A truck AEB adaptive control method based on ice and snow road recognition

Abstract

The present invention discloses a truck AEB adaptive control method based on ice and snow road surface recognition. The method overcomes the problems of poor driving safety of trucks and insufficient consideration of truck characteristics under severe weather conditions such as ice and snow in the prior art. The steps of the control method are as follows: 1) Establishing a truck longitudinal dynamics and load transfer model, and estimating the road adhesion coefficient under ice and snow scenes by using the slope method and the recursive least squares method based on genetic factors; 2) Conducting a risk assessment on the road conditions and collision speed under ice and snow weather conditions, calculating the comprehensive risk coefficient, and adaptively adjusting the warning distance threshold; 3) Adaptively calibrating the braking distance threshold according to the estimated value of the road adhesion coefficient and the driver's reaction time, and determining the expected deceleration in the two stages of partial braking and full braking; 4) Calculating the distance between the truck and the obstacle ahead, and formulating an AEB control strategy according to its relationship with the warning distance threshold and the braking distance threshold.

Description

Technical Field

[0001] The present invention relates to a control method belonging to the technical field of automobile active safety, and more specifically, to a truck AEB adaptive control method based on ice and snow road recognition. Background Art

[0002] In extreme weather conditions, especially on snow-covered roads, trucks face severe challenges in driving safety due to their high center of mass, large size, and complex articulated structure. Snowy roads not only significantly reduce the adhesion of the road surface, causing the vehicle to easily slip and lose control, but the characteristics of the truck itself also increase the complexity and danger of driving. Therefore, in response to the special challenges of trucks on snowy roads, it has become an urgent need to develop an efficient and accurate truck active safety system to ensure road traffic safety.

[0003] In active emergency collision avoidance technology, the Automatic Emergency Braking (AEB) system is a crucial safety feature that reduces the probability and severity of accidents by predicting and automatically responding to potential collision threats. Especially for heavy vehicles such as trucks, their large size, high center of mass, and complex articulation make the performance requirements of the AEB system more stringent. However, traditional AEB control strategies are mainly based on the vehicle dynamics and collision warning models of passenger cars. The warning and braking strategies are relatively simple, and they do not take into account the characteristics of trucks and the complex road conditions they face. They lack dynamic adjustment capabilities, which will lead to the inability to effectively respond to sudden dangerous situations, or lead to unnecessary braking interventions, affecting driving safety and comfort.

[0004] Different road conditions (such as wet, icy, dry, etc.) will cause significant changes in the friction between the tire and the ground, which in turn affects the braking distance and braking response time of the AEB system. Therefore, an effective AEB control strategy needs to be able to sense the changes in the road adhesion coefficient in real time and adjust the braking parameters accordingly to ensure that sufficient braking performance can be provided under various road conditions. Existing AEB systems are mainly used in passenger cars. Although they have improved the safety of vehicles to a certain extent, most of them have failed to fully consider the impact of the road adhesion coefficient and truck characteristics. As a result, in complex and changeable actual traffic environments, the performance of the AEB system may be seriously affected, and may even cause system failure or malfunction. Summary of the invention

[0005] The technical problem to be solved by the present invention is to overcome the problems of poor driving safety of trucks and insufficient consideration of truck characteristics in severe weather conditions such as ice and snow in the prior art, and to provide a truck AEB adaptive control method based on ice and snow road recognition.

[0006] The truck motion parameters are integrated by sensors to build its longitudinal dynamics and load transfer model. k The slope method and the recursive least square method optimized by genetic factors accurately estimate the road adhesion coefficient. At the same time, risk assessment is carried out on road conditions and collision speed, and the risk is quantified by grade. According to the road adhesion coefficient and comprehensive risk, the warning threshold is intelligently adjusted, and the braking threshold is dynamically calibrated according to the adhesion coefficient and driving reaction time, and the deceleration expectation of each stage is set. Real-time distance measurement, based on the relationship between distance and threshold, accurately formulates AEB control strategy to ensure driving safety.

[0007] In order to solve the above technical problems, the present invention is implemented by adopting the following technical solutions: the truck AEB adaptive control method based on ice and snow road recognition comprises the following steps:

[0008] 1) Establish a truck longitudinal dynamics and load transfer model, and use the slope method and genetic factor-based recursive least squares method to estimate the road adhesion coefficient in ice and snow scenarios;

[0009] 2) Conduct risk assessment on road conditions and collision speed under icy and snowy weather conditions, calculate the comprehensive risk factor, and adaptively adjust the warning distance threshold;

[0010] 3) Adaptively calibrate the braking distance threshold based on the estimated road adhesion coefficient and the driver's reaction time, and determine the expected deceleration in the two stages of partial braking and full braking;

[0011] 4) Calculate the distance between the truck and the obstacle ahead, and formulate the AEB control strategy based on its relationship with the warning distance threshold and braking distance threshold.

