An adaptive cruise control method using a composite model predictive controller

Abstract

The present invention discloses an adaptive cruise control method using a composite model predictive controller, the method comprising: determining an adaptive cruise system working mode according to vehicle motion state information and road information; outputting a vehicle desired longitudinal acceleration based on an adaptive cruise system engineering model in combination with vehicle motion state information; determining a motor working mode according to the vehicle desired longitudinal acceleration in combination with the acceleration-velocity relationship during the natural deceleration of the vehicle; determining a vehicle longitudinal dynamics model according to the motor working mode, and calculating a desired motor torque in combination with the vehicle desired longitudinal acceleration to obtain a control signal. By using the present invention, it is possible to calculate the vehicle desired longitudinal acceleration in real time, adjust the distance between the front and rear vehicles, and realize brake energy recovery. As an adaptive cruise control method using a composite model predictive controller, the present invention can be widely used in the field of autonomous driving electric vehicle control.

Description

Technical Field

[0001] The present invention relates to the field of autonomous driving electric vehicle control, and in particular to an adaptive cruise control method using a composite model predictive controller. Background Art

[0002] Adaptive cruise control is one of the key functions of self-driving cars. According to the distance and relative speed between the vehicle and the vehicle in front, the longitudinal speed is adjusted by controlling the acceleration and deceleration of the vehicle to keep the appropriate distance between the vehicle and the vehicle in front. Its performance will directly affect the safety of the vehicle during driving. As a complex integrated system, the on-board computer of the self-driving car needs to process multiple information such as perception, planning, communication, and control at the same time. The computing power allocated to each functional module is limited. At present, various adaptive cruise controllers based on optimization methods require a large number of complex calculations such as matrix decomposition, which takes up a lot of computing power. Conventional adaptive cruise systems focus on a single performance indicator and do not comprehensively consider relevant factors such as tracking performance, safety, comfort and economy. Compared with fuel vehicles, electric vehicles have a shorter range. Conventional adaptive cruise control systems usually ignore energy management, which is not conducive to improving energy efficiency. Summary of the invention

[0003] In order to solve the above technical problems, the purpose of the present invention is to provide an adaptive cruise control method using a composite model predictive controller, which can calculate the vehicle's expected longitudinal acceleration in real time, adjust the distance between the front and rear vehicles, and realize braking energy recovery.

[0004] The first technical solution adopted by the present invention is: an adaptive cruise control method using a composite model predictive controller, comprising the following steps:

[0005] Determine the working mode of the adaptive cruise system according to the vehicle motion state information and road information;

[0006] Based on the adaptive cruise system engineering model and combined with the vehicle motion state information, the vehicle's expected longitudinal acceleration is output;

[0007] Determine the motor operation mode based on the expected longitudinal acceleration of the vehicle and the acceleration-speed relationship during the natural deceleration of the vehicle;

[0008] The vehicle longitudinal dynamics model is determined according to the motor working mode, and the expected motor torque is calculated in combination with the expected longitudinal acceleration of the vehicle to obtain a control signal.

[0009] Furthermore, the vehicle motion state information includes the longitudinal speed of the vehicle, the longitudinal acceleration of the vehicle, the longitudinal speed of the preceding vehicle, the longitudinal acceleration of the preceding vehicle and the distance between the vehicle and the preceding vehicle, and the road information includes the road slope angle.

[0010] Furthermore, the step of determining the working mode of the adaptive cruise control system according to the vehicle motion state information and the road information specifically includes:

[0011] Calculate the minimum safe distance based on the vehicle's motion status information and determine the adaptive cruise system operating mode based on road information;

[0012] The adaptive cruise system operating modes include a cruise mode and a following mode.

