Vehicle path tracking and body posture collaborative control method

Abstract

The application discloses a vehicle path tracking and vehicle body posture cooperative control method, and specifically comprises the following steps: step 1, starting a path tracking and vehicle body posture cooperative controller of an intelligent vehicle; step 2, starting a vehicle body posture monitoring device, and monitoring a vehicle body posture in real time; and step 3, according to a real-time monitoring result of step 2, the intelligent vehicle adopts a corresponding control strategy, evaluates the cooperative control strategy, updates a vehicle state, further controls the vehicle correspondingly, and the path tracking function is ended until then. The application designs corresponding working condition driving control logic, and improves driving safety and riding comfort of the intelligent vehicle.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent vehicle road driving safety, and particularly relates to a path tracking and vehicle body posture coordinated control method for intelligent vehicles. BACKGROUND

[0002] With the development of the new four modernizations of automobiles, intelligent vehicles have become a research hotspot at present and in the future. Intelligent vehicles have many advantages such as predictable driving behavior, reduced traffic accident rate, and improved commuting efficiency, and will have wide application in future intelligent traffic methods. Motion control, as one of the three core key technologies of intelligent vehicles, directly affects the accuracy of path tracking and driving quality. The path tracking research of intelligent vehicles aims to ensure driving safety and ride comfort, and to make the vehicle travel along a pre-planned path through steering control. The goal of path tracking is to eliminate the angle deviation and lateral deviation between the actual position and the expected position of the vehicle during driving, so as to achieve accurate path tracking.

[0003] However, most of the path tracking algorithm strategies of intelligent vehicles at present only consider path tracking accuracy, without considering the influence of changes in vehicle body posture on passenger comfort and the consequences of vehicle rollover caused by further deterioration of vehicle body posture in the process of ensuring tracking accuracy, and ignoring the mutual coupling and constraint relationship between vertical vehicle body posture and lateral path tracking accuracy.

[0004] The purpose of path tracking is to control the vehicle to travel accurately along the expected path, that is, to reduce the driving deviation of the vehicle as much as possible during path tracking. Most of the current path tracking control optimization methods use optimization preview distance to achieve this, but generally, the preview distance is expressed through a mathematical model or a preview distance adaptive strategy is designed, which to some extent improves the path tracking effect, but this technical route still has limitations. The selection of preview distance is affected by many factors, and the present application adopts the idea of controlling the front wheel steering angle to compensate and optimize the path tracking accuracy.

[0005] In actual application scenarios, intelligent vehicles not only need to realize path tracking function, but also need to ensure the stability of the vehicle itself, especially when subjected to external interference, which has a greater impact on the lateral stability of the vehicle. The lateral motion control system should be able to correct the posture of the vehicle in time to ensure the driving safety of the vehicle. SUMMARY

[0006] Based on the shortcomings of the prior art, the present application proposes a path tracking and vehicle body posture coordinated control method for comprehensively considering tracking accuracy, ride comfort, and driving safety during vehicle driving, which specifically includes the following steps:

[0007] Step 1, the intelligent vehicle opens the path tracking in the vehicle body posture coordination controller, and the expected path is a curve with known coordinates;

[0008] Step 2, open the vehicle state monitoring device, real-time monitor the path tracking accuracy, steering stability metric value, comfort metric value of the vehicle; the driving situation of the intelligent vehicle is divided into three categories: 1. good path tracking accuracy; 2. path tracking accuracy is not enough, and there is a lane changing condition; 3. path tracking accuracy is not enough, and there is no lane changing condition; the specific eight driving situations are as follows:

[0009] 8) when the path tracking accuracy, steering stability metric value, and comfort metric value are all lower than the corresponding set threshold, the vehicle is normally driven, and the vehicle is normally path tracked;

[0010] 9) when the path tracking accuracy is lower than the set threshold, but the steering stability and comfort metric values do not meet the set threshold or do not meet the set threshold, the vehicle body posture compensation is performed on the vehicle;

[0011] 10) when the path tracking accuracy does not meet the set threshold and there is a lane changing condition, the steering stability and comfort of the vehicle are prioritized, and the vehicle body posture compensation is performed on the vehicle;

[0012] 11) when the path tracking accuracy does not meet the set threshold and there is a lane changing condition, but the lane changing has a rollover danger, the safety of the vehicle is prioritized, the vehicle is decelerated and corrected before lane changing, and if the correction is successful, the vehicle is lane changed and the vehicle body posture compensation is performed;

[0013] 12) if the correction fails, the vehicle cannot change lanes, at which time the automatic emergency braking system AEB is started and the driver is warned to intervene;

[0014] 13) when the path tracking accuracy does not meet the set threshold and there is no lane changing condition, the vehicle is path tracked and compensated, and if the steering stability is not deteriorated after the path tracking compensation, the vehicle is controlled according to this logic;

[0015] 14) when the path tracking accuracy does not meet the set threshold and there is no lane changing condition, the vehicle is path tracked and compensated, and after the path tracking compensation, the steering stability is deteriorated, at which time the vehicle body posture compensation control is also needed;

[0016] 15) when the vehicle body posture compensation and path tracking accuracy compensation control cannot effectively correct the vehicle state, emergency braking, warning, and driver intervention operation are needed;

[0017] Step 3, according to the real-time monitoring results of the eight driving situations in step 2, the intelligent vehicle adopts the corresponding control strategy, evaluates the coordination control strategy, updates the vehicle state, and further controls the vehicle accordingly until the path tracking function ends.

