CN112034859A - Self-adaptive dynamic planning method of anti-interference CACC system - Google Patents

Self-adaptive dynamic planning method of anti-interference CACC system Download PDF

Info

Publication number
CN112034859A
CN112034859A CN202010959308.4A CN202010959308A CN112034859A CN 112034859 A CN112034859 A CN 112034859A CN 202010959308 A CN202010959308 A CN 202010959308A CN 112034859 A CN112034859 A CN 112034859A
Authority
CN
China
Prior art keywords
vehicle
controller
model
matrix
following
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN202010959308.4A
Other languages
Chinese (zh)
Other versions
CN112034859B (en
Inventor
高振宇
安会爽
郭戈
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Northeastern University Qinhuangdao Branch
Original Assignee
Northeastern University Qinhuangdao Branch
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Northeastern University Qinhuangdao Branch filed Critical Northeastern University Qinhuangdao Branch
Priority to CN202010959308.4A priority Critical patent/CN112034859B/en
Publication of CN112034859A publication Critical patent/CN112034859A/en
Application granted granted Critical
Publication of CN112034859B publication Critical patent/CN112034859B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
    • G05D1/02—Control of position or course in two dimensions
    • G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
    • G05D1/0212—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory
    • G05D1/0223—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory involving speed control of the vehicle
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
    • G05D1/02—Control of position or course in two dimensions
    • G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
    • G05D1/0212—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory
    • G05D1/0214—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory in accordance with safety or protection criteria, e.g. avoiding hazardous areas
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
    • G05D1/02—Control of position or course in two dimensions
    • G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
    • G05D1/0212—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory
    • G05D1/0221—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory involving a learning process
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
    • G05D1/02—Control of position or course in two dimensions
    • G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
    • G05D1/0276—Control of position or course in two dimensions specially adapted to land vehicles using signals provided by a source external to the vehicle
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
    • G05D1/02—Control of position or course in two dimensions
    • G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
    • G05D1/0287—Control of position or course in two dimensions specially adapted to land vehicles involving a plurality of land vehicles, e.g. fleet or convoy travelling
    • G05D1/0291—Fleet control
    • G05D1/0295—Fleet control by at least one leading vehicle of the fleet

Landscapes

  • Engineering & Computer Science (AREA)
  • Aviation & Aerospace Engineering (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Remote Sensing (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Automation & Control Theory (AREA)
  • Feedback Control In General (AREA)

Abstract

The invention provides an anti-interference self-adaptive dynamic programming method of a CACC system, and relates to the technical field of heterogeneous fleet control. The invention solves the motorcade stability problem under the condition of simultaneous delay of an actuator, external interference and front vehicle interference by constructing a model and providing the ADP motorcade cooperative control method based on data driving, and the method can enable the motorcade to quickly reach a stable state. In the analysis of the stability of the vehicle, the stability of the vehicle is proved by analyzing the magnitude of the cost function updated by the controller. The cost function is bounded and is smaller than a minimum value, which proves that the state of the vehicle and the control input reach a stable state.

