AEB control method for semi-trailer truck based on load state adaptation and stability correction

Abstract

The present application belongs to the field of vehicle brake control system, and relates to a semi-trailer AEB control method based on load state self-adaptation and stability correction. The method identifies semi-trailer mass and center of mass parameters, calculates axle load distribution under braking conditions, combines load state self-adaptation to correct expected braking deceleration within a given braking level, further combines dynamic axle load, kingpin vertical load and wheel slip rate to correct braking force distribution, obtains final ideal braking torque, and finally generates braking control instructions corresponding to the tractor and semi-trailer according to the determined final ideal braking torque of each axle, executes braking, and performs closed-loop correction on the braking control instructions during braking execution, thereby realizing safe, efficient and stable control during the automatic emergency braking process of the semi-trailer truck.

Description

Technical Field

[0001] This invention belongs to the field of vehicle braking control systems, and relates to automatic emergency braking (AEB) control for commercial vehicles, specifically to an AEB control method for semi-trailer trucks based on load state adaptation and stability correction. Background Technology

[0002] With the continuous development of my country's highway freight system, semi-trailer trucks (semi-trailer truck-trailers) have become an important carrier for medium- and long-distance trunk line transportation and bulk cargo transportation. Their operational safety is directly related to road traffic safety and freight efficiency. Compared with passenger cars, semi-trailer truck-trailers have significant differences in dynamic characteristics, braking performance characteristics, and load variation patterns. In the event of a rear-end collision or other accident, the consequences are often more severe than with ordinary passenger cars. Especially during actual transportation, vehicles often operate under conditions of being empty, fully loaded, partially loaded, and with varying cargo distributions, causing significant changes in the vehicle's mass, center of gravity, axle load distribution, and vertical load on the drawbar. Furthermore, semi-trailer truck-trailers generally use pneumatic braking systems, which, compared to the hydraulic braking systems commonly used in passenger cars, exhibit significant response lag, slower pressure build-up, and are more susceptible to the influence of loading conditions during operation. When AEB control strategies designed for passenger cars or commercial vehicles that consider load variations are still used, the braking intensity may not match the actual needs, and there may be insufficient coordination of braking forces between the tractor and trailer. In severe cases, this may even induce instability risks such as excessive use of trailer side adhesion, vehicle folding, or fishtailing. Therefore, it is necessary to study more targeted AEB control methods that take into account the structural characteristics, load variation features, and pneumatic braking performance of semi-trailer trucks.

[0003] Existing research focuses on two main areas: one is commercial vehicle braking control or vehicle stability control itself, mainly on improving braking performance, maintaining directional stability during braking, and analyzing the instability mechanism of articulated vehicles, with particular attention to the longitudinal, lateral, and yaw response characteristics of vehicles under braking conditions; the other is on risk assessment, warning triggering, and graded intervention of automatic emergency braking systems, with a focus on improving the timeliness of collision recognition and the effectiveness of braking triggering.

[0004] Regarding automatic emergency braking (AEB) decision-making, existing technologies have begun to incorporate changes in vehicle status or external environmental factors into the adjustment of AEB control parameters. For example, Chinese patent CN119329508A adjusts autonomous emergency braking based on changes in vehicle mass; CN120270236A incorporates road surface factors into the graded AEB control decision-making of electric semi-trailers. Additionally, CN120382873A adjusts the braking acceleration of the AEB system by obtaining the vehicle's center of gravity position to consider yaw stability during braking. These methods improve AEB braking decision-making from the perspective of vehicle stability, but they primarily focus on the single parameter of center of gravity position, and the specific identification mechanism and modeling process for obtaining the center of gravity position are not further disclosed. Overall, these methods mainly modify AEB decisions based on single factors such as mass changes, road conditions, or center of gravity position, and their comprehensive utilization of multiple load parameters under complex loading conditions of semi-trailer trucks remains relatively limited.

[0005] In terms of vehicle state recognition and parameter estimation, accurate acquisition of vehicle operating status is a crucial foundation for achieving active safety control and assisted driving functions. Due to limitations such as the high cost of deploying onboard sensors and the difficulty in directly measuring certain critical states, relevant parameters typically require online acquisition using estimation algorithms. Existing technologies include research on semi-trailer train mass estimation or detection. For example, Chinese patent CN117885747A primarily focuses on real-time estimation of the total mass of the tractor and semi-trailer combination, emphasizing the acquisition of the combined vehicle mass value itself. CN119023044A detects the semi-trailer mass based on the vertical load and acceleration of the semi-trailer via the tractor seat, further estimating the tractor and overall vehicle mass; its focus is also on the detection and estimation of semi-trailer and overall vehicle mass. In contrast, existing research largely remains at the level of semi-trailer or overall vehicle mass detection and estimation, with relatively insufficient research on further solving for the longitudinal center of gravity position and height of the semi-trailer, as well as calculating axle load distribution and vertical load of the traction pin under braking conditions.

[0006] In the area of ​​braking coordination and stability control of semi-trailer truck trains, existing research has focused on maintaining vehicle combination stability and distributing braking force. For example, Chinese patent CN120363899A mainly achieves active stability control of the tractor-trailer under all working conditions by solving for the distribution of driving or braking torque through additional yaw moment; CN111824091A mainly reduces the risk of folding and fishtailing in multi-axle articulated vehicles by adjusting the braking sequence of each axle. Overall, these methods focus more on vehicle combination stability control or braking force distribution itself, mainly addressing issues such as yaw stability, attitude control, and braking coordination. They rarely incorporate dynamic axle loads, vertical loads of the traction pins, the relationship between the axle loads of the tractor and trailer, and wheel slippage states to coordinate and correct the braking force of each axle. Therefore, during AEB emergency braking of semi-trailer truck trains, there may still be problems with unreasonable braking force distribution and insufficient vehicle combination stability.

[0007] In the field of air pressure braking execution and control for commercial vehicles, existing technologies have addressed issues such as response lag, pressure build-up lag, and wheel slip control in air braking systems. For example, Chinese patent CN109501790A proposed a method for joint control of trailer brake pressure delay compensation and wheel slip ratio. By combining brake pressure delay compensation with ABS slip ratio control, it aims to reduce trailer brake response time and improve braking stability. This type of method plays a positive role in improving the performance of air pressure braking response and slip control at the lower level of commercial vehicles. However, its research focus is mainly on brake pressure compensation and slip ratio control itself, with insufficient consideration of the overall coordination between the upper-level AEB triggering logic, emergency braking requirements, and the lower-level air pressure execution process. Furthermore, the unified closed-loop correction of the pressurization, pressure holding, and depressurization processes, as well as the system linkage between air pressure lag compensation, pressure build-up constraints, and target braking force tracking, are still inadequate. Therefore, in the AEB execution layer of commercial vehicles, problems such as target braking force execution lag, insufficient pressure tracking accuracy, and increased risk of wheel lock-up due to excessive pressure build-up may still exist.

[0008] In summary, while existing research on AEB (Automatic Emergency Braking) for semi-trailer trucks has made some progress in braking decision-making, vehicle state recognition, braking coordination and stability control, and commercial vehicle pneumatic braking execution, the following shortcomings still exist: First, the loading state of the semi-trailer truck and the resulting changes in key parameters such as mass, center of gravity position, and axle load distribution are not fully utilized, resulting in insufficient matching between AEB braking decisions and actual braking capabilities; second, during emergency braking, the linkage between braking force distribution and vehicle combination stability control is not close enough, and the impact of dynamic axle load, traction pin vertical load, and wheel slippage state on braking coordination is not fully considered; third, at the commercial vehicle pneumatic braking execution level, there is a lack of effective connection between the upper-level AEB braking demand and the lower-level pneumatic execution process, making it difficult to balance response timeliness, pressure tracking accuracy, and wheel slippage constraints. Summary of the Invention

[0009] In view of the shortcomings and deficiencies of the existing technology, the purpose of this invention is to provide a semi-trailer truck AEB control method based on load state adaptation and stability correction. By identifying the mass and center of gravity parameters of the semi-trailer, calculating the axle load distribution under braking conditions, and combining load state adaptive correction within a predetermined braking level, the desired deceleration is further corrected. Furthermore, the braking force distribution is corrected for stability by combining dynamic axle load, traction pin vertical load, and wheel slip ratio. Finally, through valve control execution and closed-loop feedback correction for commercial vehicle pneumatic braking systems, safe, efficient, and stable control of semi-trailer trucks (semi-trailer truck trains) during automatic emergency braking is achieved.

[0010] AEB control method for semi-trailer trucks based on load state adaptation and stability correction includes the following steps: Step S1: Obtain loading information, vehicle measurement data and known vehicle parameters, identify the semi-trailer mass, longitudinal center of gravity position and center of gravity height, and calculate the dynamic axle load of the front and rear axles of the tractor, the dynamic axle load of the semi-trailer axle group and the vertical load of the traction pin under braking conditions. Step S2: Establish an AEB control strategy, assess the collision risk, and combine the current vehicle load status and axle load distribution obtained in Step S1 to calculate the expected braking deceleration under the current braking level given by the AEB. Step S3: Based on the desired braking deceleration output by AEB in Step S2, and combined with the dynamic axle loads of the front and rear axles of the tractor and the dynamic axle loads of the semi-trailer axle group in Step S1, calculate the ideal braking force distribution of the front axle of the tractor, the rear axle of the tractor, and the semi-trailer axle group; at the same time, combine the axle load ratio of the tractor and the semi-trailer and the vertical load of the traction pin to perform stability correction on the ideal braking torque, and then combine the wheel slip ratio feedback of each axle to perform end correction on the ideal braking torque after stability correction to obtain the final ideal braking torque; Step S4: Based on the final ideal braking torque of each axis output in step S3, determine the target braking air pressure for each braking circuit, and perform advance compensation and rate constraint processing on the target braking air pressure to obtain the reference braking air pressure used for the underlying valve control execution. Then, based on the actual braking air pressure fed back by the pressure sensor, a pressure tracking error is constructed, and the duration of pressurization, depressurization, or pressure holding within the current control cycle is adaptively adjusted according to the magnitude of the pressure tracking error.

