An attack angle control method for a vehicle active spoiler based on fuzzy logic

Abstract

The application discloses a vehicle active tail wing attack angle control method based on fuzzy logic. The application adopts double active controller design, the input of the main fuzzy controller is vehicle speed and acceleration request value, the output is active tail wing attack angle, the membership function is used for processing, fuzzy rules are made, and the fuzzy quantity of the output of the main fuzzy controller is obtained through fuzzy reasoning; the output of the slave fuzzy controller is tail wing attack angle correction, the input is the difference between the actual and reference of the front and rear wheel slip rate difference and its change amount; the fuzzy rules of the slave fuzzy controller are made, and the fuzzy quantity of the output of the slave fuzzy controller is obtained through fuzzy reasoning; the output of the main and slave fuzzy controllers is de-fuzzied by using the barycentric method, the accurate quantity of the main and slave fuzzy controllers is obtained, finally, the attack angle of the active tail wing can be obtained by combining the output of the main and slave fuzzy controllers, and the slip rate of each tire and vehicle speed are measured as feedback.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of vehicles. In particular, it relates to a method for controlling the attack angle of a vehicle active spoiler based on fuzzy logic. BACKGROUND

[0002] During driving, a vehicle will be affected by air force in addition to friction force from the ground. Especially during high-speed driving, the air force cannot be ignored and has a significant impact on the driving of the vehicle. In order to improve the impact of air force on the movement of the vehicle, increase the beneficial air force, and reduce the adverse air force, vehicle aerodynamics has emerged to optimize the aerodynamic performance of the vehicle. In addition to the optimization design of the overall vehicle aerodynamic layout, the aerodynamic components of the vehicle have become a major focus of vehicle aerodynamic research in recent years. The vehicle spoiler is one of the most widely used components. Reasonable use of the spoiler can effectively improve the handling performance and driving experience of the vehicle. However, the fixed spoiler cannot be adjusted according to the state of the vehicle. The attack angle of the active spoiler can be adjusted according to different driving conditions to achieve different aerodynamic characteristics under different conditions, thereby increasing the stability and safety of the vehicle. SUMMARY

[0003] In order to enable the vehicle to achieve the air performance state most suitable for the current driving state and improve the performance of the vehicle, the present application proposes a method for controlling the attack angle of a vehicle active spoiler based on fuzzy logic. The attack angle of the active spoiler can be controlled according to the operating state of the vehicle and the operation of the driver during driving of the vehicle.

[0004] A method for controlling the attack angle of a vehicle active spoiler based on fuzzy logic, comprising the following steps

[0005] Step 1: taking the obtained vehicle speed V and the desired acceleration Acc exp of the vehicle as input quantities, and performing fuzzy processing on the input quantities by a main fuzzy controller to obtain the membership degree of the input quantity fuzzy subset; wherein the desired acceleration Acc exp of the vehicle is calculated from the accelerator pedal opening Pd Gas and the opening Pd Brake of the brake pedal, and the specific calculation method is as follows:

[0006] Acc exp = Pd Gas - Pd Brake (1)

[0007] wherein the value range of the accelerator pedal opening Pd Gas and the brake pedal opening Pd Brake is [0, 1];

[0008] Step two: obtaining the fuzzy quantity of the active tail wing attack angle A according to the control rule and fuzzy inference method of the master fuzzy controller;

[0009] Step three: taking the deviation E of the tire slip ratio difference value and the error change value E c as input quantities, obtaining the membership of the input quantity fuzzy subsets by fuzzy processing the input quantities from the fuzzy controller; wherein the tire slip ratio difference value S F&R is the difference between the front and rear tire slip ratios, and the specific calculation formula is as follows:

[0010] S F&R = max{S RL , S RR}- max { S FL ,F FR} (2)

[0011] wherein S RL , S RR are the slip ratios of the left and right rear wheels, and S FL , F FR are the slip ratios of the left and right front wheels;

[0012] Step four: obtaining the fuzzy quantity of the attack angle correction A0 according to the control rule and fuzzy inference method of the slave fuzzy controller;

[0013] Step five: de-fuzzifying the fuzzy quantities of the master fuzzy controller and the slave fuzzy controller to obtain the final accurate output values;

[0014] Step six: summing the output quantities of the master fuzzy controller and the slave fuzzy controller after de-fuzzification to obtain the attack angle control quantity of the active tail wing;

[0015] Step seven: measuring the slip ratios S RL , S RR , S FL , F FR of each tire and the vehicle speed V as feedback, and returning to steps one and three.

