Automatic emergency braking layered control system and vehicle

Abstract

The invention provides an automatic emergency braking hierarchical control system and a vehicle. The system is mainly composed of a high-level rule supervision control module, a middle-level control algorithm switching module and a bottom-level control module. The high-level rule supervision control module evaluates a potential collision risk in real time based on the relative distance and a self-adaptive relative distance threshold value, and generates a sliding control demand and a corresponding target deceleration based on the grade of the potential collision risk; the middle-layer control algorithm switching module is used for switching among different sliding control modes based on the sliding control requirement, the target deceleration and the current vehicle state; and the bottom layer control module calculates the cruise speed of the vehicle in the vehicle following process or calculates the target slip rate of the vehicle in the braking process and applies the corresponding braking torque based on the target deceleration, the sliding control mode and the current vehicle state. A hierarchical control architecture is designed, different hierarchies are responsible for different control targets, and the modularization and expandability of the system are improved.

Description

Technical Field

[0001] The present application relates to the field of vehicle control technology, and in particular to an automatic emergency braking hierarchical control system and a vehicle. Background Art

[0002] Since head-on and tail-off collisions between vehicles are a common type of road traffic accidents, Automatic Emergency Braking (AEB) technology is particularly important for avoiding collisions and has become a key means to improve driving safety. The AEB system can automatically apply brakes to the vehicle when it detects a possible collision risk to reduce the possibility of a collision or reduce the severity of a collision.

[0003] At present, the existing AEB system mainly relies on fixed threshold settings and linear control algorithms. This method sets a fixed safety distance threshold. When the distance between the vehicle and the target in front is detected to be less than the threshold, the system performs braking. However, due to the large changes in safety distance under different working conditions (such as vehicle speed, road friction coefficient, behavior of the vehicle in front, etc.), the fixed threshold method is prone to false triggering or response lag in complex environments, which can easily lead to excessively long or short braking distances, affecting the braking effect. Summary of the invention

[0004] In response to the shortcomings of the prior art, the present application provides an automatic emergency braking hierarchical control system and a vehicle to at least solve the problem that the fixed threshold setting and linear control algorithm in the prior art are difficult to adapt to different working conditions, resulting in false triggering, response lag or unreasonable braking distance, thereby affecting the braking effect and driving safety.

[0005] In order to achieve the above objectives and other advantages, the present application adopts the following technical solutions:

[0006] In a first aspect, the present application provides an automatic emergency braking hierarchical control system, comprising:

[0007] A braking parameter information acquisition module is used to acquire state parameters that affect the braking control of the vehicle, wherein the state parameters include: the relative distance between the vehicle and the preceding vehicle, the longitudinal speeds of the vehicle and the preceding vehicle in the driving direction, and the road adhesion coefficient;

[0008] A relative distance threshold calculation module, used to dynamically and adaptively adjust the safe braking distance based on the longitudinal speeds of the vehicle and the preceding vehicle in the travel direction, and generate an adaptive relative distance threshold;

[0009] a high-level rule supervisory control module, configured to evaluate a potential collision risk in real time based on the relative distance and the adaptive relative distance threshold, and generate a sliding control demand and a corresponding target deceleration based on a level of the potential collision risk;

[0010] a middle-level control algorithm switching module, configured to switch between different sliding control modes based on the sliding control requirement, the target deceleration and the current vehicle state;

[0011] The bottom control module is used to calculate the cruising speed of the vehicle during the following process, or calculate the target slip rate of the vehicle during the braking process and apply the corresponding braking torque based on the target deceleration, the slip control mode and the current vehicle state.

[0012] According to an automatic emergency braking hierarchical control system provided by the present application, the relative distance threshold calculation module generates an adaptive relative distance threshold, specifically including:

[0013] Determine whether the longitudinal speed of the vehicle is greater than a preset speed threshold and whether there is a vehicle ahead on the road;

[0014] When the longitudinal speed of the vehicle is greater than the speed threshold and there is a vehicle ahead on the road, the minimum braking distance is calculated based on the longitudinal speed of the vehicle and the peak friction coefficient of the road, and the static safety distance margin is added to obtain the relative distance threshold.

[0015] According to an automatic emergency braking hierarchical control system provided by the present application, the high-level rule supervisory control module generates a braking control demand and a corresponding target deceleration, specifically including:

[0016] Combined with the speed and acceleration of the preceding vehicle detected by the sensor, the collision time calculation method is used to analyze the relative motion trend between vehicles;

[0017] When the relative distance is greater than the relative distance threshold and the collision time is greater than the set first safety threshold, the system determines that the risk is low, sends a vehicle following control request to the middle-level control algorithm switching module, and calculates the expected speed with the target deceleration of 0 as the cruising speed of the vehicle;

[0018] When the relative distance is close to the relative distance threshold and the collision time is greater than the set second safety threshold, the system determines that the risk is medium, sends a deceleration control request to the middle-level control algorithm switching module, and calculates the target deceleration based on the longitudinal speed of the vehicle and the minimum braking distance, and the second safety threshold is less than the first safety threshold;

[0019] When the relative distance is less than the relative distance threshold and the collision time is less than or equal to the second safety threshold, the system determines it as a high risk, sends an emergency braking control request to the middle-level control algorithm switching module, and calculates the target deceleration based on the longitudinal speed of the vehicle and the minimum braking distance.

[0020] According to an automatic emergency braking hierarchical control system provided by the present application, the high-level rule supervisory control module generates a braking control demand and a corresponding target deceleration, and further includes:

[0021] During the following and braking process, the relative distance between the vehicle and the front vehicle and the vehicle status are continuously monitored. If the front vehicle further decelerates, the potential collision risk is re-evaluated and the target deceleration is updated.

[0022] According to an automatic emergency braking hierarchical control system provided by the present application, the sliding control mode includes: a proportional integral derivative control mode and a sliding mode control mode.

