Loader vehicle control method and system, computer equipment and storage medium

Abstract

The invention relates to the technical field of intelligent vehicle control, in particular to a loader vehicle control method and system, computer equipment and a storage medium. The method comprises the following steps: collecting multi-modal data of the loader, wherein the multi-modal data comprises working environment data, dynamic load state data and vehicle attitude data of the loader; constructing a three-dimensional operation scene digital twinborn model based on the multi-modal data, and predicting vehicle stability influence data of the influence of the load change on the vehicle stability based on the dynamic load state data; based on the digital twinborn model of the three-dimensional operation scene and the vehicle stability influence data, a dynamic control strategy is generated by utilizing a reinforcement learning algorithm, and cooperative actions of steering, lifting and walking mechanisms are synchronously optimized; according to operation environment data and a dynamic control strategy, hydraulic system pressure and motor torque distribution are adjusted in a self-adaptive mode, and operation efficiency and safety are ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent vehicle control technology, and in particular to a loader vehicle control method, system, computer equipment and storage medium. Background Art

[0002] Loaders are earthwork machines widely used in construction projects such as roads, railways, buildings, hydropower, ports, and mines. They are mainly used to shovel bulk materials such as soil, sand, lime, and coal, and can also do light shoveling of ore and hard soil. By replacing different auxiliary working devices, they can also do bulldozing, lifting, and loading and unloading of other materials such as wood.

[0003] In road construction, especially in high-grade highway construction, loaders are used for filling and excavation of roadbed engineering, collection and loading of asphalt mixture and cement concrete material fields, etc. In addition, they can also carry out operations such as pushing soil, leveling the ground, and pulling other machinery. Because loaders have the advantages of fast operation speed, high efficiency, good maneuverability, and easy operation, they have become one of the main machines for earthwork construction in engineering construction.

[0004] The control system of traditional loaders mainly relies on preset parameters and manual experience. Most systems use fixed PID parameters or table lookup methods, which cannot adapt to dynamic load changes (such as material sliding, bucket collision) and terrain disturbances (such as sudden slope changes, ground subsidence) in real time, resulting in control lag or overshoot. In order to solve the above problems, the present invention provides a loader vehicle control method, system, computer equipment and storage medium. Summary of the invention

[0005] In order to solve the technical problems existing in the above-mentioned prior art, the present invention provides a loader vehicle control method, system, computer equipment and storage medium.

[0006] To achieve the above objectives, the embodiments of the present invention provide the following technical solutions: In a first aspect, in one embodiment provided by the present invention, a loader vehicle control method is provided, the method comprising the following steps: Collecting multimodal data of the loader, the multimodal data including operating environment data, dynamic load state data and vehicle posture data of the loader; Build a three-dimensional digital twin model of the operation scene based on multimodal data, and predict the vehicle stability impact data of the impact of load changes on vehicle stability based on dynamic load status data; Based on the 3D operation scene digital twin model and vehicle stability impact data, a reinforcement learning algorithm is used to generate a dynamic control strategy to simultaneously optimize the coordinated actions of the steering, lifting, and traveling mechanisms. According to the operating environment data and dynamic control strategies, the hydraulic system pressure and motor torque distribution are adaptively adjusted to ensure operating efficiency and safety.

[0007] As a further solution of the present invention, multimodal data can be collected by laser radar, camera, inertial measurement unit, pressure sensor and encoder.

[0008] As a further solution of the present invention, the three-dimensional operation scene digital twin model is constructed based on multimodal data, and the vehicle stability impact data for predicting the impact of load changes on vehicle stability based on dynamic load state data includes: Perform spatiotemporal registration and data fusion on multimodal data; Dynamic environment modeling is performed based on multimodal data processed by spatiotemporal registration and data fusion to obtain multimodal data to build a three-dimensional operation scene digital twin model and update the model status in real time; The three-dimensional operation scene digital twin model extracts and evaluates the load parameters to obtain vehicle stability impact data.

[0009] As a further solution of the present invention, the three-dimensional operation scene digital twin model extracts load parameters and evaluates the load parameters to obtain vehicle stability impact data, including: Calculate bucket / fork load based on pressure sensor signal; Calculate inertial forces using IMU accelerometer data; Combine the hydraulic cylinder pressure change rate to predict load mutations; The rolling angular acceleration is calculated based on the IMU angular velocity, and the sideslip risk is assessed based on the ground slope.

