Anti-rollover control method and system for balanced heavy forklift truck

Abstract

The invention discloses an anti-rollover control method and system for a balanced heavy forklift truck, and relates to the field of forklift truck intelligent control, and the method comprises the steps: collecting forklift truck working condition data through a sensor group, transmitting the data to a central controller through a CAN bus, and carrying out the preprocessing; inputting a rollover risk assessment algorithm, calculating a boundary condition of a static stability triangle and dynamic lateral moment balance, generating a rollover risk coefficient in combination with working condition data of the forklift, and outputting a decision instruction set; dynamic load adjustment is carried out through a hydraulic servo controller, and load pose parameters are obtained; a motor controller is driven to implement torque vector distribution, a braking energy recovery unit is involved, controllable slow-release braking is applied to wheels on the inner side, and adjusted wheel motion state parameters are obtained; and comparing the theoretical prediction model to carry out residual analysis, triggering a residual correction strategy, and carrying out repetitive control. The working state of the forklift is collected in real time and dynamically analyzed, so that the safety of the forklift in the operation process is improved.

Description

Technical Field

[0001] The present invention relates to the field of intelligent control of forklifts, in particular to an anti-rollover control method and system for a counterbalanced heavy forklift. Background Art

[0002] As an important logistics handling tool, forklifts are widely used in industries, warehouses, ports and other fields. With the rapid development of automation and intelligent technologies, modern forklifts are no longer just simple handling tools, and their role in improving work efficiency and safety has become increasingly important. Especially for counterbalanced heavy forklifts, the research and application of their anti-rollover control technology are directly related to the stability and safety of forklifts under complex working conditions. The traditional forklift stability guarantee system mainly relies on static steady-state models, and calculates by making simplified assumptions about factors such as forklift load, vehicle speed and driving angle, so as to obtain the stability boundary. However, with the continuous change of the forklift working environment, especially the dynamic change of the load position and the increase of external disturbances, the traditional static stability analysis model is difficult to cope with the increasingly complex operating conditions. In recent years, the intelligent monitoring and feedback control system based on sensor technology has become a research hotspot in the field of forklift anti-rollover. By collecting the working state of the forklift in real time and conducting dynamic analysis, the safety of the forklift during operation is improved.

[0003] However, the existing forklift anti-rollover control systems usually rely on a single static stability boundary model or overly simple dynamic analysis, ignoring the complex dynamic characteristics of the forklift during actual operation and the interference of the external environment. For example, the traditional static stability triangle method often cannot accurately predict the rollover risk of forklifts under dynamic loads and uneven ground, and does not consider the dynamic lateral moment change of the vehicle during driving. In addition, the existing technologies mostly focus on anti-rollover adjustment based on linear control models, lacking real-time feedback and optimization for complex working states. Such methods have a certain lag, especially when facing sudden working conditions, it is difficult to respond and adjust quickly, resulting in the system's untimely response to rapidly changing loads and environmental conditions, increasing the risk of forklift rollover. Summary of the Invention

[0004] In view of the above existing problems, the present invention is proposed.

[0005] Therefore, the present invention provides an anti-rollover control method for a counterbalanced heavy forklift to solve the problem that the traditional solution is not timely in response to rapidly changing loads and environmental conditions.

[0006] To solve the above technical problems, the present invention provides the following technical solutions: In a first aspect, the present invention provides an anti-rollover control method for a balanced heavy forklift, which includes installing a sensor group on the forklift, collecting forklift working condition data through the sensor group, transmitting it to a central controller via a CAN bus for preprocessing, and obtaining a preprocessed forklift working condition data packet; inputting the preprocessed forklift working condition data packet into a rollover risk assessment algorithm, calculating the boundary conditions of the static stability triangle and the dynamic lateral moment balance, generating a rollover risk coefficient in combination with the forklift working condition data, and outputting a decision instruction set; performing dynamic load adjustment through a hydraulic servo controller to obtain load pose parameters; the drive motor controller implements torque vector distribution, intervenes in the braking energy recovery unit, applies controllable and slow-release braking to the inner wheels, and obtains adjusted wheel motion state parameters; performing residual analysis by comparing with a theoretical prediction model, triggering a residual correction strategy, and performing repetitive control.

[0007] As a preferred solution of the anti-rollover control method for the balanced heavy forklift of the present invention, wherein: the forklift working condition data includes forklift attitude angle, wheel speed, load weight and distribution, terrain slope, and obstacle distance.

[0008] As a preferred solution of the anti-rollover control method for the balanced heavy forklift of the present invention, wherein: the steps of installing a sensor group on the forklift, collecting forklift working condition data through the sensor group, and transmitting it to a central controller via a CAN bus for preprocessing are specifically as follows: Install an inertial measurement unit, wheel encoder, fork pressure sensor, lidar, and camera on the forklift and connect them into a sensor group; Collect forklift working condition data through the sensor group and transmit it to the central controller via a CAN bus; Use the Kalman filter algorithm to denoise the forklift working condition data; Perform standardization processing on the denoised forklift working condition data and integrate it into a preprocessed forklift working condition data packet.

