forklift; Establishing a three-dimensional forklift model of the forklift according to the three-dimensional forklift data and the forklift attribute data; Establishing a forklift digital twin model of the forklift by combining the forklift three-dimensional model, the forklift historical working data, the forklift historical working environment data, the forklift driver data, and the forklift driver historical operation data; generating a basic model for cargo loading and unloading based on the forklift digital twin model; Generate a basic operation model of the driver according to the cargo loading and unloading basic model and the forklift digital twin model; Obtain the current driver data, current driving behavior data, current working environment data and current cargo data of the forklift; Generate a first loading and unloading model of the current cargo according to the current working environment data, the current cargo data and the cargo loading and unloading basic model; generating a first operation model of the current driver according to the first loading and unloading model, the current driver data and the driver basic operation model; generating an operation adjustment plan according to the current driving behavior data and the first operation model; The operating parameters of the forklift are adjusted according to the operation adjustment plan.

2. The intelligent forklift management method based on AI driving behavior analysis according to claim 1 is characterized in that: The step of establishing a forklift digital twin model of the forklift in combination with the forklift three-dimensional model, the forklift historical working data, the forklift historical working environment data, the forklift driver data, and the forklift driver historical operation data comprises: Establishing a forklift physical feature model based on the forklift three-dimensional model, including: extracting structural parameter data in the forklift three-dimensional model to establish a forklift basic structure model; mapping power system parameters, brake system parameters, and steering system parameters in the forklift attribute data to the forklift basic structure model to generate a forklift physical feature model; Establishing a forklift dynamic behavior model based on the forklift historical working data includes: performing time series analysis on the forklift historical working data to extract motion feature data of the forklift under different working conditions; training the motion feature data using a machine learning algorithm to establish a forklift dynamic behavior prediction model; associating and mapping the forklift dynamic behavior prediction model with the forklift physical feature model to form a forklift dynamic behavior model; Establishing an environmental interaction model based on the forklift's historical working environment data, including: analyzing the forklift's historical working environment data, identifying environmental characteristics and factors affecting the forklift's operation; establishing a correlation model between environmental characteristics and forklift's dynamic behavior; integrating the correlation model with the forklift's dynamic behavior model to obtain an environmental interaction model; Establishing a driver operation model based on the forklift driver data and the forklift driver historical operation data, including: extracting features from the forklift driver historical operation data to identify the driver operation mode; establishing a driver personalized operation feature model in combination with the forklift driver data; associating the driver personalized operation feature model with an environment interaction model to obtain a driver operation model; Integrate the above models to establish a forklift digital twin model, including: constructing a multi-dimensional data association matrix to realize data interaction between sub-models; establishing a unified data update mechanism to ensure the real-time performance of the model; and generating a forklift digital twin model.