[0013] Furthermore, the step of outputting the expected longitudinal acceleration of the vehicle based on the adaptive cruise system engineering mode and combining the vehicle motion state information specifically includes:

[0014] It is determined that the engineering mode of the adaptive cruise system is the cruise mode, and a single neuron PID controller is selected and the difference between the expected cruise speed and the actual longitudinal speed is used as input to obtain the expected longitudinal acceleration of the vehicle;

[0015] It is determined that the engineering mode of the adaptive cruise control system is the following mode. The incremental MPC controller based on discrete SDNN is selected and the distance between the vehicle and the vehicle in front, the speed of the vehicle in front relative to the vehicle, the longitudinal speed of the vehicle, and the longitudinal acceleration of the vehicle at the previous moment are used as inputs. The objective function is constructed and combined with the kinematic constraints to iteratively solve the expected longitudinal acceleration of the vehicle.

[0016] Furthermore, by selecting a single neuron PID controller, the expression of the vehicle's desired longitudinal acceleration is as follows:

[0017]

[0018] In the above formula, u(k) represents the desired longitudinal acceleration, K is the gain coefficient, and x i (k) is the intermediate state variable, w′ i (k) is the weight coefficient.

[0019] Furthermore, the expression of the motor working mode signal is as follows:

[0020]

[0021] In the above formula, s m represents the motor working mode signal, u represents the expected longitudinal acceleration, u n (v1) represents the natural deceleration of the vehicle when its longitudinal velocity is v1, u h Indicates the acceleration threshold, "00" means keeping the original working mode, "01" means the driving control mode, and "10" means the braking control mode.

[0022] Furthermore, the step of determining the longitudinal dynamics model of the vehicle according to the motor working mode, and calculating the expected motor torque in combination with the expected longitudinal acceleration of the vehicle to obtain the control signal specifically includes:

[0023] It is determined that the working mode of the motor is the original working mode, the original inverse dynamics model is maintained, and the expected longitudinal acceleration and longitudinal speed of the vehicle are used as input to obtain the expected motor torque;

[0024] It is determined that the working mode of the motor is the driving control mode, and an inverse driving model is used to obtain the expected motor torque by taking the expected longitudinal acceleration and longitudinal speed of the vehicle as input;

[0025] It is determined that the working mode of the motor is a braking control mode, and a reverse braking model is adopted to obtain a desired motor torque by taking the desired longitudinal acceleration and longitudinal speed of the vehicle as inputs;

[0026] The desired motor torque is combined with the actual motor torque, a control signal is output and the torque of the motor is controlled.

[0027] Furthermore, using the reverse drive model, the formula for the expected driving torque is expressed as follows:

[0028]

[0029] In the above formula, the T t (k) is the driving torque, u(k) is the desired longitudinal acceleration of the vehicle, v1 is the longitudinal velocity of the vehicle, R is the wheel radius, i t is the total transmission ratio of the transmission system, η is the mechanical efficiency of the transmission system, m is the vehicle mass, C D is the air resistance coefficient, A is the frontal area, g is the gravity constant, f is the rolling resistance coefficient, α is the road slope angle, and δ is the rotational mass conversion coefficient.

[0030] The method of the present invention has the following beneficial effects: the present invention can control the acceleration and deceleration of the vehicle to adjust the longitudinal speed and the distance between the front and rear vehicles, and realize the braking energy recovery. In addition, compared with the numerical solution commonly used in the conventional adaptive cruise control system based on the optimization method, the discrete SDNN used in the present invention has a relatively low computational complexity when solving the quadratic programming problem, and is particularly suitable for embedded platforms with limited resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 is a flow chart of steps of an adaptive cruise control method using a composite model predictive controller of the present invention;

[0032] Figure 2 Schematic diagram of the relationship between the front and rear vehicles in a specific embodiment of the present invention;

[0033] Figure 3It is an execution framework diagram of discrete SDNN iterative solution of expected longitudinal acceleration in a specific embodiment of the present invention. DETAILED DESCRIPTION

[0034] The present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. The step numbers in the following embodiments are only provided for the convenience of explanation and description, and the order between the steps is not limited in any way. The execution order of each step in the embodiment can be adaptively adjusted according to the understanding of those skilled in the art.

