Dynamic Estimation of the Center of Gravity Coordinates of an Off-road Forklift and Its Application Method, System, and Device

Abstract

The present invention discloses a method, system and device for dynamically estimating the center-of-gravity coordinates of an off-road forklift and its application. The method includes: obtaining the operation data of the off-road forklift in the current period, obtaining the center-of-gravity measurement data based on the operation data, and forming operation state data based on the center-of-gravity measurement data and the operation data; constructing a state prediction model and a state update model based on the operation state data; predicting the operation state data of the previous period based on the state prediction model to obtain the predicted value of the operation state data in the current period, and updating the predicted value of the operation state data in the current period based on the state update model to obtain the optimal predicted value of the operation state data in the current period, and further obtaining the optimal center-of-gravity coordinate data; obtaining the optimal torque of the off-road forklift based on the optimal center-of-gravity coordinate data and the center-of-gravity measurement data. The present invention adjusts the torque distribution between the front and rear wheels of the off-road forklift through the optimal torque, so that the off-road forklift can adapt to various off-road working conditions.

Description

Technical Field

[0001] The present invention relates to the field of vehicle control, and particularly to a method, system and device for dynamically estimating the center-of-gravity coordinates of an off-road forklift and its application. Background Art

[0002] When a four-wheel drive off-road forklift operates under complex off-road conditions, accurately grasping its motion state and parameter changes is crucial for improving the performance, safety and reliability of the forklift. Traditional forklift parameter monitoring and torque regulation methods often have problems such as low accuracy and untimely response under off-road conditions, and it is difficult to meet the special requirements of four-wheel drive off-road forklifts. Therefore, when facing different working scenarios such as muddy and rough roads, flat roads, and turning in narrow spaces, how to reasonably distribute torque according to different actual working requirements has become a key link in the high-efficiency transmission system of off-road forklifts.

[0003] At the same time, reasonable torque distribution can protect the transmission system of the forklift, because excessive torque can damage components such as gears and drive shafts in the transmission system. Controlling the torque within a reasonable range can avoid damaging the transmission system due to overload. On the other hand, the smooth transmission of torque can also reduce the impact and wear on the transmission system and extend its service life. In the early stage, the torque distribution of the forklift gearbox used a fixed gear ratio to transmit torque and could not be adjusted according to the actual working conditions. With the application of mechanical control and electronic technology to the gearbox, torque distribution can be achieved through mechanical devices and the ECU system, but the flexibility of torque distribution is still limited. However, as an off-road forklift with strong off-road performance, when dealing with complex road conditions and transmission power requirements, the ECU system is easily affected by environmental adaptability such as temperature, dust, water vapor, etc. During operation, it mainly relies on the mechanical execution of specific software programs. When there are loopholes or other unstable factors interfering, it will lead to incorrect torque distribution, thereby affecting the performance and safety of the forklift. Moreover, traditional forklifts cannot accurately adjust torque according to the change of the center of gravity, resulting in insufficient stability when driving on complex terrains. Summary of the Invention

[0004] The present invention aims at the deficiencies in the prior art and provides a method, system and device for dynamically estimating the center-of-gravity coordinates of an off-road forklift and its application.

[0005] To solve the above technical problems, the present invention is solved by the following technical solutions:

[0006] A method for dynamically estimating the center-of-gravity coordinates of an off-road forklift and its application includes the following steps:

[0007] Obtain the operation data of the off-road forklift in the current period. Based on the operation data, obtain the center-of-gravity measurement data, and form the operation state data based on the center-of-gravity measurement data and the operation data;

[0008] Based on the operating state data, construct a state transition matrix, a control input matrix, an observation matrix, and a Kalman gain; based on the state transition matrix and the control input matrix, construct a state prediction model; based on the observation matrix and the Kalman gain, construct a state update model;

[0009] Based on the state prediction model, predict the operating state data of the previous time period to obtain the predicted value of the operating state data of the current time period, and update the predicted value of the operating state data of the current time period based on the state update model to obtain the optimal predicted value of the operating state data of the current time period, and further obtain the optimal center of gravity coordinate data;

[0010] Adjust the operating state of the rough terrain forklift based on the optimal center of gravity coordinate data and the center of gravity measurement data, where at least the optimal torque of the rough terrain forklift is adjusted, specifically: based on the optimal center of gravity coordinate data, the center of gravity measurement data, and the actual torque, construct a torque adjustment model to obtain the optimal torque of the rough terrain forklift.

[0011] As an implementable manner, the operating data at least includes the total mass of the rough terrain forklift and the goods, the rolling resistance coefficient, the driving force, the braking force, the angular velocity, and the turning radius. Obtaining the center of gravity measurement data based on the operating data includes the following steps:

[0012] Obtain the measured value of the longitudinal acceleration of the center of gravity through the driving force, the braking force, the rolling resistance coefficient, and the total mass of the rough terrain forklift and the goods, and obtain the measured value of the longitudinal coordinate of the center of gravity through the measured value of the longitudinal acceleration of the center of gravity and the initial longitudinal velocity of the center of gravity;

[0013] Based on the angular velocity and the turning radius, obtain the measured value of the lateral acceleration of the center of gravity, and obtain the measured value of the lateral coordinate of the center of gravity through the measured value of the lateral acceleration of the center of gravity and the initial lateral velocity of the center of gravity, where the measured value of the longitudinal coordinate of the center of gravity and the measured value of the lateral coordinate of the center of gravity are the center of gravity measurement data;

[0014] Among them, the measured value of the longitudinal acceleration of the center of gravity and the measured value of the lateral acceleration of the center of gravity are respectively expressed as follows:

[0015]

[0016]

[0017] Among them, the measured value of the longitudinal coordinate of the center of gravity and the measured value of the lateral coordinate of the center of gravity are respectively expressed as follows:

[0018]

[0019]

[0020] Among them, represents the measured value of the longitudinal acceleration of the center of gravity, represents the measured value of the lateral acceleration of the center of gravity, represents the driving force, represents the braking force, represents the rolling resistance coefficient, represents the total mass of the rough terrain forklift and the goods, represents the acceleration due to gravity, represents the angular velocity, represents the turning radius, represents the measured value of the longitudinal coordinate of the center of gravity, represents the initial longitudinal coordinate of the center of gravity, represents the initial longitudinal velocity of the center of gravity, represents time, represents the measured value of the lateral coordinate of the center of gravity, represents the initial lateral coordinate of the center of gravity, represents the initial lateral velocity of the center of gravity.

[0021] As an implementable mode, the operating data at least includes rotational speed, torque, driving force, rough terrain forklift acceleration, angular velocity, step size and center of gravity data. Based on the operating state data, a state transition matrix, a control input matrix, an observation matrix and a Kalman gain are constructed; based on the state transition matrix and the control input matrix, a state prediction model is constructed; based on the observation matrix and the Kalman gain, a state update model is constructed, including the following steps:

[0022] Construct a state transition matrix through rotational speed, torque, driving force and rough terrain forklift acceleration;

[0023] Obtain the road surface anti-skid performance index, and construct a control input matrix through the road surface anti-skid performance index and the step size;

[0024] Construct a system state vector through center of gravity data, angular velocity, rough terrain forklift acceleration, torque and driving force, and then obtain the error covariance matrix of the system state vector;

[0025] Obtain the actual angular velocity, the actual rough terrain forklift acceleration, the actual torque and combine the center of gravity measurement data to construct an observation vector, and then obtain the noise covariance matrix of the observation vector. Based on the observation vector, construct an observation matrix;

[0026] Construct a Kalman gain through the observation matrix, the error covariance matrix and the noise covariance matrix;

[0027] Based on the system state vector, the state transition matrix and the control input matrix, construct a state prediction model, and then obtain the prior estimate of the system state vector;

[0028] Based on the prior estimate of the system state vector, the Kalman gain, the observation vector, and the observation matrix, construct a state update model to obtain the posterior estimate of the system state vector.

