Automobile self-adaptive cruise system, control method, control device, medium and vehicle

Abstract

The invention provides an automobile self-adaptive cruise system, a control method, a control device, a medium and a vehicle, the automobile self-adaptive cruise system comprises a road surface preview system, a road surface parameter updating system, a defensive self-adaptive cruise controller, a driving actuator, a braking actuator and sensors, and the sensors comprise a wheel speed sensor and a combined inertial navigation. The wheel speed sensor acquires the wheel rotating speed; the combined inertial navigation obtains the longitudinal speed and acceleration of the vehicle. Due to the fact that the defensive self-adaptive cruise control strategy can consider severe road surface conditions in advance, obvious speed or torque step changes are not generated when the vehicle runs in road surface transition areas with different attachment conditions, and control is smooth.

Description

Technical Field

[0001] The invention relates to the field of control technology, in particular to an adaptive cruise control of an automobile. Background Art

[0002] The Adaptive Cruise Control (ACC) system is an assisted driving system that can automatically adjust the cruising speed and maintain a safe distance from the vehicle in front. When following a vehicle at a close distance, a sudden change in road conditions is likely to cause the drive wheels to spin, causing the vehicle to become unstable and causing the performance of the adaptive cruise system to deteriorate, which may reduce the accuracy of the distance control between vehicles and even cause the risk of rear-end collision. Therefore, it is necessary to identify the road conditions more accurately and modify the ACC strategy accordingly.

[0003] At present, some studies have considered the impact of road surface parameter changes on the ACC longitudinal control system. According to different identification methods, ACC that takes into account road surface parameter changes can be classified into 1) ACC based on result identification method and 2) ACC based on cause identification method.

[0004] The result-based identification method refers to evaluating the road conditions by analyzing the changes in vehicle dynamic response caused by changes in road conditions, and using a system identification method. This method mainly establishes the relationship between the road adhesion coefficient and the vehicle dynamic response through the tire model. For example, Hu et al. (Hu J, Rakheja S, Zhang Y. Real-time estimation of tire-road friction coefficient based on lateral vehicle dynamics [J]. Proceedings of the lnstitution of Mechanical Engineers, Part D: Journal of Automobile Engineering, 2020, 234 (10-11): 2444-2457) established an extended Kalman filter and a double unscented Kalman filter cascade estimator based on a brush tire model, and Yang Xinjing (Yang Xinjing. Research on automobile adaptive cruise control strategy considering road adhesion coefficient [D]. Chongqing Jiaotong University, 2023) used the Dugoff tire model to estimate the road adhesion coefficient and used a hierarchical architecture to design the ACC control system.

[0005] The above result-based identification method has the following limitations: 1) It needs to meet the continuous excitation condition, that is, the vehicle must run on the road to be identified and cause changes in the vehicle dynamics response. Therefore, this method cannot predict the road conditions in areas where the vehicle has not entered, and lacks an assessment of potential severe road conditions. 2) When the estimated value is directly used for controller design, the fluctuation or overshoot phenomenon when the estimated value does not converge will significantly affect the longitudinal control performance; in some sudden harsh environments, it is difficult to quickly and accurately evaluate the sudden changes in road adhesion conditions, so there is a lack of adaptability to sudden scenarios. 3) The above method only considers the change in the road adhesion coefficient, but ignores the influence of the rolling resistance coefficient on the longitudinal dynamics of the vehicle.

[0006] The cause-based identification method refers to a method of evaluating road conditions by identifying the physical characteristics of the road surface through sensors such as cameras and lidar. According to the type of sensor, this method includes a single sensor-based solution and a multi-sensor fusion-based solution. The mainstream solution based on a single sensor is to use a camera to identify road features. For example, Liu Jiemei (Liu Jiemei. Research on the longitudinal speed planning and control strategy of intelligent vehicles based on road adhesion coefficient estimation [D]. Jilin University, 2023) estimates the road adhesion coefficient based on machine vision, designs an adaptive safety distance model considering the adhesion coefficient, and uses a model predictive controller (MPC) to generate the desired control law. Solutions based on multi-sensor fusion often use sensors such as cameras, lidar, and microphones to obtain multi-source information, and perform multi-source feature data fusion to identify the road adhesion coefficient, giving full play to the advantages of multiple sensors, such as the research of Holzmann et al. (Holzmann F, Bellino M, Siegwart R, et al. Predictive estimation of the road-tire friction coefficient [C] / / 2006 IEEE conference on computer aided control system design, 2006 IEEE international conference on control applications, 2006 IEEE international symposium on intelligent control. IEEE, 2006: 885-890).