[0018] Further, the first driving scene real-time monitoring result, the intelligent vehicle adopts the corresponding control strategy, which is realized by the following steps:

[0019] Step 2A-1, based on the two-degree-of-freedom vehicle dynamics equation, a two-degree-of-freedom vehicle dynamics model is established, according to the relationship between the tire side slip angle and the vehicle mass center side slip angle, yaw rate and the distance between the mass center and the front and rear shafts, combined with Newton's second law, the motion differential equation of the linear two-degree-of-freedom model is obtained:

[0020]

[0021] In the formula, m is the mass of the vehicle;

[0022] C f , Cr is the linear side stiffness of the front and rear wheels;

[0023] l f , l r is the distance between the mass center of the vehicle and the front and rear shafts;

[0024] I Z is the moment of inertia of the vehicle around the Z axis;

[0025] δ f is the front wheel steering angle;

[0026] V x is the longitudinal vehicle speed;

[0027] β is the mass center side slip angle;

[0028] r is the yaw rate;

[0029] Step 2A-2: Design of preview error system; Path tracking technology refers to that in the actual driving process of the intelligent vehicle, the corresponding road information is obtained through the sensor, and the driving direction of the vehicle is controlled by using appropriate control algorithm, so that the vehicle does not deviate from the set trajectory in the driving process, and the safety in the driving process is also guaranteed; The inputs of the preview error model include the output mass center side slip angle β and the yaw rate r of the vehicle dynamics model, as well as the external input road curvature ρ and the preview distance x e , the outputs of the preview error model are the lateral displacement deviation y e and the lateral position deviation e, and the formula of the preview error model of the vehicle is:

[0030]

[0031] In the formula, y e is the lateral displacement deviation;

[0032] v y is the lateral velocity;

[0033] v x —Longitudinal speed;

[0034] e – Lateral orientation deviation;

[0035] r—yaw rate;

[0036] ρ — Road curvature;

[0037] x e —Pre-aiming distance;

[0038] Step 2A-3: Regarding the lateral displacement deviation y e The weighted deviation e is then used as the comprehensive deviation E;

[0039] Step 2A-4: Input the comprehensive deviation E obtained in step 2A-3 into the PID controller. After a series of calculations, the front wheel steering angle δ of the vehicle is output. f As the control variable of the vehicle dynamics model, this forms a closed-loop control system for lateral motion tracking, thereby realizing the path tracking function.

[0040] Furthermore, based on the real-time monitoring results of driving scenarios 2), 3), and 4), vehicle body posture compensation is performed to reduce the vehicle's lateral acceleration and correct its lateral movement. This is achieved through the following steps:

[0041] Step 2B-1: The main method to improve vehicle posture is by controlling vehicle roll. Roll will generate a roll moment, which consists of three parts: 1. Roll moment M caused by the centrifugal force of the suspension mass. ΦrⅠ 2. The roll moment M caused by the weight of the suspended mass ΦrⅡ 3. The tilting moment M caused by the centrifugal force of the unsuspended mass. ΦrⅢ 4. In addition, during roll, the vertical load is transferred between the left and right wheels, generating a load transfer moment M. ZF M ZR When the vehicle body is in a tilted state, if the left and right suspensions can generate additional forces Δf in opposite directions under the current state, an anti-roll moment M can be formed. af This can suppress vehicle roll;

[0042] The roll moment M caused by the centrifugal force of the suspended mass ΦrⅠ for:

[0043] M ΦrⅠ =m s ·a y ·h

[0044] The roll moment M caused by the weight of the suspended mass ΦrⅡ for:

[0045] M ΦrⅡ = m s · g · e ≈ m s · g · h g · φ

[0046] Rolling moment M caused by centrifugal force of non-suspended mass ΦrⅢ is:

[0047] M ΦrⅢ = - F uy (h0 - r)

[0048] Load transfer moment M caused by load transfer of vertical load on left and right wheels during rolling ZF , M ZR is:

[0049] M ZF = (F rRF - F rLF ) · B / 2

[0050] M ZR = (F rRR - F rLR ) · B / 2

[0051] Taking moment about the longitudinal centerline of the vehicle body, we have:

[0052] M ΦrⅠ + M ΦrⅡ + M ΦrⅢ + M ZF + M ZR = M af

[0053] where m s —suspended mass;

[0054] a y —lateral acceleration;

[0055] h—distance from the center of mass of the suspended mass to the roll axis;

[0056] g—acceleration due to gravity;

[0057] h g —distance from the center of mass of the suspended mass to the ground;

[0058] φ—vehicle body roll angle;

[0059] F uy —centrifugal force generated by non-suspended mass;

[0060] h0—distance from the roll center to the ground;

[0061] r—wheel radius;

[0062] F rRF 、F rLF — roll rear front axle left and right wheel vertical load;

[0063] F rRR 、F rLR — roll rear rear axle left and right wheel vertical load;

[0064] B — wheel track;

[0065] Step 2B-2: By setting a three-axis gyroscope at the vehicle body mass center, the vertical acceleration, roll angle acceleration and roll angle signals at the vehicle body mass center are sent to the controller, the current roll moment of the vehicle is calculated, and the opposite moment is provided to the vehicle through the actuator on the active suspension to resist the roll movement of the vehicle. Through the above steps, the monitoring and control of the first, second, third and fourth driving scenarios are realized.