[0009] As a preferred solution of the anti-rollover control method for the balanced heavy forklift of the present invention, wherein: the steps of inputting the preprocessed forklift working condition data packet into a rollover risk assessment algorithm, calculating the boundary conditions of the static stability triangle and the dynamic lateral moment balance, generating a rollover risk coefficient in combination with the forklift working condition data, and outputting a decision instruction set are specifically as follows: Based on the dynamic potential energy field theory, reconstruct the stability boundary and combine it with the preprocessed forklift working condition data packet, establish a polar coordinate system with the forklift's center of mass as the origin, and define the stable domain boundary function; According to the dynamic stability boundary and the preprocessed forklift working condition data packet, through chaotic dynamics analysis, construct an improved Lorenz equation to describe the moment balance state, and extract the maximum Lyapunov characteristic exponent; Input the dynamic stability boundary, the maximum Lyapunov characteristic exponent, and the preprocessed forklift working condition data packet into the composite risk function to obtain the rollover risk coefficient; Input the rollover risk coefficient into the quantum genetic algorithm optimizer to generate a decision instruction set through the fitness function.

[0010] As a preferred solution of the rollover prevention control method for the counterbalanced heavy forklift described in the present invention, wherein: the dynamic load is adjusted through the hydraulic servo controller to obtain the load pose parameters, and the specific steps are as follows, According to the decision instruction set, adjust the horizontal center of gravity of the forklift load by adjusting the pressure and driving rate of the hydraulic cylinder, and push the forklift fork to move laterally along the guide rail direction; Through the hydraulic servo controller, start the forklift fork height adjustment function, and reduce the center of gravity height and the forklift fork height until the load height and the current vehicle speed meet the stability conditions; Continuously monitor the surrounding space constraints, and switch to the height priority adjustment mode according to the obstacle detection feedback to reduce lateral movement to avoid collisions; After performing the adjustment of the hydraulic servo controller, obtain the real-time updated load pose parameters.

[0011] As a preferred solution of the rollover prevention control method for the counterbalanced heavy forklift described in the present invention, wherein: the drive motor controller implements torque vector distribution, intervenes in the braking energy recovery unit, and applies controllable and slow-release braking to the inner wheels to obtain the adjusted wheel motion state parameters, and the specific steps are as follows, Based on the load pose parameters, calculate the front and rear torque distribution schemes and adjust the torque vector distribution; Intervene in the braking energy recovery unit, adjust the braking force of the inner wheels, and implement controllable and slow-release braking; Collect the adjusted wheel motion state parameters.

[0012] As a preferred solution of the rollover prevention control method for the counterbalanced heavy forklift described in the present invention, wherein: perform residual analysis on the comparison of the theoretical prediction model, trigger the residual correction strategy, and perform repetitive control, and the specific steps are as follows, Based on the historical preprocessed forklift working condition data packet, use the deep learning algorithm to establish a forklift power prediction model; Through working condition simulation, obtain the theoretical prediction model based on the forklift power prediction model; Perform residual analysis on the theoretical prediction model and the wheel motion state parameters, and trigger the residual correction strategy according to the residual analysis results; After implementing the residual correction strategy, perform repetitive residual analysis and correction until the residual analysis results meet the risk threshold.

[0013] In a second aspect, the present invention provides an anti-rollover control system for a balanced heavy forklift, including a data collection module, a risk assessment module, a load adjustment module, a torque distribution module, and a repeated correction module; The data collection module is configured to install a sensor group on the forklift, collect forklift working condition data through the sensor group, transmit it to the central controller via the CAN bus and perform preprocessing to obtain a preprocessed forklift working condition data packet; The risk assessment module is configured to input the preprocessed forklift working condition data packet into a rollover risk assessment algorithm, calculate the boundary conditions of the static stability triangle and the dynamic lateral moment balance, generate a rollover risk coefficient in combination with the forklift working condition data, and output a decision instruction set; The load adjustment module is configured to perform dynamic load adjustment through a hydraulic servo controller to obtain load pose parameters; The torque distribution module is configured to drive the motor controller to implement torque vector distribution, intervene in the braking energy recovery unit, apply controllable and slow-release braking to the inner wheels, and obtain adjusted wheel motion state parameters; The repeated correction module is configured to perform residual analysis by comparing with a theoretical prediction model, trigger a residual correction strategy, and perform repeated control.

[0014] In a third aspect, the present invention provides a computer device, including a memory and a processor, where the memory stores a computer program, and: when the computer program is executed by the processor, any step of the anti-rollover control method for a balanced heavy forklift as described in the first aspect of the present invention is implemented.

[0015] In a fourth aspect, the present invention provides a computer-readable storage medium, on which a computer program is stored, and: when the computer program is executed by the processor, any step of the anti-rollover control method for a balanced heavy forklift as described in the first aspect of the present invention is implemented.

[0016] The beneficial effects of the present invention are as follows: By installing a sensor group and preprocessing the working condition data, real-time and accurate data collection and transmission are ensured, providing a reliable information basis for subsequent control. By dynamically evaluating risks through a rollover risk assessment algorithm and generating a decision instruction set, the stability of the forklift under complex working conditions is optimized. By adjusting the load pose through a hydraulic servo controller, the rollover risk is further reduced, and at the same time, the center of gravity stability of the forklift is improved. Based on the load pose parameters, torque vector distribution is performed and the braking energy recovery is intervened to optimize the power distribution and energy efficiency of the forklift. Through residual analysis and correction control, self-optimization is carried out to ensure the continuous and stable operation of the forklift in a complex environment. Description of the Drawings

[0017] To more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the accompanying drawings required for the description of the embodiments. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other accompanying drawings can be obtained based on these drawings.