[0011] As a preferred embodiment of the present invention, the semi-trailer truck in step S1 consists of a tractor and a semi-trailer, and the semi-trailer is provided with three axles; the loading information is prior information on the loading distribution of the semi-trailer, including cargo loading area information and cargo loading offset direction information; the vehicle measurement data includes the vehicle longitudinal acceleration, the longitudinal contact force between the tires of the front and rear axles of the tractor and the road surface, and the longitudinal contact force between the tires of each axle of the semi-trailer and the road surface.

[0012] In a preferred embodiment of the present invention, step S1 inputs the acquired data into a longitudinal dynamics model based on straight-line braking conditions. First, the vertical tire forces of each wheel of the tractor and semi-trailer are calculated to obtain the vertical loads on the front and rear axles of the tractor and the vertical loads on each axle of the semi-trailer. Then, based on the vertical balance relationship in the overall vehicle longitudinal dynamics model, the mass of the semi-trailer is obtained. The expression is: ; in, , This represents the vertical load on the front and rear axles of the tractor unit. This represents the vertical load on each axle of the semi-trailer. It is the acceleration due to gravity. For the weight of the tractor; Subsequently, based on the pitch moment balance of the tractor vehicle around the ground projection point of the tractor pin, the longitudinal force of the tractor pin under the longitudinal acceleration excitation condition is calculated. ; Finally, based on the pitch moment balance of the semi-trailer around the ground projection point of the traction pin, the position of the semi-trailer's longitudinal center of gravity is established. and center of mass height The equation is expressed as: ; in, , and These represent the longitudinal distances from each axle of the semi-trailer to the ground projection point of the draw pin. This represents the longitudinal acceleration of the entire vehicle. The height of the traction pin; Then, data points under at least two different longitudinal acceleration conditions are selected to establish a multi-time semi-trailer pitching moment balance equation; finally, the front and rear distributed offsets are introduced. and uniform distribution deviation coefficient We construct prior constraints for the load distribution and solve them jointly using the weighted least squares method. and .

[0013] As a preferred embodiment of the present invention, in step S1, after an obstacle is detected ahead and the automatic emergency braking triggering condition is met, based on the identified semi-trailer mass and center of gravity parameters, and combined with the structural parameters of the tractor-semi-trailer combination, a quasi-static axle load transfer model of the tractor-semi-trailer combination under braking conditions is established, and the first... The dynamic axle load of the front axle of the tractor, the dynamic axle load of the rear axle of the tractor, the dynamic axle load of the semi-trailer axle group, and the vertical load at the towing pin at each sampling time.

[0014] As a preferred embodiment of the present invention, collision time TTC is selected as the collision risk indicator in step S2. Identify AEB control status based on collision risk indicators. This will output the basic AEB braking level. When AEB control status This indicates a safe state, and no AEB intervention will be performed. This indicates a Level 1 warning status, where only a collision warning signal is issued, and braking is not implemented. This indicates the first-level braking state, at which point the basic AEB braking level is reached. Basic braking deceleration is taken ; This indicates a level 2 braking condition, at which point the basic AEB braking level is... Basic braking deceleration is taken ; Then, regarding the basic AEB braking level and basic AEB braking level Two multi-input single-output adaptive neural fuzzy inference sub-models are constructed; among them, Used to solve for the deceleration correction under the first braking level. This is used to solve for the deceleration correction at the second braking level; then, based on the obtained deceleration correction and the base braking deceleration, the first braking deceleration is calculated. Candidate expected decelerations at each level at each time point; finally, boundary corrections are applied to the candidate expected decelerations to obtain the expected braking decelerations for first-level and second-level braking after boundary correction. and .

[0015] As a preferred embodiment of the present invention, in step S3, the feedforward desired braking force is first calculated based on the total mass of the semi-trailer truck and the desired braking deceleration output in step S2. Then, based on the error between the desired braking deceleration and the actual braking deceleration, feedback correction is performed, and the feedback desired braking force is calculated. Thus, the first The expected generalized braking force of the tractor and semi-trailer combination at that moment. Then, based on the expected generalized braking force and the dynamic axle load of the tractor's front axle... Dynamic axle load of the tractor rear axle and the dynamic axle load of the semi-trailer axle assembly Determine the target braking intensity This allows for the acquisition of ideal braking forces for the tractor's front axle, rear axle, and semi-trailer axle assembly. The ideal braking torque of the tractor front axle, tractor rear axle, and semi-trailer axle assembly. ; Then, stability correction is applied to the obtained ideal braking torque to obtain the first... Ideal braking torque of the tractor front axle, tractor rear axle, and semi-trailer axle assembly after stability correction at a given moment.

[0016] As a preferred embodiment of the present invention, in step S4... Target brake air pressure for each braking circuit The expression is: ; Among them, the The effective working area of ​​the brake chamber in each braking circuit is The return threshold pressure is Adjust the arm length to The equivalent radius of action of S-cam is The inner radius of the brake drum is Braking factor is , These represent the front axle of the tractor unit, the rear axle of the tractor unit, and the axle assembly of the semi-trailer, respectively. The final ideal braking torque is... ; The method for performing advance compensation and rate constraint processing on the target braking air pressure is as follows: ; ; Among them, the The equivalent hysteresis time constant of each braking circuit is The maximum pressure build-up rate and the maximum pressure relief rate are respectively and The sampling period is , No. The pressure of the hysteresis compensation target at each moment , No. The maximum allowable braking air pressure for each braking circuit is: , This is a limiting function, and the reference braking air pressure used for the underlying valve control execution is... .

[0017] As a preferred embodiment of the present invention, in step S4, the pressure tracking error of each braking circuit is constructed based on the actual braking air pressure and the reference braking air pressure. And calculate the first Valve control quantity of each braking circuit The expression is: ; The control cycle is as follows: The two error thresholds are respectively and The piecewise coefficient is , , and ; when At that time, the first The braking circuit is in a pressurized state; when It is in a pressure-holding state; when It is in a state of decompression; and with As the duration of pressurization or depressurization within the current control cycle, This refers to the duration of pressure holding.

[0018] As a further preferred embodiment of the present invention, the input vector of the adaptive neural fuzzy inference sub-model in step S2 is: ;in, For the speed of a semi-trailer truck, For the longitudinal acceleration of the semi-trailer truck, For the first The percentage of front axle load on the tractor at any given moment. For the first At any given moment, the axle load ratio between the tractor and the semi-trailer is... This indicates that the transpose is valid; Each input variable is fuzzy partitioned using the GB membership function. The membership degrees of the four input variables at time t are as follows: Based on this calculation, the first Under the first braking level Rule number 1 Trigger intensity at each moment The expression is: ; After normalization, the normalized rule weights are obtained. And calculate the first The deceleration correction amount for each braking level is expressed as follows: ; in, For the first The deceleration correction amount output by the adaptive neural fuzzy inference sub-model at each braking level. For the first The linear parameters of the consequent of a fuzzy rule.

[0019] As a further preferred embodiment of the present invention, when performing stability correction on the ideal braking torque in step S3, it is first based on the first... The stability correction factor is calculated based on the vertical load of the lead pin at each moment, the axle load ratio of the tractor and the semi-trailer, and the vertical load of the lead pin and the axle load ratio of the tractor and the semi-trailer under stable operating conditions. Then, based on the stability correction coefficient, the ideal braking torque of the semi-trailer axle assembly is corrected for stability, and the difference in braking torque is obtained from the result. Then the braking torque difference will be calculated according to the first... At any given moment, the ratio of dynamic axle load on the front axle to dynamic axle load on the rear axle of the tractor is redistributed to the front and rear axles of the tractor, resulting in the ideal braking torque of the front and rear axles of the tractor after stability correction.

[0020] As a further preferred embodiment of the present invention, when performing end-point correction on the ideal braking torque after stability correction in step S3, the first step is to calculate the... The average slip ratio of the corresponding wheels of the front axle, rear axle, and semi-trailer axle groups of the tractor at a given time is used to determine the difference in slip ratio between the rear axle and the front axle of the tractor. The difference in slip ratio between the semi-trailer axle assembly and the tractor front axle Based on the slip ratio difference, the slip ratio feedback correction torques for the tractor rear axle and semi-trailer axle groups are constructed as follows: ; ； in, , and These are the proportional coefficient, integral coefficient, and derivative coefficient for the rear axle slip ratio feedback control of the tractor vehicle; , and These are the proportional coefficient, integral coefficient, and derivative coefficient for the semi-trailer axle group slip ratio feedback control; Based on the slip ratio feedback correction torque of the tractor rear axle and semi-trailer axle assembly, the first... The final ideal braking torque of the tractor's rear axle and semi-trailer axle assembly at a given moment is then compensated to the tractor's front axle to obtain the final ideal braking torque of the tractor's front axle.