[0066] Further, the real-time monitoring results of the fifth and eighth driving scenarios, the intelligent vehicle adopts the corresponding control strategy, which is realized through the following steps:

[0067] Step 2C, emergency braking control method, in the case that path tracking compensation and vehicle body posture compensation are unsuccessful, the automatic emergency braking system AEB is started, the vehicle speed is reduced, and the vehicle state is controllable. The AEB system mainly includes three parts: information acquisition module, control module and execution module. The front camera, millimeter wave radar or laser radar and other devices are used to identify the obstacles in front of the vehicle, and then the driving state information of the vehicle and the front vehicle, or the information of the front obstacles or pedestrians is transmitted to the control module of the AEB system in real time. The control module calculates and judges the danger level of the vehicle according to the motion state information of the vehicle. When the system determines that there is a collision risk between the vehicle and the front vehicle, the driver is warned in the form of light, sound, etc. If the driver does not take effective measures, the system will automatically control the vehicle to brake in emergency.

[0068] Further, the real-time monitoring results of the sixth driving scenario, the intelligent vehicle adopts the corresponding control strategy, which is realized through the following steps:

[0069] Step 2D, path tracking compensation control method, based on PI control theory, the lateral position deviation, direction deviation and road curvature are comprehensively considered to design the turn angle compensation control law as follows:

[0070]

[0071] In the formula, δ compis the front wheel steering angle compensation for lateral deviation; R1 is the lateral position deviation proportional coefficient; Q1(k(t)) is the lateral deviation integral function; R2 is the direction deviation proportional coefficient; Q2(k(t)) is the lateral deviation proportional coefficient; k(t) is the expected path road curvature at the current time; e y is the actual lateral position deviation at the vehicle mass center at the current time; is the actual direction deviation at the vehicle mass center at the current time;

[0072] In order to reduce the overshoot of the lateral deviation and the direction deviation of the system tracking the target path under large curvature, the integral functions Q1(k(t)), Q2(k(t)) are set as functions of the road curvature:

[0073]

[0074] In the formula, ω1, ω2 are the lateral position and direction deviation integral coefficients;

[0075] R1, ω1 mainly represent the front wheel steering angle compensation of the system lateral deviation term, R2, ω2 mainly represent the front wheel steering angle compensation of the system direction deviation term, in order to make the compensation controller well adapt to the changes of system parameters and the interference of external environment, the system parameters R1, R2, ω1, ω2 in the compensation controller will be based on the idea of fuzzy control, two different fuzzy controllers are designed based on the system lateral deviation and direction deviation respectively. The lateral deviation e y and the lateral deviation change rate are inputs, the lateral deviation proportional coefficient R1 and the integral coefficient ω1 are outputs to design the lateral deviation fuzzy controller; the direction deviation and the direction deviation change rate are inputs, the direction deviation proportional coefficient R2 and the integral coefficient ω2 are outputs to design the direction deviation fuzzy controller, two inputs and two outputs are used to obtain the change law of R1, ω1; R2, ω2, and the change law is optimized;

[0076] In summary, the steering angle compensation control law is:

[0077]

[0078] Therefore, the front wheel steering angle control law of the path tracking compensation control is:

[0079]

[0080] Further, the real-time monitoring result of the 7th driving scenario, the intelligent vehicle adopts the corresponding control strategy, which is realized through the following steps:

[0081] Step 2E, tracking accuracy and vehicle body posture compensation at the same time, wherein the tracking accuracy compensation is realized through the control strategy of the 6th driving situation; the vehicle body posture compensation is realized through the control strategies of the 2nd, 3rd and 4th driving situations.

[0082] Further, after the compensation control of the cooperative controller, the state of the vehicle is monitored and judged again, the compensation control effect is evaluated in real time, and if the compensation is successful, the cooperative control is continued, otherwise the driver intervenes to take over the control of the vehicle to avoid the occurrence of danger.

[0083] The present application has the following technical effects:

[0084] 1. The present application realizes real-time monitoring of the qualitative parameters of the tracking accuracy and the vehicle body posture during driving, and real-time acquisition of the current state of the vehicle.

[0085] 2. The present application realizes different control strategies for the vehicle under different driving situations, i.e. when the vehicle state is good, the normal path tracking control strategy is implemented for the vehicle; when the vehicle state deteriorates, the path tracking and vehicle body posture cooperative control strategy is implemented for the vehicle.