[0012] The technical solution states that the truck longitudinal dynamics and load transfer model is established, and the road adhesion coefficient under ice and snow scenes is estimated using the slope method and the recursive least squares method based on genetic factors:

[0013] 1) Through the sensors installed on the truck, obtain the information of the truck and its external environment, including: vehicle speed, vehicle acceleration, road slope, vehicle mass, vehicle center of mass height and vehicle wheelbase, and establish a mathematical model of the truck braking process. The specific steps are as follows:

[0014] (1) Establishing the longitudinal dynamics model of the truck:

[0015] ma x =F x -R x -D A -mgsinα

[0016] R x =f·Fz

[0017]

[0018] Where: m is the mass of the truck, g is the acceleration due to gravity, and a x 、v x are the acceleration and speed of the truck, F x is the traction force, R x is the wheel rolling resistance, f is the vehicle rolling resistance coefficient, determined by the empirical formula f = 0.0076 + 0.000056v x , D A is the air resistance, α is the road slope, C d is the air resistance coefficient, A is the frontal area, and ρ is the air density;

[0019] (2) Establishing vehicle load transfer model:

[0020] Ignoring the load transfer caused by slight vehicle roll or uneven tire wear, the vertical forces on the left and right front wheels of the truck are assumed to be equal, and the vertical forces on the left and right rear wheels are equal;

[0021]

[0022]

[0023] Where: F z,fl 、F z,fr 、F z,rl 、F z,rr are the vertical forces of the left front wheel, right front wheel, left rear wheel, and right rear wheel, respectively, h g is the height of the truck's center of mass, l f and l r are the longitudinal distances from the front and rear axles of the truck to the center of mass, respectively;

[0024] (3) Calculate the slip rate when the vehicle brakes. The calculation formula is as follows:

[0025]

[0026] Where: ω is the wheel angular velocity, r is the wheel rolling radius

[0027] 2) In order to improve the control strategy accuracy of the truck in the entire motion state, μ-σ x The slope method and the recursive least squares method based on the forgetting factor are used to estimate the road adhesion coefficient in ice and snow scenes. x When ≤0.08, μ-σ x The curve is approximately a straight line, so the slope method is used to estimate the road adhesion coefficient when the slip rate is small: In the small slip rate range, there is a linear relationship μ=k·σ1·q, where k is μ-σ x The slope of the curve, F x and F z are respectively the longitudinal force and vertical force of the tire under the current slip rate, σ1 is the maximum slip rate in the linear region, which is taken as 0.08 here, and q is the proportional coefficient of the maximum road adhesion coefficient to the peak adhesion coefficient in the linear region, which is generally taken as 1.2-1.4;

[0028] 3) When the slip rate σ x When it is greater than 0.08, it is the saturation region. At this time, the wheels will lock or spin, and the tire force will reach saturation. The road adhesion coefficient estimation method in the linear region is no longer suitable. The vehicle traction can be expressed as F x =μ·F z =μmg, combined with the vehicle longitudinal dynamics equation, we get This can be organized into the basic form of recursive least squares method based on the forgetting factor: System output observation information System input observation information The parameter to be identified is θ(t)=μ, so as to obtain the estimated value of the road adhesion coefficient. The operation process of the recursive least squares method based on the forgetting factor is as follows:

[0029] (1) Initialize the parameter vector θ(0), the covariance matrix P(0), and set the forgetting factor λ. Usually, θ(0) is set to 0 and P(0) is set to 10. 6 , the value range of λ is 0 to 1;

[0030] (2) Based on the system input and state at the current time step, measure the system output y(t) and calculate the regression matrix

[0031] (3) Calculate the estimation error e(t), the actual system output y(t) and the estimated value θ(t-1) based on the current parameter and the regression matrix The difference between the calculated predicted outputs;

[0032] (4) Calculate the gain matrix

[0033] (5) Gradually adjust and reduce the uncertainty of parameter estimation and update the covariance matrix Where: I is the unit matrix;

[0034] (6) Update the estimated parameter vector: θ(t) = θ(t-1) + K(t)e(t);

[0035] (7) In each consecutive time step t, steps (2) to (6) are repeated to continuously update the parameter estimates.

[0036] The technical solution describes the risk assessment of road conditions and collision speed under icy and snowy weather conditions, calculation of the comprehensive risk factor, and adaptive adjustment of the warning distance threshold:

[0037] 1) According to the degree of snow accumulation, compaction and ice coverage on the icy and snowy road surface, the road conditions are divided into four levels: new snow, semi-melted, and ice plate; combined with expert experience and the historical accident data ratio RCT (Ratio of Crash in Total, RCT) of various icy and snowy road conditions, a risk assessment is conducted on each level of road conditions, and the corresponding risk level (low risk, medium risk, high risk) and weight are assigned; at the same time, according to the "C-NCAP Management Rules" and GB / T 33195-2016 standards, combined with traffic accident data and truck characteristics, the risk level and weight under different collision speeds are determined; the risk level and weight are determined as shown in the following table:

[0038] Table 1 Road surface risk level and weight determination

[0039]

[0040] Table 2 Collision speed risk level and weight determination

[0041]

[0042] 2) Comprehensively consider the impact of road conditions and collision speed on vehicle safety and introduce the comprehensive risk factor R:

[0043] R=ω r ×Risk level of road adhesion coefficient + ω c ×Risk level of collision speed

[0044] 3) Obtain the estimated value of the road adhesion coefficient and the comprehensive risk coefficient R, adaptively adjust the warning distance threshold to ensure the effectiveness of the warning mechanism; obtain the sound warning distance threshold D w,max Distance threshold D from active seat belt tightening warning w,min , the calculation steps are as follows:

[0045] (1) First, obtain the safe stopping distance, the formula is:

[0046]

[0047] (2) The two-stage warning distance threshold is obtained based on the driver's reaction time and risk coefficient to different warning forms. The calculation formula of the two-stage warning distance threshold is as follows:

[0048]

[0049] Where: t dis the driver's reaction time to different warning forms (the driver's reaction time to the sound warning is 1.2 seconds, and the reaction time to the active seat belt tightening is 0.7 seconds); t s is the brake action time; t w is the advance warning time, which is used to warn the driver of the potential collision risk ahead, and is usually about 1 second; a1 and v1 are the acceleration and speed of the vehicle ahead, respectively; ξ is the adjustment coefficient, which is used to control the influence of the risk level on the warning distance threshold, and is taken as 0.1.

[0050] The technical solution states that the braking distance threshold is adaptively calibrated according to the estimated road adhesion coefficient and the driver's reaction time, and the expected deceleration in the two stages of partial braking and full braking is determined as follows:

[0051] 1) Collect historical data on the driver's reaction time distribution under different road conditions and determine the distribution characteristics, as shown in the following table:

[0052] Table 3 Driver reaction time

[0053]

[0054] 2) Determine the time t required from the start of braking to the full reaching of the set braking pressure p According to empirical data, t p,min =0.3, t p,max =0.75;

[0055] 3) Based on the driver's reaction time and the time required for the brake to reach the set braking pressure, the braking distance threshold is adaptively adjusted to ensure the safety of the braking mechanism; the partial braking distance threshold and the full braking distance threshold are obtained, and the calculation steps are as follows:

[0056] (1) When t r,max =1.17s, t p,max =0.75, the partial braking distance threshold is:

[0057] D s,max =D b +t r,max ×v x +t p,max ×v x ;

[0058] (2) When t r,min =0.74s, t p,min =0.3, the full braking distance threshold is:

[0059] D s,min =D b +t r,min ×v x +tp,min ×v x ;

[0060] 4) Determine the expected deceleration when the two-stage braking is triggered based on existing research data, as follows:

[0061] The National Highway Traffic Safety Administration (NHTSA) of the United States evaluated the braking performance of 83 drivers on dry and flat roads in emergency situations. The braking deceleration distribution (μ = 0.85) is shown in the following table:

[0062] Table 4 Driver braking deceleration statistics

[0063]

[0064] Therefore, taking into full consideration the driving safety and riding comfort in ice and snow scenes, the expected deceleration a1 of partial braking is determined to be 1.7 m / s 2 , the expected deceleration a2 of full braking is 2.8m / s 2 .

[0065] The calculation of the distance between the truck and the obstacle ahead and the formulation of the AEB control strategy based on the relationship between the distance between the truck and the warning distance threshold and the braking distance threshold described in the technical solution refer to:

[0066] 1) Calculate the collision distance between the vehicle and the vehicle ahead based on the current estimated road adhesion coefficient and vehicle status information, as follows:

[0067] d=D b +D r +D p

[0068]

[0069]

[0070] D r =t r ×v x

[0071] D p =t p ×v x

[0072] Where: d is the relative distance between the vehicle and the obstacle vehicle in front during driving, D b is the braking distance of the vehicle to decelerate to zero, D r is the driver’s reaction time t r Distance traveled within, D p The vehicle's brake reaction time t pThe distance traveled within the vehicle; γ is the equivalent mass coefficient, which is 1.03 for passenger cars and 1.05 for trucks; η b is the braking efficiency;

[0073] 2) According to the relationship between the obtained collision distance and the warning distance threshold and the braking distance threshold, the collision risk is evaluated. When d>D w,max When the vehicle is in a safe state, the AEB system maintains silent monitoring; when D w,min <d≤D w,max When the AEB system intervenes and issues a warning sound, it gives the driver a first-level warning with an auditory signal; when D s,max <d≤D w,min When D s,min <d≤D s,max When d≤D s,min When the AEB system enters full braking mode, it brakes with full force at a braking deceleration of a2.

[0074] Compared with the prior art, the present invention has the following beneficial effects:

[0075] 1. The truck AEB adaptive control method based on ice and snow road recognition described in the present invention uses high-precision sensors to collect the motion state parameters of the truck in real time, constructs the truck longitudinal dynamics model and load transfer model, and combines μ-σ k The slope method and the recursive least squares method based on genetic factors can realize real-time estimation of the road adhesion coefficient.