[0023] According to an automatic emergency braking hierarchical control system provided by the present application, when the middle-level control algorithm switching module switches to the proportional integral differential control mode, the bottom-level control module adopts the PID control strategy to optimize the longitudinal motion control of the vehicle during the following process, specifically including:

[0024] A speed error is calculated based on the desired longitudinal speed and the actual longitudinal speed of the vehicle, and the speed error is input into a proportional-integral-differential controller. The driving torque or braking torque of the vehicle is dynamically adjusted according to an output signal of the proportional-integral-differential controller so that the actual longitudinal speed gradually converges to the desired longitudinal speed. If the speed error is positive, the driving torque is adjusted to increase the vehicle acceleration; if the speed error is negative, the braking torque is adjusted to reduce the vehicle acceleration.

[0025] According to an automatic emergency braking hierarchical control system provided by the present application, when the middle-level control algorithm switching module switches to the sliding mode control mode, the bottom-level control module adopts a sliding mode control strategy to calculate the control torque through slip rate error feedback during braking, specifically including:

[0026] Construct the sliding surface and adjust the system state toward the expected steady-state controllable slip condition by calculating the slip rate error;

[0027] A control law satisfying the steady-state controllable slip condition is derived, wherein the control law is in the form of a combination of an equivalent control torque and a switch control torque, wherein the equivalent control torque is used to maintain the stable operation of the system state on the sliding surface, and the switch control torque is used to apply a correction torque when the system deviates from the sliding surface, so as to quickly pull the system state back to the sliding surface;

[0028] During braking, the front and rear wheel braking torques are dynamically adjusted based on the control law. When the vehicle slip rate reaches the target value, the control system maintains the effect of the equivalent control torque.

[0029] According to an automatic emergency braking hierarchical control system provided by the present application, the bottom control module uses an artificial neural network to optimize the calculation of the equivalent control torque in the sliding mode control mode, specifically including:

[0030] Using the longitudinal friction force and the wheel translation speed as input data of the artificial neural network to obtain a friction force prediction value output by the artificial neural network;

[0031] Combining the friction force prediction value with known vehicle dynamics parameters to obtain the equivalent control torque;

[0032] Among them, the artificial neural network takes the longitudinal friction force and the wheel translation speed as input data, calculates the error between the friction force prediction value and the true value of the longitudinal friction force to generate a loss value, uses the back propagation optimization algorithm to update the parameters of the artificial neural network, and continuously iterates and optimizes during the supervised training process until the error converges or reaches the set training conditions until convergence, so as to obtain the trained artificial neural network.

[0033] According to an automatic emergency braking hierarchical control system provided by the present application, the underlying control module calculates the dynamic control parameters of the vehicle during sliding based on a nonlinear vehicle dynamics model. The nonlinear vehicle dynamics model is represented by a coupling of a nonlinear bicycle model and a nonlinear tire model. The nonlinear bicycle model is used to describe the longitudinal and lateral dynamic behavior of the vehicle. The nonlinear tire model is used to simulate the contact force characteristics between the tire and the road surface and calculate the longitudinal friction force and normal load.

[0034] In a second aspect, the present application provides a vehicle comprising an automatic emergency braking hierarchical control system as described in any one of the above.

[0035] The present application provides an automatic emergency braking hierarchical control system and vehicle, which is mainly composed of a high-level rule supervision control module, a middle-level control algorithm switching module and a bottom-level control module. The state parameters affecting the braking control of the self-vehicle are obtained through the braking parameter information acquisition module, and the state parameters include: the relative distance between the self-vehicle and the front vehicle, the longitudinal speed of the self-vehicle and the front vehicle in the driving direction, and the road adhesion coefficient; the relative distance threshold calculation module dynamically and adaptively adjusts the safe braking distance based on the longitudinal speed of the self-vehicle and the front vehicle in the driving direction, and generates an adaptive relative distance threshold; the high-level rule supervision control module evaluates the potential collision risk in real time based on the relative distance and the adaptive relative distance threshold, and generates a sliding control demand and a corresponding target deceleration based on the level of the potential collision risk; the middle-level control algorithm switching module switches between different sliding control modes based on the sliding control demand, the target deceleration and the current vehicle state; the bottom-level control module calculates the cruising speed of the self-vehicle during the following process, or calculates the target slip rate of the self-vehicle during the braking process and applies the corresponding braking torque based on the target deceleration, the sliding control mode and the current vehicle state. This application designs a hierarchical control architecture. The high-level rule supervision control module is mainly responsible for environmental perception and collision risk assessment to avoid unnecessary braking intervention; the middle-level control algorithm switching module is responsible for selecting the most appropriate control algorithm; the bottom-level control module specifically performs low-level operations such as braking force distribution and slip rate control to ensure the real-time responsiveness of the system. Different levels are responsible for different control objectives, which improves the modularity and scalability of the system. At the same time, it avoids response delays caused by the complexity of a single control logic, especially in autonomous driving or advanced driver assistance systems. It is not necessary to consider the computational workload of the entire system, which reduces the computational burden of the processor and improves the real-time and robustness of the system. Through this hierarchical control architecture, the AEB system can make autonomous decisions, adopt different braking strategies in different situations, and improve the level of intelligence. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other implementation methods can be obtained based on these drawings without paying any creative work.

[0037] Figure 1 is a logic diagram of an automatic emergency braking hierarchical control system provided in an embodiment of the present application;

[0038] Figure 2 It is a schematic diagram showing the relative distance between a preceding vehicle and the own vehicle provided in an embodiment of the present application. DETAILED DESCRIPTION

[0039] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the following specifically cites the preferred embodiments and describes them in detail with the accompanying drawings.

[0040] It should be noted that it is explicitly and implicitly understood by those of ordinary skill in the art that the embodiments described in this application can be combined with other embodiments without conflict. Unless otherwise defined, the technical terms or scientific terms involved in this application should be the usual meanings understood by people with ordinary skills in the technical field to which this application belongs. The words "one", "a", "a", "the" and the like involved in this application do not indicate a quantitative limitation and may represent the singular or plural. The terms "including", "comprising", "having" and any of their variations involved in this application are intended to cover non-exclusive inclusions; the terms "first", "second", "third", etc. involved in this application are merely to distinguish similar objects and do not represent a specific ordering of objects.