[0007] Although the cause-based identification method can predict the road conditions of the road ahead, it also cannot guarantee the absolute correctness of the identification of the road adhesion coefficient or rolling resistance coefficient. On the one hand, this solution can only identify the road type, but lacks accurate estimation of adhesion coefficients, etc. On the other hand, if this solution is combined with the result-based identification method, although it can achieve accurate real-time estimation of adhesion coefficients, etc. to a certain extent, once there are overshoot, fluctuation and other estimation inaccuracies in extremely harsh environments, it will significantly affect the ACC control performance, so there is still a lack of response measures for emergency events. Summary of the invention

[0008] The current ACC system is difficult to adapt to sudden harsh environmental scenarios, and lacks effective preventive measures for inaccurate identification and control instability under potential harsh road conditions. To this end, the present invention proposes a defensive adaptive cruise system, a control method, a control device, a medium and a vehicle. The defensive adaptive cruise system is mainly composed of a road surface preview system, a road surface parameter update system, and a defensive adaptive cruise controller. The defensive adaptive cruise control method proposed in the present invention always considers the impact of potential harsh road factors and road surface parameter identification errors on longitudinal control performance, which can improve the control accuracy and stability of the vehicle in harsh environmental scenarios, and can gradually reduce the conservatism of defensive driving behavior.

[0009] In order to solve the above technical problems, the technical solution adopted by the present invention is:

[0010] The present invention first provides a defensive adaptive cruise system, comprising:

[0011] Wheel speed sensor, to obtain wheel speed;

[0012] Combined inertial navigation, obtains the longitudinal velocity and acceleration of the vehicle;

[0013] The road preview system obtains real-time collected road image information and outputs a conservative estimate of the adhesion coefficient of the harsh road ahead and rolling resistance coefficient conservatively estimated The road parameter update system consists of a road parameter estimator and a parameter update mechanism; the road parameter estimator obtains the longitudinal speed v and the driving wheel speed ω of the vehicle, and outputs the estimated value of the road adhesion coefficient and rolling resistance coefficient estimates And the residual bounds of the road adhesion coefficient estimation and rolling resistance coefficient estimation residual bounds The parameter update mechanism is based on the conservative estimate of the adhesion coefficient output by the road preview system. and rolling resistance coefficient conservatively estimated and the estimated road adhesion coefficient estimated online by the road parameter estimator Estimated rolling resistance coefficient Residual bounds of road adhesion coefficient estimation and rolling resistance coefficient estimation residual bounds And combined with the areas where different road types are located, determine the worst road adhesion coefficient that should be considered at this time and rolling resistance coefficient Defensive adaptive cruise control, obtains the estimated road adhesion coefficient Estimated rolling resistance coefficient Conservative estimate of the worst road adhesion coefficient Conservative estimate of the worst rolling resistance coefficient And based on the defensive MPC control method, the desired torque T is obtained d ;

[0014] The drive actuator controls the driving torque of the driving wheels in real time to obtain the desired torque T obtained by the defensive adaptive cruise controller. d Conduct tracking;

[0015] The brake actuator controls the wheel-end braking force in real time to obtain the desired torque T of the defensive adaptive cruise controller. d Follow up.

[0016] The present invention utilizes a road preview module to obtain a conservative estimate of the potential worst road conditions, and adopts an online identification strategy that can simultaneously estimate road parameters and residual boundaries, and updates the parameters of a defensive adaptive cruise controller in sections, so that the ACC system can always reserve remedial measures for the worst road conditions and estimation errors.

[0017] The present invention can realize the joint estimation of the road adhesion coefficient and the rolling resistance coefficient, and comprehensively consider the influence of potential parameter changes on the performance of the ACC control system.