[0018] Figure 1 It is a flowchart of the anti-rollover control method for the counterbalanced heavy forklift in Embodiment 1.

[0019] Figure 2 It is a schematic diagram of the anti-rollover control system for the counterbalanced heavy forklift in Embodiment 1. Specific Embodiments

[0020] To make the above objects, features, and advantages of the present invention more obvious and understandable, the following will make a detailed description of the specific embodiments of the present invention in conjunction with the accompanying drawings of the specification.

[0021] In the following description, many specific details are set forth to fully understand the present invention. However, the present invention can also be implemented in other ways different from those described herein. Those skilled in the art can make similar extensions without departing from the connotation of the present invention. Therefore, the present invention is not limited by the specific embodiments disclosed below.

[0022] Secondly, the so-called "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that can be included in at least one implementation manner of the present invention. The "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor is it a separate or selectively exclusive embodiment from other embodiments.

[0023] Embodiment 1, referring to Figure 1 and Figure 2 , is the first embodiment of the present invention. This embodiment provides an anti-rollover control method for a counterbalanced heavy forklift, including the following steps: S1: Install a sensor group on the forklift, collect the forklift working condition data through the sensor group, transmit it to the central controller via the CAN bus and perform preprocessing to obtain the preprocessed forklift working condition data packet.

[0024] Specifically, it includes the following steps: S1.1: Install an inertial measurement unit, wheel encoders, fork pressure sensors, lidar, and cameras on the forklift and connect them into a sensor group.

[0025] Specifically, an Inertial Measurement Unit (IMU) is installed at the center of the forklift or at a position close to the center of mass of the forklift, and is used to detect the attitude angles (such as pitch angle, yaw angle, etc.), as well as acceleration and angular velocity of the forklift in real time. The IMU can accurately measure the dynamic behavior of the forklift, provide attitude change data, and help evaluate the stability of the forklift during movement.

[0026] Wheel encoders are installed on each wheel to detect the rotational speed of the wheels. The wheel rotational speed is important data for judging whether there are abnormal slips or overloads during the driving process of the forklift. Through the wheel rotational speed data fed back by the encoders, the motion state of the forklift and the relative ground slip can be accurately calculated.

[0027] Fork pressure sensors are installed at the forks to measure the load pressure on the forks, and then estimate the load weight. The fork pressure sensors can monitor the load situation of the forklift in real time and give early warnings for overloaded or unevenly distributed loads.

[0028] LiDAR is installed at the front or top of the forklift to scan the surrounding environment in real time and obtain the distance information of obstacles. LiDAR can detect static and dynamic obstacles in the environment and transmit the obstacle distance information in real time, providing a basis for the driving path planning of the forklift.

[0029] Cameras are installed at the front and rear of the forklift to provide real-time image data for assisting in obstacle detection and monitoring of the vehicle's surrounding environment. The cameras can be used in combination with LiDAR to optimize the obstacle recognition accuracy, especially showing good performance in complex environments.

[0030] S1.2: Collect the forklift working condition data through the sensor group and transmit it to the central controller via the CAN bus.

[0031] Specifically, the sensor group transmits the data collected by various sensors to the central controller of the forklift via the CAN bus (Controller Area Network). As a standard communication protocol, the CAN bus has strong anti-interference ability and high transmission rate, and can ensure the real-time and stability of data transmission during the operation of the forklift.

[0032] S1.2.1: The forklift working condition data includes forklift attitude angles, wheel rotational speeds, load weight and distribution, terrain slope, and obstacle distance.

[0033] S1.3: Use the Kalman filter algorithm to denoise the forklift working condition data.

[0034] Specifically, the Kalman filter is a filter based on a recursive algorithm, which is suitable for processing noise data in linear or near-linear systems. The Kalman filter can predict and correct the forklift state provided by the sensor data, and remove the random noise and errors in the measurement process.

[0035] S1.4: Standardize the forklift working condition data after denoising and integrate it into a preprocessed forklift working condition data packet.

[0036] Specifically, use Z-score standardization to convert various types of data into a unified scale so that they have the same weight in subsequent processing and calculations.

[0037] After standardization, the various forklift working condition data are integrated to form a unified preprocessed forklift working condition data packet. This data packet contains all the forklift working condition data collected by sensors (such as attitude angle, wheel speed, load, terrain slope, etc.), and has undergone denoising and standardization to ensure the accuracy and consistency of the forklift working condition data.

[0038] Preferably, by installing an inertial measurement unit, wheel encoder, fork pressure sensor, lidar, and camera on the forklift, it is possible to collect real-time forklift working condition data such as forklift attitude, wheel speed, load condition, terrain slope, and obstacle distance, and transmit it to the central controller through the CAN bus. Use the Kalman filter algorithm to denoise the forklift working condition data, removing noise and errors to ensure the accuracy of the forklift working condition data. Subsequently, use the Z-score standardization method to uniformly process the forklift working condition data and integrate it into a preprocessed forklift working condition data packet. These steps effectively improve the accuracy and consistency of the forklift working condition data, providing a reliable basis for subsequent rollover prevention control, thereby enhancing the safety, stability, and real-time response ability of the forklift in complex environments.