[0029] As an implementable manner, the state transition matrix is expressed as follows:

[0030]

[0031] The pavement skid resistance performance index and the control input matrix are respectively expressed as follows:

[0032]

[0033]

[0034] The observation matrix is expressed as follows:

[0035]

[0036] The Kalman gain is expressed as follows:

[0037]

[0038] The state prediction model is expressed as follows:

[0039]

[0040] The state update model is expressed as follows:

[0041]

[0042] Wherein, represents the state transition matrix, represents the step size, represents the rotational speed, represents the torque, represents the driving force, represents the acceleration of the rough terrain forklift, represents a function of the operating state data, represents the pavement skid resistance performance index, represents the actual lateral force coefficient value, represents the minimum lateral force coefficient value, represents the magnitude of the center of gravity speed in the center of gravity measurement data, , represents the magnitude of the longitudinal center of gravity speed in the center of gravity measurement data, represents the magnitude of the lateral center of gravity speed in the center of gravity measurement data, represents the direction of the center of gravity speed in the center of gravity measurement data, , respectively represent coefficients, represents the control input matrix, represents the observation matrix, both represent coefficients related to the observation vector, represents the Kalman gain at the current time period, represents the current time period, represents the error covariance matrix of the prior estimate of the system state vector at the current time period, , represents the state transition matrix at the current time period, represents the error covariance matrix of the posterior estimate of the system state vector at the previous time period, represents the transpose of the matrix, represents the system noise covariance matrix at the current time period, represents the noise covariance matrix of the observation vector at the current time period, represents the prior estimate of the system state vector at the current time period, represents the posterior estimate of the system state vector at the previous time period, represents the control input matrix at the current time period, represents the control input vector at the current time period, represents the posterior estimate of the system state vector at the current time period, represents the observation vector at the current time period.

[0043] As an implementable manner, the state prediction model predicts the operation state data of the previous time period to obtain the predicted value of the operation state data of the current time period, and updates the predicted value of the operation state data of the current time period based on the state update model to obtain the optimal predicted value of the operation state data of the current time period, and further obtains the optimal center of gravity coordinate data, including the following steps:

[0044] Based on the posterior estimate of the system state vector of the previous time period, the state transition matrix of the current time period, and the control input matrix of the current time period, combined with the state prediction model, obtain the prior estimate of the system state vector of the current time period;

[0045] Based on the error covariance matrix of the posterior estimate of the system state vector of the previous time period, obtain the error covariance matrix of the prior estimate of the system state vector of the current time period;

[0046] Through the error covariance matrix of the prior estimate of the system state vector of the current time period, the observation matrix of the current time period, and the noise covariance matrix of the observation vector of the current time period, obtain the Kalman gain of the current time period, and further obtain the error covariance matrix of the posterior estimate of the system state vector of the current time period;

[0047] Based on the prior estimate of the system state vector in the current time period, the Kalman gain in the current time period, the observation vector in the current time period, and the observation matrix in the current time period, combined with the state update model, obtain the posterior estimate of the system state vector in the current time period;

[0048] Until the iteration termination condition is satisfied, the obtained posterior estimate of the final system state vector is the optimal system state vector, and then the optimal centroid coordinate data is obtained.

[0049] As an implementable manner, constructing a torque adjustment model based on the optimal centroid coordinate data, centroid measurement data, and actual torque to obtain the optimal torque of the rough terrain forklift includes the following steps:

[0050] Obtain the actual rear-wheel torque and the actual front-wheel torque of the rough terrain forklift respectively;

[0051] Based on the operation data, obtain the adjustment coefficient;

[0052] Based on the centroid measurement data, optimal centroid coordinate data, actual rear-wheel torque, actual front-wheel torque, and adjustment coefficient, construct a torque adjustment model, where the torque adjustment model includes a rear-wheel torque adjustment model and a front-wheel torque adjustment model;

[0053] Optimize the rear-wheel torque adjustment model and the front-wheel torque adjustment model to obtain the optimal adjustment coefficient, and then obtain the optimal rear-wheel torque and the optimal front-wheel torque of the rough terrain forklift;

[0054] Among them, the rear-wheel torque adjustment model is expressed as follows:

[0055]

[0056] The front-wheel torque adjustment model is expressed as follows:

[0057]

[0058] Among them, represents the optimal rear-wheel torque, represents the actual rear-wheel torque, represents the optimal front-wheel torque, represents the actual front-wheel torque, represents the longitudinal centroid coordinate of the centroid measurement data, represents the lateral centroid coordinate of the centroid measurement data, represents the optimal longitudinal centroid coordinate of the optimal centroid coordinate data, represents the optimal lateral centroid coordinate of the optimal centroid coordinate data, represents the preset longitudinal centroid forward movement threshold, represents the preset lateral centroid forward movement threshold, represents the preset longitudinal centroid backward movement threshold, represents a preset threshold for the lateral rearward shift of the center of gravity both represent adjustment coefficients and are both positive numbers

[0059] As an implementable manner, the adjustment coefficient is obtained through the following steps

[0060] Obtain the lateral force, tire load, steering wheel angle, and the mass of the goods carried by the off-road forklift through sensors

[0061] Based on the lateral force and the tire load, obtain the lateral force coefficient value. Based on the lateral force coefficient value and the center of gravity measurement data, obtain the road surface anti-skid performance index, where the lateral force coefficient value = lateral force / tire load

[0062] Based on the road surface anti-skid performance index, the steering wheel angle, and the weight of the goods carried by the off-road forklift, obtain the adjustment coefficient

[0063] Among them, the adjustment coefficient is expressed as follows

[0064]

[0065] Among them represents the road surface anti-skid performance index represents a preset first road surface anti-skid index threshold represents a preset second road surface anti-skid index threshold, and , represents the steering wheel angle represents a preset angle threshold represents the mass of the goods carried by the off-road forklift represents a preset load mass threshold both represent adjustment coefficients under different conditions

[0066] As an implementable manner, the optimal adjustment coefficient is obtained through the following steps

[0067] Randomly generate an initial particle swarm, obtain the position and velocity of each particle in the initial particle swarm, and further obtain the initial individual optimal position of each particle and the initial global optimal position of the particle swarm, where the position of the particle is the adjustment coefficient to be solved

[0068] Based on the position of the particle, combine the torque adjustment model, obtain the torque prediction value and combine the actual torque, construct a fitness model, and based on the fitness model, obtain the fitness of the current position of each particle

[0069] If the fitness of the current position of a particle is greater than the fitness of the individual optimal position, then update the individual optimal position. If the fitness of the current position of any particle is greater than the fitness of the global optimal position, then update the global optimal position, and update the position and velocity of the particle to obtain the updated particle velocity and the updated particle position;

[0070] Until the termination condition is met, then output the final global optimal position as the optimal adjustment coefficient;

[0071] Among them, the fitness model is expressed as follows:

[0072]

[0073] The updated particle velocity is expressed as follows:

[0074]

[0075] The updated particle position is expressed as follows:

[0076]

[0077] Among them, represents fitness, respectively represent the predicted values of the rear-wheel torque and the front-wheel torque, represents the total quantity of the predicted rear-wheel torque and the front-wheel torque, represents the updated particle velocity, represents the inertia weight, represents the particle velocity of the previous iteration, represents the individual optimal position of the particle of the previous iteration, represents the particle position of the previous iteration, represents the global optimal position of the previous iteration, represents the learning factor, represents a random number, represents the updated particle position.

[0078] An off-road forklift center-of-gravity coordinate dynamic estimation and application system includes a data acquisition module, a model construction module, a center-of-gravity estimation module, and a state adjustment module;

[0079] The data acquisition module acquires the operation data of the off-road forklift in the current period, based on the operation data, obtains the center-of-gravity measurement data, and forms the operation state data based on the center-of-gravity measurement data and the operation data;

[0080] The model construction module constructs a state transition matrix, a control input matrix, an observation matrix, and a Kalman gain based on the operating state data; constructs a state prediction model based on the state transition matrix and the control input matrix; and constructs a state update model based on the observation matrix and the Kalman gain.