[0035] like Figure 1 As shown, the present invention provides an adaptive cruise control method using a composite model predictive controller, the method comprising the following steps:

[0036] S1, determining the working mode of the adaptive cruise control system according to the vehicle motion state information and road information;

[0037] Specifically, the vehicle motion state information includes the longitudinal speed of the vehicle, the longitudinal acceleration of the vehicle, the longitudinal speed of the preceding vehicle, the longitudinal acceleration of the preceding vehicle, and the distance between the vehicle and the preceding vehicle, and the road information includes the road slope angle.

[0038] According to the motion state information, the expected longitudinal distance is calculated, and its specific expression is:

[0039]

[0040] In the above formula, the d des is the expected longitudinal distance, v1 is the longitudinal speed of the vehicle, a 1max is the maximum longitudinal acceleration of the vehicle, t r is the driver's braking reaction time, t a is the braking coordination time, t s is the deceleration growth time, d stop For safe parking distance.

[0041] According to the motion state information, the minimum safety distance is calculated, and its specific expression is:

[0042]

[0043] In the above formula, the d safe is the longitudinal safety distance, t′ r is the driver's emergency braking reaction time, v2 is the longitudinal speed of the front vehicle, a 2max is the maximum longitudinal acceleration of the front vehicle, and the other parameters have the same meaning as formula (1). The relative position relationship between the front and rear vehicles can be seen in Figure 2 .

[0044] S1.1. Calculate the minimum safe distance according to the vehicle motion state information, and determine the working mode of the adaptive cruise system in combination with the road information. The working mode of the adaptive cruise system includes a cruising mode and a following mode.

[0045] Specifically, according to the distance and relative speed between the front and rear vehicles, the working mode of the vehicle's adaptive cruise control system is determined, and the system working mode signal is output. "0" represents the cruise mode and "1" represents the following mode. The specific expression is:

[0046]

[0047] In the above formula, s is the system working mode signal, d is the distance between the front and rear vehicles, and d des is the expected longitudinal distance, d vpt is the relative distance threshold, v r =v2-v1 is the relative speed of the two vehicles, v1 is the longitudinal speed of this vehicle, and v2 is the longitudinal speed of the front vehicle.

[0048] Specifically, when the distance and relative speed between the front and rear vehicles are greater than a set threshold, the adaptive cruise control system enters the cruise mode, otherwise, it enters the following mode.

[0049] S2: Based on the adaptive cruise system engineering model and combined with the vehicle motion state information, output the vehicle's expected longitudinal acceleration;

[0050] S2.1. Determine that the engineering mode of the adaptive cruise system is the cruise mode, select a single neuron PID controller and use the difference between the expected cruise speed and the actual longitudinal speed as input to obtain the expected longitudinal acceleration of the vehicle;

[0051] Specifically, a single neuron PID controller is selected, and the expression of the vehicle's desired longitudinal acceleration is as follows:

[0052]

[0053] In the above formula, u(k) represents the desired longitudinal acceleration, K is the gain coefficient, and x i (k) is the intermediate state variable, w′ i (k) is the weight coefficient.

[0054] Specifically, the expression of the intermediate state variable is:

[0055] x i (k)=e(k)+Δe(k) (5)

[0056] Where, e(k) = v ref (k)-v1(k), Δe(k)=e(k)-e(k-1), v ref(k) is the desired cruising speed, and v1(k) is the actual longitudinal speed of the vehicle.

[0057] The weight coefficients are updated iteratively, namely:

[0058]

[0059] In the above formula, the η P , η I , η D is the learning rate parameter.

[0060] S2.2. Determine that the engineering mode of the adaptive cruise control system is the following mode, select the incremental MPC controller based on discrete SDNN, and use the distance between the vehicle and the vehicle in front, the speed of the vehicle in front relative to the vehicle, the longitudinal speed of the vehicle, and the longitudinal acceleration of the vehicle at the previous moment as inputs, construct the objective function and combine it with kinematic constraints to iteratively solve the expected longitudinal acceleration of the vehicle.