3. The intelligent forklift management method based on AI driving behavior analysis according to claim 2 is characterized in that: The step of generating a basic model for cargo loading and unloading according to the forklift digital twin model comprises: Constructing a fork-to-pallet sub-model, including: extracting fork motion parameters and power parameters from the forklift digital twin model; analyzing historical pallet-to-pallet successful case data based on a machine learning algorithm to extract optimal pallet-to-pallet trajectory features; constructing a fork motion trajectory prediction model based on the fork motion parameters, power parameters and optimal pallet-to-pallet trajectory features; establishing a fork-to-pallet position adaptive alignment mechanism based on the fork motion trajectory prediction model and environmental perception data to obtain a fork-to-pallet sub-model; Establishing a pallet balance detection sub-model, including: extracting fork load-bearing data and pressure distribution data from the forklift digital twin model; establishing a pallet force analysis model based on the fork load-bearing data and pressure distribution data to calculate the force state of the pallet at different positions; building a pallet stability evaluation index system in combination with the pallet force analysis model; designing a pallet balance state real-time monitoring algorithm based on the pallet force analysis model and the pallet stability evaluation index system to obtain a pallet balance detection sub-model; Generating a center of gravity change prediction model, including: extracting forklift load status data from the forklift digital twin model; analyzing the influence of different cargo types on the forklift center of gravity based on the forklift load status data and the forklift three-dimensional model, and establishing a correlation model between cargo weight and forklift center of gravity position; constructing a forklift-cargo system center of gravity real-time prediction algorithm to generate a center of gravity change prediction model; Constructing a forklift driving sub-model, including: extracting forklift motion feature data from the forklift digital twin model; combining the forklift motion feature data and the center of gravity change prediction model, calculating optimal driving parameters under different load conditions, establishing a speed-steering-load safety constraint model, and generating an adaptive driving path planning algorithm to obtain a forklift driving sub-model; Establishing a pallet unloading sub-model, including: extracting fork descent control parameters from the forklift digital twin model, analyzing the spatial constraints of the target position, establishing a precise pallet placement trajectory model, and constructing a dynamic adjustment mechanism for the unloading process to obtain the pallet unloading sub-model; Integrate each sub-model to generate a complete basic model for cargo loading and unloading, including: establishing a data interaction interface between sub-models; designing a sub-model collaborative operation mechanism; and building an overall model optimization and adjustment algorithm.

4. The intelligent forklift management method based on AI driving behavior analysis according to claim 3 is characterized in that: The step of generating a basic driver operation model according to the cargo handling basic model and the forklift digital twin model comprises: Building a standard operation sequence model based on the cargo handling basic model includes: extracting key operation nodes of each sub-model from the cargo handling basic model; sorting the key operation nodes in time sequence to form a basic operation chain; setting a safety parameter threshold and an operation tolerance range for each operation node; and establishing a logical association and dependency relationship between the operation nodes; Building an environmental adaptation model based on the forklift digital twin model, including: extracting environmental characteristic parameters from the forklift digital twin model; analyzing the influence of different environmental characteristics on operation based on the environmental characteristic parameters; establishing a mapping relationship between environmental characteristics and operating parameters; and building an environmental adaptive adjustment mechanism for operating parameters; Constructing a driver capability assessment model, including: extracting operational features from historical driving data; establishing a driver skill level assessment system; constructing a driver operational habit model; and generating driver personalized operational preference features; Establish a cargo feature recognition model, including: analyzing the loading and unloading characteristics of different cargo types; establishing a correlation model between cargo attributes and operational requirements; building a real-time cargo status monitoring mechanism; and generating cargo-related operational constraints; Generate an operation paradigm template, including: combining the environmental adaptation model to adjust the standard operation sequence; setting personalized operation parameters according to the driver's ability assessment results; integrating cargo feature constraints to form specific operation guidance; and establishing a dynamic optimization mechanism for the operation paradigm; Construct a basic operation model for drivers, including: integrating operation paradigm templates and establishing a multi-scenario operation rule library; designing a real-time adjustment algorithm for the operation model; building an operation feedback evaluation mechanism; and realizing the self-optimization function of the operation model.

5. The intelligent forklift management method based on AI driving behavior analysis according to claim 4 is characterized in that: The step of generating a first loading and unloading model of the current cargo according to the current working environment data, the current cargo data and the cargo loading and unloading basic model comprises: Analyze the current working environment data and generate environmental constraints, including: identifying the spatial restriction parameters of the current working area; extracting environmental feature data such as ground conditions and lighting conditions; detecting the distribution of surrounding obstacles; and generating an environmental constraint parameter set; Analyze current cargo data and extract cargo characteristic parameters, including: obtaining basic parameters such as cargo weight, size, and shape; identifying the center of gravity and force characteristics of cargo; determining cargo loading and unloading requirements and precautions; generating cargo characteristic parameter sets; Selecting an adaptive sub-model from the cargo loading and unloading basic model includes: matching the corresponding forklift pallet picking sub-model according to cargo characteristic parameters; selecting an appropriate pallet balance detection sub-model based on cargo weight; configuring a center of gravity change prediction model based on cargo characteristics; selecting an appropriate forklift driving sub-model and pallet unloading sub-model based on environmental constraints; Optimize the parameters of the selected sub-models, including: adjusting the operating parameters of each sub-model according to environmental constraints; modifying the threshold setting of the model based on cargo characteristics; optimizing the collaborative configuration between sub-models; and generating an optimized model parameter set; Construct the first loading and unloading model of the current cargo, including: integrating the optimized sub-models; establishing a data interaction mechanism between models; setting up a real-time adjustment strategy for the model; and generating a complete loading and unloading model.