[0081] The centroid estimation module predicts the operating state data of the previous time period based on the state prediction model to obtain a predicted value of the operating state data of the current time period, and updates the predicted value of the operating state data of the current time period based on the state update model to obtain an optimal predicted value of the operating state data of the current time period, and further obtains optimal centroid coordinate data.

[0082] The state adjustment module adjusts the operating state of the rough terrain forklift based on the optimal centroid coordinate data and the centroid measurement data, where at least the optimal torque of the rough terrain forklift is adjusted. Specifically, a torque adjustment model is constructed based on the optimal centroid coordinate data, the centroid measurement data, and the actual torque to obtain the optimal torque of the rough terrain forklift.

[0083] A computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the following method is implemented:

[0084] Obtain the operating data of the rough terrain forklift in the current time period, obtain the centroid measurement data based on the operating data, and form operating state data based on the centroid measurement data and the operating data.

[0085] Based on the operating state data, construct a state transition matrix, a control input matrix, an observation matrix, and a Kalman gain; construct a state prediction model based on the state transition matrix and the control input matrix; construct a state update model based on the observation matrix and the Kalman gain.

[0086] Predict the operating state data of the previous time period based on the state prediction model to obtain a predicted value of the operating state data of the current time period, and update the predicted value of the operating state data of the current time period based on the state update model to obtain an optimal predicted value of the operating state data of the current time period, and further obtain optimal centroid coordinate data.

[0087] Adjust the operating state of the rough terrain forklift based on the optimal centroid coordinate data and the centroid measurement data, where at least the optimal torque of the rough terrain forklift is adjusted. Specifically, a torque adjustment model is constructed based on the optimal centroid coordinate data, the centroid measurement data, and the actual torque to obtain the optimal torque of the rough terrain forklift.

[0088] A rough terrain forklift centroid coordinate dynamic estimation and application device includes a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, the following method is implemented:

[0089] Obtain the operation data of the rough terrain forklift in the current period. Based on the operation data, obtain the center of gravity measurement data, and form the operation state data based on the center of gravity measurement data and the operation data;

[0090] Based on the operation state data, construct the state transition matrix, control input matrix, observation matrix and Kalman gain; Based on the state transition matrix and the control input matrix, construct the state prediction model; Based on the observation matrix and the Kalman gain, construct the state update model;

[0091] Based on the state prediction model, predict the operation state data of the previous period to obtain the predicted value of the operation state data in the current period, and update the predicted value of the operation state data in the current period based on the state update model to obtain the optimal predicted value of the operation state data in the current period, and then obtain the optimal center of gravity coordinate data;

[0092] Adjust the operation state of the rough terrain forklift based on the optimal center of gravity coordinate data and the center of gravity measurement data. Among them, at least include adjusting the optimal torque of the rough terrain forklift. Specifically: Based on the optimal center of gravity coordinate data, the center of gravity measurement data and the actual torque, construct a torque adjustment model to obtain the optimal torque of the rough terrain forklift.

[0093] Since the present invention adopts the above technical solutions, it has remarkable technical effects: The present invention provides a method, system and device for dynamically estimating the center of gravity coordinates of a rough terrain forklift and its application. The present invention obtains the center of gravity measurement data through the operation data of the rough terrain forklift, constructs a state prediction model and a state update model based on the operation data and the center of gravity measurement data, predicts and updates the operation data and the center of gravity measurement data of the previous period to obtain the optimal center of gravity coordinate data in the current period, and based on the optimal center of gravity coordinate data and the center of gravity measurement data, obtains the optimal rear wheel torque and the optimal front wheel of the rough terrain forklift, so as to adjust the torque distribution between the front and rear wheels of the rough terrain forklift, so that the rough terrain forklift can maintain good power performance and passability under various off-road conditions, greatly improving the stability and safety of the rough terrain forklift. BRIEF DESCRIPTION OF THE DRAWINGS

[0094] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0095] Figure 1 It is the overall schematic diagram of the method of the present invention;

[0096] Figure 2 It is the overall schematic diagram of the system of the present invention. Detailed implementation manners

[0097] The present invention will be further described in detail below in conjunction with embodiments. The following embodiments are explanations of the present invention, and the present invention is not limited to the following embodiments. Without conflict, the features in the following embodiments can be combined with each other.

[0098] Embodiment 1:

[0099] A method for dynamically estimating the center of gravity coordinates of an off-road forklift and its application, as Figure 1 shown, includes the following steps:

[0100] S100: Obtain the operation data of the off-road forklift in the current period. Based on the operation data, obtain the center of gravity measurement data, and form the operation state data based on the center of gravity measurement data and the operation data;

[0101] S200: Based on the operation state data, construct the state transition matrix, control input matrix, observation matrix and Kalman gain; based on the state transition matrix and the control input matrix, construct the state prediction model; based on the observation matrix and the Kalman gain, construct the state update model;

[0102] S300: Based on the state prediction model, predict the operation state data of the previous period to obtain the predicted value of the operation state data in the current period, and update the predicted value of the operation state data in the current period based on the state update model to obtain the optimal predicted value of the operation state data in the current period, and then obtain the optimal center of gravity coordinate data;

[0103] S400: Adjust the operation state of the off-road forklift based on the optimal center of gravity coordinate data and the center of gravity measurement data. Among them, at least include adjusting the optimal torque of the off-road forklift. Specifically: based on the optimal center of gravity coordinate data, the center of gravity measurement data and the actual torque, construct a torque adjustment model to obtain the optimal torque of the off-road forklift.

[0104] In S100, obtain the operation data of the off-road forklift in the current period. The operation data at least includes the total mass of the off-road forklift and the goods, the rolling resistance coefficient, the driving force, the braking force, the angular velocity and the turning radius. Based on the operation data, obtain the center of gravity measurement data, and form the operation state data based on the center of gravity measurement data and the operation data, including the following steps:

[0105] S110: Through sensors, obtain the operation data of the off-road forklift in the current period. The operation data at least includes the total mass of the off-road forklift and the goods , the rolling resistance coefficient , the driving force , the braking force , the angular velocity and the turning radius . The sensors installed on the off-road forklift can obtain the operation data of the gearbox and vehicle parameters of the off-road forklift under different off-road working conditions at any time. The sensors include, but are not limited to: position sensors (for obtaining the position of the forklift), rotational speed sensors, angular velocity sensors, acceleration sensors, driving force sensors, and center of gravity coordinate sensors (for verifying and calibrating the calculation results of dynamic center of gravity coordinates), and inertial sensors (acceleration, speed, direction, angular velocity, etc.).

[0106] S120: Obtain the measured value of the longitudinal acceleration of the center of gravity through the driving force, braking force, rolling resistance coefficient, and the total mass of the off-road forklift and the goods. Obtain the measured value of the longitudinal coordinate of the center of gravity through the measured value of the longitudinal acceleration of the center of gravity and the initial longitudinal velocity of the center of gravity;

[0107] According to Newton's second law, the relationship between the operation data and the measured value of the longitudinal acceleration of the center of gravity is expressed as follows:

[0108]

[0109] Therefore, the measured value of the longitudinal acceleration of the center of gravity is expressed as follows:

[0110]

[0111] Among them, represents the total mass of the off-road forklift and the goods, represents the measured value of the longitudinal acceleration of the center of gravity, represents the driving force, represents the braking force, represents the rolling resistance coefficient, is a dimensionless parameter, affected by various factors such as tire structure, material, air pressure, road surface conditions, and vehicle speed, and is usually obtained through experiments or by referring to relevant manuals, represents the acceleration due to gravity.

[0112] The measured value of the longitudinal coordinate of the center of gravity of the off-road forklift (the position of the center of gravity in the axis direction (longitudinal)) is expressed as follows:

[0113]

[0114] Among them, represents the measured value of the longitudinal coordinate of the center of gravity, represents the initial longitudinal coordinate of the center of gravity, represents the initial longitudinal velocity of the center of gravity, represents time. The initial longitudinal coordinate of the center of gravity and the initial longitudinal velocity of the center of gravity are both known quantities and are obtained through sensors.