[0086] 3. The present application realizes the cooperative control of the path tracking accuracy and the vehicle body posture, improves the path tracking effect on the basis of ensuring the stable driving of the vehicle, and realizes the comprehensive improvement of the tracking accuracy and the vehicle body stability. BRIEF DESCRIPTION OF DRAWINGS

[0087] Figure 1 Intelligent vehicle road driving situation diagram; a) good path tracking accuracy; b) insufficient path tracking accuracy and lane changing condition; c) insufficient path tracking accuracy and no lane changing condition;

[0088] Figure 2 is a vehicle two-degree-of-freedom dynamics model diagram;

[0089] Figure 3 is a preview error model diagram;

[0090] Figure 4 is a principle diagram of roll moment generation;

[0091] Figure 5 is a vehicle body posture compensation effect diagram;

[0092] Figure 6 is a path tracking (direction deviation and longitudinal position) compensation effect diagram;

[0093] Figure 7 is a path tracking (lateral deviation and longitudinal position) compensation effect diagram;

[0094] Figure 8The vehicle path tracking and vehicle body posture collaborative control logic roadmap. DETAILED DESCRIPTION

[0095] The driving scenarios of the intelligent vehicle are mainly divided into three categories: 1) good path tracking accuracy; 2) insufficient path tracking accuracy and lane changing condition; and 3) insufficient path tracking accuracy and no lane changing condition. As shown in the following table. Figure 1

[0096] Step 1, the intelligent vehicle starts the path tracking and vehicle body posture collaborative controller, and the expected path is a curve with known coordinates;

[0097] Step 2, the vehicle state monitoring device is started, and the path tracking accuracy, steering stability metric value and comfort metric value of the vehicle are monitored in real time. The specific driving scenarios of the vehicle are as follows:

[0098] When the path tracking accuracy, steering stability metric value and comfort metric value are all lower than the corresponding set threshold, the vehicle is normally driven, and the vehicle is normally path tracked; when the path tracking accuracy is lower than the set threshold, but the steering stability and comfort metric values do not meet the set threshold, the vehicle body posture compensation is performed on the vehicle; when the path tracking accuracy does not meet the set threshold and there is a lane changing condition, the steering stability and comfort of the vehicle are prioritized, and the vehicle body posture compensation is performed on the vehicle; when the path tracking accuracy does not meet the set threshold and there is no lane changing condition, the path tracking compensation is performed on the vehicle, and the steering stability is not deteriorated after the path tracking compensation, then the vehicle is controlled according to this logic, and the steering stability is deteriorated after the path tracking compensation, then the vehicle body posture compensation control is further needed; when the vehicle body posture compensation and path tracking accuracy compensation control cannot effectively correct the vehicle state, the vehicle needs to be controlled by deceleration, warning, driver intervention operation, etc.

[0099] Step 2A-1, a two-degree-of-freedom vehicle dynamics model is established based on a two-degree-of-freedom vehicle dynamics equation. As shown in the following figure, the two-degree-of-freedom vehicle dynamics model is as follows: Figure 2

[0100] According to the relationship between the tire side slip angle and the vehicle mass center side slip angle, the yaw rate and the distance from the mass center to the front and rear axles, and further organizing according to Newton's second law, the motion differential equation of the linear two-degree-of-freedom model is obtained as follows:

[0101]

[0102] In the formula, m is the mass of the vehicle;

[0103] C f , Cr are the front and rear wheel linear side stiffness;

[0104] l f , l r ​​— distance of vehicle center of mass from front and rear axles;

[0105] I Z — moment of inertia of the whole vehicle about the Z axis;

[0106] δ f — front wheel steering angle;

[0107] V x — longitudinal vehicle speed;

[0108] β— center of mass side slip angle;

[0109] r— yaw rate;

[0110] Step 2A-2: Design of the preview error system; Path following technology refers to that in the actual driving process, the intelligent vehicle obtains the corresponding road information through sensors, and uses appropriate control algorithm to control the driving direction of the vehicle, so that the vehicle does not deviate from the set trajectory in the driving process, and also ensures the safety in the driving process. The input of the preview error model includes the output of the vehicle dynamics model, the center of mass side slip angle and the yaw rate, as well as the external input road curvature ρ and preview distance x e , and the output of the preview error model is the lateral displacement deviation y e and the lateral orientation deviation e, and the preview error model is shown in Figure 3 :

[0111] Vehicle preview error model formula:

[0112]

[0113] In the formula, y e — lateral displacement deviation;

[0114] v y — lateral velocity;

[0115] v x — longitudinal vehicle speed;

[0116] e— lateral orientation deviation;

[0117] r— yaw rate;

[0118] ρ— road curvature;

[0119] x e — preview distance;

[0120] Step 2A-3: After the lateral displacement deviation y e and the lateral orientation deviation e are weighted and allocated, they are used as the comprehensive error;

[0121] Step 2A-4: Input the comprehensive deviation obtained in Step 2-3 into the PID controller. After a series of calculations, the front wheel steering angle δ of the vehicle is output. f As the control variable of the vehicle dynamics model, this forms a closed-loop control system for lateral motion tracking, thereby realizing the path tracking function.