6. The intelligent forklift management method based on AI driving behavior analysis according to claim 5 is characterized in that: The step of generating a first operation model of the current driver according to the first loading and unloading model, the current driver data and the driver basic operation model comprises: Analyze current driver data and establish a personalized feature model, including: extracting the driver's historical operation data and behavioral characteristics; analyzing the driver's skill level and expertise; identifying the driver's operating habits and preferences; and establishing the driver's current status evaluation index; Model matching is performed based on the basic operation model of the driver, including: selecting an operation paradigm that matches the current driver's characteristics from the basic operation model of the driver; adjusting the operation parameter threshold according to the driver's skill level; correcting the operation sequence in combination with the driver's operation habits; and generating a preliminary personalized operation plan; Performing operation optimization in combination with the first loading and unloading model includes: comparing the personalized operation plan with the requirements of the first loading and unloading model; identifying possible operation risk points and difficulties; adjusting operation parameters according to loading and unloading requirements; and establishing an operation safety protection mechanism; Build a real-time adaptation mechanism, including: design a dynamic adjustment algorithm for operating parameters; establish an operation feedback evaluation system; build an emergency response plan; generate an operation correction strategy; Generate the first operation model of the current driver, including: integrating the optimized operation plan; establishing a real-time monitoring and feedback mechanism; setting personalized operation prompts and warnings; and forming a complete operation guidance model.

7. The intelligent forklift management method based on AI driving behavior analysis according to claim 6 is characterized in that: The step of generating an operation adjustment scheme according to the current driving behavior data and the first operation model comprises: Real-time analysis of current driving behavior data, including: collecting the driver's real-time operating parameters, including steering angle, acceleration, and braking force; extracting current driving state features, including speed control, path selection, and fork operation; identifying driving behavior anomalies, including sharp turns, sudden braking, and fork shaking; generating a current driving behavior feature set; Performing difference analysis with the first operation model, including: comparing the current driving behavior characteristics with the first operation model in real time; calculating the deviation value of each operation parameter; evaluating the degree of influence of the deviation on the operation safety and efficiency; generating a deviation characteristic report; Conduct risk assessment based on deviation characteristics, including: quantifying safety risks of deviation characteristics; predicting the possible consequences of deviation behavior; determining the operational items that need to be adjusted first; and generating risk level assessment reports; Formulate targeted adjustment strategies, including: designing adjustment priorities based on risk levels; generating specific improvement suggestions for each item to be adjusted; designing progressive adjustment steps; and establishing evaluation criteria for adjustment effects; Generate an operational adjustment plan, including: integrating adjustment strategies to form a complete adjustment plan; designing a time schedule for the execution of the plan; establishing a feedback mechanism for the execution of the plan; and building a dynamic optimization mechanism for the plan.