[0115] S130: Based on the angular velocity and turning radius, obtain the measured value of the lateral acceleration of the center of gravity. Through the measured value of the lateral acceleration of the center of gravity and the initial lateral velocity of the center of gravity, obtain the measured value of the lateral coordinate of the center of gravity.

[0116] In the lateral direction, mainly consider the centripetal force acting on the rough terrain forklift during turning , and the centripetal force will affect the position of the center of gravity of the rough terrain forklift in the lateral direction, which is expressed as follows:

[0117]

[0118] Therefore, the measured value of the lateral acceleration of the center of gravity is expressed as follows:

[0119]

[0120] Among them, represents the centripetal force of the rough terrain forklift, represents the angular velocity of the rough terrain forklift, represents the turning radius of the rough terrain forklift, represents the measured value of the lateral acceleration of the center of gravity.

[0121] The measured value of the lateral coordinate of the center of gravity of the rough terrain forklift (the position of the center of gravity in the axis direction (lateral direction)) is expressed as follows:

[0122]

[0123] Among them, represents the measured value of the lateral coordinate of the center of gravity of the rough terrain forklift, represents the initial lateral coordinate of the center of gravity, represents the initial lateral velocity of the center of gravity. Both the initial lateral coordinate of the center of gravity and the initial lateral velocity of the center of gravity are known quantities and are obtained through sensors.

[0124] The measured value of the longitudinal coordinate of the center of gravity and the measured value of the lateral coordinate of the center of gravity are the center of gravity measurement data. The center of gravity measurement data is the center of gravity coordinate of the rough terrain forklift before adjustment, which is a value obtained based on theoretical calculation and there are still many noise interferences, and there is still a large error from the actual center of gravity coordinate. Therefore, noise reduction intervention is required, and after iterative calculation, a forklift center of gravity coordinate closer to the actual value is obtained.

[0125] In S200, the operating data at least includes rotational speed, torque, driving force, rough terrain forklift acceleration, angular velocity, step size, and center of gravity data. The operating data is obtained through sensors on the rough terrain forklift. Based on the operating state data, construct a state transition matrix, a control input matrix, an observation matrix, and a Kalman gain; based on the state transition matrix and the control input matrix, construct a state prediction model; based on the observation matrix and the Kalman gain, construct a state update model, the following steps:

[0126] S210: State Transition Matrix of Rough Terrain Forklift The motion characteristics of the rough terrain forklift and the influence of the transmission parameters need to be considered. Since the horizontal motion of the rough terrain forklift is linearly related to the rotational speed , torque and driving force , and the vertical motion is linearly related to the acceleration of the forklift , torque and driving force , therefore, the state transition matrix is expressed as follows:

[0127]

[0128] Among them, represents the state transition matrix, represents the step size, represents the rotational speed, represents the torque. It should be noted that if not specified explicitly, the torque in the present invention refers to the total output torque of the transmission. It can be approximately considered that the total output torque of the rough terrain forklift is the sum of the torques distributed to the front and rear wheels through the transmission system (before considering losses), represents the driving force, represents the acceleration of the rough terrain forklift, represents a function of the operating state data.

[0129] S220: Obtain the road surface skid resistance performance index through the center of gravity measurement data, and construct a control input matrix through the road surface skid resistance performance index and the step size. The road surface skid resistance performance index and the control input matrix are respectively expressed as follows:

[0130]

[0131]

[0132] Among them, represents the road surface skid resistance performance index. During the operation of the rough terrain forklift, the influence of the magnitude and direction of the center of gravity speed on the road surface skid resistance performance needs to be considered. When the center of gravity speed is relatively large, the stability of the rough terrain forklift is more affected, represents the actual lateral force coefficient value, = Lateral force / Tire load, represents the minimum lateral force coefficient value, which is generally determined according to the road grade and design requirements, represents the magnitude of the center of gravity speed in the center of gravity measurement data, , represents the magnitude of the longitudinal center of gravity speed in the center of gravity measurement data, represents the magnitude of the lateral velocity of the center of gravity in the center of gravity measurement data, represents the direction of the center of gravity velocity in the center of gravity measurement data, , respectively represent coefficients, represents model parameters, usually taken according to experience, and can be defined under off-road conditions the value is 2, respectively represent the coefficients of the magnitude and direction of the center of gravity velocity of the rough terrain forklift, and can be defined under off-road conditions the values are all 1, comprehensively reflects the magnitude of the frictional force provided by the road surface to the vehicle tires and the strength of the anti-skid ability under wet, water or other adverse conditions. A higher SRI value indicates that the road surface has better anti-skid performance, which can reduce the possibility of dangerous situations such as skidding and sideslipping of the vehicle during driving, thereby improving the safety and stability of driving. represents the control input matrix, and the control input matrix includes external input factors of off-road conditions, represents the transpose of the matrix.

[0133] S230: Construct a system state vector through the center of gravity data, angular velocity, rough terrain forklift acceleration, torque and driving force, and then obtain the error covariance matrix of the system state vector. The system state vector is expressed as follows:

[0134]

[0135] wherein, represents the system state vector, represents the longitudinal center of gravity coordinate in the center of gravity data, represents the lateral center of gravity coordinate in the center of gravity data, represents the longitudinal center of gravity velocity in the center of gravity data, represents the lateral center of gravity velocity in the center of gravity data, represents the angular velocity, represents the rough terrain forklift acceleration, represents the torque data, represents the driving force. It should be noted that these variables of the system state vector are all obtained through estimation.

[0136] S240: Obtain the actual angular velocity, actual rough terrain forklift acceleration and actual torque, and construct an observation vector in combination with the center of gravity measurement data. Then obtain the noise covariance matrix of the observation vector. Based on the observation vector, construct an observation matrix, wherein the observation vector is expressed as follows:

[0137]

[0138] The observation matrix is expressed as follows:

[0139]

[0140] Among them, represents the observation vector of the current time period ( time period), respectively represent the measured value of the longitudinal coordinate of the center of gravity in the current time period and the measured value of the transverse coordinate of the center of gravity in the current time period, respectively represent the longitudinal velocity of the center of gravity of the center of gravity measurement data in the current time period and the transverse velocity of the center of gravity of the center of gravity measurement data in the current time period, represents the actual angular velocity in the current time period, represents the actual acceleration of the rough terrain forklift in the current time period, represents the actual torque in the current time period. In the present invention, the center of gravity measurement data is used as the actual center of gravity data of the rough terrain forklift, and the actual angular velocity, the actual acceleration of the rough terrain forklift, and the actual torque are obtained through the sensors of the rough terrain forklift. represents the observation matrix, all represent the coefficients related to the observation vector.

[0141] S250: Construct the Kalman gain through the observation matrix, the error covariance matrix, and the noise covariance matrix. The Kalman gain is expressed as follows:

[0142]

[0143] Among them, represents the Kalman gain in the current time period. The Kalman gain can automatically adjust the weights of the prior estimate and the observation vector, and generate anti-interference ability for the measured target variable and system uncertainty. represents the current time period, represents the error covariance matrix of the prior estimate of the system state vector in the current time period, , represents the state transition matrix in the current time period, represents the error covariance matrix of the posterior estimate of the system state vector in the previous time period, represents the transpose of the matrix, represents the system noise covariance matrix in the current time period, represents the noise covariance matrix of the observation vector in the current time period.

[0144] S260: Based on the system state vector, the state transition matrix, and the control input matrix, construct a state prediction model, and then obtain the prior estimate of the system state vector. The state prediction model is expressed as follows:

[0145]

[0146] Among them, represents the prior estimate of the system state vector for the current time period. The prior estimate is the estimate of the system state vector without considering the observation vector for the current time period. represents the time period. represents the state transition matrix for the current time period. represents the posterior estimate of the system state vector for the previous time period. represents the control input matrix for the current time period. represents the control input vector for the current time period. The control input vector contains external information that affects the change of the system state, including the road surface anti-slip performance. The initial system The posterior estimate of the system state vector for the previous time period can be directly set through the physical characteristics or prior knowledge of the system.