[0016] Preferably, the method for the master fuzzy controller to perform fuzzy processing on the input quantities in step one is as follows:

[0017] The value range of the desired acceleration Acc exp is [-1, 1], and the fuzzy domain value range is [-1, 1]; the fuzzy subsets of the desired acceleration Acc exp are set as {BL (reduce large), BM (reduce medium), BS (reduce small), 0, GS (add small), GM (add medium), GL (add large)}, and Acc expThe membership function distribution of the fuzzy set is set as follows: two subsets BL and GL adopt a mixed Gaussian membership function, and the specific parameters are as follows:

[0018] BL: y = gauss2mf(x, [1-1 0.1-0.85]);

[0019] GL: y = gauss2mf(x, [0.1 0.85 1 1]);

[0020] wherein y = gauss2mf(x, [σ1c1σ2c2]) represents a double Gaussian mixed function; y is an output value, x is an input value, σ1 and c1 are respectively a standard deviation and a mean value of a left Gaussian function, and σ2 and c2 are respectively a standard deviation and a mean value of a right Gaussian function; other subsets adopt a Gaussian membership function, and the specific parameters are as follows:

[0021] BM: y = gaussmf(x, [0.06-0.5]);

[0022] BS: y = gaussmf(x, [0.06-0.25]);

[0023] 0: y = gaussmf(x, [0.05 0]);

[0024] GS: y = gaussmf(x, [0.06 0.25]);

[0025] GM: y = gaussmf(x, [0.06 0.5]);

[0026] wherein y = gaussmf(x, [σc]) represents a Gaussian function, y is an output value, x is an input value, σ and c are respectively a standard deviation and a mean value of the Gaussian function.

[0027] Another input is vehicle speed V, which has a value range of [0 km / h, 180 km / h], and the fuzzy domain has a value range of [0, 180], and the fuzzy subsets are {VS (large slow), MS (medium slow), SS (small slow), M (moderate), SF (small fast), MF (medium fast), VF (large fast)};

[0028] The membership function distribution of the V fuzzy set is set as follows: VS and VF adopt a joint Gaussian membership function, and the parameters thereof are respectively

[0029] VS: y = gauss2mf(x, [6 0 8 30]);

[0030] VF: y = gauss2mf(x, [3 160 10 180]);

[0031] Other fuzzy subsets adopt Gaussian membership functions, whose parameters are as follows:

[0032] MS: y = gaussmf(x, [8 60]);

[0033] SS: y = gaussmf(x, [6 90]);

[0034] M: y = gaussmf(x, [5 110]);

[0035] SF: y = gaussmf(x, [4 130]);

[0036] MF: y = gaussmf(x, [4 148]);

[0037] Preferably, the control rule of the main fuzzy controller in step two is as follows:

[0038] According to the nature of the active tail, the output of the control strategy is the attack angle of the tail. The attack angle of the active tail is A, and its adjustable range is [-30°, 30°]. The fuzzy domain value is [-30, 30], and the fuzzy subsets thereof are set as {NL (negative large), NM (negative medium), NS (negative small), 0, PS (positive small), PM (positive medium), PL (positive large)}. According to the above settings and the control logic of the tail, the attack angle fuzzy control rule of the active tail is shown in the table.

[0039] Table 1

[0040]

[0041] The membership functions of the attack angle A fuzzy set adopt Gaussian and joint Gaussian membership functions, and the distribution is set as follows: wherein the two fuzzy subsets NL and PL adopt joint Gaussian membership functions, and the parameters thereof are as follows

[0042] NL: y = gauss2mf(x, [1-30 2-25]);

[0043] PL: y = gauss2mf(x, [2 25 1 30]);

[0044] The other subsets adopt Gaussian membership functions, and the parameters thereof are as follows:

[0045] NM: y = gaussmf(x, [2.5-17]);

[0046] NS: y = gaussmf(x, [2-8]);

[0047] 0: y = gaussmf(x, [2 0]);

[0048] PS: y = gaussmf(x, [2 8]);

[0049] PM: y = gaussmf(x, [2.5 17]);

[0050] Preferably, the fuzzy reasoning process of the main fuzzy controller in step two adopts the minimum-maximum-barycenter method, which is as follows:

[0051] First, the current detected information Acc exp and the membership degree of V are brought into the rule table 1 respectively, and the membership function of the output fuzzy set corresponding to each rule can be obtained; the rule whose two inputs are not zero is triggered, and the minimum calculation is performed on the membership degrees of the two inputs of the same rule that has been triggered, and the result of each rule is called the premise credibility of the rule; the intersection operation of the premise credibility of each rule and the membership function of the fuzzy set output by each rule is performed, and the result is the fuzzy value finally output by each rule; finally, the fuzzy values output by all rules are calculated by performing the union set operation, and the main movable tail wing attack angle A fuzzy quantity output by the main fuzzy controller at this time can be obtained;

[0052] Preferably, the method for the slave fuzzy controller to perform fuzzy processing on the input quantity in step three is as follows:

[0053] The input E of the slave fuzzy controller is the difference between the actual value and the expected value of the slip rate difference S F&R of the front and rear wheels of the vehicle, wherein the expected value is 0, the fuzzy domain value is [-1, 1], and the fuzzy subsets are set as {FL (front large), FM (front medium), FS (front small), 0, RS (rear small), RM (rear medium), RL (rear large)}; the membership degree functions of the E fuzzy subsets are set as follows: the two subsets FL and RL adopt a mixed Gaussian membership degree function, and the specific parameters are as follows:

[0054] FL: y = gauss2mf(x, [0.58-1 0.05-0.8]);

[0055] RL: y = gauss2mf(x, [0.06 0.5 0.6 1]);

[0056] The other subsets adopt a Gaussian membership degree function, and the specific parameters are as follows:

[0057] FM: y = gaussmf(x, [0.08-0.5]);

[0058] FS: y = gaussmf(x, [0.05-0.2]);

[0059] 0: y = gaussmf(x, [0.04 0]);