[0122] Step 2B-1: This invention primarily improves vehicle posture by controlling vehicle roll. Rolling generates a roll moment, which consists of three parts: 1. Roll moment M caused by the centrifugal force of the suspension mass. ΦrⅠ 2. The roll moment M caused by the weight of the suspended mass ΦrⅡ 3. The tilting moment M caused by the centrifugal force of the unsuspended mass. ΦrⅢ 4. In addition, during roll, the vertical load is transferred between the left and right wheels, generating a load transfer moment M. ZF M ZR A schematic diagram of the roll effect, as shown below. Figure 4 As shown. When the vehicle body is in a tilted state, if the left and right suspensions can generate additional forces Δf in opposite directions under the current state, an anti-roll moment M can be formed. af It can suppress vehicle tilt.

[0123] The roll moment M caused by the centrifugal force of the suspended mass ΦrⅠ for:

[0124] M ΦrⅠ =m s ·a y ·h

[0125] The roll moment M caused by the weight of the suspended mass ΦrⅡ for:

[0126] M ΦrⅡ =m s ·g·e≈m s ·g·h g ·φ

[0127] The roll moment M caused by the centrifugal force of the unsuspended mass ΦrⅢ for:

[0128] M ΦrⅢ =-F uy (h0-r)

[0129] When tilting, the vertical load is transferred between the left and right wheels, generating a load transfer moment M. ZF M ZR for:

[0130] M ZF =(F rRF -F rLF )·B / 2

[0131] M ZR = (F rRR -F rLR ) · B / 2

[0132] Taking moment about the longitudinal centerline of the vehicle body, we have:

[0133] M ΦrⅠ + M ΦrⅡ + M ΦrⅢ + M ZF + M ZR = M af

[0134] where m s — the suspended mass;

[0135] a y — lateral acceleration;

[0136] h— the distance from the center of mass of the suspended mass to the roll axis;

[0137] g— the acceleration due to gravity;

[0138] h g — the distance from the center of mass of the suspended mass to the ground;

[0139] φ— the roll angle of the vehicle body;

[0140] F uy — centrifugal force generated by the non-suspended mass;

[0141] h0— the distance from the roll center to the ground;

[0142] r— the wheel radius;

[0143] F rRF , F rLF — vertical load on the left and right wheels of the front axle after roll;

[0144] F rRR , F rLR — vertical load on the left and right wheels of the rear axle after roll;

[0145] B— the track;

[0146] Step 2B-2: The present application obtains the vertical acceleration, roll angle acceleration and roll angle signal at the center of mass of the vehicle body by setting a three-axis gyroscope at the center of mass of the vehicle body, sends them to the controller, calculates the current roll moment of the vehicle, and provides a counteracting moment to the vehicle through the actuator on the active suspension to resist the roll movement of the vehicle.

[0147] Step 2C, path tracking compensation control method. Based on PI control theory, the lateral position deviation, direction deviation, and road curvature are considered to design the turn angle compensation control law as follows:

[0148]

[0149] where δ comp is the front wheel turn angle compensation; R1 is the lateral position deviation proportional coefficient; Q1(k(t)) is the lateral deviation integral function; R2 is the direction deviation proportional coefficient; Q2(k(t)) is the lateral deviation proportional coefficient; K is the desired path road curvature at the current time; e y is the actual lateral position deviation of the vehicle mass center at the current time; is the actual direction deviation of the vehicle mass center at the current time.

[0150] where Q1(k(t)), Q2(k(t)) are set as functions of road curvature to reduce the lateral deviation overshoot and direction deviation overshoot of the system tracking the target path under large curvature:

[0151]

[0152] where ω1, ω2 are the lateral position and direction deviation integral coefficients.

[0153] R1, ω1 mainly represent the front wheel turn angle compensation of the system lateral deviation term, and R2, ω2 mainly represent the front wheel turn angle compensation of the system direction deviation term. In order to make the compensation controller well adapt to the changes of system parameters and external environmental disturbances, the system parameters R1, R2, ω1, ω2 in the compensation controller will be based on the idea of fuzzy control, and two different fuzzy controllers will be designed based on the system lateral deviation and direction deviation, respectively. The lateral deviation e y and the lateral deviation change rate are taken as inputs, and the lateral deviation proportional coefficient R1 and the integral coefficient ω1 are taken as outputs to design the lateral deviation fuzzy controller; the direction deviation and the direction deviation change rate are taken as inputs, and the direction deviation proportional coefficient R2 and the integral coefficient ω2 are taken as outputs to design the direction deviation fuzzy controller, and two-input two-output is used to obtain the change law of R1, ω1; R2, ω2, and the law is optimized.

[0154] Therefore, the front wheel turn angle control law of the path tracking compensation control is:

[0155]

[0156] Therefore, the front wheel turn angle control law of the path tracking compensation control is:

[0157]

[0158] Step 2D, emergency braking control method. In the case of path tracking compensation and body posture compensation, the automatic emergency braking system (AEB) is started, the vehicle speed is reduced, and the vehicle state is controllable. The AEB system mainly includes three parts: information acquisition module, control module and execution module. Through the front camera, millimeter wave radar or laser radar and other devices to identify the obstacles in front of the vehicle, and then the information of the vehicle and the front vehicle driving state or the front obstacle or pedestrian is transmitted to the control module of the AEB system in real time. The control module calculates and judges the danger level of the vehicle according to the motion state information of the vehicle. When the system determines that there is a collision risk between the vehicle and the front vehicle, the driver is warned in the form of light, sound and other ways. If the driver does not take effective measures, the system will automatically control the vehicle to brake in emergency.