[0076] 2. The truck AEB adaptive control method based on ice and snow road recognition described in the present invention is based on expert experience and historical data, conducts risk assessment on road adhesion and collision speed, divides risk levels and determines weights, and dynamically and adaptively adjusts the warning and braking distance thresholds according to the road adhesion coefficient and the comprehensive risk coefficient. The sound warning distance threshold and the active seat belt tightening warning distance threshold are used as benchmarks to achieve differentiated safety warning strategies.

[0077] 3. The truck AEB adaptive control method based on ice and snow road recognition described in the present invention further calibrates the distance thresholds of partial braking and full braking according to the historical data of road adhesion coefficient and driver reaction time, and clarifies the braking deceleration of each stage of braking, ensuring that the braking process is both rapid and stable, overcoming the problem of easy loss of control when braking on ice and snow roads. Finally, by calculating the distance between the truck and the obstacle ahead in real time, an adaptive automatic emergency braking (AEB) control strategy is formulated to comprehensively improve the driving safety and braking efficiency of trucks in ice and snow weather. BRIEF DESCRIPTION OF THE DRAWINGS

[0078] The present invention will be further described below in conjunction with the accompanying drawings:

[0079] Figure 1 A flowchart of a truck AEB adaptive control method based on ice and snow road recognition according to the present invention;

[0080] Figure 2 A schematic diagram of the area division of the whole vehicle driving process of a truck AEB adaptive control method based on ice and snow road recognition according to the present invention;

[0081] Figure 3 is an estimated value of the road adhesion coefficient when the road adhesion coefficient of the truck AEB adaptive control method based on ice and snow road recognition described in the present invention is 0.2;

[0082] Figure 4 It is an estimated value of the road adhesion coefficient when the road adhesion coefficient of the truck AEB adaptive control method based on ice and snow road recognition described in the present invention is 0.4. DETAILED DESCRIPTION

[0083] The present invention is described in detail below in conjunction with the accompanying drawings:

[0084] See also Figure 1 The truck AEB adaptive control method based on ice and snow road recognition comprises the following steps:

[0085] 1. Establish a truck longitudinal dynamics model and a load transfer model, and use the slope method and the recursive least squares method based on genetic factors to estimate the road adhesion coefficient in ice and snow scenarios; the steps are as follows:

[0086] 1) Through the sensors installed on the truck, obtain the information of the truck and its external environment, including: vehicle speed, vehicle acceleration, road slope, vehicle mass, vehicle center of mass height and vehicle wheelbase, and establish a mathematical model of the truck braking process. The specific steps are as follows:

[0087] (1) Establishing the longitudinal dynamics model of the truck:

[0088] ma x =F x -R x -D A -mgsinα

[0089] R x =f·F z

[0090]

[0091] Where: m is the mass of the truck, g is the acceleration due to gravity, and a x 、v x are the acceleration and speed of the truck, F x is the traction force, R x is the wheel rolling resistance, f is the vehicle rolling resistance coefficient, determined by the empirical formula f = 0.0076 + 0.000056v x , D A is the air resistance, α is the road slope, C d is the air resistance coefficient, A is the frontal area, and ρ is the air density.

[0092] (2) Establishing vehicle load transfer model:

[0093] Ignoring the load transfer caused by slight vehicle roll or uneven tire wear, the vertical forces on the left and right front wheels of the truck are assumed to be equal, and the vertical forces on the left and right rear wheels are equal;

[0094]

[0095]

[0096] Where: F z,fl 、F z,fr 、F z,rl 、F z,rr are the vertical forces of the left front wheel, right front wheel, left rear wheel, and right rear wheel, respectively, h g is the height of the truck's center of mass, l f and l r are the longitudinal distances from the front and rear axles of the truck to the center of mass, respectively;

[0097] (3) Calculate the slip rate when the vehicle brakes. The calculation formula is as follows:

[0098]

[0099] Where: ω is the wheel angular velocity, r is the wheel rolling radius

[0100] 2) In order to improve the control strategy accuracy of the truck in the entire motion state, μ-σ x The slope method and the recursive least squares method based on the forgetting factor are used to estimate the road adhesion coefficient in ice and snow scenes. x When ≤0.08, μ-σ x The curve is approximately a straight line, so the slope method is used to estimate the road adhesion coefficient when the slip rate is small: In the small slip rate range, there is a linear relationship μ=k·σ1·q, where k is μ-σ x The slope of the curve, F x and F zare respectively the longitudinal force and vertical force of the tire under the current slip rate, σ1 is the maximum slip rate in the linear region, which is taken as 0.08 here, and q is the proportional coefficient of the maximum road adhesion coefficient to the peak adhesion coefficient in the linear region, which is generally taken as 1.2-1.4;

[0101] 3) When the slip rate σ x When it is greater than 0.08, it is the saturation region. At this time, the wheels will lock or spin, and the tire force will reach saturation. The road adhesion coefficient estimation method in the linear region is no longer suitable. The vehicle traction can be expressed as F x =μ·F z =μmg, combined with the vehicle longitudinal dynamics equation, we get This can be organized into the basic form of recursive least squares method based on the forgetting factor: System output observation information System input observation information The parameter to be identified is θ(t)=μ, so as to obtain the estimated value of the road adhesion coefficient. The operation process of the recursive least squares method based on the forgetting factor is as follows:

[0102] (1) Initialize the parameter vector θ(0), the covariance matrix P(0), and set the forgetting factor λ. Usually, θ(0) is set to 0 and P(0) is set to 10. 6 , the value range of λ is 0 to 1.