[0010] As a further solution of the present invention, the three-dimensional operation scene digital twin model extracts load parameters and evaluates the load parameters to obtain vehicle stability impact data, including building a prediction model, which is an LSTM model or a Transformer model.

[0011] As a further solution of the present invention, the three-dimensional operation scene digital twin model and vehicle stability impact data are used to generate a dynamic control strategy using a reinforcement learning algorithm to simultaneously optimize the coordinated actions of the steering, lifting, and walking mechanisms, including: Based on the vehicle stability impact data and the three-dimensional operation scene digital twin model, define the state space, define the action space, design the reward function, and determine the reinforcement learning algorithm.

[0012] As a further solution of the present invention, the working environment data includes terrain slope and ground adhesion coefficient.

[0013] In a second aspect, in another embodiment provided by the present invention, a loader vehicle control system is provided, the system comprising: a data acquisition module, a model building module, a control strategy generation module and an adaptive control module.

[0014] The data acquisition module is used to acquire multimodal data of the loader, and the multimodal data includes the loader's operating environment data, dynamic load status data and vehicle posture data.

[0015] In an embodiment of the present invention, the data acquisition module includes laser radar, camera, inertial measurement unit, pressure sensor and encoder.

[0016] The model building module is used to build a three-dimensional operation scene digital twin model based on multimodal data, and predict vehicle stability impact data of the impact of load changes on vehicle stability based on dynamic load state data.

[0017] The control strategy generation module is used to generate a dynamic control strategy based on the three-dimensional operation scene digital twin model and vehicle stability impact data using a reinforcement learning algorithm to simultaneously optimize the coordinated actions of the steering, lifting, and walking mechanisms.

[0018] The adaptive control module is used to adaptively adjust the hydraulic system pressure and motor torque distribution according to the operating environment data and dynamic control strategy to ensure operating efficiency and safety.

[0019] In a third aspect, in another embodiment provided by the present invention, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the loader vehicle control method when loading and executing the computer program.

[0020] In a fourth aspect, in yet another embodiment provided by the present invention, a storage medium is provided, storing a computer program, and when the computer program is loaded and executed by a processor, the steps of the loader vehicle control method are implemented.

[0021] The technical solution provided by the present invention has the following beneficial effects: The present invention provides a loader vehicle control method, system, computer equipment and storage medium, the method comprising the following steps: collecting multimodal data of the loader, the multimodal data including the loader's operating environment data, dynamic load status data and vehicle posture data; constructing a three-dimensional operating scene digital twin model based on the multimodal data, and predicting vehicle stability impact data of load changes on vehicle stability based on the dynamic load status data; generating a dynamic control strategy based on the three-dimensional operating scene digital twin model and the vehicle stability impact data using a reinforcement learning algorithm to synchronously optimize the coordinated actions of the steering, lifting and walking mechanisms; and adaptively adjusting the hydraulic system pressure and motor torque distribution according to the operating environment data and the dynamic control strategy to ensure operating efficiency and safety.

[0022] The three-dimensional digital twin model constructed by the present invention can predict the impact of load changes on the vehicle's overturning moment and center of gravity offset in real time, and combined with adaptive hydraulic pressure adjustment, effectively avoid safety accidents such as skidding and rollover caused by sudden load changes or terrain slope. The reinforcement learning algorithm previews the safety boundaries of different control strategies in the digital twin simulation environment to avoid high-risk actions before the actual vehicle is executed.

[0023] These and other aspects of the present invention will be more clearly understood in the following description of the embodiments. It should be understood that the above general description and the following detailed description are only exemplary and explanatory and cannot limit the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention. For ordinary technicians in this field, other embodiments can be obtained based on these drawings without paying creative work.

[0025] Figure 1 The present invention is a flowchart of a loader vehicle control method according to an embodiment of the present invention.

[0026] Figure 2 This is a specific flow chart of S20 in the loader vehicle control method according to an embodiment of the present invention.

[0027] Figure 3 The structure block diagram of a loader vehicle control system according to an embodiment of the present invention.

[0028] Figure 4 The figure is a structural block diagram of a computer device according to an embodiment of the present invention.

[0029] In the figure: data acquisition module-100, model building module-200, control strategy generation module-300, adaptive control module-400. DETAILED DESCRIPTION

[0030] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0031] The flowcharts shown in the accompanying drawings are only examples and do not necessarily include all the contents and operations / steps, nor must they be executed in the order described. For example, some operations / steps may also be decomposed, combined or partially merged, so the actual execution order may change according to actual conditions.