[0061] Specifically, based on the kinematic relationship between the front and rear workshops, an incremental prediction equation is established:

[0062]

[0063] In the above formula, x(k)=[d(k),v r (k), v1(k), u(k-1)] are state variables, Δu(k) = u(k)-u(k-1) is the input variable, representing the longitudinal acceleration increment, y(k) is the output variable, e d (k) is the compensation vector, is the state matrix, is the input matrix, W e is the compensation matrix, is the output matrix.

[0064] The objective optimization function is:

[0065]

[0066] In the above formula, the weighted term w d (d(k+i|k)-d des (k+i|k)) 2 +w r v r (k+i|k) 2 、w u u(k+i|k) 2 +w Δu Δu(k+i|k) 2 They correspond to tracking performance and economy respectively.

[0067] Combined with the safety and comfort constraints, the expected longitudinal acceleration is solved by the following quadratic programming problem:

[0068] min·J

[0069] st·max(d safe , Δd min +d des (k+i|k))≤d(k+i|k)≤Δd max +d des (k+i|k)

[0070] ·v rmin ≤v r (k+i|k)≤v rmax

[0071] ·u min ≤u(k+j|k)≤u max

[0072] ·Δu min ≤Δu(k+j|k)≤Δu max (9)

[0074] In the above formula, the first constraint reflects tracking performance and safety, the second constraint reflects tracking performance, and the last two constraints reflect comfort. min is the lower limit of the spacing difference, Δd max is the upper limit of the spacing difference, v rmin is the minimum relative speed, v rmax is the maximum relative speed, u min is the expected minimum longitudinal acceleration, u max is the maximum expected longitudinal acceleration increment, Δu min is the minimum expected longitudinal acceleration, Δu max is the maximum value of the expected longitudinal acceleration increment, {i=1,2,...N p , j = 1, 2, ... N c},N p is the prediction time domain, N c To control the time domain.

[0075] Specifically, according to the basic principle of SDNN, the first-order forward difference processing is used to obtain:

[0076]

[0077] Where ξ is the scaling factor, is a piecewise function, and its expression is:

[0078]

[0079] The corresponding relationship between the discrete SDNN parameters and the quadratic programming parameters in equation (9) is:

[0080]

[0081] in, y min =[max(d safe (k), Δd min +d des (k)), v rmin , v2(k)-v rmax ,u min ] T ,y max =[Δd max +d des (k), v rmax , v2(k)-v rmin ,u max ] T , E1=ψx(k)-Y ref (k)+W ve e d (k) ΔU(k)=[Δu(k) T ,…,Δu(k+N c -1) T ] T , y ref (k+i|k)=[d des (k), 0, v2(k), 0] T is the reference vector, Q and R are weight coefficients [w d , w r ,0,w u ]、w Δu The extended diagonal weight matrix, N x is the number of state variables, N u N is the number of control variables. y is the number of output variables, N c To control the number of time domain steps, N p To predict the number of time domain steps, ε is the relaxation factor, ε max is the maximum value of the loose factor, ρ is the weight coefficient of the loose factor, and the meanings of the other parameters are the same as those of formula (3) and formula (9).

[0082] According to the discrete SDNN model:

[0083]

[0084] Where P = W -1 E T , s = W -1 c.

[0085] Specifically, the detailed steps of the model predictive control algorithm based on discrete SDNN are as follows:

[0086] Input: the longitudinal speed of the vehicle v1, the relative distance d between the front and rear vehicles, the relative speed v between the front and rear vehicles r .

[0087] Step 1: Sampling time k = 0. Determine whether to initialize, if yes, initialize and jump to Step 3; if not, jump to Step 2.

[0088] Step 2: Update the state variables and output variables in equation (7) and perform feedback correction.

[0089] Step 3: Unconstrained optimization, solving analytical solution:

[0090] ΔU M (k)=(θ T Qθ+R) -1 θ T R(ψx(k)-Y ref (k)+W re e d (k))=(θ T Qθ+R) -1 θ T RE1.

[0091] Step 4: Determine the analytical solution ΔU obtained by unconstrained optimization M (k) Check whether all constraints are met. If so, go to Step 7; if not, go to Step 5.

[0092] Step 5: Set the iteration time n = 1 for discrete SDNN to iteratively solve QP.