[0018] The defensive adaptive cruise controller designed by the present invention has a driving anti-skid function, and can take into account both longitudinal control accuracy and vehicle stability.

[0019] The present invention also provides a defensive adaptive cruise control method, comprising:

[0020] Obtain real-time collected road image information and output it as a conservative estimate of the adhesion coefficient of the harsh road section ahead and a conservative estimate of the rolling resistance coefficient Obtain the vehicle's longitudinal speed v and drive wheel speed ω, and output the estimated road adhesion coefficient and rolling resistance coefficient estimate And the residual bounds of the road adhesion coefficient estimation and rolling resistance coefficient estimation residual bounds Conservative estimate based on output adhesion coefficient and a conservative estimate of the rolling resistance coefficient And the estimated road adhesion coefficient estimated online Estimated rolling resistance coefficient Residual bounds of road adhesion coefficient estimation and rolling resistance coefficient estimation residual bounds And combined with the areas where different road types are located, determine the worst road adhesion coefficient that should be considered at this time and the worst rolling resistance coefficient Get the estimated value of the road adhesion coefficient Estimated rolling resistance coefficient Conservative estimate of the worst road adhesion coefficient Conservative estimate of the worst rolling resistance coefficient And based on the defensive MPC control method, the desired torque T is obtained d ;

[0021] The desired torque T d Track and according to the expected torque T d Control the speed of the vehicle.

[0022] The present invention also provides a defensive adaptive cruise control device, comprising a processor and a memory; the memory stores a program or instruction, and the program or instruction is loaded and executed by the processor to implement the steps of the defensive adaptive cruise control method provided above.

[0023] The present invention also provides a computer-readable storage medium, on which a program or instruction is stored. When the program or instruction is executed by a processor, the steps of the defensive adaptive cruise control method provided above are implemented.

[0024] The present invention also provides a vehicle having the defensive adaptive cruise control system provided above or adopting the defensive adaptive cruise control method for cruise control.

[0025] Beneficial effects:

[0026] The adaptive cruise control system designed by the present invention has the ability of forward-looking pre-aiming and online identification of road conditions, and can always take into account the potential impact of the worst road conditions or estimated residuals on the system control performance. When driving on roads with good adhesion, the control system can maximize the control accuracy and smoothness, and always reserve avoidance strategies for emergency events; when poor road conditions are detected, the system can take timely evasive actions to ensure the driving safety and stability of the vehicle. Moreover, since the proposed defensive adaptive cruise control strategy can consider the poor road conditions in advance, driving in the transition area of ​​roads with different adhesion conditions will not produce obvious speed or torque step changes, and the control is relatively smooth. In addition, the designed rolling resistance coefficient and road adhesion coefficient joint estimation strategy has a decreasing residual boundary, so the constructed defensive controller can gradually reduce the conservatism of the emergency branch. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 Defensive adaptive cruise control system architecture;

[0028] Figure 2 Schematic diagram of situation 1;

[0029] Figure 3 Schematic diagram of situation 2;

[0030] Figure 4 Schematic diagram of situation three;

[0031] Figure 5 Schematic diagram of situation 4. DETAILED DESCRIPTION

[0032] Example 1

[0033] This embodiment provides an adaptive cruise control system, which is mainly composed of a road preview system, a road parameter update system, a defensive adaptive cruise control controller, a drive actuator, a brake actuator and a sensor. Figure 1 The sensor includes a wheel speed sensor and a combined inertial navigation system. The wheel speed sensor obtains the wheel speed; the combined inertial navigation system obtains the longitudinal speed and acceleration of the vehicle.