[0032] It should be understood that the terms used in this specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. As used in the specification of the present invention and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include plural forms.

[0033] Specifically, the embodiments of the present invention are further described below in conjunction with the accompanying drawings.

[0034] See also Figure 1 , Figure 1 is a flow chart of a loader vehicle control method provided by an embodiment of the present invention, such as Figure 1 As shown, the loader vehicle control method includes steps S10 to S40. The method breaks through the technical bottlenecks of traditional loader control systems in terms of dynamic load adaptability, multi-mechanism coordination efficiency and terrain adaptive capability through the deep integration of multi-modal perception fusion, high-fidelity digital twins and safety constraint reinforcement learning, and provides an innovative solution for the intelligentization of construction machinery.

[0035] S10, collecting multimodal data of the loader, wherein the multimodal data includes operating environment data, dynamic load state data and vehicle posture data of the loader; In an embodiment of the present invention, multimodal data can be collected by laser radar, camera, inertial measurement unit, pressure sensor, encoder and the like.

[0036] S20. Build a three-dimensional operation scene digital twin model based on the multimodal data, and predict vehicle stability impact data of load changes on vehicle stability based on the dynamic load state data.

[0037] In an embodiment of the present invention, the S20, constructing a three-dimensional operation scene digital twin model based on multimodal data, and predicting vehicle stability impact data of the impact of load changes on vehicle stability based on dynamic load state data, includes: S201, performing spatiotemporal registration and data fusion on multimodal data; S202, performing dynamic environment modeling based on the multimodal data after spatiotemporal registration and data fusion processing, so as to obtain multimodal data to construct a three-dimensional operation scene digital twin model, and update the model status in real time; S203. The three-dimensional operation scene digital twin model extracts and evaluates the load parameters to obtain vehicle stability impact data.

[0038] In an embodiment of the present invention, the three-dimensional operation scene digital twin model extracts load parameters and evaluates the load parameters to obtain vehicle stability impact data, including: Calculates bucket / fork load based on pressure sensor signal.

[0039] The bucket / fork load is calculated as follows: F=P / A +mg.

[0040] In the formula, F=P / A +mg is the bucket / fork load; P is the pressure, and A is the force-bearing area.

[0041] Calculate inertial forces using IMU accelerometer data.

[0042] Combine the hydraulic cylinder pressure change rate to predict load mutations.

[0043] The rolling angular acceleration is calculated based on the IMU angular velocity, and the sideslip risk is assessed based on the ground slope.

[0044] The S203, 3D operation scene digital twin model extracts and evaluates the load parameters to obtain vehicle stability impact data, including building a prediction model, which is an LSTM model or a Transformer model. The LSTM model or the Transformer model can predict the load change trend within the next 3 seconds. The LSTM model or the Transformer model inputs historical pressure / IMU data and outputs: .

[0045] In an embodiment of the present invention, step S20 and the digital twin model can realize a full-link closed loop from multi-source data collection → dynamic scene reconstruction → load impact prediction, and ultimately provide a high-precision, low-latency simulation environment and physical constraints for the reinforcement learning algorithm, ensuring the safety and efficiency of the loader's operation in complex terrain.

[0046] In an embodiment of the present invention, the change in the center of mass position caused by the dynamic load is calculated using the weighing sensor and IMU data, and the rollover risk is evaluated in combination with the wheel spacing (x) and the center of mass height (y).

[0047] In an embodiment of the present invention, a safety factor (such as >1.5 for stability) is generated based on the ratio of the overturning moment output by the digital twin model simulation to the vehicle anti-overturning moment.

[0048] In an embodiment of the present invention, the vehicle stability influencing data includes the center of gravity offset and the overturning moment.

[0049] The three-dimensional digital twin model of the operation scene predicts the center of mass offset through the hydraulic cylinder pressure change and fork weight data, and evaluates the stability risk based on the terrain slope (calculated by lidar point cloud segmentation and IMU data fusion).

[0050] S30. Based on the three-dimensional operation scene digital twin model and vehicle stability impact data, a reinforcement learning algorithm is used to generate a dynamic control strategy to simultaneously optimize the coordinated actions of the steering, lifting, and walking mechanisms.