[0159] In the process of emergency braking, the anti-lock braking system in the execution module is one of the key subsystems for longitudinal motion control of intelligent vehicles, which also plays a crucial role. The hydraulic ABS is selected as the research object, and the dynamic model and control model of the hydraulic ABS are established based on the original ABS system theory, and finally the ABS motion equation is obtained.

[0160] Based on Newton's second law of motion, ignoring the rolling resistance and air resistance of the wheels, the wheel motion equation, vehicle motion equation and wheel longitudinal friction equation are as follows:

[0161]

[0162]

[0163] F x =μF z

[0164] In the formula, I w , ω w , μ, R, F x , F z and T b are the moment of inertia of the wheel, the angular velocity of the wheel, the tire adhesion coefficient, the wheel radius, the vertical load of the wheel, the longitudinal friction of the wheel and the braking torque of the wheel. The tire adhesion coefficient and the slip ratio between the wheel have a certain nonlinear relationship, and the relationship between the tire adhesion coefficient and the slip ratio of the Dugoff tire model used in this paper.

[0165] The mathematical expression of the longitudinal adhesion coefficient / slip ratio bilinear curve can be derived as follows:

[0166]

[0167] where S0 is the vehicle slip rate corresponding to the peak longitudinal adhesion coefficient, μ h , μ g respectively refer to the peak longitudinal adhesion coefficient of the tire and the longitudinal adhesion coefficient of the wheel when the wheel is fully locked (slip rate is 1).

[0168] In the actual vehicle braking process, the ground braking torque is determined by the brake braking torque, but is restricted by the ground adhesion coefficient. When the braking torque is greater than the maximum ground braking torque, the wheel appears to be locked and dragged, causing the actual inequality of the vehicle speed and the wheel speed, i.e., the slip phenomenon. The longitudinal slip rate of the vehicle can be expressed as:

[0169]

[0170] where V car , V w are the vehicle speed and the wheel speed, respectively. Among them, V w = ω w · R.

[0171] According to the slip rate-longitudinal adhesion coefficient relationship, the rate of change of the tire adhesion coefficient and the rate of change of the slip rate are linearly related under local conditions, so that:

[0172]

[0173] where k μ-s is the correlation coefficient, and based on the relationship curve between the tire adhesion coefficient and the wheel slip rate, k μ-s is positive or negative under different conditions.

[0174] Further derivation of the relationship between the rate of change of the tire adhesion coefficient and the wheel angular acceleration is as follows:

[0175]

[0176] According to the working principle of ABS, the braking system controls the slip rate by adjusting the pressure regulator (solenoid valve) in real time, so that it is kept near the optimal slip rate. Therefore, the change rate of the system corresponding to the braking torque is proportional to the fluid flow rate in the ABS structure, and the liquid flow rate is proportional to the opening of the solenoid valve, so it can be concluded that the change rate of the braking torque is proportional to the control valve command rate. In addition, it needs to be pointed out that the change rate of the braking torque and the control valve command rate are not linearly related, but are monotonically related, i.e., the partial derivative of the system pressure to the fluid is positively related, i.e.

[0177] On this basis, the ABS controller is designed, the application is based on Choi's ABS control related theory, the best slip rate is obtained through the tracking of the wheel target rotation angle, the integral control algorithm has a compensation effect on the time delay ABS system, the subsequent ABS time delay stability research has a certain degree of interference, therefore, the application selects proportion differentiation (PD) control algorithm, and the specific expression is as follows:

[0178]

[0179] The above formula is synthesized, and the ABS motion equation is as follows:

[0180]

[0181] After the best slip rate is obtained, the expected brake pressure on each brake wheel cylinder is obtained, the vehicle is stably and reliably decelerated, and stability is restored.

[0182] Step 3, according to the real-time monitoring result of step 2, the intelligent vehicle adopts the corresponding control strategy, effectively intervenes the driving state of the vehicle, and improves the safety, steering stability and smoothness of the intelligent vehicle driving until the path tracking function ends.

[0183] Figure 4 is a schematic diagram of the roll moment; Figure 5 is a vehicle body posture compensation effect diagram; Figures 6-7 is a path tracking compensation effect diagram; Figure 8 is a vehicle path tracking and vehicle body posture cooperative control logic route diagram.