[0147] S270: Based on the prior estimate of the system state vector, the Kalman gain, the observation vector, and the observation matrix, construct a state update model to obtain the posterior estimate of the system state vector. The state update model is expressed as follows:

[0148]

[0149] where represents the posterior estimate of the system state vector for the current time period. represents the observation vector for the current time period.

[0150] In S300, based on the state prediction model, predict the operating state data for the previous time period to obtain the predicted value of the operating state data for the current time period, and update the predicted value of the operating state data for the current time period based on the state update model to obtain the optimal predicted value of the operating state data for the current time period, and further obtain the optimal centroid coordinate data, including the following steps:

[0151] S310: Based on the posterior estimate of the system state vector for the previous time period, the state transition matrix for the current time period, and the control input matrix for the current time period, combine with the state prediction model to obtain the prior estimate of the system state vector for the current time period;

[0152] S320: Based on the error covariance matrix of the posterior estimate of the system state vector for the previous time period, obtain the error covariance matrix of the prior estimate of the system state vector for the current time period;

[0153] S330: Through the error covariance matrix of the prior estimate of the system state vector for the current time period, the observation matrix for the current time period, and the noise covariance matrix of the observation vector for the current time period, obtain the Kalman gain for the current time period, and further obtain the error covariance matrix of the posterior estimate of the system state vector for the current time period;

[0154] S340: Based on the prior estimate of the system state vector in the current time period, the Kalman gain in the current time period, the observation vector in the current time period, and the observation matrix in the current time period, combined with the state update model, obtain the posterior estimate of the system state vector in the current time period;

[0155] S350: Until the iteration termination condition is satisfied, the obtained posterior estimate of the final system state vector is the optimal system state vector, and then the optimal centroid coordinate data is obtained. The iteration termination condition includes satisfying the convergence condition or reaching a predetermined number of iterations. The convergence condition includes that the error of the posterior estimate of the system state vector is less than a preset state threshold and / or the diagonal elements of the error covariance matrix are less than a preset matrix threshold.

[0156] In S400, adjust the operating state of the rough terrain forklift based on the optimal centroid coordinate data and the centroid measurement data. Among them, at least include adjusting the optimal torque of the rough terrain forklift. Specifically: Based on the optimal centroid coordinate data, the centroid measurement data, and the actual torque, construct a torque adjustment model to obtain the optimal torque of the rough terrain forklift, including the following steps:

[0157] S410: Through sensors, respectively obtain the actual torque of the rear wheels and the actual torque of the front wheels of the rough terrain forklift.

[0158] S420: Based on the operating data, obtain an adjustment coefficient. The adjustment coefficient is obtained through the following steps:

[0159] (1) Through sensors, obtain the lateral force, tire load, steering wheel angle, and the mass of the goods carried by the rough terrain forklift.

[0160] (2) Based on the lateral force and the tire load, obtain the lateral force coefficient value. The lateral force coefficient value = lateral force / tire load. Based on the lateral force coefficient value and the centroid measurement data, obtain the road surface anti-slip performance index. The method for obtaining the road surface anti-slip performance index is the same as that in the previous text and will not be elaborated here.

[0161] (3) Based on the road surface anti-slip performance index, the steering wheel angle, and the weight of the goods carried by the rough terrain forklift, obtain the adjustment coefficient. The adjustment coefficient is expressed as follows:

[0162]

[0163] Among them, represents the adjustment coefficient, represents the road surface anti-slip performance index, represents the preset first road surface anti-slip index threshold, represents the preset second road surface anti-slip index threshold, and , represents the steering wheel angle, represents the preset angle threshold, represents the mass of the goods carried by the rough terrain forklift, represents the preset threshold of the load mass, both represent the adjustment coefficients under different conditions. Among them, the road surface anti-slip performance index the smaller it is, the larger the adjustment coefficient is, that is , , and , the relationships between the adjustment coefficients are respectively: , and . Regardless of the previous situation, when the mass of the goods carried by the rough terrain forklift exceeds the preset threshold of the load mass, the adjustment coefficient should be increased as a whole, and the relationship between the adjustment coefficients is .

[0164] When the steering wheel angle exceeds the preset angle threshold, it is necessary to comprehensively adjust the adjustment coefficient according to the specific situation, and the relationship between the adjustment coefficients is: .

[0165] According to experience, can take 0.8, can take 0.5, and the angle threshold can be set near the maximum value of the conventional angle range. For example, the angle threshold is set to 80% of the maximum value of the conventional angle , and the preset threshold of the load mass can be set to 80% of the maximum load of the rough terrain forklift . Based on this, an example of the adjustment coefficient is shown as follows:

[0166] .

[0167] S430: Based on the center of gravity measurement data, the optimal center of gravity coordinate data, the actual torque of the rear wheels, the actual torque of the front wheels and the adjustment coefficient, construct a torque adjustment model. The concept of the torque adjustment model is as follows: when the center of gravity state is that the center of gravity moves forward, increase the torque of the rear wheels to improve stability; when the center of gravity state is that the center of gravity moves backward, increase the torque of the front wheels to maintain balance; when the center of gravity state is that the center of gravity is stable, keep the torque balance of the front and rear wheels, and overall realize the dynamic distribution of torque by controlling the shifting strategies of the first to fifth clutches. Among them, the torque adjustment model includes a rear wheel torque adjustment model and a front wheel torque adjustment model, and the rear wheel torque adjustment model and the front wheel torque adjustment model are respectively expressed as follows:

[0168]

[0169]

[0170] Among them, represents the optimal rear wheel torque, represents the actual rear wheel torque, represents the optimal front-wheel torque, represents the actual front-wheel torque, represents the longitudinal coordinate of the center of gravity of the center-of-gravity measurement data, represents the lateral coordinate of the center of gravity of the center-of-gravity measurement data, represents the optimal longitudinal coordinate of the center of gravity of the optimal center-of-gravity coordinate data, represents the optimal lateral coordinate of the center of gravity of the optimal center-of-gravity coordinate data, represents the preset threshold for the forward shift of the longitudinal center of gravity, represents the preset threshold for the forward shift of the lateral center of gravity, represents the preset threshold for the backward shift of the longitudinal center of gravity, represents the preset threshold for the backward shift of the lateral center of gravity, All represent adjustment coefficients and are all positive numbers.

[0171] S440: Optimize the rear-wheel torque adjustment model and the front-wheel torque adjustment model to obtain the optimal adjustment coefficient, and then obtain the optimal rear-wheel torque and the optimal front-wheel torque of the rough-terrain forklift. The optimal adjustment coefficient can be obtained through artificial intelligence algorithms. The artificial intelligences that can be used to obtain the optimal adjustment coefficient include: machine learning algorithms such as non-linear regression, support vector machine (SVM), decision tree, and random forest; deep learning algorithms such as feedforward neural network, convolutional neural network (CNN), recurrent neural network (RNN), and deep learning frameworks; ensemble learning algorithms such as Bagging and Boosting, and reinforcement learning algorithms, etc. In this embodiment, the particle swarm algorithm is used to obtain the adjustment coefficient, which includes the following steps:

[0172] (1) Randomly generate an initial particle swarm, obtain the position and velocity of each particle in the initial particle swarm, and then obtain the initial individual optimal position of each particle and the initial global optimal position of the particle swarm, where the position of the particle is the adjustment coefficient to be solved.

[0173] (2) Based on the position of the particle, combine the torque adjustment model, obtain the torque prediction value and combine the actual torque, construct a fitness model, and based on the fitness model, obtain the fitness of the current position of each particle. The fitness model is expressed as follows:

[0174]

[0175] Wherein, represents the fitness, respectively represent the rear-wheel torque prediction value and the front-wheel torque prediction value, represents the total number of predicted rear-wheel torque and front-wheel torque.

[0176] (3) If the fitness of the particle's current position is greater than the fitness of the individual optimal position, then update the individual optimal position. If the fitness of any particle's current position is greater than the fitness of the global optimal position, then update the global optimal position, and update the position and velocity of the particle to obtain the updated particle velocity and the updated particle position. The updated particle velocity and the updated particle position are respectively expressed as follows:

[0177]

[0178]

[0179] Wherein, represents the updated particle velocity, represents the inertia weight, represents the particle velocity of the previous iteration, represents the individual optimal position of the particle of the previous iteration, represents the particle position of the previous iteration, represents the global optimal position of the previous iteration, represents the learning factor, represents a random number, represents the updated particle position.