Description

Self-adaptive dynamic planning method of anti-interference CACC system
Technical Field
The invention relates to the technical field of heterogeneous fleet control, in particular to an adaptive dynamic planning method for an anti-interference CACC system.
Background
In recent years, research on heterogeneous vehicle fleets is still limited, and a control method based on data driving is rarely applied to vehicle fleet Coordinated Adaptive Cruise Control (CACC). With the rapid development of the technology in the field of artificial intelligence, jiang et al y.vehicular and z. -p.j., comprehensive adaptive optimal control for continuous-time linear systems with complex un-known dynamics, "automotive, vol.48, No.10, pp.2699-2704,2012. and y.j.j.z. -p.j., road adaptive dynamic programming and feedback stabilization of non-linear systems," IEEE trans.neural net.leirn.25, vol.25, No.5, pp.882-893, may2014. In Gao et al work, Gao W, Jiang ZP, Ozbay K.Data-driven adaptive optimal control of connected vehicles.IEEE Trans inner Transp Syst.2017; 18(5):1122 & 1133. and K.J.Malakorn and B.park, "Assessment of mobility, energy, and environmental impacts of Intelligent-based adaptive cruise control and internal traffic control," in Proc.IEEE. Symp.Sustain.Syhnol, May2010, pp.1-6. ADP are applied to the adaptive optimal control of the data-driven interconnected vehicle. In the aspect of cooperative adaptive fleet control, communication interference problems and fleet stability control in an interference environment currently face challenges.
In the above studies, input skew due to engine process delay was not considered in the controller design process, which greatly limited its application in practical heterogeneous fleet control. In addition, interference in the networked vehicle system comes from different aspects, such as unpredictable acceleration or deceleration of the front vehicle, signal interference in the communication process and the like.
Due to the second-order fleet model adopted by the controller design, the second-order fleet model cannot well capture dynamic characteristics inside the vehicle, and various interference problems in the travelling process of the fleet need to be considered. Furthermore, it is also important to analyze and verify the stability of each vehicle, i.e. to ensure that the inter-vehicle distance between adjacent vehicles does not continuously enlarge from lead vehicle to the last vehicle.
Disclosure of Invention
Aiming at the defects of the prior art, the invention provides an adaptive dynamic planning method of an anti-interference CACC system, which constructs a three-order dynamics model of a fleet covering various interference factors, realizes cooperative control of the fleet by reasonable inter-vehicle distance, ensures that vehicles run quickly and stably, and strictly deduces and proves the stability of each vehicle in the fleet.
In order to solve the technical problems, the technical scheme adopted by the invention is as follows:
an adaptive dynamic planning method of an anti-interference CACC system comprises the following steps:
step 1, constructing a fixed time interval strategy according to the dynamic performance of a vehicle, and establishing a longitudinal dynamic model of the vehicle by adopting a one-way communication structure;
the vehicle longitudinal dynamics model does not contain a lead vehicle, and the inter-vehicle distance error of the ith vehicle is defined to be delta pi(t) having:
Δpi(t)＝pi-1(t)-pi(t)-ds-L,i＝1,2,...,n
wherein d iss＝hivi(t)+riIs the desired inter-vehicle distance, hiIs a constant headway, vi(t) is the speed of the ith vehicle at time t, riFixed inter-vehicle distance, pi(t) is the position of the ith vehicle at time t, and L is the vehicle length;
the dynamical model of the ith vehicle is the following nonlinear differential equation:
Figure BDA0002679850150000021
wherein, Δ vi(t) is the speed error of the ith vehicle at time t, ai(t) is the acceleration of the ith vehicle at time t, ci(t) is the feedback linearization control law, hiConstant headway is constant, fi(vi,ai) Is a non-linear dynamic model of the vehicle,
Figure BDA0002679850150000022
is a constant;
for a lead car, since it has no front car, its kinematic model is defined as follows:
Figure BDA0002679850150000023
Figure BDA0002679850150000024
where σ is the specific mass of air, τiIs the mechanical time constant, Ai,cdi,dmiAnd miThe cross-sectional area, the drag coefficient, the mechanical resistance and the mass of the ith vehicle are respectively;
linearizing the model to obtain a feedback linearization control law as follows:
Figure BDA0002679850150000025
the following linearized models were obtained:
Figure BDA0002679850150000026
wherein u isi(t) is an additional control input signal;
adding disturbance xi to the linearized model due to communication failure or external disturbance in the networking environmenti(t); the linearization model after adding the disturbance is:
Figure BDA0002679850150000031
wherein
Figure BDA0002679850150000032
ξ i(t) and
Figure BDA0002679850150000033
are respectively disturbance xii(t) a lower bound and an upper bound;
step 2, setting the whole fleet to be composed of 1 leading vehicle and n following vehicles, respectively constructing controller models of the leading vehicle and the following vehicles, and providing a controller optimization algorithm based on data driving;
the controller model of the lead vehicle is as follows:
u0(t)＝-K0x0(t)
wherein x0(t)＝[p0(t) v0(t) a0(t)]T，p0(t)、v0(t) and a0(t) is the position, speed and acceleration of the lead vehicle at the moment t, and the feedback control rate K of the lead vehicle0＝[k01 k02 k03]Wherein k is01，k02，k03Respectively, the position, the speed and the acceleration of the lead vehicle.