[0060] RS:y=gaussmf(x,[0.05 0.2]);

[0061] RM:y=gaussmf(x,[0.08 0.5]);

[0062] From the input E of the fuzzy controller c It is the difference in slip ratio between the front and rear wheels of the vehicle, S F&R The change in the difference between the actual value and the expected value is denoted by E. The fuzzy universe of discourse takes the value [-1, 1], and its fuzzy subsets are set as {NL (negative large), NM (negative medium), NS (negative small), 0, PS (positive small), PM (positive medium), PL (positive large)}. The membership function distribution of the fuzzy subset E is set as follows: The two subsets NL and PL adopt a Gaussian mixture membership function, with the following specific parameters:

[0063] NL:y=gauss2mf(x,[0.6-1 0.06-0.5]);

[0064] PL:y=gauss2mf(x,[0.06 0.5 0.6 1]);

[0065] Other subsets use a Gaussian membership function with the following parameters:

[0066] VM:y=gaussmf(x,[0.05-0.3]);

[0067] VS:y=gaussmf(x,[0.04-0.15]);

[0068] 0: y = gaussmf(x, [0.04 0]);

[0069] PS:y=gaussmf(x,[0.04 0.15]);

[0070] PM:y=gaussmf(x,[0.05 0.3]);

[0071] Preferably, the control rules from the fuzzy controller in step four are as follows:

[0072] The output of the fuzzy control is the angle of attack correction A0 of the tail fin, which is adjustable in the range of [-10°, 20°] and the value of the fuzzy universe of discourse is [-10, 20]. Its fuzzy subset is set as {NL (negative large), NS (negative small), 0, PSS (positive small small), PS (positive small), PM (positive center), PL (positive large)}. Based on the above settings, the fuzzy rules of the fuzzy controller are summarized as shown in Table 2.

[0073] Table 2

[0074]

[0075] The membership function of the attack angle correction amount A0 fuzzy set adopts a Gaussian type and a joint Gaussian type membership function, and the distribution is set as follows: wherein the two fuzzy subsets NL and PL adopt a joint Gaussian type membership function, and the parameters are as follows respectively:

[0076] NL: y = gauss2mf (x, [0.5-15 1.2-8]);

[0077] PL: y = gauss2mf (x, [1 18 0.5 20]);

[0078] The other subsets adopt a Gaussian type membership function, and the parameters are as follows:

[0079] NS: y = gaussmf (x, [1-4]);

[0080] 0: y = gaussmf (x, [0.8 0]);

[0081] PSS: y = gaussmf (x, [1 4]);

[0082] PS: y = gaussmf (x, [1.2 8]);

[0083] PM: y = gaussmf (x, [1.5 13]);

[0084] Preferably, the minimum-maximum-barycenter method is adopted in the fuzzy reasoning process of the fuzzy controller in step four, and the specific process is as follows:

[0085] Firstly, the membership degrees of the current input information E and E c are brought into the rule table 2 respectively, and the membership functions of the output fuzzy sets corresponding to each rule can be obtained; the rule in which both inputs are not zero is triggered, and the minimum calculation is performed on the membership degrees of the two inputs of the same rule that has been triggered, and the result of each rule is called the premise credibility of the rule; the intersection operation of the premise credibility of each rule and the membership function of the output fuzzy set of each rule is performed, and the result is the final output fuzzy value of each rule; finally, the union set calculation is performed on the fuzzy values output by all rules, and the attack angle correction amount A0 fuzzy amount output by the fuzzy controller at this time can be obtained.

[0086] Preferably, the barycenter method is adopted for defuzzification when the fuzzy amounts of the master fuzzy controller and the slave fuzzy controller are defuzzified in step five.

[0087] The beneficial effects of the present application are as follows:

[0088] This invention presents a method for controlling the angle of attack of an active vehicle rear wing based on fuzzy logic. In different driving scenarios, this method enables the vehicle to achieve the aerodynamic performance state that best matches the current driving condition, thereby improving vehicle performance. Attached Figure Description

[0089] Figure 1 This is a structural diagram of the dual fuzzy controller proposed in this invention;

[0090] Figure 2 Force analysis diagram of the tail fin for this invention;

[0091] Figure 3 A schematic diagram of the fuzzy rules for the main fuzzy controller;

[0092] Figure 4 A fuzzy diagram of fuzzy rules;

[0093] Figure 5 The input diagram is for braking;

[0094] Figure 6 for Figure 5 The speed comparison graph shown is for different braking inputs.

[0095] Figure 7 Input the steering wheel angle;

[0096] Figure 8 For vehicles in Figure 7 The driving trajectory diagram based on the steering wheel angle input. Detailed Implementation

[0097] The proposed angle-of-attack control method for the active tail fin is further elaborated and explained below with reference to the accompanying drawings.

[0098] like Figure 1 As shown, this invention proposes a method for angle-of-attack control of a vehicle's active rear wing based on a fuzzy controller, comprising the following steps:

[0099] Step 1: Establish as follows Figure 2 The aerodynamic body of the vehicle is modeled with a six-degree-of-freedom aerodynamic body, and the influence model of the tail wing on the vehicle's aerodynamic six-component forces and the force analysis are derived. Figure 2 The forces and torques shown are all generated by the action of air.