[0180] (4) Until the termination condition is satisfied, then output the final global optimal position as the optimal adjustment coefficient. The termination condition can be reaching a preset number of iterations, or the fitness change being less than a certain threshold, or other forms.

[0181] Embodiment 2:

[0182] In this embodiment, specific data is used to further illustrate a dynamic estimation and application method for the center-of-gravity coordinates of an off-road forklift provided by the present invention:

[0183] S1: Obtain the operation data of the off-road forklift in the current period. Based on the operation data, obtain the center-of-gravity measurement data, and form the operation state data based on the center-of-gravity measurement data and the operation data. The main calculation process is as follows:

[0184] The operation parameters obtain relevant data through sensors installed on the off-road forklift. The initial data (t = 0s) is known: the total mass of the off-road forklift and the goods is , the rolling resistance coefficient of the off-road forklift is 0.05, the normal load , the driving force is , the braking force is , the angular velocity is is the turning radius For 。

[0185] Then the longitudinal force of the rough terrain forklift is:

[0186]

[0187]

[0188]

[0189] It is known that when t = 1s, , , then the measured value of the longitudinal coordinate of the center of gravity of the rough terrain forklift at this time is:

[0190]

[0191] Then the lateral force and acceleration of the rough terrain forklift are:

[0192]

[0193]

[0194] It is known that when t = 1s, , , then the measured value of the lateral coordinate of the center of gravity of the rough terrain forklift at this time is:

[0195]

[0196] Finally, the measured value of the center of gravity coordinates of the rough terrain forklift at t = 1s is (1.07 m , 1.409 m ), but there are still many noise interferences in the center of gravity coordinates obtained at this time, so denoising intervention is required.

[0197] S2: Based on the operating state data, construct the state transition matrix, control input matrix, observation matrix and Kalman gain; based on the state transition matrix and control input matrix, construct the state prediction model; based on the observation matrix and Kalman gain, construct the state update model.

[0198] Define the functional relationship in the state transition matrix as follows:

[0199]

[0200]

[0201]

[0202]

[0203]

[0204]

[0205]

[0206]

[0207]

[0208]

[0209]

[0210]

[0211]

[0212]

[0213]

[0214]

[0215] Define the coefficients in the observation matrix as follows:

[0216] .

[0217] Therefore, the noise covariance matrix of the observation vector is:

[0218]

[0219] The error covariance matrix of the initial estimated system state vector is:

[0220]

[0221] S3: Based on the state prediction model, predict the operating state data of the previous time period to obtain the predicted value of the operating state data of the current time period, and update the predicted value of the operating state data of the current time period based on the state update model to obtain the optimal predicted value of the operating state data of the current time period, and then obtain the optimal center of gravity coordinate data. Given that the center of gravity coordinate of the rough terrain forklift at the initial time period (at 0 s) is , and the center of gravity coordinate of the forklift at time 1 s is , the initial speed is , , the initial angular velocity , the initial acceleration of the rough terrain forklift , the initial torque , the initial driving force .

[0222] The torque and related parameter data output by the gearbox at different times are as follows:

[0223] When t = 1s,

[0224] When t = 2s,

[0225] When t = 3s,

[0226] The observed vector of other parameters of the off-road forklift obtained by the sensor is as follows:

[0227] t = 1 s When, ;

[0228] t = 2 s When, ;

[0229] t = 3 s When, ;

[0230] The iterative calculation steps are as follows:

[0231] At t = 1 s When, the prior estimate where Calculate the elements of the state transition matrix The following is only for demonstration, and the calculations are not shown one by one:

[0232]

[0233]

[0234] Then, at t = 1 s When, the prior estimate of the system state vector is:

[0235]

[0236] At t = 1 s When, the Kalman gain is: .

[0237] Where, , represents the known value of the system noise covariance matrix.

[0238] At t = 1 sThe posterior estimate of the system state vector at this time is:

[0239] wherein, it is calculated that t = 1 s the observation vector at .

[0240] Similarly, after t = 2 s and t = 3 s through iterative calculations, the estimated values of the center of gravity coordinates of the rough terrain forklift at different time periods can finally be obtained. For example, assume that at t = 1 s the system state vector obtained through calculation is expressed as:

[0241]

[0242] wherein, the first two elements and are the optimal center of gravity coordinate data of the rough terrain forklift at this time.

[0243] S4: Adjust the operating state of the rough terrain forklift based on the optimal center of gravity coordinate data and the center of gravity measurement data, wherein at least the optimal torque of the rough terrain forklift is adjusted, specifically: based on the optimal center of gravity coordinate data, the center of gravity measurement data and the actual torque, construct a torque adjustment model to obtain the optimal torque of the rough terrain forklift. Now, it is calculated that the road surface anti-skid performance index ( SRI ) in a certain time period is 0.3, that is, the anti-skid performance index is less than 0.5, indicating that the road surface anti-skid performance is low and the center of gravity is unstable. When the center of gravity moves forward or backward, in order to maintain the stability of the rough terrain forklift, it is necessary to adjust the torque output of the front and rear wheels.

[0244] Embodiment 3:

[0245] A dynamic estimation and application system for the center of gravity coordinates of a rough terrain forklift, as Figure 2 shown, includes a data acquisition module, a model construction module, a center of gravity estimation module and a state adjustment module;

[0246] The data acquisition module acquires the operating data of the rough terrain forklift in the current time period, based on the operating data, obtains the center of gravity measurement data, and forms the operating state data based on the center of gravity measurement data and the operating data;

[0247] The model construction module constructs a state transition matrix, a control input matrix, an observation matrix and a Kalman gain based on the operating state data; constructs a state prediction model based on the state transition matrix and the control input matrix; constructs a state update model based on the observation matrix and the Kalman gain;

[0248] The center-of-gravity estimation module predicts the operating state data of the previous period based on the state prediction model to obtain the predicted value of the operating state data of the current period, and updates the predicted value of the operating state data of the current period based on the state update model to obtain the optimal predicted value of the operating state data of the current period, and further obtains the optimal center-of-gravity coordinate data;

[0249] The state adjustment module adjusts the operating state of the rough-terrain forklift based on the optimal center-of-gravity coordinate data and the center-of-gravity measurement data. At least, it includes adjusting the optimal torque of the rough-terrain forklift. Specifically, based on the optimal center-of-gravity coordinate data, the center-of-gravity measurement data, and the actual torque, a torque adjustment model is constructed to obtain the optimal torque of the rough-terrain forklift.

[0250] All changes and variations made without departing from the spirit and scope of the present invention, and all equivalent technical solutions also fall within the scope of the present invention.

[0251] Each embodiment in this specification is described in a progressive manner. The key point of each embodiment is to illustrate the differences from other embodiments. For the same or similar parts among the embodiments, reference can be made to each other.

[0252] Those skilled in the art should understand that the embodiments of the present invention can be provided as methods, devices, or computer program products. Therefore, the present invention can be implemented in the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can be implemented in the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0253] The present invention is described with reference to the flowcharts and / or block diagrams of methods, terminal devices (systems), and computer program products according to the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, and the combination of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to the processors of general-purpose computers, special-purpose computers, embedded processors, or other programmable data processing terminal devices to generate a machine, so that the instructions executed by the processors of the computer or other programmable data processing terminal devices generate means for implementing the functions specified in one process Figure 1 one process or multiple processes and / or blocks Figure 1 or multiple blocks to specify the functions of the device.

[0254] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing terminal device to work in a specific manner, such that the instructions stored in the computer-readable memory produce a manufactured article including an instruction device that implements the functions specified in one process Figure 1 or more processes and / or blocks Figure 1 or more blocks.