[0021] As a further preferred embodiment of the present invention, in step S4, during the braking process, the valve control quantity of each braking circuit is corrected in a closed loop based on the pressure tracking error and the wheel slip state. The corrected valve control quantity... The expression is: ; in, For the first The pressure reduction ratio coefficient of each braking circuit, and The slip ratio protection threshold and the decompression threshold are respectively and , No. The braking circuit is in the first The valve control quantity calculated from the pressure tracking error at each sampling time is: .

[0022] The advantages and beneficial effects of this invention are as follows: (1) To address the problem that existing AEB systems for semi-trailer trucks do not adequately represent the loading status, typically only roughly considering changes in the total vehicle mass and failing to reflect differences in mass distribution, center of gravity position, and braking force characteristics under different loading conditions, this invention first establishes a method for identifying the load status and solving mechanical parameters for semi-trailer trucks. By acquiring prior information on the loading distribution and measurement data under linear braking conditions, the semi-trailer mass, longitudinal center of gravity position, and center of gravity height are identified, and the axle loads of the tractor's front and rear axles, the axle loads of the semi-trailer (trailer trailer) axle group, and the vertical load of the towing pin are further calculated under braking conditions. Compared with existing methods that only focus on mass detection or vehicle mass estimation, this invention can further obtain key parameters reflecting the combined mass distribution characteristics and coupled force state of the vehicle, making the AEB system's representation of the current loading status of the semi-trailer truck more complete, detailed, and accurate. This provides a more sufficient state basis for subsequent AEB braking decision correction, braking force distribution, and execution control, improving the system's adaptability to unloaded, fully loaded, off-center loaded, and different cargo distribution conditions.

[0023] (2) In view of the problem that the existing AEB system does not fully consider the dynamic axle load and the vertical load of the traction pin in the emergency braking process of semi-trailer trains, which is prone to instability risks such as folding and tail swing, the present invention adopts a braking force distribution and stability correction method based on dynamic axle load in step S3. The braking torque of each axle is corrected by combining the axle load ratio of the tractor and the trailer, the vertical load of the traction pin and the wheel slip ratio feedback, so that the front axle of the tractor can give priority to the road surface adhesion conditions, thus solving the problems of unreasonable braking force distribution and insufficient directional stability and coupling stability during the braking process of semi-trailer trains.

[0024] (3) To address the problem that existing AEB systems often output a fixed deceleration after the braking level is determined, making it difficult to further refine the braking intensity according to the actual loading status and braking capacity changes of the semi-trailer truck, this invention adopts a decision-making method of basic braking level determination and adaptive correction of expected deceleration within the level. First, a collision risk assessment is performed based on TTC, outputting the basic AEB control state, basic braking level, and basic braking deceleration. Then, within the established braking level, combined with the vehicle's current speed, longitudinal deceleration state, the axle load ratio of the tractor's front axle, and the axle load ratio between the tractor and trailer, the deceleration correction amount under different levels is solved using a graded ANFIS sub-model, and boundary constraints are applied to the candidate expected deceleration to obtain the expected braking deceleration adapted to the current load state. Thus, the problem of existing AEB systems outputting a fixed deceleration and lacking continuous correction capability is solved, making the braking decision no longer a fixed value, but adaptively adjustable according to the loading, operating status, and braking capacity, thereby improving the matching and effectiveness between the braking decision and the actual braking capacity.

[0025] (4) In view of the fact that existing AEB systems for semi-trailer trucks often only use fixed ratios, empirical rules, or general stability control methods to distribute braking force among axles after automatic emergency braking is triggered, failing to fully combine the braking demand output by AEB with the dynamic axle load changes and wheel slip state during the vehicle braking process for coordinated adjustment, thus easily leading to unreasonable distribution of braking force among axles and premature approach of the adhesion limit of some axles, this invention proposes a dynamic braking force distribution and slip ratio correction method for the AEB automatic emergency braking process. Compared with existing methods, this invention does not distribute braking force in isolation, but directly connects the expected braking demand output by the upper layer of AEB with the lower layer of axle braking force distribution process, and further combines dynamic axle load and wheel slip feedback to correct the distribution results, so that the braking force distribution results of each axle can simultaneously reflect the automatic emergency braking intensity demand, the actual force state of the vehicle combination, and the tire adhesion utilization, avoiding the problem of excessive braking of some axles or insufficient utilization of braking resources, thereby improving the rationality of the braking force distribution among axles, the coordination of adhesion utilization, and the directional stability during the AEB automatic emergency braking process of semi-trailer trucks.

[0026] (5) To address the problem that existing AEB systems rarely consider changes in the coupling relationship between the tractor and trailer during the braking process of semi-trailer trains, and fail to fully utilize parameters such as the vertical load of the traction pin and the axle load relationship between the tractor and trailer to correct the stability of the braking force distribution, thus easily inducing instability risks such as vehicle folding and tail-swing during high-intensity braking, this invention further introduces a coupling stability correction mechanism based on the ideal braking force distribution. By constructing a stability correction coefficient, the ideal braking torque of the semi-trailer axle group is appropriately reduced, and the released braking torque is redistributed to the front and rear axles of the tractor according to the axle load ratio of the tractor's front and rear axles, so that the braking force distribution result can be dynamically adjusted according to changes in the vertical load of the traction pin and changes in the axle load relationship between the tractor and trailer. Thus, this invention can not only reflect the changes in the vehicle combination coupling state during the emergency braking process of semi-trailer trains, but also actively suppress excessive use of trailer side adhesion while maintaining a constant total braking torque, improve the tractor's guiding ability for the vehicle combination, reduce the risk of folding, tail-swing, and deterioration of directional stability, thereby enhancing the coupling stability and driving safety of semi-trailer trains during emergency braking.

[0027] (6) In view of the problem that the existing commercial vehicle AEB execution layer does not fully consider the response lag of the air pressure braking system, pressure building constraints, solenoid valve switching characteristics and wheel slip protection, resulting in the difficulty of timely, accurate and stable execution of the target braking force, this invention constructs a method to back-calculate the target braking air pressure from the final ideal braking torque of each axle, and then combines the equivalent lag time constant of the air pressure system and the maximum pressure building / depressurization rate to perform advance compensation and rate constraint on the target pressure; at the same time, in view of the problem that the on / off type solenoid valve is difficult to directly and continuously adjust the pressure, a segmented SPWM pressure closed-loop control method is adopted to adaptively adjust the duration of pressure increase, pressure holding and pressure reduction according to the pressure tracking error, and to perform closed-loop correction of the valve control execution process in combination with the wheel slip state. Compared with existing methods, this invention does not only make local improvements to the bottom-level pressure control or slip control, but directly connects the upper-level braking demand of AEB with the lower-level air pressure braking execution process. It also comprehensively considers the impact of air pressure hysteresis compensation, pressure build-up rate constraints and wheel slip feedback on the braking execution process, so that the actual braking air pressure of each braking circuit can more stably and timely track the target demand. At the same time, it suppresses excessive pressure build-up when the slip rate is too large, reducing the risk of wheel lock-up. This improves the timeliness of execution, pressure tracking accuracy, braking smoothness and overall safety of the automatic emergency braking process of semi-trailer trucks. Attached Figure Description

[0028] Figure 1 Flowchart of the AEB control method for semi-trailer trucks based on load state adaptation and stability correction provided by the present invention; Figure 2 This is a yz-plane view of the tractor axle; Figure 3 This is a plan view of the semi-trailer's axle (xz plane). Figure 4 This is a force diagram of the entire vehicle (semi-trailer truck); Figure 5 Force diagram of the tractor unit; Figure 6 Force diagram of a semi-trailer; Figure 7 This is a diagram of the AEB control strategy. Detailed Implementation

[0029] To enable those skilled in the art to better understand the technical solutions and advantages of the present invention, the present application will be described in detail below with reference to the accompanying drawings, but this is not intended to limit the scope of protection of the present invention.

[0030] like Figure 1 As shown in the figure, this embodiment provides an AEB control method for semi-trailer trucks based on load state adaptation and stability correction. The method includes the following steps:

[0031] Step S1: Obtain load information and calculate vehicle mechanical parameters; The system acquires loading information, vehicle measurement data, and known vehicle parameters to identify the semi-trailer's mass, longitudinal center of gravity position, and center of gravity height. It then calculates the dynamic axle loads of the tractor's front and rear axles, the dynamic axle loads of the semi-trailer's axle assembly, and the vertical load of the traction pin under braking conditions, providing a basis for subsequent AEB triggering decision correction and braking force distribution.

[0032] Step S2: Assess the collision risk and calculate the expected deceleration of AEB; First, a basic AEB control strategy is established, and the vehicle's driving status information and the information of the target obstacle ahead are obtained. Based on the above information, the collision risk index is calculated. Then, combined with the vehicle's current load status and axle load distribution parameters obtained in step S1, the expected braking deceleration under the braking level given by the current AEB is calculated, and the AEB braking decision result adapted to the current loading state is determined.

[0033] Step S3: Distribute braking force and correct stability based on load information; Based on the desired braking deceleration output by AEB in step S2, and combined with the dynamic axle loads of the front and rear axles of the tractor, the dynamic axle loads of the semi-trailer axle group, and the vertical load of the traction pin in step S1, the ideal braking force distribution of the front and rear axles of the tractor and the semi-trailer axle group is calculated. Furthermore, the ideal braking torque is corrected for stability by combining the axle load ratio of the tractor and the semi-trailer and the vertical load of the traction pin. Then, the ideal braking torque after stability correction is corrected by combining the wheel slip ratio feedback of each axle, so as to obtain the final ideal braking torque, thereby improving the directional stability and coupling stability of the semi-trailer truck train (semi-trailer freight truck) during the braking process.