[0103] (2) Based on the system input and state at the current time step, measure the system output y(t) and calculate the regression matrix

[0104] (3) Calculate the estimation error e(t), the actual system output y(t) and the estimated value θ(t-1) based on the current parameter and the regression matrix The difference between the calculated predicted outputs;

[0105] (4) Calculate the gain matrix

[0106] (5) Gradually adjust and reduce the uncertainty of parameter estimation and update the covariance matrix Where: I is the unit matrix.

[0107] (6) Update the estimated parameter vector: θ(t) = θ(t-1) + K(t)e(t);

[0108] (7) Repeat steps (2) to (6) in each consecutive time step t, thereby continuously updating the parameter estimates;

[0109] 2. See Figure 2, conduct risk assessment on road conditions and collision speed under icy and snowy weather conditions, calculate comprehensive risk coefficient, and adaptively adjust warning distance threshold; the steps are as follows:

[0110] 1) According to the degree of snow accumulation, compaction and ice coverage on the icy and snowy road surface, the road conditions are divided into four levels: new snow, semi-melted, and ice plate; combined with expert experience and the historical accident data ratio RCT (Ratio of Crash in Total, RCT) of various icy and snowy road conditions, a risk assessment is conducted on each level of road conditions, and the corresponding risk level (low risk, medium risk, high risk) and weight are assigned; at the same time, according to the "C-NCAP Management Rules" and GB / T 33195-2016 standards, combined with traffic accident data and truck characteristics, the risk level and weight under different collision speeds are determined; the risk level and weight are determined as shown in the following table:

[0111] Table 1 Road surface risk level and weight determination

[0112]

[0113] Table 2 Collision speed risk level and weight determination

[0114]

[0115] 2) Comprehensively consider the impact of road conditions and collision speed on vehicle safety and introduce the comprehensive risk factor R:

[0116] R=ω r ×Risk level of road adhesion coefficient + ω c ×Risk level of collision speed

[0117] 3) Obtain the estimated value of the road adhesion coefficient and the comprehensive risk coefficient R, adaptively adjust the warning distance threshold to ensure the effectiveness of the warning mechanism; obtain the sound warning distance threshold D w,max Distance threshold D from active seat belt tightening warning w,min , the calculation steps are as follows:

[0118] (1) First, obtain the safe stopping distance, the formula is:

[0119]

[0120] (2) The two-stage warning distance threshold is obtained based on the driver's reaction time and risk coefficient to different warning forms. The calculation formula of the two-stage warning distance threshold is as follows:

[0121]

[0122] Where: t dis the driver's reaction time to different warning forms (the driver's reaction time to the sound warning is 1.2 seconds, and the reaction time to the active seat belt tightening is 0.7 seconds); t s is the brake action time; t w is the advance warning time, which is used to warn the driver of the potential collision risk ahead, and is usually about 1 second; a1 and v1 are the acceleration and speed of the front vehicle respectively; ξ is the adjustment coefficient, which is used to control the influence of the risk level on the warning distance threshold, and is taken as 0.1;

[0123] 3. Adaptively calibrate the braking distance threshold according to the estimated road adhesion coefficient and the driver's reaction time, and determine the expected deceleration in the two stages of partial braking and full braking; the steps are as follows:

[0124] 1) Collect historical data on the driver's reaction time distribution under different road conditions and determine the distribution characteristics, as shown in the following table:

[0125] Table 3 Driver reaction time

[0126]

[0127] 2) Determine the time t required from the start of braking to the full reaching of the set braking pressure p According to empirical data, t p,min =0.3, t p,max =0.75.

[0128] 3) Based on the driver's reaction time and the time required for the brake to reach the set braking pressure, the braking distance threshold is adaptively adjusted to ensure the safety of the braking mechanism; the partial braking distance threshold and the full braking distance threshold are obtained, and the calculation steps are as follows:

[0129] (1) When t r,max =1.17s, t p,max =0.75, the partial braking distance threshold is:

[0130] D s,max =D b +t r,max ×v x +t p,max ×v x ;

[0131] (2) When t r,min =0.74s, t p,min =0.3, the full braking distance threshold is:

[0132] D s,min =D b +t r,min ×v x +tp,min ×v x ;

[0133] 4) Determine the expected deceleration when the two-stage braking is triggered based on existing research data, as follows:

[0134] The National Highway Traffic Safety Administration (NHTSA) of the United States evaluated the braking performance of 83 drivers on dry and flat roads in emergency situations. The braking deceleration distribution (μ = 0.85) is shown in the following table:

[0135] Table 4 Driver braking deceleration statistics

[0136]