[0039] S2: Input the preprocessed forklift working condition data packet into the rollover risk assessment algorithm, calculate the boundary conditions of the static stability triangle and the dynamic lateral moment balance, generate a rollover risk coefficient in combination with the forklift working condition data, and output a decision instruction set.

[0040] Specifically, it includes the following steps: S2.1: Reconstruct the stability boundary based on the dynamic potential field theory and combine it with the preprocessed forklift working condition data packet. Establish a polar coordinate system with the forklift's center of mass as the origin, and define the stable domain boundary function, the expression of which is: ; where, is the dynamic stability boundary, is the azimuth angle parameter, is the current timestamp, is the total number of pressure sensor grids, is the grid index coefficient, is the th grid's real-time pressure value, is the natural logarithm base, is the distance attenuation coefficient, is the the Euclidean distance from a grid to the virtual boundary in the direction, is the influence factor of the fork height change rate, is the fork height change rate.

[0041] Preferably, by combining the pre - processed forklift working condition data packet with the dynamic potential field theory, the dynamic stability boundary of the forklift under different working conditions can be accurately calculated, considering multiple factors such as pressure and fork height, improving the accuracy and real - time performance of risk assessment.

[0042] S2.2: Based on the dynamic stability boundary and the pre - processed forklift working condition data packet, through chaotic dynamics analysis, an improved Lorenz equation is constructed to describe the moment balance state, and the maximum Lyapunov characteristic exponent is extracted. The expression is: ; ; ; where, is the differential identifier, is the dimensionless lateral moment, is the moment dissipation rate, is the external disturbing moment, is the acceleration coupling coefficient, is the lateral acceleration of the forklift, is the gravitational acceleration, is the critical instability angle, is the phase offset, is the phase attenuation coefficient.

[0043] Preferably, through chaotic dynamics analysis, the moment balance state and dynamic stability of the forklift under complex working conditions can be more accurately reflected. Especially when considering dynamic changes and disturbances, the potential risk of rollover can be captured in a timely manner.

[0044] S2.3: Input the dynamic stability boundary, the maximum Lyapunov characteristic exponent and the pre - processed forklift working condition data packet into the composite risk function to obtain the rollover risk coefficient. The expression of the composite risk function is: ; where, is the rollover risk coefficient, is the environment - dynamics coupling factor, is the maximum Lyapunov characteristic exponent, is the dynamic stability boundary at the critical instability angle, is the maximum - taking identifier, is the partial derivative identifier, is the lidar terrain data, is the abscissa of the terrain data, is the ordinate of the terrain data.

[0045] Preferably, the risk function fuses multiple factors (such as torque, stability boundary, terrain information, etc.) to provide a more comprehensive and accurate risk assessment, ensuring the stability of the forklift under different environmental conditions.

[0046] S2.4: Input the rollover risk coefficient into the quantum genetic algorithm optimizer to generate a decision instruction set through the fitness function. The expression of the fitness function is: ; where, is the fitness function, is the target risk weight, is the control variable weight, is the target risk value, is the change range of the control variable, is the control variable index.

[0047] Specifically, the quantum chromosome encoding contains 20 qubits, representing: the lateral displacement of the load (8 bits), the torque distribution ratio (6 bits), the active steering angle (4 bits), and the braking energy recovery rate (2 bits).

[0048] Through the quantum rotation gate update strategy, iterate 50 generations within 5 ms to output the Pareto optimal solution set. The decision instruction set includes the load adjustment vector, torque difference distribution, and active steering angle.

[0049] It should be understood that the fitness function drives the quantum genetic algorithm to screen out the load adjustment vector, torque difference distribution, and active steering angle parameter combination that maximizes in the 20-dimensional solution space by quantitatively evaluating the quality of the control strategy corresponding to each chromosome.

[0050] It should be noted that the quantum genetic algorithm is an intelligent optimization algorithm that combines the principles of quantum computing and biological evolution mechanisms. Its core uses quantum bit encoding for chromosomes, enabling a single chromosome to carry multiple potential solutions simultaneously through the characteristics of quantum superposition states, greatly improving the search efficiency. The quantum genetic algorithm simulates the process of natural selection, uses the quantum rotation gate to dynamically adjust the chromosome state, retains and optimizes the solutions with high fitness in the form of enhanced probability amplitudes, and maintains the population diversity by means of quantum entanglement, effectively avoiding the defect that the traditional genetic algorithm is prone to falling into local optima. During the iteration process, the parallel evolution ability of the quantum state can quickly explore the solution space and screen out the global optimal solution in combination with the evaluation of the fitness function.

[0051] Furthermore, the role of the fitness function can be divided into the following three levels: Target-oriented role: The numerator term Directly quantify the current rollover risk coefficient The deviation from the target value is weighted to control the priority of risk convergence; the numerator term penalizes the sum of squares of the change in the control amount and restricts the drastic fluctuation of the control command through weighting ; the denominator is normalized to ensure the fitness value is convenient for the algorithm to compare the quality of solutions.

[0052] Chromosome evaluation mechanism: Each quantum chromosome (20-bit encoding) generates a decision instruction set containing parameters such as load displacement and torque difference distribution after decoding; by calculating the value in real time, the balance effect between risk control and execution stability of the current instruction set is evaluated; individuals with high fitness ( approaching 1) indicate that the instruction set can both reduce the rollover risk ( small) and maintain smooth control ( small).