The vehicle-following kinetic model is:
Figure BDA0002679850150000034
wherein,
Figure BDA0002679850150000035
i＝1,2,…,n，xi(t)＝[Δpi(t) Δvi(t) ai(t)]T；Δpi(t) is the position error of the ith vehicle from the target position at time t, Δ vi(t) is the speed error between the ith vehicle and the preceding vehicle at time t, and the controller ui(t)＝-Kixi(t)，Ki＝[ki1 ki1 ki2]The feedback control laws respectively follow the position, speed and acceleration of the vehicle. The controller after the unfolding is obtained as follows:
ui(t)＝-ki1Δpi(t)-ki2Δvi(t)-ki3ai(t)
the vehicle-following dynamics model is rewritten as:
Figure BDA0002679850150000036
wherein,
Figure BDA0002679850150000037
j＝0,1,…,n，ωi(t)＝Dixi-1(t)+Iiξi(t)，Aijis a coefficient matrix of the state inputs of i cars after iteration j times, BiIs a coefficient matrix of control inputs for i cars, DiIs a coefficient matrix of the status inputs of the i-1 vehicle,
Figure BDA0002679850150000038
is the control gain for i cars after j iterations.
The controller ui(t) the optimization algorithm comprises the steps of:
step S1: selecting a minimum loss function according to a control target:
Figure BDA0002679850150000039
wherein Q is blockdiag (Q)1,Q2,…,Qn)，
Figure BDA00026798501500000310
R＝blockdiag(R1,R2,…,Rn),
Figure BDA00026798501500000311
Figure BDA0002679850150000041
Is the set of states at time t of all vehicles, u ═ u1,u2,…,un]TIs the set of control inputs of all vehicles at the time t, the control gain of all vehicles is K*＝R-1BTP*Wherein P is*Solved by algebraic Riccati equation.
Step S2: obtaining the iteration value of the feedback gain of the step j +1 of the ith vehicle according to the Riccati equation
Figure BDA0002679850150000042
Wherein
Figure BDA0002679850150000043
The intermediate value of step j of the ith vehicle;
step S3: since there is no way to predict the parameter matrix A of the vehicles in the fleetiAnd BiThen, the solution is solved by the following equation, i.e., [ t, t + t [ ]]The difference in time is then the minimum loss function:
Figure BDA0002679850150000044
where is the signal sample interval time.
According to the kronecker product, and
a＝[vecv(a(t1))-vecv(a(t0)),…,vecv(a(ts))-vecv(a(ts-1))]T,
Figure BDA0002679850150000045
the formula is simplified to obtain:
Figure BDA0002679850150000046
Figure BDA0002679850150000047
Figure BDA0002679850150000048
wherein,
Figure BDA0002679850150000049
is a process matrix and satisfies the column full rank,
Figure BDA00026798501500000410
in the form of a matrix of processes,
Figure BDA00026798501500000411
is a process variable matrix;
step 3, obtaining a data-driven CACC control method according to a data-driven-based controller optimization algorithm;
step 3.1: selecting an initial controller gain K for a follower vehicle0And a desired threshold σ > 0;
step 3.2: the initial controller model inputs u (t) ═ K including interference information0x (t) + e (t) at time intervals of [ t [, ]0,ts]；
Step 3.3: calculating a process variable
Figure BDA0002679850150000051
Satisfy the requirement of
Figure BDA0002679850150000052
Let j ← 0;
step 3.4: p pair by using the formula in step S3(j)，K(j+1)Solving is carried out;
step 3.5: j ← j +1, until | P(j)-P(j-1)|＜σ；
Step 3.6: updating the controller model so that u (t) is equal to-K(j)x(t)；
Step 3.7: returning the updating result of the controller u (t);
when in use
Figure BDA0002679850150000053
When the algorithm converges, i.e. it is
Figure BDA0002679850150000054
And
Figure BDA0002679850150000055
respectively converge to the optimum intermediate value P*And an optimal feedback control gain K*；
Step 4, constructing a following vehicle differential equation according to the dynamic performance of the vehicle, and proving the stability of the following vehicle;
step 4.1: firstly, constructing a following vehicle dynamic equation:
Figure BDA0002679850150000056
wherein [ delta ] | is not more than rho is interfered when the vehicle runs, and is obtained according to the steps of 3.1-3.7, and exists
Figure BDA0002679850150000057
Get the estimated optimal controller of the vehicle > 0
Figure BDA0002679850150000058
Step 4.2: when T is less than or equal to s is less than or equal to T + T,
Figure BDA0002679850150000059
when there is a minimum valueiTime, matrix
Figure BDA00026798501500000510
Is always a Hulviz matrix and finds the constant βi,λiIs greater than 0, and satisfies:
Figure BDA00026798501500000511
wherein i is 1,2, …, n;
step 4.3: for the lead car:
Figure BDA00026798501500000512
β0,λ0is constant, thus proving that the closed loop system of the vehicle is exponentially stable.
Step 5, constructing a transfer matrix to prove the optimality of the controller;
defining a minimum loss function J according to the minimum loss function proof⊙＝xT(0)P*x (0), obtained in step 3, the state transition matrix phi (tau, t) satisfies that phi (tau, t) is less than or equal to beta eλ(τ-t),
Figure BDA00026798501500000514
Definition of
Figure BDA00026798501500000513
Presence of c1,c2>0 is a constant number of times, and,
Figure BDA0002679850150000061
wherein
Figure BDA0002679850150000062
Is semi-positive and continuously differentiable, and has an upper limit of
Figure BDA0002679850150000063
Wherein
Figure BDA0002679850150000064
Φ (τ, τ) ═ I, according to the law of the leibuctz integral, we obtain:
Figure BDA0002679850150000065
distributed minimum loss function
Figure BDA0002679850150000066
Satisfy the requirement of
Figure BDA0002679850150000067
Wherein λ ismax(Q) is the maximum eigenvalue of the Q matrix, λmin(P) is the minimum eigenvalue of the P matrix, μ is a constant.