[0093] Step 6: Discrete SDNN iteratively solves QP.

[0094] a) Calculate the current iteration time according to the discrete SDNN model

[0095] b) Calculate the relative error:

[0096]

[0097] in

[0098]

[0099] c) Determine error≤ε QP Is it finished? If satisfied, jump to step 7; if not satisfied, jump to d);

[0100] d) n = n + 1, jump to a);

[0101] Step 7: Get the longitudinal acceleration increment Δu(k) and calculate the longitudinal acceleration u(k).

[0102] Step 8: k=k+1, jump to Step 2.

[0103] The algorithm flow chart is shown in Figure 3 When the MPC is turned on, it is initialized first and then enters the MPC real-time control. The algorithm can solve QP in an embedded platform with limited resources and can achieve parallel computing through a multi-core processor.

[0104] S3, determining the motor operation mode according to the expected longitudinal acceleration of the vehicle and the acceleration-speed relationship during the natural deceleration of the vehicle;

[0105] S3.1. According to the longitudinal speed of the vehicle, query the acceleration-speed relationship table of the vehicle's natural deceleration process to obtain the corresponding natural deceleration;

[0106] S3.2. Determine the relative magnitude relationship between the expected longitudinal acceleration and natural deceleration of the vehicle, determine the motor working mode, and output the motor working mode signal. "00" represents maintaining the original working mode, "01" represents the driving control mode, and "10" represents the braking control mode.

[0107] Specifically, the expression of the motor working mode signal is as follows:

[0108]

[0109] In the above formula, s m is the motor working mode signal, u n (v1) is the natural deceleration when the longitudinal velocity of the vehicle is v1, u h is the acceleration threshold.

[0110] S4. Determine the longitudinal dynamics model of the vehicle according to the motor working mode, and calculate the expected motor torque in combination with the expected longitudinal acceleration of the vehicle to obtain a control signal.

[0111] S4.1. Determine that the motor working mode is the original working mode, maintain the original inverse dynamics model, take the vehicle's expected longitudinal acceleration and longitudinal speed as input, and obtain the expected motor torque;

[0112] S4.2, judging that the working mode of the motor is the driving control mode, adopting the reverse driving model, taking the expected longitudinal acceleration and longitudinal speed of the vehicle as input, and obtaining the expected motor torque;

[0113] Specifically, if the reverse drive model is adopted, the expected driving torque is:

[0114]

[0115] In the above formula, the T t (k) is the driving torque, u(k) is the desired longitudinal acceleration of the vehicle, v1 is the longitudinal velocity of the vehicle, R is the wheel radius, i t is the total transmission ratio of the transmission system, η is the mechanical efficiency of the transmission system, m is the vehicle mass, C D is the air resistance coefficient, A is the frontal area, g is the gravity constant, f is the rolling resistance coefficient, α is the road slope angle, and δ is the rotational mass conversion coefficient.

[0116] S4.3, judging that the motor working mode is the braking control mode, adopting the reverse braking model, taking the expected longitudinal acceleration and longitudinal speed of the vehicle as input, and obtaining the expected motor torque;

[0117] Specifically, if the reverse braking model is adopted, the expected braking torque is:

[0118]

[0119] In the above formula, the T t (k) is the absolute value of the braking torque, and the meanings of the other parameters are the same as those in formula (16).

[0120] S4.4. Combine the expected motor torque with the actual motor torque, output a control signal and perform torque control on the motor.

[0121] S4.4.1. If drive control is performed, the difference between the desired drive torque and the measured actual motor torque is used as input to perform direct torque control of the motor through a PID controller.

[0122] The braking mode is determined according to the state of charge and braking intensity, namely:

[0123]

[0124] In the above formula, the s b is the braking mode signal, SOC is the state of charge, and z is the braking intensity. "00" corresponds to full mechanical braking; "01" corresponds to a combination of regenerative braking and mechanical braking; "10" corresponds to full regenerative braking.