[0051] In an embodiment of the present invention, S30, based on the three-dimensional operation scene digital twin model and vehicle stability impact data, a dynamic control strategy is generated using a reinforcement learning algorithm to simultaneously optimize the coordinated actions of the steering, lifting, and walking mechanisms, including: Based on the vehicle stability impact data and the three-dimensional operation scene digital twin model, define the state space, define the action space, design the reward function, and determine the reinforcement learning algorithm.

[0052] The reinforcement learning algorithm of the present invention can be based on high-precision simulation and real-time stability data of digital twins.

[0053] S40, adaptively adjusts hydraulic system pressure and motor torque distribution according to operating environment data and dynamic control strategies to ensure operating efficiency and safety.

[0054] The S40, according to the operating environment data and the dynamic control strategy, adaptively adjusts the hydraulic system pressure and the motor torque distribution to ensure the operating efficiency and safety, including: When a sudden increase in bucket load (such as material slippage) is detected, the pressure compensation algorithm is triggered; When working on a slope, adjust the hydraulic system pressure according to the slope; Calculate the maximum allowable driving torque based on the ground adhesion coefficient μ and the current vehicle speed v; When lifting and steering actions occur simultaneously, resources are dynamically allocated based on stability indicators.

[0055] The present invention S40 can dynamically adjust the distribution of hydraulic pressure and motor torque according to the terrain slope, load changes and stability requirements.

[0056] In an embodiment of the present invention, the working environment data includes terrain slope and ground adhesion coefficient.

[0057] In an embodiment of the present invention, the terrain slope and the ground adhesion coefficient are calculated by laser radar point cloud segmentation and IMU attitude fusion.

[0058] The present invention can predict the impact of load changes on the vehicle's overturning moment and center of gravity offset in real time through the constructed three-dimensional digital twin model, and the adaptive hydraulic pressure adjustment can effectively avoid safety accidents such as skidding and rollover caused by sudden load changes or terrain slope. The reinforcement learning algorithm previews the safety boundaries of different control strategies through the digital twin simulation environment to avoid high-risk actions before the actual vehicle is executed. The reinforcement learning algorithm eliminates the timing conflicts of traditional staged control and shortens the operation cycle time (measured efficiency improvement ≥20%) by learning the coupling relationship between steering, lifting, and walking mechanisms end-to-end. The hydraulic pressure and motor torque are dynamically allocated according to the terrain adhesion coefficient to avoid overload or redundant output of the power system and reduce the ineffective energy consumption of the hydraulic pump and motor (energy saving rate in typical scenarios ≥15%). The multi-modal sensor data fusion system can automatically switch between flat, sloped, rugged and other operation modes to ensure stable operation on slippery, soft and other low-adhesion surfaces. Through the closed-loop architecture of "perception-modeling-decision-execution", the present invention realizes safe, efficient and adaptive control of the loader under dynamic loads and complex terrain, solves the pain points of traditional loaders such as single control strategy, poor environmental adaptability and high energy consumption, and has significant technical and economic value.

[0059] It should be understood that, although described in a certain order, these steps are not necessarily performed in sequence in the above order. Unless there is clear explanation in this article, the execution of these steps does not have strict order restriction, and these steps can be performed in other orders. Moreover, a part of the steps of the present embodiment may include a plurality of steps or a plurality of stages, and these steps or stages are not necessarily performed at the same time, but can be performed at different times, and the execution order of these steps or stages is not necessarily performed in sequence, but can be performed in turn or alternately with at least a part of the steps or stages in other steps or other steps.

[0060] In one embodiment, see Figure 3As shown, in an embodiment of the present invention, a loader vehicle control system is also provided, which includes a data acquisition module 100, a model building module 200, a control strategy generation module 300 and an adaptive control module 400.

[0061] The data acquisition module 100 is used to acquire multimodal data of the loader, and the multimodal data includes the working environment data, dynamic load state data and vehicle posture data of the loader.

[0062] In an embodiment of the present invention, the data acquisition module 100 includes a laser radar, a camera, an inertial measurement unit, a pressure sensor, an encoder and the like.

[0063] The model building module 200 is used to build a three-dimensional operation scene digital twin model based on multimodal data, and predict vehicle stability impact data of the impact of load changes on vehicle stability based on dynamic load state data.

[0064] In an embodiment of the present invention, the model construction module 200 is used to perform spatiotemporal registration and data fusion on multimodal data, and perform dynamic environment modeling based on the multimodal data after spatiotemporal registration and data fusion processing to obtain multimodal data to construct a three-dimensional work scene digital twin model, and update the model status in real time; the three-dimensional work scene digital twin model extracts load parameters and evaluates the load parameters to obtain vehicle stability impact data.