[0255] These computer program instructions can also be loaded onto a computer or other programmable data processing terminal device, such that a series of operational steps are performed on the computer or other programmable terminal device to produce a computer-implemented process, and thus the instructions executed on the computer or other programmable terminal device provide steps for implementing the functions specified in one process Figure 1 or more processes and / or blocks Figure 1 or more blocks.

[0256] It should be noted that:

[0257] As used in the specification, the phrase "one embodiment" or "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one embodiment of the present invention. Thus, the phrases "one embodiment" or "an embodiment" that appear throughout the specification do not necessarily all refer to the same embodiment.

[0258] In addition, it should be noted that for the specific embodiments described in this specification, the shapes, names, etc. of the components may be different. Any equivalent or simple changes made according to the structure, features, and principles described in the inventive concept of the present invention are included within the protection scope of the present invention. Those skilled in the art to which the present invention pertains can make various modifications or supplements to the specific embodiments described or use similar means for substitution, as long as they do not deviate from the structure of the present invention or exceed the scope defined by the claims, and they should all fall within the protection scope of the present invention.

Claims

1. A method for dynamic estimation of center of gravity coordinates of an off-road forklift and its application, characterized in that: The following steps are involved: Obtaining the operating data of the off-road forklift in the current period, obtaining the center of gravity measurement data based on the operating data, and forming the operating status data based on the center of gravity measurement data and the operating data; Based on the operating status data, the state transfer matrix, control input matrix, observation matrix and Kalman gain are constructed; based on the state transfer matrix and control input matrix, the state prediction model is constructed; Based on the observation matrix and Kalman gain, a state update model is constructed; Based on the state prediction model, the running state data of the previous period is predicted to obtain the predicted value of the running state data of the current period, and the predicted value of the running state data of the current period is updated based on the state update model to obtain the optimal predicted value of the running state data of the current period, and then the optimal center of gravity coordinate data is obtained; The operating state of the off-road forklift is adjusted based on the optimal center of gravity coordinate data and the center of gravity measurement data, wherein at least the optimal torque of the off-road forklift is adjusted, specifically: based on the optimal center of gravity coordinate data, the center of gravity measurement data and the actual torque, a torque adjustment model is constructed to obtain the optimal torque of the off-road forklift; The method of constructing a torque adjustment model based on the optimal center of gravity coordinate data, the center of gravity measurement data and the actual torque to obtain the optimal torque of the off-road forklift includes the following steps: The actual torque of the rear wheels and the actual torque of the front wheels of the off-road forklift are obtained respectively; Based on the operating data, an adjustment coefficient is obtained; Based on the center of gravity measurement data, the optimal center of gravity coordinate data, the actual torque of the rear wheel, the actual torque of the front wheel and the adjustment coefficient, a torque adjustment model is constructed, wherein the torque adjustment model includes a rear wheel torque adjustment model and a front wheel torque adjustment model; The rear wheel torque adjustment model and the front wheel torque adjustment model are optimized to obtain the optimal adjustment coefficient, and then the optimal rear wheel torque and the optimal front wheel torque of the off-road forklift are obtained; The rear wheel torque adjustment model is expressed as follows: The front wheel torque adjustment model is expressed as follows: in, represents the optimal rear wheel torque, Indicates the actual rear wheel torque, represents the optimal front wheel torque, Indicates the actual front wheel torque, The longitudinal coordinate of the center of gravity representing the center of gravity measurement data, The horizontal coordinate of the center of gravity representing the center of gravity measurement data, The optimal center of gravity longitudinal coordinate representing the optimal center of gravity coordinate data, The optimal center of gravity lateral coordinate representing the optimal center of gravity coordinate data, Indicates the preset longitudinal forward displacement threshold of the center of gravity. Indicates the preset lateral forward displacement threshold of the center of gravity. Indicates the preset longitudinal rearward shift threshold of the center of gravity. Indicates the preset lateral rearward shift threshold of the center of gravity. Both represent adjustment coefficients and are positive numbers.

2. The method for dynamic estimation and application of center of gravity coordinates of an off-road forklift according to claim 1, characterized in that: The operation data at least includes the total mass of the off-road forklift and the cargo, the rolling resistance coefficient, the driving force, the braking force, the angular velocity and the turning radius. The obtaining of the center of gravity measurement data based on the operation data includes the following steps: The longitudinal acceleration measurement value of the center of gravity is obtained through the driving force, braking force, rolling resistance coefficient and the total mass of the off-road forklift and the cargo. The longitudinal coordinate measurement value of the center of gravity is obtained through the longitudinal acceleration measurement value of the center of gravity and the longitudinal initial velocity of the center of gravity. Based on the angular velocity and the turning radius, the lateral acceleration measurement value of the center of gravity is obtained, and the lateral coordinate measurement value of the center of gravity is obtained through the lateral acceleration measurement value of the center of gravity and the lateral initial velocity of the center of gravity, wherein the longitudinal coordinate measurement value of the center of gravity and the lateral coordinate measurement value of the center of gravity are the center of gravity measurement data; The longitudinal acceleration measurement value of the center of gravity and the lateral acceleration measurement value of the center of gravity are respectively expressed as follows: The longitudinal coordinate measurement value of the center of gravity and the transverse coordinate measurement value of the center of gravity are respectively expressed as follows: in, represents the longitudinal acceleration measurement of the center of gravity, represents the measured value of the lateral acceleration of the center of gravity, Indicates driving force, Indicates the braking force, is the rolling resistance coefficient, Indicates the total mass of the off-road forklift and the cargo. represents the acceleration due to gravity, represents the angular velocity, Indicates the turning radius, represents the longitudinal coordinate measurement value of the center of gravity, represents the initial longitudinal coordinate of the center of gravity, represents the initial longitudinal velocity of the center of gravity, Indicates time, represents the measured value of the lateral coordinate of the center of gravity, represents the initial lateral coordinate of the center of gravity, represents the initial lateral velocity of the center of gravity.

3. The method for dynamic estimation and application of center of gravity coordinates of an off-road forklift according to claim 1, characterized in that: The operation data at least includes speed, torque, driving force, off-road forklift acceleration, angular velocity, step length and center of gravity data; based on the operation state data, a state transfer matrix, a control input matrix, an observation matrix and a Kalman gain are constructed; based on the state transfer matrix and the control input matrix, a state prediction model is constructed; Based on the observation matrix and Kalman gain, a state update model is constructed, which includes the following steps: The state transfer matrix is ​​constructed through the speed, torque, driving force and acceleration of the off-road forklift; Obtain the road surface anti-skid performance index, and construct a control input matrix through the road surface anti-skid performance index and step length; The system state vector is constructed through the center of gravity data, angular velocity, off-road forklift acceleration, torque and driving force, and then the error covariance matrix of the system state vector is obtained; The actual angular velocity, the actual acceleration of the off-road forklift, and the actual torque are obtained and combined with the center of gravity measurement data to construct an observation vector, and then the noise covariance matrix of the observation vector is obtained. Based on the observation vector, an observation matrix is ​​constructed; Construct the Kalman gain through the observation matrix, error covariance matrix and noise covariance matrix; Based on the system state vector, state transfer matrix and control input matrix, a state prediction model is constructed to obtain a priori estimation of the system state vector; Based on the prior estimate of the system state vector, Kalman gain, observation vector and observation matrix, a state update model is constructed to obtain the posterior estimate of the system state vector.