Thus, the controller minimum loss function proves to have optimal performance.
Adopt the produced beneficial effect of above-mentioned technical scheme to lie in:
the invention provides an adaptive dynamic planning method of an anti-interference CACC system, which is used for constructing a model and providing an ADP fleet cooperative control algorithm based on data driving, solving the problem of fleet stability under the condition that actuator delay, external interference and front vehicle interference exist simultaneously, and enabling a fleet to reach a stable state quickly. In the analysis of the stability of the vehicle, the stability of the vehicle is proved by analyzing the magnitude of the cost function updated by the controller. The cost function is bounded and is smaller than a minimum value through analysis, and the vehicle state and the control input can be proved to reach a stable state.
The invention establishes a third-order linear fleet dynamics model on the basis of a fleet model with a second order adopted by the cooperative braking control of the traditional interconnected vehicles. Compared with a second-order fleet model, the third-order model can better capture dynamic characteristics inside the vehicle. Aiming at the condition that disturbance exists in the heterogeneous fleet system environment, a controller of a leading fleet vehicle and a cooperative controller of a following fleet vehicle are respectively designed, and an ADP self-adaptive cooperative control algorithm based on data driving is designed, so that the problem of fleet stability under the condition that various disturbances exist simultaneously is solved. The invention provides a data-driven fleet cooperative braking control method, which is used for evaluating the convergence of vehicles by using a cost function. The simulation result verifies the effectiveness of the method.
Drawings
Fig. 1 is a flowchart of an adaptive dynamic programming method of an anti-interference CACC system according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of a heterogeneous fleet and communication topology used in an embodiment of the present invention;
FIG. 3 is a schematic diagram of an optimization iteration process according to an embodiment of the present invention;
wherein diagrams (a) -PijMatrix optimization iterative process, graph (b) -optimization iterative process of error feedback gain
FIG. 4 is a schematic diagram of errors between a following vehicle and a preceding vehicle of a straight lane when a lead vehicle is disturbed according to an embodiment of the invention;
wherein, the following vehicle v1 and the preceding vehicle error condition when the vehicle in the straight lane runs are shown in the graph (a), the following vehicle v2 and the preceding vehicle error condition when the vehicle in the straight lane runs are shown in the graph (b), and the following vehicle v3 and the preceding vehicle error condition when the vehicle in the straight lane runs are shown in the graph (c);
FIG. 5 is a schematic diagram of errors between a following vehicle and a preceding vehicle of a straight lane when a lead vehicle is undisturbed according to the embodiment of the invention;
wherein, figure (a) -the following vehicle v1 and the preceding vehicle error condition when the vehicle runs on the straight lane; (b) following vehicle v2 versus preceding vehicle error condition while driving in straight lane; (c) following vehicle v3 versus preceding vehicle error while driving in straight lane
Detailed Description
The following detailed description of embodiments of the present invention is provided in connection with the accompanying drawings and examples. The following examples are intended to illustrate the invention but are not intended to limit the scope of the invention.
The heterogeneous fleet model provided by the embodiment comprehensively considers actuator faults and external interference, adopts an ADP algorithm based on data driving, and designs the cooperative controller, so that the convergence of each vehicle in the whole braking process of the fleet can be realized, and the queue stability of the whole fleet can be realized.
An adaptive dynamic planning method for an anti-interference CACC system, as shown in fig. 1, includes the following steps:
step 1, constructing a fixed time interval strategy according to the dynamic performance of a vehicle, and establishing a longitudinal dynamic model of the vehicle by adopting a one-way communication structure;
the vehicle longitudinal dynamics model does not contain a lead vehicle, and the inter-vehicle distance error of the ith vehicle is defined to be delta pi(t) having:
Δpi(t)＝pi-1(t)-pi(t)-ds-L,i＝1,2,...,n
wherein d iss＝hivi(t)+riIs the desired inter-vehicle distance, hiIs a constant headway, vi(t) is the speed of the ith vehicle at time t, riFixed inter-vehicle distance, pi(t) is the position of the ith vehicle at time t, and L is the vehicle length;
the dynamical model of the ith vehicle is the following nonlinear differential equation:
Figure BDA0002679850150000071
wherein, Δ vi(t) is the speed error of the ith vehicle at time t, ai(t) is the acceleration of the ith vehicle at time t, ci(t) is the feedback linearization control law, hiConstant headway is constant, fi(vi,ai) Is a non-linear dynamic model of the vehicle,
Figure BDA0002679850150000072
is a constant;
for a lead car, since it has no front car, its kinematic model is defined as follows:
Figure BDA0002679850150000073
Figure BDA0002679850150000074