[0041] Reference Figure 1 As shown, the present application provides an automatic emergency braking hierarchical control system, comprising:

[0042] The braking parameter information acquisition module is used to obtain the state parameters that affect the braking control of the own vehicle. The state parameters include: the relative distance between the own vehicle and the front vehicle, the longitudinal speed of the own vehicle and the front vehicle in the driving direction, and the road adhesion coefficient.

[0043] It should be explained that the relative distance refers to the physical distance between the vehicle (the vehicle equipped with AEB, referred to as the EGO vehicle) and the vehicle in front (the target vehicle in front, referred to as the GVT vehicle), usually expressed as Δx = X GVT -X EGO Indicates. EGO Indicates that the EGO vehicle is in X I Position coordinates in the direction, X GVT Indicates that the GVT vehicle is at X I Position coordinates in a direction, such as Figure 2 The relative distance can be measured by a laser radar, millimeter wave radar, camera or ultrasonic sensor and updated in real time.

[0044] Longitudinal velocity refers to the speed of the vehicle in the current direction of travel, which can be obtained by devices such as the on-board inertial measurement unit (IMU), wheel speed sensor, GPS or millimeter wave radar.

[0045] The road adhesion coefficient indicates the friction between the tire and the road surface, and determines the available range of the maximum braking force. For example, the adhesion coefficient of a dry asphalt road surface is greater than that of a wet road surface, and the adhesion coefficient of a wet road surface is greater than that of an icy road surface. The road adhesion coefficient can be estimated by on-board sensors (such as road friction sensors) or external data (such as weather sensors, road information systems). The road adhesion coefficient determines the vehicle's braking force strategy. The higher the road adhesion coefficient, the higher the braking force available to the vehicle and the shorter the braking distance. The lower the road adhesion coefficient, the softer the braking strategy is required to avoid tire locking or slipping.

[0046] The state parameters obtained by this module (such as relative distance, longitudinal speed, and road adhesion coefficient) directly affect the collision risk assessment, braking decision-making, and execution effect of the AEB system.

[0047] The relative distance threshold calculation module is used to dynamically and adaptively adjust the safe braking distance based on the longitudinal speed of the vehicle and the preceding vehicle in the driving direction, and generate an adaptive relative distance threshold.

[0048] In this embodiment, the relative distance threshold calculation module generates an adaptive relative distance threshold, specifically including:

[0049] Determine whether the longitudinal speed of the vehicle is greater than a preset speed threshold and whether there is a vehicle ahead on the road;

[0050] When the longitudinal speed of the vehicle is greater than the speed threshold and there is a vehicle ahead on the road, the minimum braking distance is calculated based on the longitudinal speed of the vehicle and the peak friction coefficient of the road, and the static safety distance margin is added to obtain the relative distance threshold.

[0051] Specifically, in order to calculate a reasonable adaptive relative distance threshold, the following steps are usually adopted:

[0052] Step 1: Determine whether it is necessary to calculate the minimum braking distance; set the actual longitudinal speed of the EGO vehicle to if If the speed is less than or equal to the set speed threshold (such as low speed driving), emergency braking is not required and the relative distance threshold can be set to a smaller value. And if no vehicle is detected in front (GVT vehicle), the minimum braking distance does not need to be calculated.

[0053] Step 2: If If the speed is greater than the set speed threshold and a GVT vehicle is detected in front, the minimum braking distance is calculated.

[0054] Its judgment logic can be implemented by the following algorithm:

[0055] "Algorithm 1:targetXdotdotGenActivation

[0056] Input: bool isLeadingVehicleDeceted,μ peak

[0057] while &&(isLeadingVehicleDetected==1)do

[0058] targetDecelerationGenerator(Δx, μ peak )

[0059] endwhile

[0060] end”

[0061] Minimum braking distance x for EGO vehicles br,min The definition is as follows:

[0062]

[0063] In the formula, is the actual longitudinal velocity of the EGO vehicle, -ug is the achievable peak deceleration, and g is the acceleration due to gravity.

[0064] Step 3: According to equation (1), the relative distance threshold is a function of speed. The minimum braking distance plus the additional static safety distance margin is used to obtain the relative distance threshold Δx Thres :

[0065] Δx Thres =x br,min +margin (2)

[0066] In the formula, x br,min is the minimum braking distance, and margin is the static safety distance margin.

[0067] The relative distance threshold can be dynamically and adaptively adjusted according to the vehicle's speed at a certain moment, the relative motion relationship between the vehicle and the vehicle in front, the road adhesion conditions, etc., to ensure that the safe braking distance always meets the current driving conditions. Compared with the traditional fixed safety distance setting method, the adaptive relative distance threshold can enable the AEB system to be triggered at a more appropriate time, avoiding false triggering or delayed response, and improving safety and comfort.

[0068] In this embodiment, the system is mainly composed of a high-level rule supervision control module, a middle-level control algorithm switching module and a bottom-level control module. The system adopts hierarchical control (high-level supervision control, middle-level algorithm adjustment, bottom-level control torque execution) to ensure that automatic emergency braking can provide effective collision avoidance and optimize vehicle stability. The following is a detailed description of the three major modules of this system.

[0069] The high-level rule supervisory control module is used to evaluate the potential collision risk in real time based on the relative distance and the adaptive relative distance threshold, and to generate the sliding control demand and the corresponding target deceleration based on the level of the potential collision risk.