4. The method for dynamic estimation and application of center of gravity coordinates of an off-road forklift according to claim 3, characterized in that: The state transfer matrix is ​​expressed as follows: The road surface anti-skid performance index and the control input matrix are respectively expressed as follows: The observation matrix is ​​expressed as follows: The Kalman gain is expressed as follows: The state prediction model is expressed as follows: The state update model is expressed as follows: in, represents the state transfer matrix, represents the step length, Indicates the speed, Indicates torque, Indicates driving force, Indicates the acceleration of the off-road forklift, Represents a function about the running status data, Represents the road skid resistance index. Indicates the actual lateral force coefficient value, represents the minimum lateral force coefficient value, Indicates the magnitude of the center of gravity velocity in the center of gravity measurement data, , Indicates the magnitude of the longitudinal velocity of the center of gravity in the center of gravity measurement data, Indicates the magnitude of the lateral velocity of the center of gravity in the center of gravity measurement data. Indicates the direction of the center of gravity velocity in the center of gravity measurement data, , Respectively represent the coefficients, represents the control input matrix, represents the observation matrix, Both represent coefficients related to the observation vector, represents the Kalman gain of the current period, Indicates the current period. represents the error covariance matrix of the prior estimate of the system state vector in the current period, , represents the state transfer matrix of the current period, represents the error covariance matrix of the posterior estimate of the system state vector in the previous period, represents the transpose of a matrix, represents the system noise covariance matrix of the current period, represents the noise covariance matrix of the observation vector in the current period, represents the prior estimate of the system state vector in the current period, represents the posterior estimate of the system state vector in the previous period, Represents the control input matrix of the current period, represents the control input vector for the current period, represents the a posteriori estimate of the system state vector for the current period, Represents the observation vector of the current period.

5. The method for dynamic estimation of center of gravity coordinates of an off-road forklift according to claim 1, characterized in that: The method predicts the running state data of the previous period based on the state prediction model to obtain the predicted value of the running state data of the current period, and updates the predicted value of the running state data of the current period based on the state update model to obtain the optimal predicted value of the running state data of the current period, and then obtains the optimal center of gravity coordinate data, including the following steps: Based on the a posteriori estimate of the system state vector in the previous period, the state transfer matrix in the current period and the control input matrix in the current period, combined with the state prediction model, the a priori estimate of the system state vector in the current period is obtained; Based on the error covariance matrix of the posterior estimation of the system state vector in the previous period, the error covariance matrix of the a priori estimation of the system state vector in the current period is obtained; The Kalman gain of the current period is obtained through the error covariance matrix of the prior estimation of the system state vector of the current period, the observation matrix of the current period, and the noise covariance matrix of the observation vector of the current period, and then the error covariance matrix of the posterior estimation of the system state vector of the current period is obtained; Based on the prior estimate of the system state vector of the current period, the Kalman gain of the current period, the observation vector of the current period and the observation matrix of the current period, combined with the state update model, the posterior estimate of the system state vector of the current period is obtained; Until the iteration termination condition is met, the final a posteriori estimate of the system state vector is obtained, which is the optimal system state vector, and then the optimal center of gravity coordinate data is obtained.

6. The method for dynamic estimation and application of center of gravity coordinates of an off-road forklift according to claim 1, characterized in that: The adjustment coefficient is obtained by the following steps: Through sensors, the lateral force, tire load, steering wheel angle and the mass of the cargo carried by the off-road forklift are obtained; Based on the lateral force and the tire load, a lateral force coefficient value is obtained, and based on the lateral force coefficient value and the center of gravity measurement data, a road surface anti-skid performance index is obtained, wherein the lateral force coefficient value = lateral force / tire load; Based on the road surface anti-skid performance index, the steering wheel angle and the weight of the cargo carried by the off-road forklift, an adjustment coefficient is obtained; Wherein, the adjustment coefficient is expressed as follows: in, Represents the road skid resistance index. represents the preset first road surface anti-skid index threshold, represents a preset second road surface anti-skid index threshold, and , Indicates the steering wheel angle, Indicates the preset turning angle threshold. Indicates the mass of cargo carried by the off-road forklift. Indicates the preset load mass threshold, They all represent the adjustment coefficients under different conditions.

7. The method for dynamic estimation of center of gravity coordinates of an off-road forklift according to claim 1, characterized in that: The optimal adjustment coefficient is obtained by the following steps: Randomly generate an initialized particle swarm, obtain the position and velocity of each particle in the initialized particle swarm, and then obtain the initial individual optimal position of each particle and the initial global optimal position of the particle swarm, where the position of the particle is the adjustment coefficient to be solved; Based on the particle position, combined with the torque adjustment model, the torque prediction value is obtained and combined with the actual torque to build a fitness model. Based on the fitness model, the fitness of each particle's current position is obtained; If the fitness of the current position of the particle is greater than the fitness of the individual optimal position, the individual optimal position is updated. If the fitness of the current position of any particle is greater than the fitness of the global optimal position, the global optimal position is updated, and the position and speed of the particle are updated to obtain the updated particle speed and updated particle position. Until the termination condition is met, the final global optimal position is output as the optimal adjustment coefficient; Wherein, the fitness model is expressed as follows: The updated particle velocity is expressed as follows: The updated particle position is expressed as follows: in, represents fitness, Respectively represent the rear wheel torque prediction value and the front wheel torque prediction value, represents the total amount of predicted rear wheel torque and front wheel torque, represents the updated particle velocity, represents the inertia weight, represents the particle velocity of the last iteration, represents the optimal position of the individual particle in the last iteration, represents the particle position of the last iteration, represents the global optimal position of the last iteration, represents the learning factor, Represents a random number, Indicates the updated particle position.

8. A dynamic estimation and application system of the center of gravity coordinates of an off-road forklift, characterized in that: It includes a data acquisition module, a model building module, a center of gravity estimation module and a state adjustment module; The data acquisition module acquires the operating data of the off-road forklift in the current period, obtains the center of gravity measurement data based on the operating data, and forms the operating status data based on the center of gravity measurement data and the operating data; The model building module builds a state transfer matrix, a control input matrix, an observation matrix and a Kalman gain based on the operating state data; and builds a state prediction model based on the state transfer matrix and the control input matrix; Based on the observation matrix and Kalman gain, a state update model is constructed; The center of gravity estimation module predicts the running state data of the previous period based on the state prediction model to obtain the predicted value of the running state data of the current period, and updates the predicted value of the running state data of the current period based on the state update model to obtain the optimal predicted value of the running state data of the current period, and then obtains the optimal center of gravity coordinate data; The state adjustment module adjusts the operating state of the off-road forklift based on the optimal center of gravity coordinate data and the center of gravity measurement data, which at least includes adjusting the optimal torque of the off-road forklift, specifically: based on the optimal center of gravity coordinate data, the center of gravity measurement data and the actual torque, a torque adjustment model is constructed to obtain the optimal torque of the off-road forklift; The method of constructing a torque adjustment model based on the optimal center of gravity coordinate data, the center of gravity measurement data and the actual torque to obtain the optimal torque of the off-road forklift includes the following steps: The actual torque of the rear wheels and the actual torque of the front wheels of the off-road forklift are obtained respectively; Based on the operating data, an adjustment coefficient is obtained; Based on the center of gravity measurement data, the optimal center of gravity coordinate data, the actual torque of the rear wheel, the actual torque of the front wheel and the adjustment coefficient, a torque adjustment model is constructed, wherein the torque adjustment model includes a rear wheel torque adjustment model and a front wheel torque adjustment model; The rear wheel torque adjustment model and the front wheel torque adjustment model are optimized to obtain the optimal adjustment coefficient, and then the optimal rear wheel torque and the optimal front wheel torque of the off-road forklift are obtained; The rear wheel torque adjustment model is expressed as follows: The front wheel torque adjustment model is expressed as follows: in, represents the optimal rear wheel torque, Indicates the actual rear wheel torque, represents the optimal front wheel torque, Indicates the actual front wheel torque, The longitudinal coordinate of the center of gravity representing the center of gravity measurement data, The horizontal coordinate of the center of gravity representing the center of gravity measurement data, The optimal center of gravity longitudinal coordinate representing the optimal center of gravity coordinate data, The optimal center of gravity lateral coordinate representing the optimal center of gravity coordinate data, Indicates the preset longitudinal forward displacement threshold of the center of gravity. Indicates the preset lateral forward displacement threshold of the center of gravity. Indicates the preset longitudinal rearward shift threshold of the center of gravity. Indicates the preset lateral rearward shift threshold of the center of gravity. Both represent adjustment coefficients and are positive numbers.

9. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

10. A device for dynamically estimating the center of gravity coordinates of an off-road forklift and its application, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the computer program, the method according to any one of claims 1 to 7 is implemented.

Images