[0125] S4.4.2. If braking control is performed, regenerative braking is used to improve energy utilization efficiency. The difference between the expected motor braking torque and the measured actual motor torque is used as input, and the motor direct torque control is performed through a PID controller. The motor braking torque is combined with the mechanical braking torque to perform braking.

[0126] Specifically, the following analysis is for front-wheel drive models, but it is still applicable to other drive modes. When regenerative braking is fully adopted, the expression of regenerative braking torque is:

[0127] T r (k) = T b (k)

[0128] In the above formula, the T r (k) is the expected regenerative braking torque, T b (k) is the desired braking torque.

[0129] When regenerative braking is combined with mechanical braking, a fuzzy controller is used to obtain the ratio of the expected regenerative braking torque to the expected braking torque. The fuzzy subsets of the fuzzy controller input signal are: braking intensity E(z) = {L, M, H}, ranging from 0 to 1; state of charge E(SOC) = {L, M, H}, ranging from 0 to 1; longitudinal speed of the vehicle E(v1) = {L, M, H}, ranging from 0 to 100 km / h. The ratio of the regenerative braking torque to the expected braking torque is the output signal of the fuzzy controller, and its fuzzy subset is E(p) = {L, M, H}, ranging from 0 to 1.

[0130] The expression of the expected regenerative braking torque is:

[0131] T r (k) = min(pT b (k), T m (k))

[0132] In the above formula, the T r (k) is the expected regenerative braking torque, T b (k) is the expected braking torque, T m (k) is the front axle braking torque under ideal braking conditions, and the mechanical braking torque T h (k) = T b (k)-T r (k).

[0133] When mechanical braking is fully adopted, the expected regenerative braking torque is T r (k)=0.

[0134] The difference between the desired regenerative braking torque and the measured actual motor torque is used as input to perform direct torque control of the motor through a PID controller.

[0135] An adaptive cruise control system using a composite model predictive controller, comprising:

[0136] A system working mode selection module, used to determine the working mode of the adaptive cruise system according to the vehicle motion state information and road information;

[0137] The acceleration calculation module outputs the vehicle's expected longitudinal acceleration based on the adaptive cruise system engineering model and combined with the vehicle's motion state information;

[0138] A motor operation mode selection module is used to determine the motor operation mode according to the expected longitudinal acceleration of the vehicle and the acceleration-speed relationship during the natural deceleration of the vehicle;

[0139] The control signal output module is used to determine the vehicle longitudinal dynamics model according to the motor working mode, and calculate the expected motor torque in combination with the expected longitudinal acceleration of the vehicle to obtain a control signal.

[0140] The contents of the above method embodiments are all applicable to the present system embodiments. The functions specifically implemented by the present system embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0141] An adaptive cruise control device using a composite model predictive controller:

[0142] at least one processor;

[0143] at least one memory for storing at least one program;

[0144] When the at least one program is executed by the at least one processor, the at least one processor implements the adaptive cruise control method using a composite model predictive controller as described above.

[0145] The contents of the above method embodiments are all applicable to the present device embodiments. The functions specifically implemented by the present device embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0146] A storage medium stores processor-executable instructions, wherein the processor-executable instructions are used to implement the above-mentioned adaptive cruise control method using a composite model predictive controller when executed by the processor.

[0147] The contents of the above method embodiments are all applicable to the present storage medium embodiments. The functions specifically implemented by the present storage medium embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0148] The above is a specific description of the preferred implementation of the present invention, but the invention is not limited to the embodiments. Those skilled in the art may make various equivalent modifications or substitutions without violating the spirit of the present invention. These equivalent modifications or substitutions are all included in the scope defined by the claims of this application.