[0034] Step S4: Braking command generation, execution, and closed-loop feedback correction; Based on the final ideal braking torque of each axle determined in step S3, braking control commands corresponding to the tractor and semi-trailer are generated and braking is executed. During the braking process, vehicle deceleration, wheel status and braking response information are acquired in real time, and the braking control commands are corrected in a closed loop to improve braking safety and take into account ride comfort.

[0035] Furthermore, in this embodiment, the semi-trailer truck consists of a tractor and a semi-trailer, and the semi-trailer has three axles. Step S1 specifically includes the following steps: Step S11: Obtain the known vehicle parameters of the tractor-trailer combination and the prior information on the semi-trailer loading distribution, wherein the known vehicle parameters include at least the tractor mass. Longitudinal position of the center of gravity of the tractor Tractor center of gravity height traction pin height Longitudinal distance from the front and rear axles of the tractor to the ground projection point of the towing pin and Longitudinal distance from each axle of the semi-trailer to the ground projection point of the traction pin , and Mass of each axle of the tractor , tractor axles around Moment of inertia of the shaft Wheelbase Horizontal installation spacing of air springs Lateral installation spacing of vibration dampers unsprung mass of semi-trailer unsprung mass of semi-trailer Moment of inertia of the shaft And the geometric parameters in the side view model of the single-side suspension of the semi-trailer. , , and .

[0036] The prior information on loading distribution is used to characterize the arrangement of goods in the longitudinal direction of the semi-trailer, including at least the information on the loading area and the loading offset direction, and the distribution ratio of goods in the front, middle and rear areas. The vehicle measurement data includes at least the vehicle's longitudinal acceleration. Longitudinal contact force between the front and rear axle tires of the tractor and the road surface , And the longitudinal contact force between the tires of each axle of the semi-trailer and the road surface. , , .

[0037] Furthermore, in this embodiment, the prior information on the semi-trailer loading distribution is obtained from the semi-trailer cargo compartment zoning weighing information. Specifically, the semi-trailer cargo loading area is longitudinally divided into a front area, a middle area, and a rear area, and the cargo distribution ratio in the corresponding area is denoted as follows: , and And satisfy:

[0038] To characterize the loading offset features of cargo in the longitudinal direction, a front-to-back distribution offset is defined. for: Among them, when Time indicates that the distribution of goods is biased towards the front, when Time indicates that the distribution of goods is biased towards the rear, when This indicates that the distribution of goods along the longitudinal direction is relatively balanced.

[0039] To characterize the degree of deviation of cargo loading distribution from a uniform loading state, a uniform distribution deviation coefficient is defined. for: ; Output the known vehicle parameters, prior information on the semi-trailer loading distribution, measurement data of straight-line braking conditions, and loading distribution characteristics. and This will be used as a basis for subsequent solutions to the semi-trailer mass based on the longitudinal dynamics model. Longitudinal centroid position and center of mass height Input and constraints.

[0040] Step S12: In this embodiment, the longitudinal dynamics model used satisfies the following assumptions: the vehicle combination (tractor and semi-trailer combination) is under an approximately longitudinal acceleration excitation condition, and the vehicle combination along... For axial motion, the lateral acceleration, steering angle, and yaw rate are approximately zero; the longitudinal acceleration of the vehicle may take different values ​​at different sampling times, but is approximately constant within a single sampling time; the tractor mass... The longitudinal, vertical, and lateral positions of the tractor's center of gravity are considered known and do not change with load; the lateral center of gravity position of the semi-trailer is considered known, based on the cargo's position relative to the load. The assumption of planar symmetry loading is applied; air resistance and tire rolling resistance are relatively small compared to the longitudinal forces generated by braking and traction, and are therefore negligible. The axial longitudinal force of each axle is also considered. And vertical force (vertical load) These are the sum of the forces exerted by the left and right wheels on the same axle, respectively. , Based on the above assumptions, and considering the longitudinal and vertical forces on each axle, as well as the force acting at the drawbar, the semi-trailer's mass is identified. Longitudinal center of gravity position of semi-trailer and semi-trailer center of gravity height .

[0041] Specifically, first, the vertical tire forces on the left and right wheels of each axle of the tractor unit are calculated. The tractor unit axles are in... A schematic diagram of the plan is shown below. Figure 2 Let the first... The vehicle body, the first The vertical tire forces of the left and right wheels of the axle are respectively and ,in Indicates tractor unit. If the vehicle is a semi-trailer, then the vertical balance and axial balance of its axle are... Shaft roll balance satisfies:

[0042]

[0043] in, and These are the mass and twist of the corresponding axle, respectively. moment of inertia of the shaft The wheelbase is the distance between the wheels. and These refer to the lateral mounting spacing of the air spring and the shock absorber, respectively. For air spring force, For the force of the vibration damper, and These are the vertical acceleration and roll acceleration of the axle, respectively. Solving these equations simultaneously yields the vertical tire forces on the left and right wheels of each axle of the tractor.

[0044] Furthermore, the semi-trailer axle is in A schematic diagram of the plan is shown below. Figure 3 For each wheel of the semi-trailer, the air spring force, shock absorber force, and axle torsional force are first converted to the wheel center according to the principle of torque equivalence. Specifically, the rotational reference point of the unsprung mass of the semi-trailer is used. As the torque equivalent reference, let the first... The first vehicle body root axle From the wheel center to the rotation reference point The equivalent arm is The lines of action of the air spring and the damper to the rotational reference point The lever arms are respectively and The torsional effect of the axle is equivalent to the rotation around a reference point. Torsional moment The equivalent air spring force referred to the wheel center is then... Equivalent vibration damper force Equivalent vertical force of axle torsion They are respectively: ,

[0045] in, To correspond to the actual force of the air spring, To correspond to the actual force applied by the vibration damper, The torsional force of the axle acts on the rotation reference point. The equivalent torque generated at that point. When a certain force is relative to the rotational reference point. When the direction of the moment is opposite to the direction of the moment of the equivalent vertical force at the wheel center, the stress arm or equivalent force is taken with a negative sign, and the final equivalent suspension force at the wheel center is... :

[0046] In obtaining Then, using the unsprung mass of the semi-trailer around a rotation reference point... To determine the rotational balance, solve for the vertical tire forces on each wheel. :

[0047] in, For corresponding unsprung mass winding Moment of inertia of the shaft The angular acceleration of the unsprung portion. For the longitudinal force of the wheel, and These are the vertical acceleration and longitudinal acceleration at the suspension's center of gravity, respectively. Indicates the tire contact point relative to the rotation reference point. The longitudinal distance, Indicates the tire contact point relative to Vertical distance of the point Indicates the center of mass of the unsprung mass Compared to Vertical distance of points Indicates the center of mass of the unsprung mass Compared to Vertical distance of the point For the first The first vehicle body root axle The mass of the non-sprung portion corresponds to the side. This is the acceleration due to gravity.

[0048] After obtaining the vertical tire forces of each wheel of the tractor and semi-trailer, the vertical tire forces of the left and right wheels on the same axle are added together to obtain the vertical loads of the front and rear axles of the tractor. , and the vertical loads on each axle of the semi-trailer. , and .

[0049] It should be noted that the vertical loads obtained in this step are mainly used as intermediate variables in the identification of the semi-trailer's mass and center of gravity parameters. They are used to establish the subsequent vertical balance relationship of the entire vehicle, the pitching moment balance relationship of the tractor, and the pitching moment balance relationship of the semi-trailer, thereby further determining the semi-trailer's mass, longitudinal center of gravity position, and center of gravity height. This part of the vertical load is not the final axle load distribution result in the AEB control stage.

[0050] After obtaining the vertical loads of each axle, the vertical balance relationship in the longitudinal dynamics model of the whole vehicle (such as...) is used. Figure 4 Solve for the mass of the semi-trailer. :

[0051] Subsequently, based on the pitch moment balance of the tractor vehicle around the ground projection point of the tractor pin (such as... Figure 5 Solve for the longitudinal force of the traction pin under approximate longitudinal acceleration excitation conditions. :

[0052] in, and These are the longitudinal position and height of the tractor's center of gravity, respectively. and These are the longitudinal distances from the front and rear axles of the tractor to the ground projection points of the towing pins, respectively. This represents the longitudinal acceleration of the entire vehicle. This refers to the height of the traction pin.

[0053] Finally, based on the pitch moment balance of the semi-trailer around the ground projection point of the traction pin (such as... Figure 6 Establish the longitudinal center of gravity position of the semi-trailer. and center of mass height The equation:

[0054] in, , and These represent the longitudinal distances from each axle of the semi-trailer to the ground projection point of the drawbar. The unknown quantity is the location of the semi-trailer's longitudinal center of gravity. and center of mass height Therefore, data points under at least two different longitudinal acceleration conditions were selected to establish a multi-time semi-trailer pitching moment balance equation. Finally, the front and rear distributed offsets were introduced. and uniform distribution deviation coefficient We construct prior constraints for the load distribution and solve them jointly using the weighted least squares method. and .