[0053] Evolution driving effect: The quantum rotation gate adjusts the phase of the quantum bit according to the value, enhancing the probability amplitude of the gene state of excellent individuals; under the constraint of 5ms / 50 generations, the population converges to the Pareto front through iterative update, and finally outputs a multi-dimensional optimal solution set; the weight ratio dynamically adjusts the optimization direction, focusing on risk avoidance when and on control smoothness otherwise.

[0054] Preferably, by combining the dynamic potential field theory, chaotic dynamics analysis, composite risk function and quantum genetic algorithm, the rollover risk of the forklift can be evaluated in real time and accurately. The calculation of the dynamic stability boundary and chaotic moment balance combined with the preprocessing working condition data makes the risk assessment closer to the actual working conditions and can capture potential risks that are difficult to identify by traditional methods. The composite risk function provides a more comprehensive assessment by integrating multiple factors (such as the stability boundary, the largest Lyapunov characteristic exponent and terrain information), while the quantum genetic algorithm optimizes the decision-making strategy and can quickly generate highly adaptable control instructions, thus ensuring the stability and safety of the forklift in complex environments.

[0055] S3: Dynamically adjust the load through the hydraulic servo controller to obtain the load pose parameters.

[0056] Specifically, it includes the following steps: S3.1: According to the decision instruction set, adjust the horizontal center of gravity of the forklift load by adjusting the pressure and driving rate of the hydraulic cylinder, and push the forklift forks to move horizontally along the guide rail direction.

[0057] Specifically, according to the decision instruction set, the hydraulic cylinder pressure gradient is determined through the load adjustment vector (the low / middle / high gears correspond to displacement amounts of ±5 cm, 5 - 15 cm, >15 cm), and the differential pressure between the left and right hydraulic cylinders is adjusted by combining the torque difference to balance the load torque. At the same time, the active steering angle sensor real-time feeds back the fork azimuth deviation, and the servo valve is driven to control the lateral displacement of the guide rail in a two-speed mode (fine adjustment 0.1 m / s / fast movement 0.5 m / s). The hydraulic pump adjusts the flow through the ECO valve, and under the coordinated action of the planetary gear set and the DC synchronous motor, according to the center of gravity offset monitored by the triaxial inclination sensor.

[0058] S3.2: Through the hydraulic servo controller, start the fork height adjustment function to reduce the center of gravity height and the fork height until the load height and the current vehicle speed meet the stability condition, and the expression is: ; Wherein, is the center of gravity height, is the vehicle speed, is the friction coefficient.

[0059] Specifically, start the fork height adjustment function, and adjust the height of the fork through the hydraulic servo controller to reduce the center of gravity height of the forklift. The hydraulic controller monitors the vehicle speed in real time and adjusts the fork height to ensure that the center of gravity height of the forklift matches the vehicle speed to meet the stability condition.

[0060] S3.3: Continuously monitor the surrounding space constraints, and switch to the height priority adjustment mode according to the obstacle detection feedback to reduce lateral movement to avoid collisions.

[0061] Specifically, continuously monitor the surrounding space and obtain the feedback information of obstacles in real time. If an obstacle or a narrow space is detected, switch to the "height priority adjustment mode". In this mode, the lateral movement will be reduced, and the fork height will be adjusted preferentially to avoid collisions with obstacles.

[0062] S3.4: After performing the adjustment of the hydraulic servo controller, obtain the real-time updated load pose parameters.

[0063] Specifically, after the adjustment by the hydraulic servo controller, the central control unit will obtain the real-time updated load pose parameters, including the lateral displacement, longitudinal displacement, fork height, attitude angle, etc. of the load. All these parameters will be fed back to the central control unit for the next risk assessment and control adjustment.

[0064] Preferably, through precise hydraulic servo control and real-time feedback mechanisms, dynamic load regulation and stability control of the forklift are achieved. By adjusting the hydraulic cylinder pressure and driving speed, the horizontal center of gravity of the load is finely adjusted to ensure the forklift remains stable during dynamic operation. The real-time adjustment of the fork height is matched with the vehicle speed to avoid the tipping risk caused by too high a center of gravity. At the same time, through obstacle detection and the "height priority adjustment mode", collisions are effectively avoided to ensure the safe operation of the forklift in narrow spaces. The real-time updated load pose parameter feedback provides a basis for dynamic adjustment by the central control unit, further optimizing the stability and safety of the forklift. These measures greatly improve the operation efficiency and safety of the forklift in complex environments.

[0065] S4: The drive motor controller implements torque vector distribution, intervenes in the braking energy recovery unit, and applies controllable and slow-release braking to the inner wheels to obtain the adjusted wheel motion state parameters.

[0066] Specifically, it includes the following steps: S4.1: Based on the load pose parameters, calculate the front and rear torque distribution schemes and adjust the torque vector distribution.