[0034] The road preview system obtains the road image information collected in real time, and after denoising and dedistortion processing, it is transmitted to the trained road adhesion parameter recognition network, and the road state area division and type are output. The expert knowledge base refers to the prior mapping relationship table between road parameters and road types, and its output is the conservative estimate of the adhesion coefficient and rolling resistance coefficient of the bad road section ahead, which are recorded as and

[0035] The road surface parameter update system consists of a road surface parameter estimator and a parameter update mechanism. The road surface parameter estimator obtains the longitudinal speed v of the vehicle and the driving wheel speed ω, and outputs the estimated values ​​of the road surface adhesion coefficient and rolling resistance coefficient. and and their estimated residual bounds and The designed parameter update mechanism can be based on the output of the road preview system and and online estimation information of road surface parameters and And combined with the areas where different road types are located, the worst road parameters that should be considered at this time are determined and

[0036] Defensive adaptive cruise control can obtain and information, and obtain the desired torque T based on the defensive MPC control method d Finally, the drive and brake actuators can track the desired torque.

[0037] The pavement parameter estimator designed in the pavement parameter updating system of the present invention can realize the joint estimation of the rolling resistance coefficient and the adhesion coefficient, and simultaneously obtain the estimation residual boundary that decreases with time.

[0038] Establishing the vehicle longitudinal dynamics model:

[0039]

[0040] Where m is the total mass of the vehicle, v and Represent the vehicle longitudinal velocity and its derivative with respect to time, F x represents the ground tangential reaction force on the tire, ρ is the air density, C d is the air resistance coefficient, A is the frontal area, g is the acceleration of gravity, and θ is the road slope.

[0041] Establish the driving wheel rolling dynamics:

[0042]

[0043] Where I is the wheel moment of inertia, is the derivative of the wheel angular velocity, T d is the wheel torque, and R is the effective radius of the wheel.

[0044] The Dugoff tire model is used to calculate the ground tangential reaction force on the tire:

[0045]

[0046] Among them, C x is the longitudinal stiffness of the wheel, is the nonlinear characteristic boundary value of the tire longitudinal force, κ is the speed influence factor, and s is the driving slip rate or braking slip rate, which is calculated as follows:

[0047]

[0048] In order to estimate the pavement parameters, the following observation equation is defined:

[0049] y k =h k ξ+w k (25)

[0050] Among them, y k With h k are all system information observed at time k, ξ is the parameter to be estimated, and w k is the unknown but bounded system uncertainty. Assume that w k The border Known, satisfied Specifically, for the estimation of the rolling resistance coefficient, the observation equation is constructed as follows:

[0051]

[0052] For the estimation of the road adhesion coefficient, the observation equation is constructed as follows:

[0053]

[0054] The set of road surface parameter estimation residuals is defined as in Refers to the estimated residual bound at time k and The set membership estimation strategy based on least squares is adopted as follows:

[0055]

[0056] Among them, P k represents the estimated covariance of parameters at time k, r k is the adaptive coefficient at time k, is the estimated value of the road surface parameters at time k, Reflects the residual boundary information related to the covariance at time k, is the estimated residual boundary of the road surface parameters. This estimation strategy has the following two characteristics: 1) is bounded and non-increasing.

[0057] The present invention updates the parameters in different regions according to the spatial position of the vehicle and the sensor sensing range. The details are as follows:

[0058] 1) When the vehicle is driving on a road with good adhesion, the road preview system does not detect potential bad road conditions, such as Figure 2 As shown. At this time, the conservative estimate of the worst road parameters and The updates are as follows:

[0059]

[0060] Among them, the estimated residual boundaries of the road adhesion coefficient and the rolling resistance coefficient are respectively expressed as and At this point, the conservative estimate and is obtained through a joint estimator of road surface parameters.

[0061] 2) When the vehicle is driving on a road with good adhesion, the road surface preview system has detected a potential bad road surface, such as Figure 3 As shown. At this time, the conservative estimate of the worst road parameters and The updates are as follows:

[0062]

[0063] At this time, the conservative estimate and It is obtained by identifying low-adhesion road sections through the road surface preview system.

[0064] 3) When the vehicle is driving on a low-adhesion road, the road preview system does not detect a good adhesion road surface, such as Figure 4 As shown. At this time, the conservative estimate of the worst road parameters and The updates are as follows:

[0065]

[0066] At this time, the conservative estimate and It is obtained through online identification of low-adhesion road surfaces by a road surface parameter joint estimator.