[0100] like Figure 2 As shown, in the established aerodynamic six-degree-of-freedom model of the vehicle body, the X direction is the forward direction of the vehicle, the Y direction points to the right side of the vehicle, and the Z direction is vertically upward.

[0101] Regarding the influence of the active rear wing on the six aerodynamic components, the aerodynamic forces and moments exerted by the active rear wing on the vehicle body in the X, Y, and Z directions are as follows:

[0102] F AeroX =-F Drag (1)

[0103] F AreoY =0 (2)

[0104] F AreoZ =-F Down (3)

[0105] M AreoX =0 (4)

[0106] M AreoY =-F Drag hF Down l (5)

[0107] M AreoZ =0 (6)

[0108] Where F AeroX F AeroY F AeroZ The aerodynamic forces in the X, Y, and Z directions are respectively, M AeroX M AeroY M AeroZ The aerodynamic moments F about X, Y, and Z are respectively. Down For the downforce generated by the active tail fin, F Drag denoted as , where h is the air resistance generated by the active tail wing, h is the height distance from the point of action of the active tail wing to the vehicle's center of gravity, and l is the longitudinal distance from the point of action of the active tail wing to the vehicle's center of gravity.

[0109] Step 2: Input and output quantities of the main fuzzy controller and fuzzification processing. The input of the main fuzzy controller is determined to be the vehicle speed and the desired acceleration. The desired acceleration is calculated by the opening of the accelerator pedal and the brake pedal (7). The output is the tail wing angle of attack. The input quantities are fuzzified using the joint Gaussian membership function and the Gaussian membership function.

[0110] Acc exp =Pd Gas - Pd Brake (7)

[0111] Expected acceleration Acc exp The value range of is [-1, 1], and the value range of the fuzzy universe of discourse is [-1, 1]; the desired acceleration Acc is set. exp The fuzzy subset is {BL (decrease), BM (decrease), BS (decrease), 0, GS (decrease), GM (decrease), GL (decrease)}, Acc exp The membership function distribution of the fuzzy set is set as follows: The two subsets BL and GL use a Gaussian mixture membership function, with specific parameters as follows:

[0112] BL: y = gauss2mf(x, [1 -1 0.1 -0.85]);

[0113] GL: y = gauss2mf(x, [0.1 0.85 1 1]);

[0114] where y = gauss2mf(x, [σ1 c1 σ2 c2]) represents a double Gaussian mixture function; y is the output value, x is the input value, σ1 and c1 are the standard deviation and mean of the left Gaussian function, respectively, and σ2 and c2 are the standard deviation and mean of the right Gaussian function, respectively; other subsets use Gaussian membership functions, and the specific parameters are as follows:

[0115] BM: y = gaussmf(x, [0.06 -0.5]);

[0116] BS: y = gaussmf(x, [0.06 -0.25]);

[0117] 0: y = gaussmf(x, [0.05 0]);

[0118] GS: y = gaussmf(x, [0.06 0.25]);

[0119] GM: y = gaussmf(x, [0.06 0.5]);

[0120] where y = gaussmf(x, [σ c]) represents a Gaussian function, y is the output value, x is the input value, σ and c are the standard deviation and mean of the Gaussian function, respectively.

[0121] Another input is the vehicle speed V, which has a value range of [0 km / h, 180 km / h], and the fuzzy domain has a value range of [0, 180], and its fuzzy subsets are {VS (large slow), MS (medium slow), SS (small slow), M (moderate), SF (small fast), MF (medium fast), VF (large fast)};

[0122] The membership function distribution of the V fuzzy set is set as follows: VS and VF use joint Gaussian membership functions, and their parameters are respectively

[0123] VS: y = gauss2mf(x, [6 0 8 30]);

[0124] VF: y = gauss2mf(x, [3 160 10 180]);

[0125] Other fuzzy subsets use Gaussian membership functions, and their parameters are as follows:

[0126] MS: y = gaussmf(x, [8 60]);

[0127] SS:y = gaussmf(x, [6 90]);

[0128] M:y = gaussmf(x, [5 110]);

[0129] SF:y = gaussmf(x, [4 130]);

[0130] MF:y = gaussmf(x, [4 148]);

[0131] Step three: Obtain the fuzzy quantity of the active tail wing attack angle A according to the control rule and fuzzy inference method of the main fuzzy controller;

[0132] The main fuzzy controller considers that the main influence of the active tail wing on the aerodynamics of the vehicle is to provide additional downforce and drag, and the main factors affecting the size of the downforce and drag are the tail wing attack angle and the vehicle speed. Considering that the tail wing provides resistance to increase the acceleration during braking, but will produce additional resistance during acceleration to reduce the acceleration performance, the difference between the acceleration request value, i.e., the throttle and the brake, is considered when formulating the fuzzy rule, so as to increase the tail wing angle when the driver brakes to provide additional resistance to help braking, and reduce the tail wing angle when the driver accelerates to reduce the impact of additional resistance on the acceleration process. The stability of the vehicle during high-speed driving is crucial, and the downforce plays a crucial role in the stability of the vehicle, so the speed of the vehicle is referred to, and when the speed increases, the attack angle of the tail wing is appropriately increased to provide more additional downforce, enhance the vehicle grip, and improve the vehicle stability performance. The completed main fuzzy control rule is shown in Figure 3 ;

[0133] The attack angle of the active tail wing is A, and its adjustable range is [-30°, 30°], the fuzzy domain value is [-30, 30], and the fuzzy subsets are set as {NL (negative large), NM (negative medium), NS (negative small), 0, PS (positive small), PM (positive medium), PL (positive large)}; According to the above settings and the control logic of the tail wing, the attack angle fuzzy control rule of the active tail wing is shown in the table.