[0055] Specifically, let the reference position of the longitudinal center of gravity of the semi-trailer under uniform loading be . The longitudinal distances from the geometric centers of the front loading area and the rear loading area to the ground projection point of the traction pin are respectively and Then, based on the forward and backward distribution offsets The corrected prior position of the longitudinal centroid of the semi-trailer is:

[0056] in, The prior position of the longitudinal centroid of the semi-trailer is determined by the load distribution characteristics.

[0057] At the same time, using the uniform distribution deviation coefficient Adjust the prior constraint weights for load distribution and define the prior constraint weights for load distribution. , The basic prior constraint weights are set as follows: Considering that the semi-trailer pitch moment balance residual is related to the semi-trailer gravity term, the basic prior constraint weights in this embodiment are set as follows: .

[0058] Let the first The residual terms obtained from the semi-trailer pitching moment balance equation at each sampling time are: The dynamic residuals from multiple sampling times, together with the prior constraints on the loading distribution, constitute the following weighted least squares objective function:

[0059] in, The number of sampling moments involved in the solution process. The longitudinal centroid position of the semi-trailer is obtained by solving the above weighted least squares problem. and semi-trailer center of gravity height .

[0060] Step S13: When the vehicle detects an obstacle ahead and the automatic emergency braking triggering conditions are met, the vehicle assembly enters the straight-line braking phase. In this phase, the vehicle undergoes approximately pure deceleration along the longitudinal direction, with lateral acceleration, steering angle, and yaw rate approximately zero; within a single sampling moment, the longitudinal acceleration is approximately constant.

[0061] In this embodiment, after identifying the semi-trailer mass, longitudinal center of gravity position, and center of gravity height in step S12, a quasi-static axle load transfer model is established for the current braking condition after AEB triggering. Combining the longitudinal deceleration and quasi-static axle load transfer relationship under the current AEB braking condition, the dynamic axle load used for desired deceleration correction and stability correction is recalculated, i.e., the [model name missing]. The dynamic axle load of the front axle of the tractor, the dynamic axle load of the rear axle of the tractor, the dynamic axle load of the semi-trailer axle group, and the vertical load at the towing pin at each sampling time.

[0062] The three axles of the semi-trailer are considered as an equivalent axle group support point, with the position of the axle group support point taken as the geometric center of the three axles. The longitudinal distance from the equivalent axle group support point of the semi-trailer to the ground projection point of the drawbar is defined as:

[0063] Definition of the first Braking intensity of vehicle combination at a given moment for:

[0064] When the vehicle is in a braking and deceleration state, ,therefore ; For the first The adhesion utilization coefficient of the semi-trailer axle assembly at a given time. For the dynamic longitudinal force of the equivalent axle group of the three axles of the semi-trailer, The dynamic vertical load of the equivalent axle group of the three axles of the semi-trailer.

[0065] The semi-trailer three-axle assembly is in the The quasi-static equilibrium equation at time n can be written as:

[0066]

[0067]

[0068] Based on the longitudinal force balance and vertical force balance of the semi-trailer, the longitudinal force and vertical load of the drawbar can be expressed as follows:

[0069]

[0070] in, For the first The longitudinal force at the traction pin at each sampling moment For the first Vertical load on the traction pin at each sampling moment.

[0071] Substituting the above two equations into the pitching moment balance equation of the semi-trailer about the ground projection point of the traction pin, and... Substituting, we can obtain The dynamic vertical load of the equivalent axle group of the three axles of the semi-trailer (dynamic axle load of the semi-trailer axle group) at any given time satisfies:

[0072] Under the conditions of resistance and tire rolling resistance, the tractor unit in the first... The quasi-static equilibrium equation at time n can be written as:

[0073]

[0074]

[0075] In seeking and Then, by solving the equations simultaneously, the dynamic axle loads of the front and rear axles of the tractor are obtained as follows:

[0076]

[0077] Finally, the number was obtained The axle load distribution parameters of the vehicle combination at a given moment under the current braking condition, i.e., the dynamic axle load of the tractor's front axle. Dynamic axle load of the tractor rear axle Dynamic axle load of semi-trailer axle assembly and vertical load at the traction pin .

[0078] Furthermore, in this embodiment, step S2 specifically includes the following steps: Step S21: Collision risk assessment and basic braking level determination; In the At each sampling time point, the speed of the vehicle (semi-trailer truck) is obtained. Longitudinal acceleration Longitudinal distance between this vehicle and the target obstacle ahead Relative velocity Relative acceleration This invention selects the time to collision (TTC) as the collision risk indicator and uses it to output the basic AEB braking level. Considering that the forward collision scenario addressed by this invention mainly manifests as a longitudinal approach process, and that the longitudinal acceleration of both the vehicle and the target obstacle can be approximated as constant within a single sampling moment, the two satisfy a constant acceleration relative motion relationship.

[0079] Let the first The effective collision margin at each sampling time is: ;in, This is the minimum safe distance.

[0080] When the remaining effective collision margin between the vehicle and the target obstacle decays to 0, the collision boundary is considered reached. The longitudinal relative motion equation between the vehicle and the target obstacle can then be written as:

[0081] in, Represents time.

[0082] Therefore, the first is defined The collision time TTC at each sampling time is the smallest positive real root that satisfies the above equation, i.e.:

[0083] in, .

[0084] In this embodiment, This indicates that the current target distance is no greater than the minimum safe distance; This indicates that there is no current trend of an effective collision approaching or that a longitudinal collision will not occur under the current relative motion state.

[0085] Furthermore, the TTC threshold is determined by combining forward collision warning research, the phased control law of commercial vehicle AEBS, and the emergency braking requirements of heavy vehicles. Considering that the triggering TTC of typical FCW (Forward Collision Warning) systems is mostly in the range of 2.7s to 2.0s, and that commercial vehicle AEBS (Automatic Emergency Braking System) needs to reserve time for air pressure braking response hysteresis and collision warning, while the automatic braking deceleration of heavy vehicles during emergency braking should reach a level of not less than 4m / s², this invention sets a warning threshold. Level 1 braking threshold and secondary braking threshold They are respectively: , , And let the basic braking decelerations corresponding to the first-level braking and the second-level braking be respectively: , ;in, Used for gentle braking when the risk of collision is higher than the warning level but has not yet reached the level for emergency braking. Used for emergency braking when a collision is imminent.

[0086] Furthermore, to meet the requirements of the first level of the AEB control strategy for early warning, the first level is defined as follows: AEB control status at each sampling time Basic AEB braking level and basic braking deceleration Specific AEB control strategies include: Figure 7 Its formula is expressed as:

[0087] in, This indicates a safe state, and the system will not perform AEB intervention. This indicates a Level 1 warning state; the system only issues a collision warning signal and does not apply braking. This indicates the first-level braking state, at which point the basic AEB braking level is reached. Basic braking deceleration is taken ; This indicates a level 2 braking condition, at which point the basic AEB braking level is... Basic braking deceleration is taken .along with As the threshold decreases, the system control state gradually transitions from a safe state to a first-level warning, a first-level braking, and a second-level braking. The final output is... Collision risk index at each sampling time AEB control status Basic AEB braking level and basic braking deceleration ;in, As the gating signal for the corresponding level ANFIS (Adaptive Neuro-Fuzzy Inference System) sub-model in step S22, As input for subsequent load distribution-sensitive deceleration correction.

[0088] Step S22: Solve for the desired deceleration using a load distribution-sensitive ANFIS based on braking level gating; In the At each sampling time, read the current basic AEB braking level output in step S21. ,in Simultaneously obtain the vehicle's speed. Longitudinal acceleration and the dynamic axle load of the tractor front axle obtained in step S13. Dynamic axle load of the tractor rear axle Dynamic axle load of semi-trailer axle assembly Vertical load at the traction pin The load information is used only for adaptive correction of the desired braking deceleration within the AEB braking level and does not involve the braking force distribution between the tractor and the semi-trailer.

[0089] To characterize the load distribution characteristics of the vehicle combination under the current braking condition, the first... The percentage of front axle load on the tractor at any given time is:

[0090] Definition of the first The axle load ratio of the tractor to the semi-trailer at that moment is:

[0091] Based on this, construct the first The input vector of the ANFIS model at each time step is: .

[0092] Furthermore, to avoid overlapping deceleration mapping relationships under different braking levels, two multi-input single-output adaptive neurofuzzy inference sub-models are constructed for first-level braking and second-level braking, respectively. Among them, Used to solve for the deceleration correction under the first braking level. Used to solve for the deceleration correction under the second braking level.

[0093] For the The braking level is (i.e., its corresponding braking level is 1). Fuzzy rules for:

[0094]

[0095] in, The corresponding input variables are in the th The fuzzy set corresponding to a fuzzy rule; For the linear parameters of the consequent of this rule; for At this moment Under the first braking level The output of the fuzzy rules.

[0096] Each input variable is fuzzy-partitioned using a generalized bell-shaped membership function, i.e., the GB membership function. For the... Under the first braking level Rule 1, record input variables GB membership for:

[0097] in, These correspond to four input variables; This is the width parameter of the GB membership function. For shape parameters, The central parameter is used. Therefore, The membership degrees of the four input variables at time t are written as follows:

[0098] in, Indicates the first Under the first braking level, the first In a fuzzy rule, the input variables The degree to which it belongs to the corresponding fuzzy set; , and The physical meaning is the same, and its value range is... .