where σ is the specific mass of air, τiIs the mechanical time constant, Ai,cdi,dmiAnd miThe cross-sectional area, the drag coefficient, the mechanical resistance and the mass of the ith vehicle are respectively;
linearizing the model to obtain a feedback linearization control law as follows:
Figure BDA0002679850150000081
the following linearized models were obtained:
Figure BDA0002679850150000082
wherein u isi(t) is an additional control input signal to enable the closed loop system to satisfy the stable reactancePerformance indicators of interference; with this controller, the following two objectives are achieved: 1) linearization of ith vehicle dynamics; 2) the system model is simplified by excluding some characteristic parameters in the vehicle dynamics, such as mechanical resistance, mass and air resistance.
Adding disturbance xi to the linearized model due to communication failure or external disturbance in the networking environmenti(t); the linearization model after adding the disturbance is:
Figure BDA0002679850150000083
wherein
Figure BDA0002679850150000084
ξ i(t) and
Figure BDA0002679850150000085
are respectively disturbance xii(t) a lower bound and an upper bound;
step 2, setting the whole fleet to be composed of 1 leading vehicle and n following vehicles, respectively constructing controller models of the leading vehicle and the following vehicles as shown in fig. 2, and providing a controller optimization algorithm based on data driving;
the controller model of the lead vehicle is as follows:
u0(t)＝-K0x0(t)
wherein x0(t)＝[p0(t) v0(t) a0(t)]T，p0(t)、v0(t) and a0(t) is the position, speed and acceleration of the lead vehicle at the moment t, and the feedback control rate K of the lead vehicle0＝[k01 k02 k03]Wherein k is01，k02，k03Respectively, the position, the speed and the acceleration of the lead vehicle.
The vehicle-following kinetic model is:
Figure BDA0002679850150000086
wherein,
Figure BDA0002679850150000087
i＝1,2,…,n，xi(t)＝[Δpi(t) Δvi(t) ai(t)]T；Δpi(t) is the position error of the ith vehicle from the target position at time t, Δ vi(t) is the speed error between the ith vehicle and the preceding vehicle at time t, and the controller ui(t)＝-Kixi(t)，Ki＝[ki1 ki1 ki2]The feedback control laws respectively follow the position, speed and acceleration of the vehicle. The controller after the unfolding is obtained as follows:
ui(t)＝-ki1Δpi(t)-ki2Δvi(t)-ki3ai(t)
the vehicle-following dynamics model is rewritten as:
Figure BDA0002679850150000091
wherein,
Figure BDA0002679850150000092
j＝0,1,…,n，Aijis a coefficient matrix of the state inputs of i cars after iteration j times, BiIs a coefficient matrix of control inputs for i cars, DiIs a coefficient matrix of the status inputs of the i-1 vehicle,
Figure BDA00026798501500000911
is the control gain for i cars after j iterations.
The controller ui(t) the optimization algorithm comprises the steps of:
step S1: selecting a minimum loss function according to a control target:
Figure BDA0002679850150000093
wherein Q is blockdiag (Q)1,Q2,…,Qn)，
Figure BDA0002679850150000094
R＝blockdiag(R1,R2,…,Rn),
Figure BDA0002679850150000095
Figure BDA0002679850150000096
Is the set of states at time t of all vehicles, u ═ u1,u2,…,un]TIs the set of control inputs of all vehicles at the time t, the control gain of all vehicles is K*＝R-1BTP*Wherein P is*Solved by algebraic Riccati equation.
Step S2: obtaining the iteration value of the feedback gain of the step j +1 of the ith vehicle according to the Riccati equation
Figure BDA0002679850150000097
Wherein
Figure BDA0002679850150000098
The intermediate value of step j of the ith vehicle;
step S3: since there is no way to predict the parameter matrix A of the vehicles in the fleetiAnd BiThen, the solution is solved by the following equation, i.e., [ t, t + t [ ]]The difference in time is then the minimum loss function:
Figure BDA0002679850150000099
where is the signal sample interval time.
According to the kronecker product, and
a＝[vecv(a(t1))-vecv(a(t0)),…,vecv(a(ts))-vecv(a(ts-1))]T,
Figure BDA00026798501500000910
the above formula can be simplified to obtain:
Figure BDA0002679850150000101
Figure BDA0002679850150000102
Figure BDA0002679850150000103
wherein,
Figure BDA0002679850150000104
is a process matrix and satisfies the column full rank,
Figure BDA0002679850150000105
in the form of a matrix of processes,
Figure BDA0002679850150000106
is a process variable matrix;
step 3, obtaining a data-driven CACC control method according to a data-driven-based controller optimization algorithm;
step 3.1: selecting an initial controller gain K for a follower vehicle0And a desired threshold σ > 0;
step 3.2: the initial controller model inputs u (t) ═ K including interference information0x (t) + e (t) at time intervals of [ t [, ]0,ts]；
Step 3.3: calculating a process variable
Figure BDA0002679850150000107
Satisfy the requirement of
Figure BDA0002679850150000108
Let j ← 0;
step 3.4: p pair by using the formula in step S3(j)，K(j+1)Solving is carried out;
step 3.5: j ← j +1, until | P(j)-P(j-1)|＜σ；
Step 3.6: updating the controller model so that u (t) is equal to-K(j)x(t)；
Step 3.7: returning the updating result of the controller u (t);
when in use
Figure BDA0002679850150000109
When the algorithm converges, i.e. it is
Figure BDA00026798501500001010
And
Figure BDA00026798501500001011
respectively converge to the optimum intermediate value P*And an optimal feedback control gain K*；
Step 4, constructing a following vehicle differential equation according to the dynamic performance of the vehicle, and proving the stability of the following vehicle;
step 4.1: firstly, constructing a following vehicle dynamic equation:
Figure BDA00026798501500001012