Claims

1. A vehicle path tracking and body posture cooperative control method characterized by comprising: Comprising the following steps: Step 1, the intelligent vehicle opens the path tracking in the vehicle body posture coordination controller, and the expected path is a curve with known coordinates; Step 2, open the vehicle state monitoring device, real-time monitoring of vehicle path tracking accuracy, steering stability measure, comfort measure; The driving situation of intelligent vehicle is divided into three categories:

1. Good path tracking accuracy; 2. Path tracking accuracy is not enough, and there is a lane changing condition; 3. Path tracking accuracy is not enough, and there is no lane changing condition; The specific eight driving situations are as follows: 1) When the path tracking accuracy, steering stability measure and comfort measure are all lower than the corresponding set threshold, the vehicle is normally driven, and the vehicle is normally path tracked; 2) When the path tracking accuracy is lower than the set threshold, but the steering stability and comfort measure do not meet the set threshold or do not meet the set threshold, the vehicle body posture compensation is performed on the vehicle; 3) When the path tracking accuracy does not meet the set threshold and there is a lane changing condition, the steering stability and comfort of the vehicle are prioritized, and the vehicle body posture compensation is performed on the vehicle; 4) When the path tracking accuracy does not meet the set threshold and there is a lane changing condition, but the lane changing has a rollover danger, the safety of the vehicle is prioritized, the vehicle is decelerated and corrected before lane changing, and the vehicle is compensated for vehicle body posture if the correction is successful; 5) If the correction fails, the vehicle cannot change lanes, at which time the AEB (automatic emergency braking system) is started and the driver is warned to intervene; 6) When the path tracking accuracy does not meet the set threshold and there is no lane changing condition, the vehicle is path tracked, and the steering stability is not deteriorated after path tracking compensation, then the vehicle is controlled according to this logic; 7) When the path tracking accuracy does not meet the set threshold and there is no lane changing condition, the vehicle is path tracked, and the steering stability is deteriorated after path tracking compensation, at which time the vehicle body posture compensation control is needed; 8) When the vehicle body posture compensation and path tracking accuracy compensation control cannot effectively correct the vehicle state, emergency braking deceleration, warning and driver intervention operation control are needed; Step 3, according to the real-time monitoring results of the eight driving situations in step 2, the intelligent vehicle adopts the corresponding control strategy, and evaluates the coordinated control strategy, updates the vehicle state, and further controls the vehicle accordingly until the path tracking function ends; The real-time monitoring results of the first driving situation, the intelligent vehicle adopts the corresponding control strategy, which is realized by the following steps: Step 2A-1, based on the two-degree-of-freedom vehicle dynamics equation, a two-degree-of-freedom vehicle dynamics model is established, according to the relationship between the tire side slip angle and the vehicle mass center side slip angle, the yaw rate and the distance between the mass center and the front and rear axles, combined with Newton's second law, the motion differential equation of the linear two-degree-of-freedom model is further arranged as follows: ； In the formula, m is the mass of the vehicle; C f 、Cr — front and rear wheel linear cornering stiffnesses; l f , l r — distance of the vehicle's center of mass from the front and rear axles; I z - the moment of inertia of the whole vehicle about the Z axis; δ f front wheel steering angle; V x - longitudinal vehicle speed; β - a centroid side slip angle; r - yaw angular velocity; Step 2A-2: Design of preview error system; path tracking technology refers to that in the actual driving process of the intelligent vehicle, the corresponding road information is obtained through the sensor, and the driving direction of the vehicle is controlled by using a suitable control algorithm, so that the vehicle does not deviate from the set trajectory in the driving process, and the safety in the driving process is also ensured; the inputs of the preview error model include the output of the vehicle dynamics model, i.e. the mass center side slip angle β and the yaw rate r , and the external inputs, i.e. the road curvature ρ and the preview distance x e The outputs of the preview error model are the lateral displacement deviation y e and the lateral orientation deviation e The formula of the preview error model of the vehicle is: ； In the formula y e lateral displacement deviation; v y - transverse velocity; v x - longitudinal vehicle speed; e - lateral azimuth deviation; r - yaw angular velocity; ρ - road curvature; x e — preview distance; Step 2A-3: Weighting the lateral displacement bias y e and the lateral orientation bias e as the combined bias E after weighting Step 2A-4: The integrated error E from step 2A-3 is input to the PID controller, which after a series of calculations outputs the front wheel steering angle of the vehicle δ f As the control variable of the vehicle dynamics model, the closed loop control system of the lateral motion of path tracking is formed, and the path tracking function is realized.

2. The vehicle path tracking and body posture cooperative control method according to claim 1, characterized by, The real-time monitoring results of the second, third and fourth driving situations, the vehicle body posture compensation is performed on the vehicle to weaken the lateral acceleration of the vehicle and correct the lateral motion of the vehicle, which is realized by the following steps: Step 2B-1: To improve the body attitude mainly by controlling the vehicle roll, roll moment will be generated after roll happens, roll moment is composed of three parts:

1. Roll moment caused by centrifugal force of suspension mass M ΦrⅠ 2. Roll moment caused by gravity of suspension mass M ΦrⅡ 3. Roll moment caused by centrifugal force of non-suspension mass M ΦrⅢ 4. In addition, load transfer moment will be generated when vertical load shifts between left and right wheels during roll M ZF , M ZR When the body is in roll state, if the left and right suspensions can generate additional forces Δ f in opposite directions under the current state, anti-roll moment M af will be formed to suppress the roll of the vehicle; Rolling moment due to centrifugal force of suspended mass M ΦrⅠ is: ； Rolling moment due to gravity of suspended mass M ΦrⅡ is: ； Rolling moment caused by centrifugal force of non-suspended mass M ΦrⅢ is: ； When the vehicle leans, the vertical load is transferred between the left and right wheels, generating a load transfer moment M ZF , M ZR is: ， ； The moment of the vehicle body longitudinal center line is taken, that is: ； In the formula m s - suspended mass; a y - lateral acceleration; h - the distance of the center of mass of the suspended mass to the roll axis; g - gravitational acceleration; h g - the distance of the center of mass of the suspended mass to the ground; ϕ - the body roll angle; F uy - centrifugal force generated by the non-suspended mass; h 0 - the distance of the roll center from the ground; r - wheel radius; F rRF 、 F rLF - side roll rear front axle left and right wheel vertical load; F rRR 、 F rLR - side roll rear rear axle left and right wheel vertical load; B — wheelbase; Step 2B-2: By setting a three-axis gyroscope at the vehicle body mass center, the vertical acceleration, roll angle acceleration and roll angle signals at the vehicle body mass center are obtained and sent to the controller, the current roll moment of the vehicle is calculated, and the opposite moment is provided to the vehicle through the actuator on the active suspension to resist the roll movement of the vehicle. The above steps realize the monitoring and control of the first, second, third and fourth driving scenarios.

3. The vehicle path tracking and body posture cooperative control method according to claim 1, characterized by, The fifth and eighth driving scenarios are monitored in real time, and the intelligent vehicle adopts the corresponding control strategy through the following steps: Step 2C, emergency braking control method, in the case that path tracking compensation and vehicle body posture compensation are unsuccessful, the automatic emergency braking system AEB is started, the vehicle speed is reduced, and the vehicle state is controlled. The AEB system mainly includes three parts: information acquisition module, control module and execution module. The front camera, millimeter wave radar and laser radar device identify the obstacles in front of the vehicle, and then the information of the vehicle and the front vehicle driving state or the front obstacle or the pedestrian is transmitted to the control module of the AEB system in real time. The control module calculates and judges the danger level of the vehicle according to the motion state information of the vehicle. When the system determines that there is a collision risk between the vehicle and the front vehicle, the driver is warned by light and sound. If the driver does not take effective measures, the system will automatically control the vehicle to brake in emergency.

4. The vehicle path tracking and body posture cooperative control method according to claim 1, characterized by, The sixth driving scenario is monitored in real time, and the intelligent vehicle adopts the corresponding control strategy through the following steps: Step 2D, path tracking compensation control method, based on PI control theory, the lateral position deviation, direction deviation and road curvature are considered to design the turn angle compensation control law as follows: ； wherein So, the front wheel turn angle control law of path tracking compensation control is obtained: comp is the front wheel steering angle compensation; R 1is the lateral position deviation proportional coefficient; Q 1( k ( t )) is the lateral deviation integral function; R 2is the directional deviation proportional coefficient; Q 2( k ( t )) is the lateral deviation proportional coefficient; k ( t ) is the desired path road curvature at the current time instant; e y is the actual lateral position deviation at the vehicle center of mass at the current time instant; e φ is the actual directional deviation at the vehicle center of mass at the current time instant; wherein to reduce the overshoot of lateral deviation and the overshoot of directional deviation of the system in tracking the target path with large curvature, the integral function Q 1( k ( t ))、 Q 2( k ( t )) is set as a function of the road curvature: ； In the formula, The seventh driving scenario is monitored in real time, and the intelligent vehicle adopts the corresponding control strategy through the following steps: 1、 Step 2E, the vehicle is compensated for tracking accuracy and vehicle body posture at the same time. The compensation for tracking accuracy is realized through the control strategy of the sixth driving scenario; the vehicle body posture compensation is realized through the control strategies of the first, second, third and fourth driving scenarios. 2 is a lateral position, a direction deviation integral coefficient; R 1、 In addition, after the compensation control of the cooperative controller, the state of the vehicle is monitored and judged again, and the compensation control effect is evaluated in real time. If the compensation is successful, the cooperative control is continued, otherwise the driver intervenes and takes over the control of the vehicle to avoid danger. 1 The front wheel steering angle compensation quantity mainly representing the lateral deviation of the system, R 2、 ​ 2 The front wheel steering angle compensation quantity mainly representing the directional deviation of the system, in order to make the compensation controller well adapt to the change of the system parameters and the disturbance of the external environment, the system parameters in the compensation controller R 1、 R 2、 ​ 1、 ​ 2 Based on the idea of fuzzy control, two different fuzzy controllers are designed respectively based on the lateral deviation and the directional deviation of the system, and the lateral deviation e y and the change rate of the lateral deviation are taken as the inputs, the proportional coefficient R 1 of the lateral deviation and the integral coefficient ​ 1 are taken as the outputs to design the lateral deviation fuzzy controller; the directional deviation e φ and the change rate of the directional deviation are taken as the inputs, the proportional coefficient R 2 of the directional deviation and the integral coefficient ​ 2 are taken as the outputs to design the directional deviation fuzzy controller, and the change law of R 1、 ​ 1、 R 2、 ​ 2 is obtained respectively by using two inputs and two outputs, and the two are optimized. ​ ； ​ 。 5. The vehicle path tracking and body posture cooperative control method according to claim 1, characterized by, ​ ​ 6. The vehicle path tracking and body posture cooperative control method according to claim 1, characterized by, ​

Images