[0070] In this embodiment, the high-level rule supervisory control module generates a braking control requirement and a corresponding target deceleration, specifically including:

[0071] Combined with the speed and acceleration of the preceding vehicle detected by the sensor, the collision time calculation method is used to analyze the relative motion trend between vehicles;

[0072] When the relative distance is greater than the relative distance threshold and the collision time is greater than the set first safety threshold, the system determines that the risk is low, sends a following vehicle control request to the middle-level control algorithm switching module, and calculates the expected speed with a target deceleration of 0 as the cruising speed of the vehicle;

[0073] When the relative distance is close to the relative distance threshold and the collision time is greater than the set second safety threshold, the system determines that the risk is medium, sends a deceleration control request to the middle-level control algorithm switching module, and calculates the target deceleration based on the longitudinal speed of the vehicle and the minimum braking distance. The second safety threshold is less than the first safety threshold;

[0074] When the relative distance is less than the relative distance threshold and the collision time is less than or equal to the second safety threshold, the system determines it as a high risk, sends an emergency braking control request to the middle-level control algorithm switching module, and calculates the target deceleration based on the longitudinal speed of the vehicle and the minimum braking distance.

[0075] Specifically, the relative motion state of the two vehicles is determined by combining the speed and acceleration of the preceding vehicle detected by the sensor. The Time-to-Collision (TTC) calculation method is used to evaluate the relative motion trend between the EGO vehicle and the GVT vehicle and calculate the collision risk. The Time-to-Collision (TTC) = Δv / Δx, where Δv is the relative speed of the two vehicles and Δx is the relative distance between the two vehicles.

[0076] According to the relative distance Δx and the adaptive relative distance threshold Δx ThresBased on the comparison results and the relationship between the collision time TTC and the set safety threshold, the potential collision risks are divided into three situations: low risk, medium risk and high risk, and different control strategies are adopted respectively.

[0077] If the relative distance Δx is greater than the adaptive relative distance threshold Δx Thres , and the collision time TTC is greater than the first safety threshold T1, indicating that the vehicle and the vehicle in front maintain a safe enough distance, the system will not make unnecessary interventions to reduce false triggering. At this time, the system determines that the risk is low, the system does not need to slow down, and can maintain normal cruising mode. Send a following control request to the middle-level control algorithm switching module, and do not trigger braking.

[0078] If the relative distance Δx is close to the adaptive relative distance threshold Δx Thres (If the absolute value of the difference between the two is within the set range), and the collision time TTC is between the first safety threshold T1 and the second safety threshold T2, it means entering the danger zone, but there is still buffer space. At this time, the system determines that it is a medium risk and enters the deceleration mode. Send a deceleration control request to the middle-level control algorithm switching module and calculate the target deceleration. Use the longitudinal speed of the vehicle and the minimum braking distance to make adjustments to make the vehicle decelerate smoothly and avoid sudden braking that affects the driving experience.

[0079] If the relative distance Δx is less than the adaptive relative distance threshold Δx Thres , and the collision time TTC is less than or equal to the second safety threshold T2, indicating that the collision risk is extremely high and emergency braking is required. At this time, the system determines that it is a high risk, sends an emergency braking control request to the middle-level control algorithm switching module, and calculates the target deceleration. When the system detects that the front vehicle brakes suddenly or encounters an unexpected obstacle, the AEB system immediately intervenes and applies the maximum safety braking force to avoid or reduce the impact of the collision as much as possible.

[0080] If the current Δx is less than or equal to Δx Thres , then the EGO vehicle needs to decelerate, and its target deceleration Defined as:

[0081]

[0082] In the formula, is the actual longitudinal velocity of the EGO vehicle, x br,min is the minimum braking distance.

[0083] Its judgment logic can be implemented by the following algorithm:

[0084]

[0085]

[0086] Braking parameter information (relative distance, longitudinal speed, road adhesion coefficient, etc.) is a direct input for collision risk assessment and target deceleration calculation in the high-level rule supervision and control module. The braking parameter information acquisition module and the relative distance threshold calculation module are integrated into the high-level rule supervision and control module. This makes the relative distance threshold calculation essentially a part of the high-level supervision and control module, and the two can be implemented in the same module to improve calculation consistency. In this way, data is directly processed within the high-level supervision and control module, reducing data exchange between modules and improving real-time performance.

[0087] The middle-level control algorithm switching module is used to switch between different sliding control modes based on sliding control requirements, target deceleration and current vehicle status. The sliding control modes include: proportional integral derivative control mode and sliding mode control mode.

[0088] Specifically, according to the calculated target deceleration Decide which brake control method to use, PID control mode (proportional integral derivative control mode) or sliding mode control mode. Usually the following steps are used:

[0089] Step 1: Initialize control variables: The control variables include the front and rear axle wheel slip control torque (torqueWSCf, torqueWSCr), and the front and rear axle drive torque (torqueDrvf, torqueDrvr);

[0090] Step 2: Determine whether emergency braking is required: If Not equal to 0, activate the sliding mode controller and turn off the PID speed controller; if When equal to 0, the sliding mode controller is turned off and the PID speed controller is enabled.

[0091] The middle-level control algorithm switching module intelligently switches between different control modes according to the actual driving environment and braking requirements. When the sliding control demand is low (such as low risk or medium risk level), the PID control mode is selected to smoothly adjust the longitudinal movement of the vehicle to achieve cruising or light braking. When the sliding control demand is high (such as high risk level), it switches to the sliding mode control mode to cope with complex braking conditions and enhance the system's robustness to interference.

[0092] The implementation logic of the middle-level control algorithm switching module can be implemented by the following algorithm:

[0093]

[0094] In this embodiment, the high-level rule supervisory control module generates a braking control demand and a corresponding target deceleration, and further includes:

[0095] During the following - distance and braking processes, continuously monitor the relative distance between the host vehicle and the leading vehicle and the vehicle state. If the leading vehicle further decelerates, re - evaluate the potential collision risk and update the target deceleration.

[0096] Specifically, during the following - distance and braking processes, continuously obtain the driving state of the host vehicle and the motion state of the leading vehicle. When the leading vehicle further decelerates or the traffic environment changes, re - evaluate the potential collision risk. For example, when the speed of the leading vehicle decreases, the relative distance Δx between the two vehicles begins to decrease, and the TTC drops to T2 < TTC < T1, the risk level upgrades from low risk to medium risk. The system sends a deceleration request to the middle - layer control algorithm switching module to adapt to the new speed of the leading vehicle and prevent the relative distance from continuing to shrink. Or, the relative distance Δx between the two vehicles decreases instantaneously, and the TTC rapidly drops to TTC ≤ T2, the risk level upgrades to high risk. Immediately send an emergency braking request to the middle - layer control algorithm switching module and switch to the sliding - mode control mode. The bottom - layer control module adopts a sliding - mode control strategy, calculates the braking torque according to the slip - rate error, applies an emergency braking force to ensure safe stopping or reduce the collision risk.