wherein [ delta ] | is not more than rho is interfered when the vehicle runs, and is obtained according to the steps of 3.1-3.7, and exists
Figure BDA00026798501500001013
Get the estimated optimal controller of the vehicle > 0
Figure BDA00026798501500001014
Step 4.2: when T is less than or equal to s is less than or equal to T + T,
Figure BDA0002679850150000111
when there is a minimum valueiTime, matrix
Figure BDA0002679850150000112
Is always a Hulviz matrix and finds the constant βi,λiIs greater than 0, and satisfies:
Figure BDA0002679850150000113
wherein i is 1,2, …, n;
step 4.3: for the lead car:
Figure BDA0002679850150000114
β0,λ0being constant, it is thus demonstrated that the closed loop system of the vehicle is exponentially stable.
Step 5, constructing a transfer matrix to prove the optimality of the controller;
defining a minimum loss function J according to the minimum loss function proof⊙＝xT(0)P*x (0), as shown in step 3, the state transition matrix phi (tau, t) satisfies that phi (tau, t) is less than or equal to beta eλ(τ-t),
Figure BDA0002679850150000115
Definition of
Figure BDA0002679850150000116
Presence of c1,c2>0 is a constant number of times, and,
Figure BDA0002679850150000117
wherein
Figure BDA0002679850150000118
Is semi-positive and continuously differentiable, and has an upper limit of
Figure BDA0002679850150000119
Wherein
Figure BDA00026798501500001110
Φ (τ, τ) ═ I, according to the law of the leibuctz integral, we obtain:
Figure BDA00026798501500001111
setting up
Figure BDA00026798501500001112
According to
Figure BDA00026798501500001113
Distributed minimum loss function
Figure BDA00026798501500001114
Satisfy the requirement of
Figure BDA00026798501500001115
Wherein λ ismax(Q) is the maximum eigenvalue of the Q matrix, λmin(P) is the minimum eigenvalue of the P matrix, μ is a constant.
Thus, the controller minimum loss function proves to have optimal performance.
In this embodiment, assume that a leading vehicle and 3 following vehicles travel in a straight line on a lane, and in order to study and analyze the influence of disturbance on performance, two cases are considered: the lead vehicle is undisturbed, the lead vehicle is disturbed, and all vehicles of the fleet are disturbed by the sampling signal in the two conditions. The sampling interval is set to 0.2s therein. The initial position is set to p (0) [ -3,8,20,31]Tm, initial velocity v (0) [5,8,9,7 ]]Tm/s。
Disturbance xii(t) can be specifically classified into the following two forms:
case 1: without disturbance
ξi(t)＝0，i＝0,...n.
Case 2: the lead car is disturbed
ξi(t)＝10*3sin(3*t)，i＝0.
All vehicles being disturbed
ξi(t)＝0.3∑sin(random(-1000,1000)*t)，i＝0,...,n.
In the simulation, the initial estimation of the disturbance upper bound and the disturbance lower bound of the lead vehicle are respectively
Figure BDA0002679850150000121
And
Figure BDA0002679850150000122
the initial estimation of the disturbance upper bound and the disturbance lower bound of the following vehicle are respectively
Figure BDA0002679850150000123
And
Figure BDA0002679850150000124
the fleet dynamics parameters were set as follows: time constant of engine (tau)i＝[0.08,0.12,0.14,0.08]) Time constant hi＝[0.5,0.49,0.51,0.43]Initial feedback control gain set to K0＝[-0.5,-0.5,0]. In the simulation, the length of the vehicle was ignored.
Based on the parameters, simulation verification is carried out on the motorcade cooperative braking control method based on the data-driven ADP control theory, and the simulation verification is shown in 3-5. Wherein figure 3 shows K under the proposed controllerij,PijAnd (5) an iterative process. As can be seen from FIG. 3, the intermediate value PijAnd a feedback control gain KijThe stability is achieved. FIG. 4 shows the variation of the position error, velocity error and self-acceleration of the following vehicle and the preceding vehicle; it can be seen from fig. 4 that the position error of each vehicle in the platoon may slowly converge smoothly to zero after the controller updates. This process takes approximately 5s or so. We can clearly see that the disturbance is not gradually amplified in the following vehicles and the fleet is able to maintain a steady state. Fig. 5 shows the position error, speed error and self acceleration of the following vehicle and the front vehicle when the head vehicle in the fleet has no disturbance. From FIG. 5, it can be seen thatUnder the condition of no disturbance, the distance between vehicles always keeps a reasonable safe distance, thereby avoiding the occurrence of rear-end accidents. Further, the pitch error converges to 0 around 5s, and the magnitude of the pitch error decreases as the vehicle index increases in the fleet. The data-driven ADP cooperative brake controller based on the motorcade can not only ensure the stability of each vehicle, but also ensure the stability of the motorcade. And, the ADP controller based on data driving is robust to disturbances.
The foregoing description is only exemplary of the preferred embodiments of the disclosure and is illustrative of the principles of the technology employed. It will be appreciated by those skilled in the art that the scope of the invention in the embodiments of the present disclosure is not limited to the specific combination of the above-mentioned features, but also encompasses other embodiments in which any combination of the above-mentioned features or their equivalents is made without departing from the inventive concept as defined above. For example, the above features and (but not limited to) technical features with similar functions disclosed in the embodiments of the present disclosure are mutually replaced to form the technical solution.