[0067] 4) When the vehicle is driving on a low-adhesion road, the road preview system has detected a good adhesion road surface, such as Figure 5 As shown. At this time, the conservative estimate of the worst road parameters and The updates are as follows:

[0068]

[0069] At this time, the conservative estimate and It is obtained by identifying good road sections through the road preview system.

[0070] Defensive Adaptive Cruise Control

[0071] Establish the vehicle longitudinal kinematic model:

[0072]

[0073] Combine the vehicle longitudinal dynamics (21), the driving wheel rolling dynamics (22), the tire model (23) and the kinematic equation (33), and select the state variable as X = [xv ω] T , the control quantity is u=T d Constructed into a unified discrete state space equation:

[0074] χ k+1 =g(χ k ,u k )| μ，f (34)

[0075] The model is affected by the road adhesion coefficient μ and the rolling resistance coefficient f.

[0076] The present invention uses a defensive MPC controller to implement adaptive cruise control, which can always reserve countermeasures for potential adverse road conditions. Different from the standard MPC, this strategy includes two open-loop trajectory branches, namely the state sequence branch under the nominal condition and state sequence branching in emergency situations Correspondingly, there are also action sequence branches in the nominal case and action sequence branching in emergency situations The nominal branch indicates that the road surface parameter estimation value is obtained by using the road surface parameter estimator at time k. and Update the vehicle model (34), and recursively move forward N steps based on this model to obtain the nominal trajectory; the emergency branch indicates that the worst-case estimate of the road surface is adopted at time k and Perform open-loop prediction to obtain the emergency trajectory in the prediction time domain N. These two trajectories are connected by the state and action at time k, that is,

[0077]

[0078] Among them, χ kis the measured state at time k. Equation (35) means that the trajectories of the nominal branch and the emergency branch intersect at time k, and the action decisions at this time are the same. Therefore, the action decision at time k can take into account both nominal operation and emergency avoidance, while the action decision in the time domain from k+1 to k+N-1 makes the nominal trajectory and the emergency trajectory differentiated. MPC is a rolling time domain optimization strategy that only applies the control action at time k to the controlled object. Therefore, the or At the same time, it has the tendency to evolve towards nominal operation and emergency risk avoidance.

[0079] Design a cost function of the following form:

[0080]

[0081] Among them, Q1, Q2, Q3 and R1 are all weight matrices. n represents the cost function under the nominal branch, L is the vehicle length, d safe For the ideal safe following distance, x ref represents the longitudinal displacement of the vehicle in front, and the first term on the right side of the equation represents the speed tracking accuracy cost (v ref is the reference longitudinal speed output by the planning layer), the second term represents the cost of maintaining the vehicle spacing accuracy, and the third term represents the cost of control smoothness; J e represents the cost function under the emergency branch. The first term on the right side of the equation is the slip rate (or slip rate) and the reference value s ref The weighted bi-norm of the bias, the second term is the cost of the slack variable ε.

[0082] Design driving safety constraints in emergency situations as follows:

[0083]

[0084] Among them, d min The minimum safe distance to prevent rear-end collision.

[0085] Design the driving stability constraints in emergency situations as follows:

[0086]

[0087] Among them, Δs max is the slip rate (or slip ratio) deviation from the reference value s ref The maximum permissible value of ε k+i ≥0 is the slack variable of the stability constraint.

[0088] Design the following speed limit and torque limit constraints:

[0089]

[0090] Among them, v min and v max are the minimum and maximum speeds allowed on the current road section, T min and T max is the minimum and maximum torque at the current speed.

[0091] In summary, the designed defensive adaptive cruise controller is as follows:

[0092]