[0065] The control strategy generation module 300 is used to generate a dynamic control strategy based on the three-dimensional operation scene digital twin model and vehicle stability impact data using a reinforcement learning algorithm to simultaneously optimize the coordinated actions of the steering, lifting, and walking mechanisms.

[0066] The adaptive control module 400 is used to adaptively adjust the hydraulic system pressure and motor torque distribution according to the operating environment data and the dynamic control strategy to ensure operating efficiency and safety.

[0067] In one embodiment, see Figure 4 As shown, in an embodiment of the present invention, a computer device is also provided, including a processor 501, a communication interface 502, a memory 503 and a communication bus 504, wherein the processor 501, the communication interface 502, and the memory 503 communicate with each other through the communication bus 504.

[0068] Memory 503, used for storing computer programs; The processor 501 is used to execute the computer program stored in the memory 503 to execute the loader vehicle control method. When the processor executes the instructions, the steps in the above method embodiment are implemented: S10, collecting multimodal data of the loader, wherein the multimodal data includes operating environment data, dynamic load state data and vehicle posture data of the loader; S20, constructing a three-dimensional operation scene digital twin model based on the multimodal data, and predicting vehicle stability impact data of load changes on vehicle stability based on the dynamic load state data; S30, based on the three-dimensional operation scene digital twin model and vehicle stability impact data, use the reinforcement learning algorithm to generate dynamic control strategies and simultaneously optimize the coordinated actions of the steering, lifting and walking mechanisms; S40, adaptively adjusts hydraulic system pressure and motor torque distribution according to operating environment data and dynamic control strategies to ensure operating efficiency and safety.

[0069] The communication bus mentioned in the above terminal can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. The communication bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, only one thick line is used in the figure, but it does not mean that there is only one bus or one type of bus.

[0070] The communication interface is used for communication between the above terminal and other devices.

[0071] The memory may include a random access memory (RAM) or a non-volatile memory, such as at least one disk memory. Optionally, the memory may also be at least one storage device located away from the aforementioned processor.

[0072] The above-mentioned processor can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.

[0073] The computer device includes user devices and network devices. The user devices include but are not limited to computers, smart phones, PDAs, etc. The network devices include but are not limited to a single network server, a server group consisting of multiple network servers, or a cloud consisting of a large number of computers or network servers based on cloud computing (Cloud Computing), wherein cloud computing is a type of distributed computing, a super virtual computer consisting of a group of loosely coupled computer sets. The computer device can be operated alone to implement the present invention, or it can be connected to the network and implement the present invention through interactive operations with other computer devices in the network. The network where the computer device is located includes but is not limited to the Internet, wide area network, metropolitan area network, local area network, VPN network, etc.

[0074] It should be further understood that the term “and / or” used in the present specification and the appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0075] In one embodiment of the present invention, a storage medium is further provided, on which a computer program is stored. When the computer program is executed by a processor, the steps in the above method embodiment are implemented: S10, collecting multimodal data of the loader, wherein the multimodal data includes operating environment data, dynamic load state data and vehicle posture data of the loader; S20, constructing a three-dimensional operation scene digital twin model based on the multimodal data, and predicting vehicle stability impact data of load changes on vehicle stability based on the dynamic load state data; S30, based on the three-dimensional operation scene digital twin model and vehicle stability impact data, use the reinforcement learning algorithm to generate dynamic control strategies and simultaneously optimize the coordinated actions of the steering, lifting and walking mechanisms; S40, adaptively adjusts hydraulic system pressure and motor torque distribution according to operating environment data and dynamic control strategies to ensure operating efficiency and safety.

[0076] A person skilled in the art can understand that all or part of the processes in the above-mentioned embodiment method can be implemented by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the process of the embodiment of the above-mentioned method. Among them, any reference to memory, storage, database or other media used in the embodiments provided by the present invention can include at least one of non-volatile and volatile memory.

[0077] It should be understood that, as used herein, the singular form "a" or "an" is intended to include the plural form as well, unless the context clearly supports an exception. It should also be understood that, as used herein, "and / or" refers to any and all possible combinations of one or more of the items listed in association. The serial numbers of the embodiments disclosed in the above embodiments of the present invention are for description only and do not represent the advantages and disadvantages of the embodiments.