[0134] Table 1

[0135]

[0136] The membership function of the attack angle A fuzzy set adopts the Gaussian type and the joint Gaussian type membership function, and the distribution is set as follows: wherein the two fuzzy subsets NL and PL adopt the joint Gaussian type membership function, and the parameters are as follows

[0137] NL:y = gauss2mf(x, [1-30 2-25]);

[0138] PL: y = gauss2mf(x, [2 25 1 30]);

[0139] Other subsets adopt Gaussian membership function, whose parameters are as follows:

[0140] NM: y = gaussmf(x, [2.5 -17]);

[0141] NS: y = gaussmf(x, [2 -8]);

[0142] 0: y = gaussmf(x, [2 0]);

[0143] PS: y = gaussmf(x, [2 8]);

[0144] PM: y = gaussmf(x, [2.5 17]).

[0145] The fuzzy reasoning process of the main fuzzy controller adopts the minimum-maximum-barycenter method, which is as follows:

[0146] First, the current detected information Acc exp and the membership degree of V are brought into Table 1, so that the membership function of the output fuzzy set corresponding to each rule can be obtained; the rule whose both inputs are not zero is triggered, and the minimum calculation is performed on the membership degrees of the two inputs of the same rule that has been triggered, so that the result of each rule is obtained, which is called the premise credibility of the rule; the intersection operation of the premise credibility of each rule and the membership function of the fuzzy set output by each rule is performed, so that the result is obtained, which is the final fuzzy value output by each rule; finally, the union set calculation is performed on the fuzzy values output by all rules, so that the main movable tail wing attack angle A fuzzy quantity output by the main fuzzy controller at this time can be obtained.

[0147] Step four: taking the deviation E of the tire slip ratio difference value and the error change E c as input quantities, and performing fuzzy processing on the input quantities through the fuzzy controller; wherein the tire slip ratio difference value S F&R is the difference between the front and rear tire slip ratios, and the specific calculation formula is as follows:

[0148] S F&R = max{S RL , S RR}- max { S FL , F FR} (8)

[0149] wherein S RL , S RR are the slip ratios of the left and right rear wheels, and S FL , FFR The slip ratio of the left and right front wheels;

[0150] The input E from the fuzzy controller is the difference in slip ratio between the front and rear wheels of the vehicle, S. F&R The difference between the actual value and the expected value is given by the fuzzy universe of discourse, which takes the value [-1, 1]. Its fuzzy subsets are defined as {FL (largest front), FM (middle front), FS (smallest front), 0, RS (smallest back), RM (middle back), RL (largest back)}. The membership functions of the fuzzy subsets E are set as follows: The subsets FL and RL use Gaussian mixture membership functions with the following parameters:

[0151] FL:y=gauss2mf(x,[0.58-1 0.05-0.8]);

[0152] RL:y=gauss2mf(x,[0.06 0.5 0.6 1]);

[0153] Other subsets use a Gaussian membership function with the following parameters:

[0154] FM:y=gaussmf(x,[0.08-0.5]);

[0155] FS:y=gaussmf(x,[0.05-0.2]);

[0156] 0: y = gaussmf(x, [0.04 0]);

[0157] RS:y=gaussmf(x,[0.05 0.2]);

[0158] RM:y=gaussmf(x,[0.08 0.5]);

[0159] From the input E of the fuzzy controller c It is the difference in slip ratio between the front and rear wheels of the vehicle, S F&R The change in the difference between the actual value and the expected value is denoted by E. The fuzzy universe of discourse takes the value [-1, 1], and its fuzzy subsets are set as {NL (negative large), NM (negative medium), NS (negative small), 0, PS (positive small), PM (positive medium), PL (positive large)}. The membership function distribution of the fuzzy subset E is set as follows: The two subsets NL and PL adopt a Gaussian mixture membership function, with the following specific parameters:

[0160] NL:y=gauss2mf(x,[0.6-1 0.06-0.5]);

[0161] PL:y=gauss2mf(x,[0.06 0.5 0.6 1]);

[0162] Other subsets use Gaussian membership functions with the following parameters:

[0163] VM: y = gaussmf(x, [0.05 -0.3]);

[0164] VS: y = gaussmf(x, [0.04 -0.15]);

[0165] 0: y = gaussmf(x, [0.04 0]);

[0166] PS: y = gaussmf(x, [0.04 0.15]);

[0167] PM: y = gaussmf(x, [0.05 0.3]);

[0168] Step five: obtain the fuzzy quantity of the attack angle correction A0 according to the control rules and fuzzy inference method of the slave fuzzy controller;

[0169] The slave fuzzy control rules are formulated by using force analysis of the tail wing and the slip ratio properties of the vehicle,

[0170] The main role of the slave fuzzy controller is to fine-tune the attack angle of the active tail wing according to the actual situation of the difference between the front and rear wheel slip ratios after the main controller sets the attack angle of the tail wing. From the mechanical analysis in step one, it can be concluded that the downforce of the tail wing and can change the overall downforce of the vehicle, and generate a moment that can affect the downforce distribution of the front and rear wheels. The expected difference between the front and rear wheel slip ratios should be 0. When the rear wheel slip ratio is too large, it means that the downforce of the rear wheel is insufficient, at which time the tail wing angle is increased to increase the downforce of the rear wheel. When the front wheel slip ratio is too large, it means that the downforce of the front wheel is insufficient, at which time the tail wing angle is reduced so that the front wheel obtains more downforce distribution.