[0093] Among them, the first two items in the constraint equation are the trajectory evolution dynamics constraints under different branches, the third to sixth items are the physical restriction constraints that the state and action need to meet, the seventh and eighth items are the driving safety and stability constraints that need to be met in emergency situations, and the last two items are the association constraints between the nominal branch and the emergency branch. Note that the nominal branch cost in the cost function only focuses on the accuracy and smoothness of longitudinal control, while the emergency branch cost focuses on driving stability. In addition, in addition to the necessary physical restriction constraints, the emergency branch must also follow the hard constraints of driving safety and the soft constraints of stability. The two relatively independent branches are linked together through the association constraints, so that the control volume at the kth moment takes into account the characteristics of precise and efficient control and emergency avoidance. The vehicle will not directly choose between the nominal trajectory and the emergency trajectory, but will make a balanced decision. When an emergency does not occur, the vehicle maximizes the speed tracking and spacing following accuracy, while always reserving avoidance strategies for emergency events. In addition, since the designed least squares-based set membership estimation strategy has a decreasing residual property, updating the vehicle dynamics model according to the designed parameter update mechanism can gradually reduce the conservatism of the emergency branch itself.

[0094] Example 2

[0095] This embodiment provides a defensive adaptive cruise control method, which is based on the defensive adaptive cruise control system provided in Embodiment 1 and includes:

[0096] Obtain real-time collected road image information and output it as a conservative estimate of the adhesion coefficient of the harsh road section ahead and rolling resistance coefficient conservatively estimated Obtain the vehicle's longitudinal speed v and drive wheel speed ω, and output the estimated road adhesion coefficient and rolling resistance coefficient estimates And the residual bounds of the road adhesion coefficient estimation and rolling resistance coefficient estimation residual bounds Conservative estimate based on output adhesion coefficient and rolling resistance coefficient conservatively estimated And the estimated road adhesion coefficient estimated online Estimated rolling resistance coefficient Residual bounds of road adhesion coefficient estimation and rolling resistance coefficient estimation residual bounds And combined with the areas where different road types are located, determine the worst road adhesion coefficient that should be considered at this time and the worst rolling resistance coefficient Get the estimated value of the road adhesion coefficient Estimated rolling resistance coefficient Conservative estimate of the worst road adhesion coefficient Conservative estimate of the worst rolling resistance coefficient And based on the defensive MPC control method, the desired torque T is obtained d ;

[0097] The desired torque T d Track and according to the expected torque T d Control the speed of the vehicle.

[0098] Example 3

[0099] This embodiment provides a defensive adaptive cruise control device, including a processor and a memory; the memory stores a program or instruction, and the program or instruction is loaded and executed by the processor to implement the steps of the defensive adaptive cruise control method provided in Example 2.

[0100] Example 4

[0101] This embodiment provides a computer-readable storage medium, on which a program or instruction is stored. When the program or instruction is executed by a processor, the steps of the defensive adaptive cruise control method provided above are implemented.

[0102] Example 5

[0103] This embodiment provides a vehicle having the defensive adaptive cruise control system provided in embodiment 1 or cruise control using the defensive adaptive cruise control method provided in embodiment 2. The vehicle includes a small car, a commercial vehicle, and the like.

Claims

1. A defensive adaptive cruise system, characterized in that: include: Wheel speed sensor, to obtain wheel speed; Combined inertial navigation, obtains the longitudinal velocity and acceleration of the vehicle; The road preview system obtains real-time collected road image information and outputs a conservative estimate of the adhesion coefficient of the harsh road ahead and a conservative estimate of the rolling resistance coefficient The road parameter update system consists of a road parameter estimator and a parameter update mechanism; the road parameter estimator obtains the longitudinal speed v and the driving wheel speed ω of the vehicle, and outputs the estimated value of the road adhesion coefficient and rolling resistance coefficient estimate And the residual bounds of the road adhesion coefficient estimation and rolling resistance coefficient estimation residual bounds The parameter update mechanism is based on the conservative estimate of the adhesion coefficient output by the road preview system. and rolling resistance coefficient conservatively estimated and the estimated road adhesion coefficient estimated online by the road parameter estimator Estimated rolling resistance coefficient Residual bounds of road adhesion coefficient estimation and rolling resistance coefficient estimation residual bounds And combined with the areas where different road types are located, determine the worst road adhesion coefficient that should be considered at this time and rolling resistance coefficient Defensive adaptive cruise control, obtains the estimated road adhesion coefficient Estimated rolling resistance coefficient Conservative estimate of the worst road adhesion coefficient Conservative estimate of the worst rolling resistance coefficient And based on the defensive MPC control method, the desired torque T is obtained d ; The drive actuator controls the driving torque of the driving wheels in real time to obtain the desired torque T obtained by the defensive adaptive cruise controller. d Conduct tracking; The brake actuator controls the wheel-end braking force in real time to obtain the desired torque T of the defensive adaptive cruise controller. d Follow up.