[0097] According to the new vehicle state, dynamically update the target deceleration, adjust the braking strategy to optimize the braking process. Through continuous monitoring and real - time adjustment, avoid over - braking or under - braking caused by a fixed deceleration strategy, and improve the adaptability and safety of the braking system.

[0098] The bottom - layer control module is used to calculate the cruise speed of the host vehicle during the following - distance process or calculate the target slip rate of the host vehicle during the braking process and apply the corresponding braking torque based on the target deceleration, the sliding - control mode, and the current vehicle state.

[0099] In this embodiment, the bottom - layer control module calculates the dynamic control parameters during the vehicle sliding process based on a non - linear vehicle dynamics model. The non - linear vehicle dynamics model is represented by coupling a non - linear bicycle model and a non - linear tire model. The non - linear bicycle model is used to describe the longitudinal and lateral dynamic behaviors of the host vehicle, and the non - linear tire model is used to simulate the contact - force characteristics between the tire and the road surface, calculate the longitudinal friction force and the normal load.

[0100] As an example, the dynamic model of the EGO vehicle can consist of a non - linear differential equation system (including longitudinal and lateral motion equations). Since the considered motivation is autonomous straight - line braking, the motion equation is simplified to describe the non - linear pure longitudinal motion equation. The non - linear bicycle model and the non - linear tire model are coupled to represent the dynamic behavior of the vehicle to achieve a high - precision description of vehicle dynamics. The specific content of constructing this model is as follows:

[0101] 1. On each axis, two wheels are represented by an equivalent wheel, so that the dynamic behavior of each axis can be represented by a single equivalent tire.

[0102] 2. The total mass of the vehicle is concentrated at the center of gravity of the vehicle, while the reaction force loads occur on the front axle and rear axle respectively.

[0103] 3. The longitudinal / lateral friction of the tire on each axle is estimated using the Pacejka MF model (Pacejka Magic Formula is an empirical model widely used in vehicle dynamics modeling, which can accurately describe the interaction between the tire and the road surface).

[0104] 4. This vehicle architecture can include electric prime movers on each axle, all of which have the ability to act as regenerative braking actuators.

[0105] 5. The brake actuation delay is ignored to simplify the calculation and make the control strategy more ideal.

[0106] The vehicle motion is relative to the inertial reference frame {X I ,Y I ,Z I}, while in low-level control analysis, the vehicle sliding dynamics are studied relative to the reference frame of the wheels. The motion equation of the constructed nonlinear vehicle dynamics model is:

[0107]

[0108] In the formula, is an inertial reference frame {X I ,Y I ,Z I The acceleration component of the x-axis, f ix is the longitudinal friction of the front and rear axle tires (i={F,R}, F is the front axle and R is the rear axle), m veh is the gross vehicle mass.

[0109] For the needs of wheel slip control design and control logic deployment, the actual wheel slip ratio λ i Defined as:

[0110]

[0111] In the formula, ω i is the rotation speed of the front and rear axle wheels (i={F,R}), R is the wheel radius, V ix is the translational velocity of the front and rear axle wheels.

[0112] Longitudinal friction coefficient μ estimated using the Pacejka MF model ix Given by:

[0113] μ ix=Dsin(Ctan -1 (Bs ix )) (6)

[0114] Where D is the peak adhesion coefficient of the road tire, B is the stiffness coefficient, C is the shape coefficient, and s ix It is the theoretical slip ratio of the front and rear axle wheels.

[0115] The normal load on each tire includes the static load at zero acceleration or deceleration, and the dynamic load caused by longitudinal load transfer, which is defined as follows:

[0116]

[0117] In the formula, f Fz is the normal load on the front axle (front wheel), f Rz is the normal load on the rear axle (rear wheel), g is the acceleration due to gravity, l F , l R is the distance between the front and rear wheels and the center of gravity of the vehicle, and h is the height of the center of gravity.

[0118] Longitudinal friction force f ix equal:

[0119] f ix =μ ix f iz (8)

[0120] In the formula, μ ix is the longitudinal friction coefficient, f iz is the normal load ((i={F,R}).

[0121] The rotational acceleration of the front and rear wheels is:

[0122]

[0123] In the formula, are the rotational acceleration of the front and rear wheels, J is the moment of inertia of the wheel,

[0124] T front 、T rear The control torques for the front and rear wheels are provided by the sliding mode controller when emergency braking is required and by the PID speed controller when cruising.

[0125] This modeling method is considered to find a reasonable compromise between high-fidelity, computationally intensive models (such as high-order multibody dynamics models) and low-fidelity, computationally inexpensive models (such as point mass models, kinematic models with linear tire response, etc.), thereby achieving a balance between computational complexity and computational accuracy. This can effectively simulate the effects of tire friction changes and longitudinal load transfer during braking, and can be solved in a relatively short time, making it suitable for real-time control.

[0126] In this embodiment, when the middle-level control algorithm switching module switches to the proportional-integral-derivative control mode, the bottom-level control module adopts the PID control strategy to optimize the longitudinal motion control of the vehicle during the following process, specifically including:

[0127] The speed error is calculated based on the desired longitudinal speed and the actual longitudinal speed of the vehicle, and the speed error is input into the proportional-integral-differential controller. The driving torque or braking torque of the vehicle is dynamically adjusted according to the output signal of the proportional-integral-differential controller so that the actual longitudinal speed gradually converges to the desired longitudinal speed. If the speed error is positive, the driving torque is adjusted to increase the vehicle acceleration; if the speed error is negative, the braking torque is adjusted to reduce the vehicle acceleration.