8. The intelligent forklift management method based on AI driving behavior analysis according to claim 7 is characterized in that: The step of adjusting the operating parameters of the forklift according to the operation adjustment scheme comprises: Analyze the operation adjustment plan and determine the parameter adjustment range, including: extracting specific parameter adjustment items in the operation adjustment plan; obtaining the target adjustment value and adjustment tolerance range of each parameter; determining the priority order of parameter adjustment; and establishing the association constraint relationship between parameters; Perform safety pre-checks, including: evaluating the impact of parameter adjustments on forklift stability; verifying whether the adjusted parameters are within the safety threshold range; analyzing potential risks during parameter adjustment; and generating safety assessment reports; Formulate parameter adjustment execution strategies, including: designing adjustment steps based on safety assessment results; developing progressive adjustment curves for each parameter; establishing a buffer mechanism for parameter adjustment; and setting emergency termination conditions for the adjustment process; Execute parameter adjustments, including: gradually adjust power system parameters, including maximum speed, acceleration, etc.; optimize steering system parameters, including steering sensitivity and maximum turning angle; adjust hydraulic system parameters, including fork lifting speed and tilt angle; update brake system parameters, including braking force and response time; Establish a real-time monitoring and feedback mechanism, including: real-time monitoring of parameter adjustment effects; collecting driver operation feedback; detecting equipment operating status; and generating adjustment effect evaluation reports.

9. The intelligent forklift management method based on AI driving behavior analysis according to claim 8 is characterized in that: The step of extracting key operation nodes of each sub-model from the cargo handling basic model comprises: Analyze the key operation nodes of the fork picking up pallet sub-model, including: extracting the key parameter points of the fork leveling stage, including the initial height and horizontal angle; identifying the control points of the fork insertion stage, including the insertion speed, depth, and angle; extracting the state points of the fork lifting stage, including the lifting speed and target height; and establishing a node sequence model of the pallet picking process; Extract the key nodes of the pallet balance detection sub-model, including: identifying weight distribution detection points, including left and right balance points, front and back balance points; extracting pressure sensing key points, including the pressure value of each support point; determining stability judgment nodes, including tilt angle and shaking amplitude; generating a node state matrix for balance detection; Analyze the key nodes of the center of gravity change prediction model, including: identifying static center of gravity measurement points, including empty center of gravity and fully loaded center of gravity; extracting dynamic center of gravity change points, including center of gravity offset during acceleration and steering; determining critical state points, including the maximum allowable offset; establishing a monitoring node network for center of gravity changes; Extract the key nodes of the forklift driving sub-model, including: identifying speed control nodes, including starting, cruising, and deceleration points; extracting steering control points, including steering start, maximum turning angle, and return point; determining path planning points, including obstacle avoidance points and turning points; generating a control node chain for the driving process; Analyze the key nodes of the pallet unloading sub-model, including: identifying the positioning alignment points, including horizontal position and vertical height; extracting the descent control points, including descent speed change points and buffer points; determining the release judgment points, including pallet contact points and separation points; and establishing a node sequence table for the unloading process.

10. A smart forklift management system based on AI driving behavior analysis, used to execute the smart forklift management method based on AI driving behavior analysis as claimed in any one of claims 1 to 9, characterized in that: include: Cloud servers and IoT servers; The cloud server is configured as follows: Obtaining forklift three-dimensional data, forklift attribute data, forklift historical working data, forklift historical working environment data, forklift driver data, and forklift driver historical operation data of the forklift; Establishing a three-dimensional forklift model of the forklift according to the three-dimensional forklift data and the forklift attribute data; Establishing a forklift digital twin model of the forklift by combining the forklift three-dimensional model, the forklift historical working data, the forklift historical working environment data, the forklift driver data, and the forklift driver historical operation data; generating a basic model for cargo loading and unloading based on the forklift digital twin model; Generate a basic operation model of the driver according to the cargo loading and unloading basic model and the forklift digital twin model; The IoT server is configured as follows: Obtain the current driver data, current driving behavior data, current working environment data and current cargo data of the forklift; Generate a first loading and unloading model of the current cargo according to the current working environment data, the current cargo data and the cargo loading and unloading basic model; generating a first operation model of the current driver according to the first loading and unloading model, the current driver data and the driver basic operation model; generating an operation adjustment plan according to the current driving behavior data and the first operation model; The operating parameters of the forklift are adjusted according to the operation adjustment plan.

Images