[0171] The output of the slave fuzzy control is the attack angle correction A0 of the tail wing, which can be adjusted in the range of [-10°, 20°], and the fuzzy universe is valued at [-10, 20]. The fuzzy subsets are set as {NL (negative large), NS (negative small), 0, PSS (positive small small), PS (positive small), PM (positive medium), PL (positive large)}; according to the above settings, the fuzzy rules of the slave fuzzy controller are summarized as shown in Table 2. The slave fuzzy control rule diagram is shown in Figure 4 ;

[0172] Table 2

[0173]

[0174] The membership function of the attack angle correction amount A0 fuzzy set adopts Gaussian type and joint Gaussian type membership functions, and the distribution is set as follows: wherein the two fuzzy subsets NL and PL adopt joint Gaussian type membership functions, and the parameters are as follows:

[0175] NL: y = gauss2mf(x, [0.5 -15 1.2 -8]);

[0176] PL: y = gauss2mf(x, [1 18 0.5 20]);

[0177] The other subsets adopt Gaussian type membership functions, and the parameters are as follows:

[0178] NS: y = gaussmf(x, [1 -4]);

[0179] 0: y = gaussmf(x, [0.8 0]);

[0180] PSS: y = gaussmf(x, [1 4]);

[0181] PS: y = gaussmf(x, [1.2 8]);

[0182] PM: y = gaussmf(x, [1.5 13]);

[0183] The minimum-maximum-center of gravity method is adopted in the fuzzy reasoning process of the fuzzy controller, and the details are as follows:

[0184] First, the membership degrees of the current input information E and E c are respectively brought into the rule table 2, and the membership functions of the output fuzzy sets corresponding to each rule can be obtained; the rule in which both inputs are not zero is triggered, and the minimum calculation is performed on the membership degrees of the two inputs of the same rule that has been triggered, and the result of each rule is called the premise credibility of the rule; the intersection operation of the premise credibility of each rule and the membership function of the output fuzzy set of each rule is performed, and the result is the fuzzy value finally output by each rule; and finally, the fuzzy values output by all rules are calculated by performing the union set operation, and the attack angle correction amount A0 fuzzy amount output by the fuzzy controller at this time can be obtained.

[0185] Step six: the fuzzy amounts of the master fuzzy controller and the slave fuzzy controller are respectively de-fuzzied;

[0186] The center of gravity method is preferred to be used for de-fuzzification because the center of gravity method has smoother output reasoning control, that is, the final output of reasoning generally also changes to a certain extent corresponding to the slight change of the input signal. The center of gravity method takes the center of gravity of the area surrounded by the fuzzy membership function curve and the horizontal coordinate as the final output value of fuzzy reasoning, that is, the output is:

[0187]

[0188] wherein v0 is the output result of deblurring, u(v) is the membership function, and v is the corresponding output value.

[0189] Step seven: sum the output of the master fuzzy controller and the slave fuzzy controller after deblurring to obtain the attack angle of the active tail wing.

[0190] Step eight: measure the slip rate S of each tire RL , S RR , S FL , F FR and the vehicle speed V as feedback, return to step two and step four.

[0191] The simulation experiment data of the technical solution provided by the present application is given below.

[0192] The experimental simulation environment is a joint simulation platform built using CarSim and Simulink software, different vehicle driving conditions are designed to verify the effectiveness of the active tail wing controller proposed.

[0193] Figure 6 As shown in the figure, the active tail wing significantly shortens the time to brake to zero, and the braking performance is more excellent. Figure 8 As shown in the figure, the trajectory of the simulation vehicle under the condition of a fixed steering wheel angle is shown, and it is obvious that the vehicle with an active tail wing has a smaller turning radius and better performance. It can be seen that the control method of the active tail wing based on fuzzy control proposed in the present application can improve the performance of the vehicle.