[0128] Specifically, during the cruise control process, the EGO vehicle needs to maintain a reasonable distance from the GVT vehicle in front and smoothly adjust its own speed to ensure safety and comfort when the speed of the GVT vehicle changes. To this end, when the middle-level control algorithm switching module switches to the PID control mode, the bottom-level control module adopts the PID control strategy to optimize the longitudinal motion control.

[0129] The low-level control module first calculates the speed error between the desired longitudinal speed and the actual longitudinal speed of the EGO vehicle.

[0130] Desired longitudinal speed The definition is as follows:

[0131]

[0132] The error e(t) is defined as the expected longitudinal velocity The actual longitudinal speed Difference:

[0133]

[0134] The speed error e(t) is used as input to the PID controller, and the adjusted torque control signal torquedrv is calculated based on the proportional (P), integral (I), and differential (D) terms:

[0135]

[0136] In the formula, k p is the proportional gain, k i is the integral gain, k d is the differential gain, and N is the filter coefficient acting on the differential part.

[0137] When e(t)>0, it means that the current vehicle speed is lower than the expected speed, and the driving torque needs to be increased to accelerate the vehicle to the target speed; when e(t)<0, it means that the current vehicle speed is higher than the expected speed, and the braking torque needs to be applied to gradually slow the vehicle down to the target speed.

[0138] Through PID control, the actual speed of the EGO vehicle gradually converges to the desired speed, ensuring the smoothness, comfort and responsiveness of the vehicle's longitudinal motion during the following process.

[0139] In this embodiment, when the middle-level control algorithm switching module switches to the sliding mode control mode, the bottom-level control module adopts the sliding mode control strategy, and calculates the control torque through the slip rate error feedback during the braking process, specifically including:

[0140] Construct the sliding surface and adjust the system state toward the expected steady-state controllable slip condition by calculating the slip rate error;

[0141] The control law that satisfies the steady-state controllable slip condition is derived. The control law adopts the combination of equivalent control torque and switch control torque. The equivalent control torque is used to maintain the stable operation of the system state on the sliding surface, and the switch control torque is used to apply a correction torque when the system deviates from the sliding surface to quickly pull the system state back to the sliding surface.

[0142] During braking, the front and rear wheel braking torques are dynamically adjusted based on the control law. When the vehicle slip rate reaches the target value, the control system maintains the equivalent control torque.

[0143] It should be noted that sliding mode control is a nonlinear control method that aims to push the state of a variable structure control system toward its desired value. This is achieved by changing the control action of the system dynamics by applying discontinuous signals. Discontinuous signals exhibit switching control behavior between some mathematically defined control boundaries, that is, sliding mode control adjusts the control input by high-frequency switching so that the system always evolves toward the desired state. In braking control, when it is detected that the slip rate deviates from the target value, the sliding mode controller will immediately adjust the braking torque to return the slip rate to the set range. Therefore, in the application field of wheel slip control, sliding mode control is often selected as the basic control logic due to its robustness under parameter changes and disturbances.

[0144] Specifically, let the actual slip rate λ of the actual wheel be i is the system state (λ i Including the actual slip rate λ of the front axle factualand the actual slip rate of the rear axle λr actual ), λ i,ref is the target slip ratio (λ i,ref Including the target slip ratio λf of the front axle target and the target slip ratio λr of the rear axle target ), the tracking error e'(t) is e'(t) = λ i (t)-λ i,ref (t), i={F,R}, the slip rate dynamic equation can be obtained as:

[0145]

[0146] In the formula, is the rate of change of slip rate, R is the wheel radius, J is the moment of inertia of the wheel, V ix is the translational velocity of the wheel, is the translational acceleration of the wheel, f ix is the longitudinal friction force, T i To control the torque.

[0147] The sliding mode control surface is defined as:

[0148]

[0149] Where s(λ,t) is the sliding surface function, s is the sliding surface state variable, that is, the deviation e'(t) between the actual slip rate and the target slip rate, is the rate of change of slip ratio deviation.

[0150] The sliding surface function is used to describe the state deviation in sliding mode control, and is defined as zero when the slip rate reaches the expected target value. This enables the system to adjust the tire slip rate through feedback control to optimize braking performance.

[0151] For steady-state controlled slip conditions The steady-state slip control can be obtained:

[0152]

[0153] Taking all the above factors into consideration, in order to derive the control law μ=g(x,t) to ensure e'(t)→0, for t→∞, the Lyapunov function (the form of this function is similar to the energy function, which is used to measure the degree to which the system deviates from the desired state. The goal is to make it monotonically decreasing to ensure that s converges to 0) is defined as:

[0154]

[0155] Satisfy the stability condition of the Lyapunov function:

[0156]

[0157] The sliding mode control conditions are derived and the sliding mode control law is obtained:

[0158]

[0159] When η>0, the system is in a sliding mode. The goal is to derive a control law that satisfies the stability conditions and ensures that the system does not move away from the sliding surface. Therefore, substituting equation (16) into equation (19) yields:

[0160]

[0161] Combining the slip rate dynamics equation, formula (14) and formula (20) can be obtained:

[0162]

[0163] Then the final control law can be defined as:

[0164] T i =T eq +T s (twenty two)

[0165] The control law is composed of the equivalent control torque and the switch control torque to ensure that the slip rate converges to the target value. eq is the equivalent control torque, which is used to make the system state converge to the sliding surface and ensure that the system slides stably on the sliding surface. And solve the slip rate dynamics equation (14) to obtain T eq . T s is the switch control torque, which is used to compensate when the system deviates from the sliding surface to force the system to return to the sliding surface. s This can be achieved using symbolic functions: T s =k s sign(s), where k s is the switch control gain, and sign(s) determines the direction of the correction torque.

[0166] During automatic emergency braking, the nonlinear vehicle dynamics model needs to adjust the slip ratio in a very short time to optimize the braking effect. The sliding mode control strategy can quickly adjust the slip ratio so that the tires work within the optimal adhesion range, ensuring the shortest braking distance and reducing the risk of collision. In intelligent cruise control, the nonlinear vehicle dynamics model can quickly adjust the following distance according to the control torque provided by the PID speed controller to ensure that the appropriate safety distance is maintained with the vehicle in front, without causing rear-end collision or excessive braking due to response lag.