[0067] Specifically, based on the previously obtained load pose parameters (such as the lateral position, height, vehicle speed, etc. of the load), according to the working state and dynamic requirements of the forklift, calculate the torque distribution schemes for the front and rear wheels. The system will consider factors such as load distribution, the current speed of the forklift, and road surface friction to ensure that the front and rear wheels obtain appropriate torque outputs to optimize the dynamic performance and stability of the forklift. Through torque vector distribution, adjust the torque distribution of the front and rear wheels to achieve lateral stability and vehicle steering control. The adjustment process uses a closed-loop control method to continuously correct the torque distribution to adapt to environmental changes and load changes.

[0068] S4.2: Intervene in the braking energy recovery unit, adjust the braking force of the inner wheels, and implement controllable and slow-release braking.

[0069] Specifically, during the turning or operation of the forklift, the inner wheels may slip due to uneven load or the risk of rollover. Through the braking energy recovery unit, first monitor the slip state of the inner wheels and calculate the required braking force. According to the calculation results, adjust the braking system of the inner wheels and gradually apply slow-release braking. This process controls the braking force of the inner wheels to reduce lateral slip, while achieving braking energy recovery and improving energy efficiency. The braking process is applied and released progressively to ensure wheel stability and avoid the danger caused by sudden braking.

[0070] S4.3: Collect the adjusted wheel motion state parameters.

[0071] Specifically, after torque distribution and braking adjustment, the motion state parameters of the wheels are collected in real time through wheel encoders or other motion sensors. These motion state parameters include the rotational speed, steering angle, slip ratio, etc. of the wheels. These data will be fed back to the central control unit for further adjustment of the control strategy. The collected motion state parameters of the wheels will be used to evaluate the effects of torque distribution and braking adjustment, ensure that the vehicle is in an optimal stable state, and provide a basis for the next dynamic control and risk assessment.

[0072] Preferably, by performing torque vector distribution and braking energy recovery based on the load pose parameters, the operating performance and safety of the forklift are effectively improved. By accurately calculating the front and rear torque distribution, the power output and stability of the forklift are optimized, ensuring the balance and maneuverability of the forklift under different loads and dynamic environments. At the same time, by intervening in the braking energy recovery unit, controllable slow-release braking is implemented on the inner wheels to reduce slip, improve energy efficiency and reduce the risk of accidents. Collecting the motion state parameters of the wheels in real time helps to dynamically adjust the control strategy to ensure that the forklift always maintains the best stability and safety in complex environments.

[0073] S5: Conduct residual analysis by comparing with the theoretical prediction model, trigger the residual correction strategy, and perform repetitive control.

[0074] Specifically, it includes the following steps: S5.1: Based on the historical preprocessed forklift working condition data packet, use deep learning algorithms to establish a forklift power prediction model.

[0075] Specifically, to establish a power prediction model based on the historical preprocessed forklift working condition data packet, first perform normalization processing on multi-source data such as the forklift attitude angle, wheel rotational speed, load distribution, terrain slope, and obstacle distance. Construct a time series feature matrix through a sliding time window (1-second interval). Adopt a deep neural network architecture, use the LSTM layer to extract the time series correlation between the wheel rotational speed and the attitude angle, the CNN branch processes the spatial features of the load distribution, and the fusion layer integrates the dynamic effects of the terrain slope gradient and the obstacle distance. The power prediction model is trained using the Adam optimizer, with the mean square error between the actual power output and the predicted value as the loss function, combined with the early stopping method to prevent overfitting. Finally, the power prediction model outputs the hydraulic system pressure gradient and the drive rate adjustment amount, continuously updates the weights, and realizes adaptive power matching under variable working conditions. In the verification stage, vibration sensor data is introduced for cross-verification to ensure that the prediction error is controlled within ±5%.

[0076] S5.2: Through working condition simulation and simulation, obtain the theoretical prediction model based on the forklift power prediction model.

[0077] Specifically, by using the forklift power prediction model, the forklift model is run in the working condition simulation and simulation environment to simulate the behaviors of the forklift under different loads, different vehicle speeds, and different turning radii. Through these simulations, a set of theoretical prediction models are generated, which are specifically manifested as various motion parameters of the forklift, such as wheel speed, vehicle speed, roll angle, lateral force, etc. These simulation results will serve as the theoretical performance benchmark of the forklift under ideal conditions and can be used for subsequent comparative analysis with the actual wheel motion state.

[0078] S5.3: Conduct a residual analysis on the theoretical prediction model and the wheel motion state parameters, and trigger the residual correction strategy according to the results of the residual analysis.

[0079] Specifically, during the actual operation process, the actual motion state parameters of the forklift are obtained through wheel encoders and other motion sensors, including wheel speed, lateral slip ratio, longitudinal acceleration, etc. Compare these actual parameters with the results in the theoretical prediction model and calculate the residuals (i.e., the differences between the theoretical prediction values and the actual measurement values). Residual analysis can reveal the deviations or errors existing in the actual operation, including the deviations caused by factors such as environmental changes, uneven loads, and control errors.

[0080] S5.4: After implementing the residual correction strategy, conduct repeated residual analysis and correction until the results of the residual analysis meet the risk threshold.

[0081] Specifically, once the deviation of the residual analysis results exceeds the preset threshold (more than 10%, and the threshold is set according to historical experience), the residual correction strategy will be triggered. This strategy includes the following adjustments: Adjust torque distribution: According to the results of the residual analysis, adjust the torque distribution of the front and rear wheels to correct the difference in wheel speed; Adjust braking and acceleration control: For the deviations caused by uneven loads or vehicle speed changes, adjust the response of braking or acceleration to correct the lateral stability of the vehicle; Dynamically correct the driving path: By changing the steering angle or trajectory of the forklift, correct the slip or instability phenomenon in the lateral movement.