Claims

1. A method for attack angle control of a vehicle active spoiler based on fuzzy logic, comprising the following steps: Step one: with the acquired vehicle speed V and the expected acceleration Acc of the vehicle exp is an input, the input is fuzzified by the main fuzzy controller to obtain the membership of the input fuzzy subset; wherein, Desired acceleration Acc of the vehicle exp by the accelerator pedal opening Pd Gas and the opening Pd of the brake pedal Brake is calculated as follows: Acc exp = Pd Gas - Pd Brake (1) Wherein, the value range of the throttle pedal opening Pd Gas and the brake pedal opening Pd Brake are both [0, 1]; Step two: obtaining the fuzzy quantity of the attack angle A of the active spoiler according to the control rule and fuzzy reasoning method of the master fuzzy controller; Step three: the deviation E of the difference of the tire slip ratio and the error change amount E c is an input, and the membership of the input fuzzy subset is obtained by fuzzy processing the input from the fuzzy controller; wherein the difference S of the tire slip ratio F&R is the difference of the front and rear tire slip ratios, and the specific calculation formula is as follows: S F&R = max{S RL , S RR}- max{S FL , F FR} (2) where S RL is the slip ratio of the right rear wheel, S RR is the slip ratio of the left rear wheel, S FL is the slip ratio of the right front wheel, and S FR is the slip ratio of the left front wheel. Step four: obtaining the fuzzy quantity of the attack angle correction A0 according to the control rule and fuzzy reasoning method of the slave fuzzy controller; Step five: respectively de-fuzzifying the fuzzy quantities of the master fuzzy controller and the slave fuzzy controller to obtain the final accurate output value; Step six: summing the output quantities of the master fuzzy controller and the slave fuzzy controller after de-fuzzification to obtain the attack angle control quantity of the active spoiler; Step seven: measure the slip ratio S of each tire RL , S RR , S FL , F FR and vehicle speed V as feedback, return to step one and step three.

2. The method of attack angle control for a fuzzy logic based active vehicle spoiler according to claim 1, wherein, The method for fuzzy processing of the input quantity of the master fuzzy controller in step one is as follows: Desired acceleration Acc exp The value range of the desired acceleration Acc is [-1, 1], and the fuzzy domain value range is [-1, 1]; the fuzzy subset of the desired acceleration Acc exp is set as {BL (reduce large), BM (reduce medium), BS (reduce small), 0, GS (add small), GM (add medium), GL (add large)}, and the membership function distribution of the fuzzy set Acc exp is set as follows: wherein the two subsets BL and GL adopt a mixed Gaussian membership function, and the specific parameters are as follows: BL: y = gauss2mf(x, [1-1 0.1-0.85]); GL: y = gauss2mf(x, [0.1 0.85 1 1]); Wherein y = gauss2mf(x, [σ1c1σ2c2]) represents a double Gaussian mixture function; y is the output value, x is the input value, σ1 and c1 are the standard deviation and mean value of the left Gaussian function respectively, and σ2 and c2 are the standard deviation and mean value of the right Gaussian function respectively; other subsets adopt Gaussian membership functions, and the specific parameters are as follows: BM: y = gaussmf(x, [0.06-0.5]); BS: y = gaussmf(x, [0.06-0.25]); 0: y = gaussmf(x, [0.05 0]); GS: y = gaussmf(x, [0.06 0.25]); GM: y = gaussmf(x, [0.06 0.5]); Wherein y = gaussmf(x, [σc]) represents a Gaussian function, y is the output value, x is the input value, σ and c are the standard deviation and mean value of the Gaussian function respectively; Another input is the vehicle speed V, whose value range is [0 km / h, 180 km / h], and the value range of the fuzzy domain is [0, 180], and the fuzzy subsets are {VS (large slow), MS (medium slow), SS (small slow), M (medium), SF (small fast), MF (medium fast), VF (large fast)}; The membership function distribution of the V fuzzy set is set as follows: VS and VF adopt joint Gaussian membership functions, and their parameters are respectively VS: y = gauss2mf(x, [6 0 8 30]); VF: y = gauss2mf(x, [3 160 10 180]); Other fuzzy subsets adopt Gaussian membership functions, and their parameters are respectively as follows: MS: y = gaussmf(x, [8 60]); SS: y = gaussmf(x, [6 90]); M: y = gaussmf(x, [5 110]); SF: y = gaussmf(x, [4 130]); MF: y = gaussmf(x, [4 148]).

3. The method of attack angle control for a fuzzy logic based active vehicle spoiler according to claim 2, wherein, The control rule of the master fuzzy controller in step two is as follows: According to the nature of the active tail, the output of the control strategy is the angle of attack of the tail; the angle of attack of the active tail is A, its adjustable range is [-30°, 30°], the value of the fuzzy domain is [-30, 30], and the fuzzy subsets are set as {NL (negative large), NM (negative medium), NS (negative small), 0, PS (positive small), PM (positive medium), PL (positive large)}; according to the above settings and the control logic of the tail, the fuzzy control rules of the angle of attack of the active tail are shown in Table 1; Table 1 The membership function of the angle of attack A fuzzy set adopts the Gaussian type and joint Gaussian type membership function, and the distribution is set as follows: wherein the two fuzzy subsets NL and PL adopt the joint Gaussian type membership function, and the parameters are as follows: NL: y = gauss2mf(x, [1-30 2-25]); PL: y = gauss2mf(x, [2 25 1 30]); The other subsets adopt the Gaussian type membership function, and the parameters are as follows: NM: y = gaussmf(x, [2.5-17]); NS: y = gaussmf(x, [2-8]); 0: y = gaussmf(x, [2 0]); PS: y = gaussmf(x, [2 8]); PM: y = gaussmf(x, [2.5 17]).