[0099] In this embodiment, the and The parameters were obtained through offline training. The corresponding parameters were calculated under different loading weights, different cargo distributions, different initial vehicle speeds, and different collision risk scenarios. and Combining the basic AEB graded control strategy and vehicle combined stability constraints, a sample of reference deceleration corrections under different braking levels is constructed. A hybrid learning method combining least squares and backpropagation is used to train the GB membership function parameters and consequent parameters, and finally obtains the intra-level expected deceleration solution model for first-level braking and second-level braking.

[0100] Furthermore, the first Under the first braking level Rule number 1 The trigger strength at each moment is: ; The corresponding normalized trigger strength is: ;in, For the first Each braking level corresponds to the total number of fuzzy rules in the ANFIS sub-model; Indicates the current input state for the first... The overall matching degree of the fuzzy rules, i.e., the degree of matching of the fuzzy rules. The trigger strength of a fuzzy rule; This represents the normalized rule weights.

[0101] Therefore, the first The deceleration correction for each braking level can be expressed as: ; in, For the first The deceleration correction amount output by the ANFIS model at each braking level is used to characterize the adaptive correction result of the baseline deceleration at that level under the current load and operating conditions.

[0102] To highlight the adaptive correction effect of the current load state on deceleration within different levels, this invention makes the ANFIS output the correction amount relative to the current level's base braking deceleration. Let the base braking decelerations corresponding to Level 1 braking and Level 2 braking be respectively... and Then the first The expected deceleration rates for each level at each time point are as follows:

[0103]

[0104] in, and These represent the candidate desired deceleration rates after correction for primary braking and secondary braking under the current load conditions, respectively.

[0105] To ensure that the candidate desired deceleration always remains within the allowable range of the corresponding braking level, boundary correction is applied to the candidate desired deceleration. Let the deceleration ranges corresponding to Level 1 braking and Level 2 braking be respectively... and Then we have:

[0106]

[0107] in, This is a limiting function used to constrain the input value within a range. Inside; and These represent the expected braking decelerations of the first-level braking and second-level braking after boundary correction, respectively.

[0108] During online operation, when step S21 determines that the vehicle is currently in a Level 1 or Level 2 braking state, the corresponding level of the ANFIS sub-model is activated, and the corrected expected braking deceleration at that level is output as the AEB module output. When the vehicle is in a safe state or a Level 1 warning state, the AEB module does not output automatic braking deceleration. That is:

[0109] in, For the first The desired braking deceleration output by the AEB module at each moment; The AEB control status output in step S21 represents the safety status, first-level warning status, first-level braking status, and second-level braking status, respectively. and These are the desired braking decelerations after load adaptive correction under first-level and second-level braking conditions, respectively. As the final output of step S2, it is used in the subsequent step S3 for ideal braking force distribution and stability correction.

[0110] Furthermore, in this embodiment, the specific steps of step S3 are as follows: Step S31: In the first At that moment, the dynamic axle load of the tractor front axle obtained in step S13 is read. Dynamic axle load of the tractor rear axle Dynamic axle load of semi-trailer axle assembly and vertical load of the traction pin Simultaneously read the desired braking deceleration output from the AEB module. and the actual longitudinal deceleration of the vehicle combination. .

[0111] First, based on the total mass of the vehicle combination Desired braking deceleration with AEB output Calculate the expected braking force of the feedforward for:

[0112] Then, feedback correction is performed based on the error between the desired braking deceleration and the actual braking deceleration, and the desired braking force is fed back. for:

[0113] in, , and These are the proportional coefficient, integral coefficient, and differential coefficient, respectively. For the first The error between the expected braking deceleration and the actual longitudinal deceleration at each moment, and , This represents the sampling time interval.

[0114] Thus, the first The expected generalized braking force of the vehicle combination at any given moment :

[0115] Step S32: When a heavy-duty commercial vehicle brakes on a road surface with any coefficient of adhesion, the front and rear axles of the tractor and the axles of the semi-trailer must lock simultaneously to fully utilize the road adhesion conditions and maintain good directional stability. This step uses the coefficient of adhesion to describe the braking of the tractor and semi-trailer:

[0116] in, for Time of the first Ground braking force of the axle; The first one obtained in step S13 The dynamic axle loads of the tractor's front axle, rear axle, and semi-trailer axle groups at any given time are calculated. The closer the coefficient of adhesion is to the braking intensity, the more fully the corresponding axle utilizes the road surface adhesion conditions. Under ideal braking force distribution, the coefficients of adhesion utilized by each axle should be as equal as possible and equal to the target braking intensity.

[0117] In the At any given moment, let the coefficients of adhesion for the tractor's front axle, rear axle, and semi-trailer axle assembly be respectively... , and And take all three to be equal to the target braking intensity. ,Right now:

[0118] Among them, target braking intensity Braking intensity different from that of the vehicle combination in step S13 It is the first one obtained from step S31. At that moment, we expect generalized braking force. The ratio to the total dynamic axle load of the current vehicle combination is determined, that is:

[0119] According to the result obtained in step S13 Dynamic axle load of the tractor front axle at any given moment Dynamic axle load of the tractor rear axle and the dynamic axle load of the semi-trailer axle assembly Thus, the ideal braking force of each axle can be obtained as follows:

[0120] Therefore, the first The ideal braking forces for the front axle, rear axle, and semi-trailer axle assembly of the tractor unit at each moment are as follows:

[0121]

[0122]

[0123] Furthermore, based on the ideal braking force of each axle and the corresponding wheel rolling radius, the first... The ideal braking torques of the tractor's front axle, rear axle, and semi-trailer axle assembly at each moment are as follows:

[0124]

[0125]

[0126] in, , and These are the rolling radii of the wheels corresponding to the front and rear axles of the tractor and the axle group of the semi-trailer, respectively.

[0127] Step S33: Further consider the coupling stability of the semi-trailer truck train during braking. As the vertical load of the traction pin decreases and the axle load of the semi-trailer axle group increases relative to the total axle load of the two axles of the tractor, the vehicle combination is more prone to folding or fishtailing during braking. Therefore, it is necessary to make stability correction to the ideal braking torque obtained in step S32.

[0128] Furthermore, by utilizing the axle load ratio of the tractor and the semi-trailer... And set the first The vertical load on the traction pin at any given moment is The vertical load on the traction pin and the axle load ratio of the tractor to the semi-trailer under stable operating conditions are respectively and Then define the first The stability correction coefficient at time t is:

[0129] in, For the amplitude limiting function, it is guaranteed that .

[0130] After applying stability correction to the ideal braking torque of the semi-trailer axle assembly, we obtain:

[0131] The resulting difference in braking torque for:

[0132] To keep the total braking torque constant, the difference in braking torque is calculated according to the first... At a given moment, the ratio of dynamic axle load on the front axle to dynamic axle load on the rear axle of the tractor is redistributed to the front and rear axles of the tractor. The ideal braking torques of the front and rear axles of the tractor after stability correction are then obtained as follows:

[0133]

[0134] Step S34: Obtain the first [item] from step S33. Ideal braking torque of the tractor front axle, tractor rear axle, and semi-trailer axle assembly after stability correction at a given moment. , and Subsequently, by combining the slip ratio feedback of each axle wheel, the ideal braking torque is corrected at the end to prevent the rear axle of the tractor or the axle group of the semi-trailer from approaching the adhesion limit too early, thereby further improving the directional stability and coupling stability of the vehicle combination during the braking process.

[0135] The relative motion between the wheel and the road surface during braking can be characterized by the slip ratio. Let the first... At time 1, the average slip rates of the wheels corresponding to the front axle, rear axle, and semi-trailer axle groups of the tractor were respectively: , and Then we have:

[0136] in, For the first The longitudinal velocity at the center of the wheel at that moment For the first At the [time]th moment The axle corresponds to the angular velocity of the wheel. For the first The axle corresponds to the rolling radius of the wheel.

[0137] The front axle of the tractor unit performs steering and is more suitable as a reference axle for adhesion utilization and directional stability during straight-line braking. When the slip ratio of the rear axle of the tractor unit or the semi-trailer axle assembly is higher than that of the front axle, it indicates that the rear axle or semi-trailer side approaches the adhesion limit earlier, which can easily lead to a decrease in the combined steering ability of the vehicle and induce the risk of overturning or fishtailing. Therefore, this invention uses the slip ratio of the front axle of the tractor unit as a benchmark to construct the slip ratio difference between the rear axle and the front axle of the tractor unit, as well as the slip ratio difference between the semi-trailer axle assembly and the front axle. Based on this, the ideal braking torque of the rear axle of the tractor unit and the semi-trailer axle assembly is corrected by feedback, so that the front axle of the tractor unit can preferentially utilize the road adhesion conditions, thereby improving the directional stability and coupling stability of the semi-trailer train during braking.

[0138] Based on the above analysis, to quantitatively characterize the adhesion utilization deviation of the tractor's rear axle and semi-trailer axle assembly relative to the tractor's front axle, the slip ratio difference between the tractor's rear axle and the front axle, and the slip ratio difference between the semi-trailer axle assembly and the front axle are defined as follows:

[0139]

[0140] in, Characterizes the slippage tendency of the rear axle relative to the front axle of the tractor. It characterizes the slippage tendency of the semi-trailer axle assembly relative to the front axle.

[0141] Furthermore, slip ratio feedback correction torques are constructed for the rear axle of the tractor and the axle assembly of the semi-trailer based on the slip ratio difference, as follows:

[0142] in, , and These are the proportional coefficient, integral coefficient, and derivative coefficient for the rear axle slip ratio feedback control of the tractor vehicle; , and These are the proportional coefficient, integral coefficient, and derivative coefficient for the semi-trailer axle slip ratio feedback control.