[0078] A person skilled in the art should understand that the discussion of any of the above embodiments is only exemplary and is not intended to imply that the scope of the disclosure of the embodiments of the present invention (including the claims) is limited to these examples; under the concept of the embodiments of the present invention, the technical features in the above embodiments or different embodiments can also be combined, and there are many other changes in different aspects of the above embodiments of the present invention, which are not provided in detail for the sake of simplicity. Therefore, any omissions, modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the embodiments of the present invention should be included in the protection scope of the embodiments of the present invention.

Claims

1. A loader vehicle control method, applied to a loader, characterized in that: The method includes: Collecting multimodal data of the loader, the multimodal data including operating environment data, dynamic load state data and vehicle posture data of the loader; Build a three-dimensional digital twin model of the operation scene based on multimodal data, and predict the vehicle stability impact data of the impact of load changes on vehicle stability based on dynamic load status data; Based on the 3D operation scene digital twin model and vehicle stability impact data, a reinforcement learning algorithm is used to generate a dynamic control strategy to simultaneously optimize the coordinated actions of the steering, lifting, and traveling mechanisms. Adaptively adjust the hydraulic system pressure and motor torque distribution according to the operating environment data and dynamic control strategy.

2. The loader vehicle control method according to claim 1, characterized in that: Multimodal data can be collected through lidar, cameras, inertial measurement units, pressure sensors, and encoders.

3. The loader vehicle control method according to claim 1, characterized in that: The vehicle stability impact data of building a three-dimensional operation scene digital twin model based on multimodal data and predicting the impact of load changes on vehicle stability based on dynamic load state data includes: Perform spatiotemporal registration and data fusion on multimodal data; Dynamic environment modeling is performed based on multimodal data processed by spatiotemporal registration and data fusion to obtain multimodal data to build a three-dimensional operation scene digital twin model and update the model status in real time; The three-dimensional operation scene digital twin model extracts and evaluates the load parameters to obtain vehicle stability impact data.

4. The loader vehicle control method according to claim 3, characterized in that: The three-dimensional operation scene digital twin model extracts and evaluates load parameters to obtain vehicle stability impact data, including: Calculate bucket / fork load based on pressure sensor signal; Calculate inertial forces using IMU accelerometer data; Combine the hydraulic cylinder pressure change rate to predict load mutations; The rolling angular acceleration is calculated based on the IMU angular velocity, and the sideslip risk is assessed based on the ground slope.

5. The loader vehicle control method according to claim 4, characterized in that: The three-dimensional operation scene digital twin model extracts load parameters and evaluates the load parameters to obtain vehicle stability impact data, including building a prediction model, which is an LSTM model or a Transformer model.

6. The loader vehicle control method according to claim 1, characterized in that: The three-dimensional operation scene digital twin model and vehicle stability impact data are used to generate a dynamic control strategy using a reinforcement learning algorithm to simultaneously optimize the coordinated actions of the steering, lifting, and walking mechanisms, including: Based on the vehicle stability impact data and the three-dimensional operation scene digital twin model, define the state space, define the action space, design the reward function, and determine the reinforcement learning algorithm.

7. The loader vehicle control method according to claim 1, characterized in that: The working environment data includes terrain slope and ground adhesion coefficient.

8. A loader vehicle control system, characterized in that: The system includes: a data acquisition module, a model building module, a control strategy generation module and an adaptive control module; The data acquisition module is used to collect multimodal data of the loader, wherein the multimodal data includes the working environment data, dynamic load state data and vehicle posture data of the loader; In an embodiment of the present invention, the data acquisition module includes a laser radar, a camera, an inertial measurement unit, a pressure sensor, an encoder, etc.; The model building module is used to build a three-dimensional operation scene digital twin model based on multimodal data, and predict vehicle stability impact data of the impact of load changes on vehicle stability based on dynamic load state data; The control strategy generation module is used to generate a dynamic control strategy based on the three-dimensional operation scene digital twin model and vehicle stability impact data using a reinforcement learning algorithm to simultaneously optimize the coordinated actions of the steering, lifting, and walking mechanisms; The adaptive control module is used to adaptively adjust the hydraulic system pressure and motor torque distribution according to the operating environment data and the dynamic control strategy.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor loads and executes the computer program, the steps of the loader vehicle control method according to any one of claims 1 to 7 are implemented.

10. A storage medium storing a computer program, characterized in that: When the computer program is loaded and executed by the processor, the steps of the loader vehicle control method according to any one of claims 1 to 7 are implemented.

Images