4. The method of attack angle control for a fuzzy logic based active vehicle spoiler according to claim 3, wherein, In step two, the fuzzy reasoning process of the master fuzzy controller adopts the minimum-maximum-barycenter method, and the specific process is as follows: First, the current detected information Acc exp The membership of Acc and V is brought into Table 1 respectively, and the membership function of the output fuzzy set corresponding to each rule can be obtained; the rule is triggered when both inputs are not zero, and the minimum of the membership of the two inputs of the same rule is calculated, and the result is called the premise credibility of the rule; the result of the intersection operation between the premise credibility of each rule and the membership function of the fuzzy set output by each rule is the final fuzzy value output by each rule; finally, the union set of the fuzzy values output by all rules is calculated, and the active tail wing attack angle A fuzzy quantity output by the main fuzzy controller at this time can be obtained.

5. The method of attack angle control for a fuzzy logic based active vehicle spoiler of claim 4, wherein, In step three, the method for fuzzy processing of the input quantity of the slave fuzzy controller is as follows: The input E of the fuzzy controller is the difference S of the slip rates of the front and rear wheels of the vehicle F&R the difference between the actual value and the desired value, where the desired value is 0, the fuzzy domain takes values [-1, 1], and its fuzzy subsets are set as {FL (front large), FM (front medium), FS (front small), 0, RS (rear small), RM (rear medium), RL (rear large)}; the membership functions of the E fuzzy subsets are set as follows: the two subsets FL and RL adopt a mixed Gaussian membership function, and the specific parameters are as follows: FL: y = gauss2mf(x, [0.58-1 0.05-0.8]); RL: y = gauss2mf(x, [0.06 0.5 0.6 1]); The other subsets adopt the Gaussian type membership function, and the specific parameters are as follows: FM: y = gaussmf(x, [0.08-0.5]); FS: y = gaussmf(x, [0.05-0.2]); 0: y = gaussmf(x, [0.04 0]); RS: y = gaussmf(x, [0.05 0.2]); RM: y = gaussmf(x, [0.08 0.5]); From the input E of the fuzzy controller c is the actual value of the difference between the front and rear wheel slip rates S F&R of the expected value, the fuzzy domain takes the value [-1,1], and its fuzzy subsets are set as {NL (negative large), NM (negative medium), NS (negative small), 0, PS (positive small), PM (positive medium), PL (positive large)}; The membership function distribution of E fuzzy subsets is set as follows: the two subsets NL and PL adopt a mixed Gaussian membership function, and the specific parameters are as follows: NL: y = gauss2mf(x, [0.6-1 0.06-0.5]); PL: y = gauss2mf(x, [0.06 0.5 0.6 1]); The other subsets adopt the Gaussian type membership function, and the specific parameters are as follows: VM: y = gaussmf(x, [0.05-0.3]); VS: y = gaussmf(x, [0.04-0.15]); 0: y = gaussmf(x, [0.04 0]); PS: y = gaussmf(x, [0.04 0.15]); PM: y = gaussmf(x, [0.05 0.3]).

6. The method of attack angle control for a fuzzy logic based active vehicle spoiler of claim 5, wherein, In step four, the control rules of the slave fuzzy controller are as follows: The output of the fuzzy controller is the attack angle correction A0 of the tail wing, which can be adjusted in the range of [-10°, 20°], the fuzzy domain value is [-10, 20], and the fuzzy subsets are set as {NL (negative large), NS (negative small), 0, PSS (positive small small), PS (positive small), PM (positive medium), PL (positive large)}; according to the above settings, the fuzzy rules of the slave fuzzy controller are summarized in Table 2; Table 2 The membership functions of the attack angle correction A0 fuzzy set adopt Gaussian and joint Gaussian membership functions, and the distribution is set as follows: wherein the two fuzzy subsets NL and PL adopt joint Gaussian membership functions, and the parameters are as follows: NL: y = gauss2mf(x, [0.5-15 1.2-8]); PL: y = gauss2mf(x, [1 18 0.5 20]); The other subsets adopt Gaussian membership functions, and the parameters are as follows: NS: y = gaussmf(x, [1-4]); 0: y = gaussmf(x, [0.8 0]); PSS: y = gaussmf(x, [1 4]); PS: y = gaussmf(x, [1.2 8]); PM: y = gaussmf(x, [1.5 13]).

7. The method of attack angle control for a fuzzy logic based active vehicle spoiler according to claim 6, wherein The fuzzy reasoning process of the slave fuzzy controller in step four adopts the minimum-maximum-barycenter method, which is as follows: Firstly, the membership of the current input information E and E c The membership of the output fuzzy set corresponding to each rule can be obtained by bringing the membership of the current input information E and E into rule table 2 respectively. The rule is triggered when both inputs are not zero. The membership of the two inputs of the same rule is calculated by taking the minimum value. The result is called the premise credibility of the rule. The result of the intersection operation between the premise credibility of each rule and the membership function of the output fuzzy set of each rule is the final output fuzzy value of each rule. Finally, the fuzzy values output by all rules are calculated by taking the union set, and the attack angle correction amount A0 fuzzy value output by the fuzzy controller at this time is obtained.

8. The method of attack angle control for a fuzzy logic based active vehicle spoiler of claim 7, wherein, In step five, the barycenter method is used for defuzzification when the fuzzy quantities of the master fuzzy controller and the slave fuzzy controller are defuzzified.

Images