[0143] Based on this, the stability correction result obtained in step S33 is subjected to end correction to obtain the first... The final ideal braking torques of the tractor's rear axle and the semi-trailer's axle assembly at each moment are as follows:

[0144]

[0145] To maintain a constant total braking torque, the braking torque released from the rear axle of the tractor and the semi-trailer axle assembly is compensated to the front axle of the tractor, resulting in the first... The final ideal braking torque of the front axle of the tractor at that moment is:

[0146] Furthermore, in this embodiment, step S4 specifically involves the following steps: Step S41: Based on the final ideal braking torque of each shaft output in step S3, calculate the target braking air pressure for each braking circuit. At any given moment, read the final ideal braking torque of the tractor front axle, tractor rear axle, and semi-trailer axle group output in step S34. , and Let the first... The effective working area of ​​the brake chamber in each braking circuit is The return threshold pressure is Adjust the arm length to The equivalent radius of action of S-cam is The inner radius of the brake drum is Braking factor is ,in These represent the front axle, rear axle, and semi-trailer axle assembly of the tractor unit, respectively. Based on the mechanical force transmission relationship of drum brakes, the brake shoe force is first calculated from the target braking torque. Then solve for the S-cam axial torque. With push rod force Finally, the target braking air pressure is obtained. The calculation process is as follows:

[0147]

[0148]

[0149]

[0150] Combining the above equations, we can obtain the first... The equivalent inverse relationship between the target braking torque and the target braking air pressure for each braking circuit is as follows:

[0151] Further define the first The braking circuit is in the first Equivalent braking gain at time 1 for:

[0152] Then we have:

[0153] In this embodiment, the three axles of the semi-trailer use the same brake air supply circuit, at which time... As a unified target braking air pressure for the semi-trailer axle group, the controller controls the semi-trailer's electronically controlled air pressure valve or relay valve to increase, maintain, or reduce the pressure of the shared air supply circuit based on the deviation between the unified target braking air pressure and the actual pressure of the semi-trailer axle group's air supply circuit. Since the braking chambers of the three axles of the semi-trailer are connected to the same air supply circuit, the adjusted circuit pressure is synchronously applied to the braking chambers corresponding to the three axles through the air circuit, thereby achieving synchronous control of the braking air pressure of the three axles of the semi-trailer.

[0154] Step S42: Due to the compressibility of gas, long pipeline transmission, and solenoid valve switching lag in commercial vehicle air pressure braking systems, the target braking air pressure cannot be achieved instantaneously. Therefore, the target braking air pressure obtained in step S41 needs to undergo advance compensation and rate constraint processing. Let the first... The equivalent hysteresis time constant of each braking circuit is The maximum pressure build-up rate and the maximum pressure relief rate are respectively and The sampling period is Then the first The pressure of the hysteresis compensation target at each moment It can be written as:

[0155] Based on this, the pressure change rate is further limited to obtain the reference braking air pressure used for the underlying valve control execution. :

[0156] in, For the first The maximum allowable braking air pressure for each braking circuit This is a limiting function used to ensure that the pressure command remains within the executable range. The above processing makes the braking command more consistent with the actual pressure build-up, pressure holding, and pressure release processes of commercial vehicle pneumatic braking systems, which helps to suppress response lag and pressure overshoot caused by sudden increases in command.

[0157] Step S43: In the first At any given moment, read the actual brake air pressure fed back by the brake air chamber pressure sensors of the corresponding brake circuits of the tractor's front axle, tractor's rear axle, and semi-trailer axle assembly. It is then compared with the reference braking air pressure output in step S42 to construct the pressure tracking error of each loop. :

[0158] in, For the first Reference brake air pressure for each braking circuit This is the actual brake air pressure fed back by the pressure sensor.

[0159] To address the challenge of direct and continuous pressure regulation by solenoid valve on / off actuators in commercial vehicle electronically controlled pneumatic braking systems, a segmented SPWM (Sinusoidal Pulse Width Modulation) pressure closed-loop control method is employed. This method adaptively adjusts the duration of pressure increase or decrease within the current control cycle based on the magnitude of the pressure error. Let the control cycle be... The error threshold is and The piecewise coefficient is , , and Then the first Valve control quantity of each braking circuit It can be represented as:

[0160] Among them, when At that time, the first The braking circuit is in a pressurized state; when It is in a pressure-holding state; when It is in a state of decompression. As the duration of pressurization or depressurization within the current control cycle, This serves as the pressure holding duration, enabling rapid and smooth tracking of the brake air pressure to the reference value. For normally open booster valves and normally closed pressure reducing valves, the above three states correspond to the three types of solenoid valve drive sequences: boosting, holding, and reducing pressure, respectively.

[0161] In this embodiment, the valve control parameters of each braking circuit are corrected based on the pressure tracking error and wheel slippage state. Let the... The braking circuit is in the first The valve control quantity calculated from the pressure tracking error at each sampling time is: The corresponding average wheel slip ratio The slip ratio protection threshold and the decompression threshold are respectively and The corrected valve control quantity Represented as:

[0162] in, For the first The pressure reduction ratio coefficient of each braking circuit, and .when When this occurs, it indicates that the corresponding wheel slip ratio is significantly too high, and the actuator prioritizes decompression; when And valve control quantity If the pressure is high, it indicates that the circuit originally had a pressure boosting requirement, but the wheel is close to the slip protection zone, so the pressure boosting is stopped and the circuit switches to pressure holding; otherwise, the valve control quantity calculated according to the pressure tracking error is still executed.

[0163] Corrected valve control quantity Used to determine the solenoid valve drive timing within the current control cycle. When At that time, the first Each braking circuit performs a boost, with the boost duration being... Hold pressure for the rest of the time; when When, the circuit maintains pressure; when At that time, the circuit performs pressure reduction, and the pressure reduction duration is... The pressure is maintained for the remaining time. Therefore, while ensuring that the actual braking air pressure tracks the reference braking air pressure, excessive pressure build-up can be promptly suppressed when wheel slippage is too high, reducing the risk of wheel lock-up and improving braking stability and driving safety during emergency braking of semi-trailer trucks.

[0164] The present invention also provides an electronic device, comprising: one or more processors and a memory; wherein the memory is used to store one or more programs, and when the one or more programs are executed by the one or more processors, the one or more processors implement the above-described AEB control method for semi-trailer trucks based on load state adaptation and stability correction.

[0165] The present invention also provides a computer-readable medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the aforementioned AEB control method for semi-trailer trucks based on load state adaptation and stability correction.

[0166] Those skilled in the art will understand that all or part of the functions of the various methods / modules in the above embodiments can be implemented by hardware or by computer programs. When all or part of the functions in the above embodiments are implemented by computer programs, the program can be stored in a computer-readable storage medium, which may include: read-only memory, random access memory, disk, optical disk, hard disk, etc., and the program is executed by a computer to achieve the above functions. For example, the program can be stored in the memory of a device, and when the program in the memory is executed by the processor, all or part of the above functions can be achieved.

[0167] In addition, when all or part of the functions in the above embodiments are implemented by computer programs, the programs can also be stored in storage media such as servers, other computers, disks, optical discs, flash drives, or portable hard drives. They can be downloaded or copied to the memory of the local device, or the system of the local device can be updated. When the program in the memory is executed by the processor, all or part of the functions in the above embodiments can be implemented.

[0168] The above describes specific embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A semi-trailer truck AEB control method based on load state adaptation and stability correction, characterized in that, The method includes the following steps: Step S1: Obtain loading information, vehicle measurement data and known vehicle parameters, identify the semi-trailer mass, longitudinal center of gravity position and center of gravity height, and calculate the dynamic axle load of the front and rear axles of the tractor, the dynamic axle load of the semi-trailer axle group and the vertical load of the traction pin under braking conditions. Step S2: Establish an AEB control strategy, assess the collision risk, and combine the current vehicle load status and axle load distribution obtained in Step S1 to calculate the expected braking deceleration under the current braking level given by the AEB. Step S3: Based on the desired braking deceleration output by AEB in Step S2, and combined with the dynamic axle loads of the front and rear axles of the tractor and the dynamic axle loads of the semi-trailer axle group in Step S1, calculate the ideal braking force distribution of the front axle of the tractor, the rear axle of the tractor, and the semi-trailer axle group; at the same time, combine the axle load ratio of the tractor and the semi-trailer and the vertical load of the traction pin to perform stability correction on the ideal braking torque, and then combine the wheel slip ratio feedback of each axle to perform end correction on the ideal braking torque after stability correction to obtain the final ideal braking torque; Step S4: Based on the final ideal braking torque of each axis output in step S3, determine the target braking air pressure for each braking circuit, and perform advance compensation and rate constraint processing on the target braking air pressure to obtain the reference braking air pressure used for the underlying valve control execution. Then, based on the actual braking air pressure fed back by the pressure sensor, a pressure tracking error is constructed, and the duration of pressurization, depressurization, or pressure holding within the current control cycle is adaptively adjusted according to the magnitude of the pressure tracking error.