Claims

1. An adaptive cruise control method using a composite model predictive controller, characterized in that: The following steps are involved: Determine the working mode of the adaptive cruise system according to the vehicle motion state information and road information; Based on the working mode of the adaptive cruise control system and combined with the vehicle motion status information, the expected longitudinal acceleration of the vehicle is output; Determine the motor operation mode based on the expected longitudinal acceleration of the vehicle and the acceleration-speed relationship during the natural deceleration of the vehicle; Determine the longitudinal dynamics model of the vehicle according to the motor working mode, and calculate the expected motor torque in combination with the expected longitudinal acceleration of the vehicle to obtain a control signal; The expression of the motor working mode signal is as follows: In the above formula, s m represents the motor working mode signal, u represents the expected longitudinal acceleration, u n (v1) represents the natural deceleration of the vehicle when its longitudinal velocity is v1, u h Indicates the acceleration threshold, "00" means keeping the original working mode, "01" means the driving control mode, and "10" means the braking control mode; The step of determining the vehicle longitudinal dynamics model according to the motor working mode, and calculating the expected motor torque in combination with the expected longitudinal acceleration of the vehicle to obtain the control signal specifically includes: It is determined that the motor working mode is the original working mode, the original inverse dynamics model is maintained, and the expected longitudinal acceleration and longitudinal speed of the vehicle are used as input to obtain the expected motor torque; It is determined that the motor working mode is the driving control mode, and the inverse driving model is used to obtain the expected motor torque by taking the expected longitudinal acceleration and longitudinal speed of the vehicle as input; It is determined that the motor working mode is the braking control mode, and a reverse braking model is used to obtain the expected motor torque by taking the expected longitudinal acceleration and longitudinal speed of the vehicle as input; Combine the expected motor torque with the actual motor torque, output a control signal and perform torque control on the motor; Using the reverse drive model, the expected motor torque is the expected drive torque, and the formula of the expected drive torque is as follows: In the above formula, the T t (k) is the expected driving torque, u(k) is the expected longitudinal acceleration of the vehicle, v1(k) is the longitudinal velocity of the vehicle, R is the wheel radius, i t is the total transmission ratio of the transmission system, η is the mechanical efficiency of the transmission system, m is the vehicle mass, C D is the air resistance coefficient, A is the frontal area, g is the gravity constant, f is the rolling resistance coefficient, α is the road slope angle, and δ is the rotational mass conversion coefficient.

2. The adaptive cruise control method using a composite model predictive controller according to claim 1, characterized in that: The vehicle motion state information includes the longitudinal speed of the vehicle, the longitudinal acceleration of the vehicle, the longitudinal speed of the preceding vehicle, the longitudinal acceleration of the preceding vehicle and the distance between the vehicle and the preceding vehicle, and the road information includes the road slope angle.

3. The adaptive cruise control method using a composite model predictive controller according to claim 2, characterized in that: The step of determining the working mode of the adaptive cruise control system according to the vehicle motion state information and the road information specifically includes: Calculate the minimum safe distance based on the vehicle's motion status information and determine the adaptive cruise system operating mode based on road information; The adaptive cruise system operating modes include a cruise mode and a following mode.

4. The adaptive cruise control method using a composite model predictive controller according to claim 3, characterized in that: The step of outputting the expected longitudinal acceleration of the vehicle based on the working mode of the adaptive cruise system and in combination with the vehicle motion state information specifically includes: It is determined that the working mode of the adaptive cruise system is the cruise mode, and a single neuron PID controller is selected and the difference between the expected cruise speed and the actual longitudinal speed is used as input to obtain the expected longitudinal acceleration of the vehicle; It is determined that the working mode of the adaptive cruise control system is the following mode. The incremental MPC controller based on discrete SDNN is selected and the distance between the vehicle and the vehicle in front, the speed of the vehicle in front relative to the vehicle, the longitudinal speed of the vehicle, and the longitudinal acceleration of the vehicle at the previous moment are used as inputs. The objective function is constructed and combined with the kinematic constraints to iteratively solve the expected longitudinal acceleration of the vehicle.

5. The adaptive cruise control method using a composite model predictive controller according to claim 4, characterized in that: Selecting a single neuron PID controller, the expression of the vehicle's desired longitudinal acceleration is as follows: In the above formula, u(k) represents the expected longitudinal acceleration, K is the gain coefficient, u(k-1) represents the expected longitudinal acceleration of the previous iteration, and x i (k) is the intermediate state variable, w' i (k) is the weight coefficient.

Images