2. The defensive adaptive cruise control system according to claim 1, characterized in that: The road surface preview system includes a road surface parameter recognition network and an expert knowledge base; The road parameter recognition network outputs the road condition area division and type based on the input of the road image information collected in real time; the expert knowledge base outputs a conservative estimate of the adhesion coefficient of the bad road section ahead based on the prior mapping relationship table between the road parameters and the road type. and rolling resistance coefficient conservatively estimated 3. The defensive adaptive cruise control system according to claim 1, characterized in that: The road surface parameter estimator realizes the joint estimation of rolling resistance coefficient and adhesion coefficient, and obtains the estimation residual boundary that decreases with time; Establishing the vehicle longitudinal dynamics model: Where m is the total mass of the vehicle, v and Represent the vehicle longitudinal velocity and its derivative with respect to time, F x represents the ground tangential reaction force on the tire, ρ is the air density, C d is the air resistance coefficient, A is the frontal area, g is the acceleration of gravity, and θ is the road slope; Establish the driving wheel rolling dynamics: Where I is the wheel moment of inertia, is the derivative of the wheel angular velocity, T d is the wheel torque, R is the effective radius of the wheel; The Dugoff tire model is used to calculate the ground tangential reaction force on the tire: Among them, C x is the longitudinal stiffness of the wheel, is the nonlinear characteristic boundary value of the tire longitudinal force, κ is the speed influence factor, s is the driving slip rate or braking slip rate, and s is calculated as follows: In order to estimate the pavement parameters, the following observation equation is defined: y k =h k ξ+w k (5) Among them, y k With h k are all system information observed at time k, ξ is the parameter to be estimated, and w k is the unknown but bounded system uncertainty; assuming that w k The border Known, satisfied For the estimation of rolling resistance coefficient, the observation equation is formed as follows: For the estimation of the road adhesion coefficient, the observation equation is constructed as follows: The set of road surface parameter estimation residuals is defined as in Refers to the estimated residual bound at time k and Refers to the estimated value of the road adhesion coefficient at time k and rolling resistance coefficient estimate The set membership estimation strategy based on least squares is adopted as follows: Among them, P k represents the estimated covariance of parameters at time k, r k is the adaptive coefficient at time k, It reflects the residual boundary information related to the covariance at time k.

4. The defensive adaptive cruise control system according to claim 3, characterized in that: The parameters are updated in different regions according to the spatial position of the vehicle and the sensor sensing range, as follows: 1) When the vehicle is driving on a road with good adhesion and the road preview system does not detect the potential bad road surface, the worst road parameter condition is conservatively estimated. and The updates are as follows: Among them, the estimated residual boundaries of the road adhesion coefficient and the rolling resistance coefficient are respectively expressed as and To express; 2) When the vehicle is driving on a road with good adhesion, the road preview system has detected a potential bad road surface, and the worst road parameter condition is conservatively estimated. and The updates are as follows: 3) When the vehicle is driving on a low-adhesion road and the road preview system does not detect a good adhesion road surface, the worst-case road parameter condition is conservatively estimated. and The updates are as follows: 4) When the vehicle is driving on a low-adhesion road, the road preview system has detected a good adhesion road surface, and the worst road parameter condition is conservatively estimated. and The updates are as follows: At this time, the conservative estimate and It is obtained by identifying good road sections through the road preview system.