[0167] In this embodiment, the underlying control module uses an artificial neural network to optimize the calculation of the equivalent control torque in the sliding mode control mode, specifically including:

[0168] The longitudinal friction force and the wheel translation speed are used as input data of the artificial neural network to obtain the friction force prediction value output by the artificial neural network;

[0169] The friction force prediction is combined with the known vehicle dynamics parameters to obtain the equivalent control torque;

[0170] Among them, the artificial neural network takes the longitudinal friction force and the wheel translation speed as input data, calculates the error between the predicted friction force and the true value of the longitudinal friction force to generate a loss value, and uses the back propagation optimization algorithm to update the parameters of the artificial neural network. During the supervised training process, it is continuously iterated and optimized until the error converges or the set training conditions are reached until convergence, so as to obtain a trained artificial neural network.

[0171] Specifically, an artificial neural network is used to evaluate T eq The artificial neural network consists of the following hyperparameters and features:

[0172] The input vector is x=[f ix V ix ], f ix is the longitudinal friction force, V ix is the wheel translation speed.

[0173] The feedforward structure consists of an input layer with 2 input neurons, a hidden layer with 20 hidden neurons, and an output layer with a single neuron. In the hidden layer, the hyperbolic tangent sigmoid activation function is selected to enhance the nonlinear fitting ability of the neural network. In the output layer, a linear activation function is used to output the friction prediction value.

[0174] The artificial neural network can be written in a compact form as:

[0175]

[0176] In the formula, is the friction force prediction value predicted by the neural network, x is the input vector of the neural network, V is the weight matrix from the input layer to the hidden layer, W is the weight matrix from the hidden layer to the output layer, b is the bias of the hidden layer, B is the bias of the output layer, and e is the base of the natural exponential function.

[0177] The loss function uses the mean square error (MSE) to calculate the predicted value of the neural network. The error between the true value f. The optimization algorithm uses the Levenberg-Marquardt training algorithm, which is an optimization algorithm that combines the gradient descent method with the Newton method to accelerate convergence and reduce overfitting.

[0178] The maximum number of training iterations epoch is set to 3000.

[0179] The training process is: 1. Initialize network weights and biases; 2. Calculate predictions; 3. Calculate errors; 4. Perform error backpropagation and update weights until the loss function converges or the maximum number of training times is reached.

[0180] After training, the prediction results of the artificial neural network can be used to optimize the sliding mode control strategy during braking. eq The true value f or its estimate can be With known kinetic parameters Multiply to calculate.

[0181] The true value of f can be calculated by the following formula:

[0182]

[0183] Where R is the wheel radius, J is the moment of inertia of the wheel, V ix is the translational velocity of the wheel, is the translational acceleration of the wheel, f ix is the longitudinal friction, λ i is the actual slip rate.

[0184] Compared with the traditional formula (24) to calculate the true value of friction, it is usually necessary to model the tire-road contact mechanics relationship. The model may have certain idealized assumptions and it is difficult to fully and accurately describe the friction characteristics under all working conditions. The artificial neural network model can automatically approximate the true friction characteristics by learning data under different working conditions, thereby maintaining high accuracy in more complex driving environments. Moreover, the trained artificial neural network model can directly map the input variables to the friction estimate value, which has a faster calculation speed and is suitable for real-time control applications.

[0185] The embodiment of the present application also provides a vehicle, which may specifically include the automatic emergency braking hierarchical control system. The vehicle may specifically be a passenger car, a commercial vehicle, an off-road vehicle, or a special-purpose vehicle, etc. The vehicle may be an internal combustion engine vehicle with an internal combustion engine as a driving source, an electric vehicle or a fuel cell vehicle with an electric motor as a driving source, a hybrid vehicle with the above two as driving sources, or a vehicle with other driving sources. The vehicle may also include an autonomous driving vehicle.

[0186] In summary, the embodiments of the present application provide an automatic emergency braking hierarchical control system and a vehicle, which are mainly composed of a high-level rule supervision control module, a middle-level control algorithm switching module and a bottom-level control module. A hierarchical control architecture is designed, in which the high-level rule supervision control module is mainly responsible for environmental perception and collision risk assessment to avoid unnecessary braking intervention; the middle-level control algorithm switching module is responsible for selecting the most appropriate control algorithm; the bottom-level control module specifically performs bottom-level operations such as braking force distribution and slip rate control to ensure the real-time responsiveness of the system. Different levels are responsible for different control objectives, which improves the modularity and scalability of the system. At the same time, it avoids response delays caused by the complexity of a single control logic, especially in autonomous driving or advanced driver assistance systems, without having to consider the computational complexity of the entire system, reducing the computational burden of the processor and improving the real-time and robustness of the system. Through this hierarchical control architecture, the AEB system can make autonomous decisions, adopt different braking strategies in different situations, and improve the level of intelligence.

[0187] The flow chart or block diagram in the accompanying drawings shows the possible architecture, function and operation of the equipment, method and computer program product according to various embodiments of the present application. In this regard, each box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flow chart, and the combination of the boxes in the block diagram and / or flow chart can be implemented with a dedicated system for hardware that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0188] The technical features of the above-described embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above-described embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0189] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art can easily mention changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims, and the above embodiments should be regarded as exemplary and non-restrictive.