[0137] Therefore, taking into full consideration the driving safety and riding comfort in ice and snow scenes, the expected deceleration a1 of partial braking is determined to be 1.7 m / s 2 , the expected deceleration a2 of full braking is 2.8m / s 2 ;

[0138] 4. Calculate the distance between the truck and the obstacle ahead, and formulate an AEB control strategy based on its relationship with the warning distance threshold and the braking distance threshold. The steps are as follows:

[0139] 1) Calculate the collision distance between the vehicle and the vehicle ahead based on the current estimated road adhesion coefficient and vehicle status information, as follows:

[0140] d=D b +D r +D p

[0141]

[0142]

[0143] D r =t r ×v x

[0144] D p =t p ×v x

[0145] Where: d is the relative distance between the vehicle and the obstacle vehicle in front during driving, D b is the braking distance of the vehicle to decelerate to zero, D r is the driver’s reaction time t r Distance traveled within, D p The vehicle's brake reaction time t pThe distance traveled within the vehicle; γ is the equivalent mass coefficient, which is 1.03 for passenger cars and 1.05 for trucks; η b is the braking efficiency;

[0146] 2) According to the relationship between the obtained collision distance and the warning distance threshold and the braking distance threshold, the collision risk is evaluated. When d>D w,max When the vehicle is in a safe state, the AEB system maintains silent monitoring; when D w,min <d≤D w,max When the AEB system intervenes and issues a warning sound, it gives the driver a first-level warning with an auditory signal; when D s,max <d≤D w,min When D s,min <d≤D s,max When d≤D s,min When the AEB system enters full braking mode, it brakes with full force at a braking deceleration of a2.

[0147] Example:

[0148] Based on Trucksim software, an emergency condition with the front vehicle stationary is constructed, and the proposed AEB collision avoidance algorithm is analyzed and verified by the truck dynamics model constructed in conjunction with Matlab software. The truck included in Trucksim is used as the simulation object to determine the truck information and the surrounding environment information of the vehicle. The main truck dynamics parameters are shown in the following table:

[0149] Table 5 Vehicle parameter settings

[0150]

[0151] Based on the above vehicle parameters, the vehicle dynamics model and vehicle load transfer model are established, and the formula is used To calculate the truck slip ratio.

[0152] In order to verify the effectiveness of estimating the road adhesion coefficient by the recursive least squares method based on genetic factors, simulations were performed on ice and snow roads with road adhesion coefficients of 0.2 and 0.4, respectively. The initial vehicle speed was set to 30 km / h. The estimated results of the road adhesion coefficient are shown in the figure below: Figure 3 , Figure 4 As shown, by comparing the estimated value of the road adhesion coefficient with the actual value, it can be seen that the estimated value is basically consistent with the actual value.

[0153] According to Table 1 and Table 2, the road conditions and collision speed are evaluated, the corresponding risk level and weight are determined, and the formula R = ω r×Risk level of road adhesion coefficient + ω c × The risk level of the collision speed is used to calculate the comprehensive risk factor. For example, if a truck is traveling at 40 km / h on a freshly snowed road (μ = 0.4), the comprehensive risk factor of the truck is 1.3.

[0154] Table 1 Road surface risk level and weight determination

[0155]

[0156] Table 2 Collision speed risk level and weight determination

[0157]

[0158] According to the estimated value of the road adhesion coefficient and the historical data of the driver's reaction time distribution, the braking distance threshold is adaptively calibrated to obtain the partial braking distance threshold D s,max and full braking distance threshold D s,min ; and determine the expected deceleration a1 of partial braking to be 1.7m / s 2 The expected deceleration a2 for full braking is 2.8 m / s 2 ; Calculate the distance between the truck and the obstacle ahead, and formulate an AEB control strategy based on the relationship between the obtained distance and the warning distance threshold and braking distance threshold.

[0159] The simulation conditions are constructed based on the "Technical Specifications and Performance Test Requirements for Commercial Vehicle Automatic Emergency Braking System (AEBS) Aftermarket" to verify the collision avoidance performance of the truck AEB system under different road conditions. The simulation results are shown in the table below:

[0160] Table 6 AEB intervention timing under different road conditions

[0161]

[0162] Taking the first row of data in Table 6 as an example for result analysis, at the beginning of the simulation, the initial distance between the two vehicles on the icy road (μ = 0.15) is 100m, and the truck approaches the stationary obstacle vehicle in front at a constant speed of 20km / h. When the two vehicles are 46.35m apart, the AEB system intervenes and issues a warning sound to the driver with an auditory signal. When the two vehicles are 42.68m apart, the AEB system automatically enters the second-level warning stage, and the seat belt pretensioner is quickly activated to tighten the seat belt. When the two vehicles are 14.94m apart, the AEB system starts the partial braking mode and brakes at a preset deceleration of 1.7m / s 2 Slow down; when the two vehicles are 10.16m apart, the AEB system immediately enters full braking mode with a braking deceleration of 2.8m / s 2 After full braking, the truck slowed down to zero speed with a distance of 7.43m between the two vehicles, thus successfully avoiding a collision.