[0082] After implementing the residual correction, continue to monitor the wheel motion state and conduct another residual analysis. If the corrected residuals meet the set threshold, it is considered stable; otherwise, continue to correct until the error is controlled within the threshold range.

[0083] Preferably, through the combination of deep learning and residual correction strategy, the dynamic control of the forklift is analyzed in real time to ensure the stability and safety of the forklift under various working conditions. By establishing a forklift power prediction model and working condition simulation, the behavior of the forklift can be accurately predicted to provide a theoretical benchmark; by comparing the theoretical model with the actual wheel motion state, residual analysis is carried out to timely detect and correct deviations. If the deviation exceeds the threshold, the torque distribution, braking and acceleration control, as well as the driving path are readjusted to ensure the stable operation of the forklift. The robustness, operation efficiency and energy efficiency of the forklift are improved, providing strong support for the intelligent control of the forklift.

[0084] This embodiment also provides an anti-rollover control system for a counterbalanced heavy forklift, including: a data collection module, which is used to install a sensor group on the forklift, collect forklift working condition data through the sensor group, transmit it to the central controller through the CAN bus and perform preprocessing to obtain a preprocessed forklift working condition data packet; a risk assessment module, which is used to input the preprocessed forklift working condition data packet into the rollover risk assessment algorithm, calculate the boundary conditions of the static stability triangle and the dynamic lateral moment balance, generate a rollover risk coefficient in combination with the forklift working condition data, and output a decision instruction set; a load adjustment module, which is used to perform dynamic load adjustment through a hydraulic servo controller to obtain load pose parameters; a torque distribution module, which is used to drive the motor controller to implement torque vector distribution, intervene in the braking energy recovery unit, and apply controllable slow-release braking to the inner wheels to obtain adjusted wheel motion state parameters; a repeated correction module, which is used to perform residual analysis by comparing with the theoretical prediction model, trigger the residual correction strategy, and perform repeated control.

[0085] This embodiment also provides a computer device, which is applicable to the situation of the anti-rollover control method for a counterbalanced heavy forklift, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the anti-rollover control method for a counterbalanced heavy forklift as proposed in the above embodiment.

[0086] The computer device can be a terminal, which includes a processor, a memory, a communication interface, a display screen, and an input device connected via a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The communication interface of the computer device is used to communicate with external terminals in a wired or wireless manner, and the wireless manner can be implemented through WIFI, a carrier network, NFC (Near Field Communication), or other technologies. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covering the display screen, or buttons, trackballs, or touchpads provided on the housing of the computer device, or an external keyboard, touchpad, or mouse, etc.

[0087] This embodiment also provides a storage medium, on which a computer program is stored. When the program is executed by a processor, it implements the anti-rollover control method for a balanced counterbalanced forklift as proposed in the above embodiment; the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (Static Random Access Memory, abbreviated as SRAM), electrically erasable programmable read-only memory (Electrically Erasable Programmable Read-Only Memory, abbreviated as EEPROM), erasable programmable read-only memory (Erasable Programmable Read Only Memory, abbreviated as EPROM), programmable read-only memory (Programmable Red-Only Memory, abbreviated as PROM), read-only memory (Read-Only Memory, abbreviated as ROM), magnetic memory, flash memory, a magnetic disk, or an optical disc.

[0088] In summary, the present invention: installs a sensor group and preprocesses the working condition data, ensuring real-time and accurate data collection and transmission, and providing a reliable information basis for subsequent control. Dynamically evaluates risks through a rollover risk assessment algorithm and generates a decision instruction set, optimizing the stability of the forklift under complex working conditions. Adjusts the load posture through a hydraulic servo controller, further reducing the rollover risk and improving the center of gravity stability of the forklift. Performs torque vector distribution based on the load posture parameters and intervenes in the braking energy recovery, optimizing the power distribution and energy efficiency of the forklift. Through residual analysis and correction control, self-optimization is carried out to ensure the continuous and stable operation of the forklift in a complex environment.

[0089] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention, and they should all be covered within the scope of the claims of the present invention.

Claims

1. A rollover prevention control method for a balanced heavy forklift, characterized in that: including Install a sensor group on the forklift, collect forklift working condition data through the sensor group, transmit it to the central controller via the CAN bus and perform preprocessing to obtain a preprocessed forklift working condition data packet; Input the preprocessed forklift working condition data packet into the rollover risk assessment algorithm, calculate the boundary conditions of the static stability triangle and the dynamic lateral moment balance, generate a rollover risk coefficient in combination with the forklift working condition data, and output a decision instruction set; Perform dynamic load adjustment through a hydraulic servo controller to obtain load pose parameters; The drive motor controller implements torque vector distribution, intervenes in the braking energy recovery unit, applies controllable and slow-release braking to the inner wheels, and obtains adjusted wheel motion state parameters; Conduct residual analysis by comparing with the theoretical prediction model, trigger the residual correction strategy, and perform repetitive control.

2. The anti-rollover control method of the balanced heavy forklift according to claim 1, characterized in that: The forklift working condition data includes forklift attitude angle, wheel speed, load weight and distribution, terrain slope, and obstacle distance.