2. The AEB control method for semi-trailer trucks based on load state adaptation and stability correction according to claim 1, characterized in that, The semi-trailer truck mentioned in step S1 consists of a tractor and a semi-trailer, and the semi-trailer has three axles; the loading information is the prior information of the semi-trailer loading distribution, including cargo loading area information and cargo loading offset direction information; the vehicle measurement data includes the vehicle longitudinal acceleration, the longitudinal contact force between the front and rear axle tires of the tractor and the road surface, and the longitudinal contact force between the tires of each axle of the semi-trailer and the road surface. Step S1 inputs the acquired data into the longitudinal dynamics model based on straight braking conditions. First, the vertical tire forces of each wheel of the tractor and semi-trailer are calculated to obtain the vertical loads of the front and rear axles of the tractor and the vertical loads of each axle of the semi-trailer. Then, based on the vertical balance relationship in the longitudinal dynamics model of the whole vehicle, the mass of the semi-trailer is obtained. The expression is: ； in, , This represents the vertical load on the front and rear axles of the tractor unit. , , This represents the vertical load on each axle of the semi-trailer. It is the acceleration due to gravity. For the weight of the tractor; Subsequently, based on the pitch moment balance of the tractor vehicle around the ground projection point of the tractor pin, the longitudinal force of the tractor pin under the longitudinal acceleration excitation condition is calculated. ; Finally, based on the pitch moment balance of the semi-trailer around the ground projection point of the traction pin, the position of the semi-trailer's longitudinal center of gravity is established. and center of mass height The equation is expressed as: ； in, , and These represent the longitudinal distances from each axle of the semi-trailer to the ground projection point of the draw pin. This represents the longitudinal acceleration of the entire vehicle. The height of the traction pin; Then, data points under at least two different longitudinal acceleration conditions are selected to establish a multi-time semi-trailer pitching moment balance equation; finally, the front and rear distributed offsets are introduced. and uniform distribution deviation coefficient We construct prior constraints for the load distribution and solve them jointly using the weighted least squares method. and .

3. The AEB control method for semi-trailer trucks based on load state adaptation and stability correction according to claim 1, characterized in that, In step S1, after an obstacle is detected ahead and the automatic emergency braking triggering conditions are met, based on the identified semi-trailer mass and center of gravity parameters, and combined with the structural parameters of the tractor-semi-trailer combination, a quasi-static axle load transfer model of the tractor-semi-trailer combination under braking conditions is established, and the first... The dynamic axle load of the tractor's front axle, the dynamic axle load of the tractor's rear axle, the dynamic axle load of the semi-trailer's axle assembly, and the vertical load at the towing pin at each sampling time.

4. The AEB control method for semi-trailer trucks based on load state adaptation and stability correction according to claim 1, characterized in that, In step S2, the collision time TTC is selected as the collision risk indicator. Identify AEB control status based on collision risk indicators. This will be used to output the basic AEB braking level. When AEB control status This indicates a safe state, and no AEB intervention will be performed. This indicates a Level 1 warning status, where only a collision warning signal is issued and braking is not implemented; This indicates the first-level braking state, at which point the basic AEB braking level is reached. Basic braking deceleration is taken ; This indicates a level 2 braking condition, at which point the basic AEB braking level is... Basic braking deceleration is taken ; Then, regarding the basic AEB braking level and basic AEB braking level Two multi-input single-output adaptive neural fuzzy inference sub-models are constructed; among them, Used to solve for the deceleration correction under the first braking level. This is used to solve for the deceleration correction at the second braking level; then, based on the obtained deceleration correction and the base braking deceleration, the first braking deceleration is calculated. Candidate expected decelerations at each level at each time point; finally, boundary corrections are applied to the candidate expected decelerations to obtain the expected braking decelerations for first-level and second-level braking after boundary correction. and .

5. The AEB control method for semi-trailer trucks based on load state adaptation and stability correction according to claim 1, characterized in that, In step S3, the feedforward desired braking force is first calculated based on the total mass of the semi-trailer truck and the desired braking deceleration output in step S2. ; Then, based on the error between the desired braking deceleration and the actual braking deceleration, feedback correction is performed, and the feedback desired braking force is calculated. Thus, the first The expected generalized braking force of the tractor and semi-trailer combination at that moment. Then, based on the expected generalized braking force and the dynamic axle load of the tractor's front axle... Dynamic axle load of the tractor rear axle and the dynamic axle load of the semi-trailer axle assembly Determine the target braking intensity This allows for the acquisition of ideal braking forces for the tractor's front axle, rear axle, and semi-trailer axle assembly. , , The ideal braking torque of the tractor front axle, tractor rear axle, and semi-trailer axle assembly. , , ; Then, stability correction is applied to the obtained ideal braking torque to obtain the first... Ideal braking torque of the tractor front axle, tractor rear axle, and semi-trailer axle assembly after stability correction at a given moment.

6. The AEB control method for semi-trailer trucks based on load state adaptation and stability correction according to claim 1, characterized in that, In step S4 Target brake air pressure for each braking circuit The expression is: ； Among them, the The effective working area of ​​the brake chamber in each brake circuit is The return threshold pressure is Adjust the arm length to The equivalent radius of action of S-cam is The inner radius of the brake drum is Braking factor is , These represent the front axle of the tractor unit, the rear axle of the tractor unit, and the axle assembly of the semi-trailer, respectively. The final ideal braking torque is... ; The method for performing advance compensation and rate constraint processing on the target braking air pressure is as follows: ； ； Among them, the The equivalent hysteresis time constant of each braking circuit is The maximum pressure build-up rate and the maximum pressure relief rate are respectively and The sampling period is , No. The pressure of the hysteresis compensation target at each moment , No. The maximum allowable braking air pressure for each braking circuit is: , This is a limiting function, and the reference braking air pressure used for the underlying valve control execution is... .

7. The AEB control method for semi-trailer trucks based on load state adaptation and stability correction according to claim 1, characterized in that, In step S4, the pressure tracking error of each braking circuit is constructed based on the actual braking air pressure and the reference braking air pressure. And calculate the first Valve control quantity of each braking circuit The expression is: ； The control cycle is as follows: The two error thresholds are respectively and The piecewise coefficient is , , and ; when At that time, the first The braking circuit is in a pressurized state; when It is in a pressure-holding state; when It is in a state of decompression; and with As the duration of pressurization or depressurization within the current control cycle, This refers to the duration of pressure holding.

8. The AEB control method for semi-trailer trucks based on load state adaptation and stability correction according to claim 4, characterized in that, The input vector for the adaptive neural fuzzy inference sub-model in step S2 is: ;in, For the speed of a semi-trailer truck, For the longitudinal acceleration of the semi-trailer truck, For the first The percentage of front axle load on the tractor at any given moment. For the first The axle load ratio of the tractor to the semi-trailer at any given time. This indicates that the transpose is valid; Each input variable is fuzzy partitioned using the GB membership function. The membership degrees of the four input variables at time t are as follows: Based on this calculation, the first Under the first braking level Rule No. Trigger intensity at each moment The expression is: ； After normalization, the normalized rule weights are obtained. And calculate the first The deceleration correction amount for each braking level is expressed as follows: ； in, For the first The deceleration correction amount output by the adaptive neural fuzzy inference sub-model at each braking level. For the first The linear parameters of the consequent of a fuzzy rule.

9. The AEB control method for semi-trailer trucks based on load state adaptation and stability correction according to claim 5, characterized in that, In step S3, when performing stability correction on the ideal braking torque, first based on the first The stability correction factor is calculated based on the vertical load of the lead pin at each moment, the axle load ratio of the tractor and the semi-trailer, and the vertical load of the lead pin and the axle load ratio of the tractor and the semi-trailer under stable operating conditions. Then, based on the stability correction coefficient, the ideal braking torque of the semi-trailer axle assembly is corrected for stability, and the difference in braking torque is obtained from the result. Then the braking torque difference will be calculated according to the first... At any given moment, the ratio of dynamic axle load on the front axle to dynamic axle load on the rear axle of the tractor is redistributed to the front and rear axles of the tractor, resulting in the ideal braking torque of the front and rear axles of the tractor after stability correction. In step S3, when performing end-point correction on the ideal braking torque after stability correction, the first step is to calculate the... The average slip ratio of the corresponding wheels of the front axle, rear axle, and semi-trailer axle groups of the tractor at a given time is used to determine the difference in slip ratio between the rear axle and the front axle of the tractor. The difference in slip ratio between the semi-trailer axle assembly and the tractor front axle Based on the slip ratio difference, the slip ratio feedback correction torques for the tractor rear axle and semi-trailer axle groups are constructed as follows: ； ； in, , and These are the proportional coefficient, integral coefficient, and derivative coefficient for the rear axle slip ratio feedback control of the tractor vehicle; , and These are the proportional coefficient, integral coefficient, and derivative coefficient for the semi-trailer axle group slip ratio feedback control; Based on the slip ratio feedback correction torque of the tractor rear axle and semi-trailer axle assembly, the first... The final ideal braking torque of the tractor's rear axle and semi-trailer axle assembly at a given moment is then compensated to the tractor's front axle to obtain the final ideal braking torque of the tractor's front axle.

10. The AEB control method for semi-trailer trucks based on load state adaptation and stability correction according to claim 7, characterized in that, Step S4: During braking execution, based on the pressure tracking error and wheel slip state, the valve control quantity of each braking circuit is corrected in a closed loop. The corrected valve control quantity... The expression is: ； in, For the first The pressure reduction ratio coefficient of each braking circuit, and The slip ratio protection threshold and the decompression threshold are respectively and , No. The braking circuit is in the first The valve control quantity calculated from the pressure tracking error at each sampling time is: .

Images