[0163] From the above simulation results, it can be seen that when a truck approaches a stationary target in front at different speeds under different road conditions, the AEB system designed in this study determines the timing of graded warning and autonomous braking intervention based on the current estimated road adhesion coefficient and the truck speed, ensuring that the truck decelerates to zero, has a good braking effect, and avoids the occurrence of collisions.

Claims

1. A truck AEB adaptive control method based on ice and snow road recognition, characterized in that: The truck AEB adaptive control method based on ice and snow road recognition comprises the following steps: 1) Establish a truck longitudinal dynamics and load transfer model, and use the slope method and the recursive least squares method based on the forgetting factor to estimate the road adhesion coefficient in ice and snow scenarios; 2) Conduct risk assessment on road conditions and collision speed under icy and snowy weather conditions, calculate the comprehensive risk factor, and adaptively adjust the warning distance threshold; 3) Adaptively calibrate the braking distance threshold based on the estimated road adhesion coefficient and the driver's reaction time, and determine the expected deceleration in the two stages of partial braking and full braking; 4) Calculate the distance between the truck and the obstacle ahead, and formulate the AEB control strategy based on its relationship with the warning distance threshold and braking distance threshold; The risk assessment of the road conditions and collision speed under icy and snowy weather conditions, calculation of the comprehensive risk coefficient, and adaptive adjustment of the warning distance threshold refers to: 1) According to the degree of snow accumulation, compaction and ice coverage on the icy and snowy road surface, the road conditions are divided into three levels: new snow, semi-melted, and ice plate; combined with expert experience and the historical accident data ratio RCT (Ratio of Crash in Total, RCT) of various icy and snowy road conditions, risk assessment is conducted on each level of road conditions, and corresponding risk levels and weights are assigned; at the same time, according to the "C-NCAP Management Rules" and GB / T 33195-2016 standards, combined with traffic accident data and truck characteristics, the risk levels and weights at different collision speeds are determined; the risk levels and weights are determined as shown in the following table: Table 1 Road surface risk level and weight determination Table 2 Collision speed risk level and weight determination 2) Comprehensively consider the impact of road conditions and collision speed on vehicle safety and introduce the comprehensive risk factor R: R = ω r ×Risk level of road conditions + ω c ×Risk level of collision speed 3) Obtain the estimated value of the road adhesion coefficient and the comprehensive risk coefficient R, adaptively adjust the warning distance threshold to ensure the effectiveness of the warning mechanism; obtain the sound warning distance threshold D w,max Distance threshold D from active seat belt tightening warning w,min , the calculation steps are as follows: (1) First, obtain the safe stopping distance, the formula is: (2) The two-stage warning distance threshold is obtained based on the driver's reaction time to different warning forms and the comprehensive risk coefficient. The two-stage warning distance threshold calculation formula is as follows: Where: t d is the driver's reaction time to different warning forms. The driver's reaction time to the sound warning is 1.2 seconds, and the reaction time to the active seat belt tightening is 0.7 seconds; t s is the brake action time; t w is the advance warning time, used to warn the driver of the potential collision risk ahead, and is 1 second; a1 and v1 are the acceleration and speed of the vehicle ahead, respectively; a x 、v x are the truck acceleration and speed respectively; ξ is the adjustment coefficient, which is used to control the influence of risk level on the warning distance threshold and is taken as 0.

1.