3. The anti-rollover control method of the balanced heavy forklift according to claim 2, characterized in that: The step of installing a sensor group on the forklift, collecting forklift working condition data through the sensor group, transmitting it to the central controller via the CAN bus and performing preprocessing is as follows: Install an inertial measurement unit, wheel encoder, fork pressure sensor, lidar, and camera on the forklift and connect them into a sensor group; Collect forklift working condition data through the sensor group and transmit it to the central controller via the CAN bus; Use the Kalman filter algorithm to denoise the forklift working condition data; Perform standardization processing on the denoised forklift working condition data and integrate it into a preprocessed forklift working condition data packet.

4. The anti-rollover control method of the balanced heavy forklift according to claim 3, characterized in that: The step of inputting the preprocessed forklift working condition data packet into the rollover risk assessment algorithm, calculating the boundary conditions of the static stability triangle and the dynamic lateral moment balance, generating a rollover risk coefficient in combination with the forklift working condition data, and outputting a decision instruction set is as follows: Based on the dynamic potential energy field theory, reconstruct the stability boundary and combine it with the preprocessed forklift working condition data packet, establish a polar coordinate system with the forklift's center of mass as the origin, and define the stable domain boundary function; According to the dynamic stability boundary and the preprocessed forklift working condition data packet, through chaotic dynamics analysis, construct an improved Lorenz equation to describe the moment balance state and extract the maximum Lyapunov characteristic exponent; Input the dynamic stability boundary, the maximum Lyapunov characteristic exponent, and the preprocessed forklift working condition data packet into the composite risk function to obtain the rollover risk coefficient; Input the rollover risk coefficient into the quantum genetic algorithm optimizer and generate a decision instruction set through the fitness function.

5. The anti-rollover control method of the balanced heavy forklift according to claim 4, characterized in that: The step of performing dynamic load adjustment through a hydraulic servo controller to obtain load pose parameters is as follows: According to the decision instruction set, adjust the pressure and driving rate of the hydraulic cylinder to adjust the horizontal center of gravity of the forklift load, and push the fork to move laterally along the guide rail direction; Through the hydraulic servo controller, start the fork height adjustment function, reduce the center of gravity height and fork height until the load height and the current vehicle speed meet the stability conditions; Continuously monitor the surrounding space constraints, switch to the height priority adjustment mode according to the obstacle detection feedback, and reduce the lateral movement; After performing the adjustment of the hydraulic servo controller, obtain the real-time updated load pose parameters.

6. The anti-rollover control method of the balanced heavy forklift according to claim 5, characterized in that: The drive motor controller implements torque vector distribution, intervenes in the braking energy recovery unit, applies controllable and slow-release braking to the inner wheels, and obtains the adjusted wheel motion state parameters. The specific steps are as follows: Based on the load pose parameters, calculate the front and rear torque distribution scheme and adjust the torque vector distribution; Intervene in the braking energy recovery unit, adjust the braking force of the inner wheels, and implement controllable and slow-release braking; Collect the adjusted wheel motion state parameters.

7. The anti-rollover control method of the balanced heavy forklift according to claim 6, characterized in that: Perform residual analysis on the comparison with the theoretical prediction model, trigger the residual correction strategy, and perform repetitive control. The specific steps are as follows: Based on the historical preprocessed forklift working condition data packet, use the deep learning algorithm to establish a forklift power prediction model; Through working condition simulation, obtain the theoretical prediction model based on the forklift power prediction model; Perform residual analysis on the theoretical prediction model and the wheel motion state parameters, and trigger the residual correction strategy according to the residual analysis results; After implementing the residual correction strategy, perform repeated residual analysis and correction until the residual analysis results meet the risk threshold.

8. An anti-rollover control system for a counterbalanced heavy forklift, based on the anti-rollover control method for a counterbalanced heavy forklift according to any one of claims 1 to 7, characterized in that: It includes a data collection module, a risk assessment module, a load adjustment module, a torque distribution module, and a repeated correction module; The data collection module is used to install a sensor group on the forklift, collect forklift working condition data through the sensor group, transmit it to the central controller through the CAN bus and perform preprocessing to obtain the preprocessed forklift working condition data packet; The risk assessment module is used to input the preprocessed forklift working condition data packet into the rollover risk assessment algorithm, calculate the boundary conditions of the static stability triangle and the dynamic lateral moment balance, generate a rollover risk coefficient in combination with the forklift working condition data, and output a decision instruction set; The load adjustment module is used to perform dynamic load adjustment through a hydraulic servo controller to obtain the load pose parameters; The torque distribution module is used to drive the motor controller to implement torque vector distribution, intervene in the braking energy recovery unit, apply controllable and slow-release braking to the inner wheels, and obtain the adjusted wheel motion state parameters; The repeated correction module is used to perform residual analysis on the comparison with the theoretical prediction model, trigger the residual correction strategy, and perform repetitive control.

9. A computer device, comprising a memory and a processor, the memory storing a computer program, characterized in that: When the processor executes the computer program, it realizes the steps of the anti-rollover control method of the counterbalanced heavy forklift according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, it realizes the steps of the anti-rollover control method of the counterbalanced heavy forklift according to any one of claims 1 to 7.

Images