Claims (6)

1. An adaptive dynamic planning method of an anti-interference CACC system is characterized by comprising the following steps:
step 1, constructing a fixed time interval strategy according to the dynamic performance of a vehicle, and establishing a longitudinal dynamic model of the vehicle by adopting a one-way communication structure;
step 2, setting the whole fleet to be composed of 1 leading vehicle and n following vehicles, respectively constructing controller models of the leading vehicle and the following vehicles, and providing a controller optimization algorithm based on data driving;
step 3, obtaining a data-driven CACC control method according to a data-driven-based controller optimization algorithm;
step 4, constructing a following vehicle differential equation according to the dynamic performance of the vehicle, and proving the stability of the following vehicle;
step 5, constructing a transfer matrix to prove the optimality of the controller;
according to a minimum loss functionProve, define the minimum loss function J⊙＝xT(0)P*x (0), obtained in step 3, the state transition matrix Φ (τ, t) satisfies
Figure FDA0002679850140000011
Definition of
Figure FDA0002679850140000012
Presence of c1,c2>0 is a constant number of times, and,
Figure FDA0002679850140000013
wherein
Figure FDA0002679850140000014
Is semi-positive and continuously differentiable, and has an upper limit of
Figure FDA0002679850140000015
Wherein
Figure FDA0002679850140000016
Φ (τ, τ) ═ I, according to the law of the leibuctz integral, we obtain:
Figure FDA0002679850140000017
distributed minimum loss function
Figure FDA0002679850140000018
Satisfy the requirement of
Figure FDA0002679850140000019
Wherein λ ismax(Q) is the maximum eigenvalue of the Q matrix, λmin(P) is the minimum eigenvalue of the P matrix, μ is a constant;
thus, the controller minimum loss function proves to have optimal performance.
2. The adaptive dynamic programming method for an antijam CACC system as claimed in claim 1, wherein the longitudinal dynamics model of the vehicle in step 1 does not include a lead vehicle, and the inter-vehicle distance error of the i-th vehicle is defined as Δ pi(t) having:
Δpi(t)＝pi-1(t)-pi(t)-ds-L,i＝1,2,...,n；
wherein d iss＝hivi(t)+riIs the desired inter-vehicle distance, hiIs a constant headway, vi(t) is the speed of the ith vehicle at time t, riFixed inter-vehicle distance, pi(t) is the position of the ith vehicle at time t, and L is the vehicle length;
the dynamical model of the ith vehicle is the following nonlinear differential equation:
Figure FDA0002679850140000021
wherein, Δ vi(t) is the speed error of the ith vehicle at time t, ai(t) is the acceleration of the ith vehicle at time t, ci(t) is the feedback linearization control law, hiConstant headway is constant, fi(vi,ai) Is a non-linear dynamic model of the vehicle,
Figure FDA0002679850140000022
is a constant;
for a lead car, since it has no front car, its kinematic model is defined as follows:
Figure FDA0002679850140000023
Figure FDA0002679850140000024
where σ is the specific mass of air, τiIs the mechanical time constant, Ai,cdi,dmiAnd miThe cross-sectional area, the drag coefficient, the mechanical resistance and the mass of the ith vehicle are respectively;
linearizing the model to obtain a feedback linearization control law as follows:
Figure FDA0002679850140000025
the following linearized models were obtained:
Figure FDA0002679850140000026
wherein u isi(t) is an additional control input signal;
adding disturbance xi to the linearized model due to communication failure or external disturbance in the networking environmenti(t); the linearization model after adding the disturbance is:
Figure FDA0002679850140000027
wherein
Figure FDA0002679850140000028
ξ i(t) and
Figure FDA0002679850140000029
are respectively disturbance xiiLower and upper bounds of (t).
3. The adaptive dynamic programming method of an antijam CACC system as set forth in claim 1, wherein the controller model of the lead vehicle in step 2 is as follows:
u0(t)＝-K0x0(t)
wherein x0(t)＝[p0(t) v0(t) a0(t)]T，p0(t)、v0(t) and a0(t) is the position, speed and acceleration of the lead vehicle at the moment t, and the feedback control rate K of the lead vehicle0＝[k01 k02 k03]Wherein k is01，k02，k03Respectively are feedback control laws of the position, the speed and the acceleration of the lead vehicle;
the vehicle-following kinetic model is:
Figure FDA0002679850140000031
wherein,
Figure FDA0002679850140000032
xi(t)＝[Δpi(t) Δvi(t) ai(t)]T；Δpi(t) is the position error of the ith vehicle from the target position at time t, Δ vi(t) is the speed error between the ith vehicle and the preceding vehicle at time t, and the controller ui(t)＝-Kixi(t)，Ki＝[ki1 ki1 ki2]Respectively following the feedback control laws of the position, the speed and the acceleration of the vehicle, and obtaining the controller after expansion as follows:
ui(t)＝-ki1Δpi(t)-ki2Δvi(t)-ki3ai(t)
the vehicle-following dynamics model is rewritten as:
Figure FDA0002679850140000033
wherein,
Figure FDA0002679850140000034
ωi(t)＝Dixi-1(t)+Iiξi(t)，Aijis a coefficient matrix of the state inputs of i cars after iteration j times, BiIs a coefficient matrix of control inputs for i cars, DiIs a coefficient matrix of the status inputs of the i-1 vehicle,
Figure FDA0002679850140000035
is the control gain for i cars after j iterations.
4. The adaptive dynamic programming method of an antijam CACC system as set forth in claim 1, wherein said controller u in step 2i(t) the optimization algorithm comprises the steps of:
step S1: selecting a minimum loss function according to a control target:
Figure FDA0002679850140000036
wherein
Figure FDA0002679850140000037
Figure FDA0002679850140000038
Is the set of states at time t of all vehicles, u ═ u1,u2,…,un]TIs the set of control inputs of all vehicles at the time t, the control gain of all vehicles is K*＝R-1BTP*Wherein P is*Solved by algebraic Riccati equation;
step S2: obtaining the iteration value of the feedback gain of the step j +1 of the ith vehicle according to the Riccati equation
Figure FDA0002679850140000039
Wherein P isi (j)The intermediate value of step j of the ith vehicle;
step S3: since there is no way to predict the parameter matrix A of the vehicles in the fleetiAnd BiThen, the following equation is used to solve, i.e., [ t ],t+t]The difference in time is then the minimum loss function:
Figure FDA0002679850140000041
where is the signal sample interval time;
according to the kronecker product, and
a＝[vecv(a(t1))-vecv(a(t0)),…,vecv(a(ts))-vecv(a(ts-1))]T,
Figure FDA0002679850140000042
the formula is simplified to obtain:
Figure FDA0002679850140000043
Figure FDA0002679850140000044
Figure FDA0002679850140000045
wherein,
Figure FDA0002679850140000046
is a process matrix and satisfies the column full rank,
Figure FDA0002679850140000047
in the form of a matrix of processes,
Figure FDA0002679850140000048
is a process variable matrix.
5. The adaptive dynamic planning method for an anti-interference CACC system according to claim 1, wherein step 3 specifically comprises the following steps:
step 3.1: selecting an initial controller gain K for a follower vehicle0And a desired threshold σ > 0;
step 3.2: the initial controller model inputs u (t) ═ K including interference information0x (t) + e (t) at time intervals of [ t [, ]0,ts]；
Step 3.3: calculating a process variable
Figure FDA0002679850140000049
Until it is satisfied
Figure FDA00026798501400000410
Let j ← 0;
step 3.4: p pair by using the formula in step S3(j)，K(j+1)Solving is carried out;
step 3.5: j ← j +1, until | P(j)-P(j-1)|＜σ；
Step 3.6: updating the controller model so that u (t) is equal to-K(j)x(t)；
Step 3.7: returning the updating result of the controller u (t);
when in use
Figure FDA0002679850140000051
When the algorithm converges, i.e. it is
Figure FDA0002679850140000052
And
Figure FDA0002679850140000053
respectively converge to the optimum intermediate value P*And an optimal feedback control gain K*。
6. The adaptive dynamic planning method for an anti-jamming CACC system according to claim 1, wherein step 4 specifically includes the steps of:
step 4.1: firstly, constructing a following vehicle dynamic equation:
Figure FDA0002679850140000054
wherein [ delta ] | is not more than rho is interfered when the vehicle runs, and is obtained according to the steps of 3.1-3.7, and exists
Figure FDA0002679850140000055
Get the estimated optimal controller of the vehicle > 0
Figure FDA0002679850140000056
Step 4.2: when T is less than or equal to s is less than or equal to T + T,
Figure FDA0002679850140000057
when there is a minimum valueiTime, matrix
Figure FDA0002679850140000058
Is always a Hulviz matrix and finds the constant βi,λiIs greater than 0, and satisfies:
Figure FDA0002679850140000059
wherein i is 1,2, …, n;
step 4.3: for the lead car:
Figure FDA00026798501400000510
β0,λ0being constant, it is thus demonstrated that the closed loop system of the vehicle is exponentially stable.
CN202010959308.4A 2020-09-14 2020-09-14 Self-adaptive dynamic planning method of anti-interference CACC system Active CN112034859B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010959308.4A CN112034859B (en) 2020-09-14 2020-09-14 Self-adaptive dynamic planning method of anti-interference CACC system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010959308.4A CN112034859B (en) 2020-09-14 2020-09-14 Self-adaptive dynamic planning method of anti-interference CACC system