2. A truck AEB adaptive control method based on ice and snow road recognition according to claim 1, characterized in that: The establishment of the truck longitudinal dynamics and load transfer model, and the estimation of the road adhesion coefficient in the ice and snow scene using the slope method and the recursive least squares method based on the forgetting factor refer to: 1) Through the sensors installed on the truck, obtain the information of the truck and its external environment, including: vehicle speed, vehicle acceleration, road slope, vehicle mass, vehicle center of mass height and vehicle wheelbase, and establish a mathematical model of the truck braking process. The specific steps are as follows: (1) Establishing the longitudinal dynamics model of the truck: ma x =F x -R x -D A -mgsinα R x =f·F z Where: m is the mass of the truck, g is the acceleration due to gravity, and F x is the traction force, R x is the wheel rolling resistance, f is the vehicle rolling resistance coefficient, determined by the empirical formula f = 0.0076 + 0.000056v x , D A is the air resistance, α is the road slope, C d is the air resistance coefficient, A is the frontal area, and ρ is the air density; (2) Establish a vehicle load transfer model: Ignoring the load transfer caused by slight vehicle roll or uneven tire wear, the vertical forces on the left and right front wheels of the truck are assumed to be equal, and the vertical forces on the left and right rear wheels are equal; Where: F z,fl 、F z,fr 、F z,rl 、F z,rr are the vertical forces of the left front wheel, right front wheel, left rear wheel, and right rear wheel, respectively, h g is the height of the truck's center of mass, l f and l r are the longitudinal distances from the front and rear axles of the truck to the center of mass, respectively; (3) Calculate the slip rate when the vehicle brakes. The calculation formula is as follows: Where: ω is the wheel angular velocity, r is the wheel rolling radius; 2) In order to improve the control strategy accuracy of the truck in the entire motion state, μ-σ x The slope method and the recursive least squares method based on the forgetting factor are used to estimate the road adhesion coefficient in ice and snow scenes. x When ≤0.08, μ-σ x The curve is approximately a straight line, so the slope method is used to estimate the road adhesion coefficient when the slip rate is small: In the small slip rate range, there is a linear relationship μ=k·σ1·q, where k is μ-σ x The slope of the curve, F x and F z are the longitudinal force and vertical force of the tire under the current slip rate, σ1 is the maximum slip rate in the linear region, which is taken as 0.08, and q is the proportional coefficient of the maximum road adhesion coefficient to the peak adhesion coefficient in the linear region, which ranges from 1.2 to 1.4; 3) When the slip rate σ x When it is greater than 0.08, it is the saturation region. At this time, the wheels will lock or spin, and the tire force will reach saturation. The road adhesion coefficient estimation method in the linear region is no longer suitable. The vehicle traction is expressed as F x =μ·F z =μmg, combined with the vehicle longitudinal dynamics equation, we get This can be organized into the basic form of recursive least squares method based on the forgetting factor: System output observation information System input observation information The parameter to be identified is θ(t)=μ, so as to obtain the estimated value of the road adhesion coefficient. The operation process of the recursive least squares method based on the forgetting factor is as follows: (1) Initialize the parameter vector θ(0), the covariance matrix P(0), set the forgetting factor λ, set θ(0) to 0, and P(0) to 10 6 , the value range of λ is 0 to 1; (2) Based on the system input and state at the current time step, measure the system output y(t) and calculate the regression matrix (3) Calculate the estimation error e(t), the actual system output y(t) and the estimated value θ(t-1) based on the current parameter and the regression matrix The difference between the calculated predicted outputs; (4) Calculate the gain matrix (5) Gradually adjust and reduce the uncertainty of parameter estimation and update the covariance matrix Where: I is the unit matrix; (6) Update the estimated parameter vector: θ(t) = θ(t-1) + K(t)e(t); (7) In each consecutive time step t, steps (2) to (6) are repeated to continuously update the parameter estimates.

3. A truck AEB adaptive control method based on ice and snow road recognition according to claim 2, characterized in that: The adaptive calibration of the braking distance threshold according to the estimated road adhesion coefficient and the driver's reaction time, and the determination of the expected deceleration in the two stages of partial braking and full braking, refers to: 1) Collect historical data on the driver's reaction time distribution under different road conditions and determine the distribution characteristics, as shown in the following table: Table 3 Driver reaction time 2) Determine the time t required from the start of braking to the full reaching of the set braking pressure p According to empirical data, t p,min =0.3s, t p,max =0.75s; 3) Based on the driver's reaction time and the time required for the brake to reach the set braking pressure, the braking distance threshold is adaptively adjusted to ensure the safety of the braking mechanism; the partial braking distance threshold and the full braking distance threshold are obtained, and the calculation steps are as follows: (1) When t r,max =1.17s, t p,max = 0.75s, the partial braking distance threshold is: D s,max =D b +t r,max ×v x +t p,max ×v x ; (2) When t r,min =0.74s, t p,min = 0.3s, the complete braking distance threshold is: D s,min =D b +t r,min ×v x +t p,min ×v x ; 4) Determine the expected deceleration when the two-stage braking is triggered based on existing research data, as follows: The braking deceleration distribution measured according to the braking performance of the driver in an emergency on a dry and flat road, where μ = 0.85, is shown in the following table: Table 4 Driver braking deceleration statistics Therefore, taking into full consideration the driving safety and riding comfort in ice and snow scenes, the expected deceleration a1 of partial braking is determined to be 1.7 m / s 2 , the expected deceleration a2 of full braking is 2.8m / s 2 ; The calculation of the distance between the truck and the obstacle ahead and the formulation of the AEB control strategy based on the relationship between the distance between the truck and the warning distance threshold and the braking distance threshold refer to: 1) Calculate the collision distance between the vehicle and the vehicle ahead based on the current estimated road adhesion coefficient and vehicle status information, as follows: d=D b +D r +D p D r =t r ×v x D p =t p ×v x Where: d is the relative distance between the vehicle and the obstacle vehicle in front during driving, D b is the braking distance of the vehicle to decelerate to zero, D r is the driver’s reaction time t r Distance traveled within, D p The vehicle's brake reaction time t p The distance traveled within the vehicle; γ is the equivalent mass coefficient, which is 1.03 for passenger cars and 1.05 for trucks; η b is the braking efficiency; 2) According to the relationship between the obtained collision distance and the warning distance threshold and the braking distance threshold, the collision risk is evaluated. When d>D w,max When the vehicle is in a safe state, the AEB system maintains silent monitoring; when D w,min <d≤D w,max When the AEB system intervenes and issues a warning sound, it gives the driver a first-level warning with an auditory signal; when D s,max <d≤D w,min When D s,min <d≤D s,max When d≤D s,min When the AEB system enters full braking mode, it brakes with full force at a braking deceleration of a2.

Images