Claims

1. An automatic emergency braking hierarchical control system, characterized in that: include: A braking parameter information acquisition module is used to acquire state parameters that affect the braking control of the vehicle, wherein the state parameters include: the relative distance between the vehicle and the preceding vehicle, the longitudinal speeds of the vehicle and the preceding vehicle in the driving direction, and the road adhesion coefficient; A relative distance threshold calculation module, used to dynamically and adaptively adjust the safe braking distance based on the longitudinal speeds of the vehicle and the preceding vehicle in the travel direction, and generate an adaptive relative distance threshold; a high-level rule supervisory control module, configured to evaluate a potential collision risk in real time based on the relative distance and the adaptive relative distance threshold, and generate a sliding control demand and a corresponding target deceleration based on a level of the potential collision risk; a middle-level control algorithm switching module, configured to switch between different sliding control modes based on the sliding control requirement, the target deceleration and the current vehicle state; The bottom control module is used to calculate the cruising speed of the vehicle during the following process, or calculate the target slip rate of the vehicle during the braking process and apply the corresponding braking torque based on the target deceleration, the slip control mode and the current vehicle state.

2. The automatic emergency braking hierarchical control system according to claim 1, characterized in that: The relative distance threshold calculation module generates an adaptive relative distance threshold, specifically including: Determine whether the longitudinal speed of the vehicle is greater than a preset speed threshold and whether there is a vehicle ahead on the road; When the longitudinal speed of the vehicle is greater than the speed threshold and there is a vehicle ahead on the road, the minimum braking distance is calculated based on the longitudinal speed of the vehicle and the peak friction coefficient of the road, and the static safety distance margin is added to obtain the relative distance threshold.

3. The automatic emergency braking hierarchical control system according to claim 2, characterized in that: The high-level rule supervisory control module generates a braking control requirement and a corresponding target deceleration, specifically including: Combined with the speed and acceleration of the preceding vehicle detected by the sensor, the collision time calculation method is used to analyze the relative motion trend between vehicles; When the relative distance is greater than the relative distance threshold and the collision time is greater than the set first safety threshold, the system determines that the risk is low, sends a vehicle following control request to the middle-level control algorithm switching module, and calculates the expected speed with the target deceleration of 0 as the cruising speed of the vehicle; When the relative distance is close to the relative distance threshold and the collision time is greater than the set second safety threshold, the system determines that the risk is medium, sends a deceleration control request to the middle-level control algorithm switching module, and calculates the target deceleration based on the longitudinal speed of the vehicle and the minimum braking distance, and the second safety threshold is less than the first safety threshold; When the relative distance is less than the relative distance threshold and the collision time is less than or equal to the second safety threshold, the system determines it as a high risk, sends an emergency braking control request to the middle-level control algorithm switching module, and calculates the target deceleration based on the longitudinal speed of the vehicle and the minimum braking distance.

4. The automatic emergency braking hierarchical control system according to claim 3, characterized in that: The high-level rule supervisory control module generates a braking control demand and a corresponding target deceleration, and further includes: During the following and braking process, the relative distance between the vehicle and the front vehicle and the vehicle status are continuously monitored. If the front vehicle further decelerates, the potential collision risk is re-evaluated and the target deceleration is updated.

5. The automatic emergency braking hierarchical control system according to claim 1, characterized in that: The sliding control mode includes: a proportional-integral-derivative control mode and a sliding mode control mode.

6. The automatic emergency braking hierarchical control system according to claim 1 or 5, characterized in that: When the middle-level control algorithm switching module switches to the proportional-integral-derivative control mode, the bottom-level control module adopts the PID control strategy to optimize the longitudinal motion control of the vehicle during the following process, specifically including: A speed error is calculated based on the desired longitudinal speed and the actual longitudinal speed of the vehicle, and the speed error is input into a proportional-integral-differential controller. The driving torque or braking torque of the vehicle is dynamically adjusted according to an output signal of the proportional-integral-differential controller so that the actual longitudinal speed gradually converges to the desired longitudinal speed. If the speed error is positive, the driving torque is adjusted to increase the vehicle acceleration; if the speed error is negative, the braking torque is adjusted to reduce the vehicle acceleration.

7. The automatic emergency braking hierarchical control system according to claim 1 or 5, characterized in that: When the middle-level control algorithm switching module switches to the sliding mode control mode, the bottom-level control module adopts the sliding mode control strategy to calculate the control torque through slip rate error feedback during braking, specifically including: Construct the sliding surface and adjust the system state toward the expected steady-state controllable slip condition by calculating the slip rate error; A control law satisfying the steady-state controllable slip condition is derived, wherein the control law is in the form of a combination of an equivalent control torque and a switch control torque, wherein the equivalent control torque is used to maintain the stable operation of the system state on the sliding surface, and the switch control torque is used to apply a correction torque when the system deviates from the sliding surface, so as to quickly pull the system state back to the sliding surface; During braking, the front and rear wheel braking torques are dynamically adjusted based on the control law. When the vehicle slip rate reaches the target value, the control system maintains the effect of the equivalent control torque.

8. The automatic emergency braking hierarchical control system according to claim 7, characterized in that: The bottom control module uses an artificial neural network to optimize the calculation of the equivalent control torque in the sliding mode control mode, specifically including: Using the longitudinal friction force and the wheel translation speed as input data of the artificial neural network to obtain a friction force prediction value output by the artificial neural network; Combining the friction force prediction value with known vehicle dynamics parameters to obtain the equivalent control torque; Among them, the artificial neural network takes the longitudinal friction force and the wheel translation speed as input data, calculates the error between the friction force prediction value and the true value of the longitudinal friction force to generate a loss value, uses the back propagation optimization algorithm to update the parameters of the artificial neural network, and continuously iterates and optimizes during the supervised training process until the error converges or reaches the set training conditions until convergence, so as to obtain the trained artificial neural network.

9. The automatic emergency braking hierarchical control system according to claim 1, characterized in that: The underlying control module calculates the dynamic control parameters of the vehicle during sliding based on a nonlinear vehicle dynamics model. The nonlinear vehicle dynamics model is represented by coupling a nonlinear bicycle model and a nonlinear tire model. The nonlinear bicycle model is used to describe the longitudinal and lateral dynamic behaviors of the vehicle. The nonlinear tire model is used to simulate the contact force characteristics between the tire and the road surface and calculate the longitudinal friction force and normal load.

10. A vehicle, characterized in that: It comprises an automatic emergency braking hierarchical control system as described in any one of claims 1-9.

Images