5. The defensive adaptive cruise control system according to claim 1, characterized in that: Defensive adaptive cruise controller, based on defensive MPC control method to obtain the desired torque T d ,include: Establish the vehicle longitudinal kinematic model: Combine the vehicle longitudinal dynamics model (1), the driving wheel rolling dynamics (2), the tire model (3) and the kinematic equation (13), and select the state variable as χ = [xv ω] T , the control quantity is u=T d Constructed into a unified discrete state space equation: Adaptive cruise control is implemented using a defensive MPC controller, which includes two open-loop trajectory branches: the state sequence branch under nominal conditions and state sequence branching in emergency situations Action sequence branching in the nominal case and action sequence branching in emergency situations The nominal branch indicates that the road surface parameter estimation value is obtained by using the road surface parameter estimator at time k. and Update the vehicle model (14), and recursively move forward N steps based on this model to obtain the nominal trajectory; the emergency branch indicates that the worst-case estimate of the road surface is adopted at time k and Perform open-loop prediction to obtain the emergency trajectory in the prediction time domain N. These two trajectories are connected by the state and action at time k, namely: Among them, χ k is the measured state at time k; Design a cost function of the following form: Among them, Q1, Q2, Q3 and R1 are weight matrices, J n represents the cost function under the nominal branch, L is the vehicle length, d safe For the ideal safe following distance, x ref represents the longitudinal displacement of the vehicle in front, and the first term on the right side of the equation represents the speed tracking accuracy cost (v ref is the reference longitudinal speed output by the planning layer), the second term represents the cost of maintaining the vehicle spacing accuracy, and the third term represents the cost of control smoothness; J e represents the cost function under the emergency branch, and the first term on the right side of the equation is the slip rate or slip ratio and the reference value s ref The weighted bi-norm of the deviation, the second term is the cost of the slack variable ε; Design driving safety constraints in emergency situations as follows: Among them, d min Minimum safe distance to prevent rear-end collision; Design the driving stability constraints in emergency situations as follows: Among them, Δs max is the slip rate or the slip rate deviation from the reference value s ref The maximum permissible value of ε k+i ≥0 is the slack variable of the stability constraint; Design the following speed limit and torque limit constraints: Among them, v min and v max are the minimum and maximum speeds allowed on the current road section, T min and T max is the minimum and maximum torque at the current speed.

6. The defensive adaptive cruise control system according to claim 5, characterized in that: The designed defensive adaptive cruise controller is shown below: Among them, the first two items in the constraint equation are the trajectory evolution dynamics constraints under different branches, the third to sixth items are the physical restriction constraints that the state and action must meet, the seventh and eighth items are the driving safety and stability constraints that must be met in emergency situations, and the last two items are the association constraints between the nominal branch and the emergency branch.

7. A defensive adaptive cruise control method, characterized in that: include: Obtain real-time collected road image information and output it as a conservative estimate of the adhesion coefficient of the harsh road section ahead and rolling resistance coefficient conservatively estimated Obtain the vehicle's longitudinal speed v and drive wheel speed ω, and output the estimated road adhesion coefficient and rolling resistance coefficient estimate And the residual bounds of the road adhesion coefficient estimation and rolling resistance coefficient estimation residual bounds Conservative estimate based on output adhesion coefficient and rolling resistance coefficient conservatively estimated And the estimated road adhesion coefficient estimated online Estimated rolling resistance coefficient Residual bounds of road adhesion coefficient estimation and rolling resistance coefficient estimation residual bounds And combined with the areas where different road types are located, determine the worst road adhesion coefficient that should be considered at this time and the worst rolling resistance coefficient Get the estimated value of the road adhesion coefficient Estimated rolling resistance coefficient Conservative estimate of the worst road adhesion coefficient Conservative estimate of the worst rolling resistance coefficient And based on the defensive MPC control method, the desired torque T is obtained d ; The desired torque T d Track and according to the expected torque T d Control the speed of the vehicle.

8. A defensive adaptive cruise control device, characterized in that: It comprises a processor and a memory; the memory stores a program or instruction, and the program or instruction is loaded and executed by the processor to implement the steps of the defensive adaptive cruise control method as claimed in claim 7.

9. A computer-readable storage medium, characterized in that: The readable storage medium stores a program or instruction, and when the program or instruction is executed by the processor, the steps of the defensive adaptive cruise control method as claimed in claim 7 are implemented.

10. A vehicle, characterized in that: A defensive adaptive cruise control system according to any one of claims 1 to 6 or a defensive adaptive cruise control method according to claim 7 is used for cruise control.

Images