Publications (2)

Publication Number Publication Date
CN112034859A true CN112034859A (en) 2020-12-04
CN112034859B CN112034859B (en) 2022-08-02

Family

ID=73589064

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010959308.4A Active CN112034859B (en) 2020-09-14 2020-09-14 Self-adaptive dynamic planning method of anti-interference CACC system

Country Status (1)

Country Link
CN (1) CN112034859B (en)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113034911A (en) * 2020-12-14 2021-06-25 湖南大学 Vehicle queue control method and system with parameter and structure heterogeneity
CN113147764A (en) * 2021-06-01 2021-07-23 吉林大学 Vehicle control method based on hybrid potential function of cooperative adaptive cruise system
CN113391553A (en) * 2021-06-15 2021-09-14 东北大学秦皇岛分校 Self-adaptive optimal control method of heterogeneous CACC system with actuator saturation
CN114253274A (en) * 2021-12-24 2022-03-29 吉林大学 Data-driven-based online hybrid vehicle formation rolling optimization control method

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP0828071A2 (en) * 1996-09-05 1998-03-11 Toyota Jidosha Kabushiki Kaisha Acceleration control apparatus for a diesel engine provided with supercharger
CN108973998A (en) * 2018-07-11 2018-12-11 清华大学 A kind of heterogeneous vehicle platoon distribution energy-saving control method based on MPC
CN109144076A (en) * 2018-10-31 2019-01-04 吉林大学 A kind of more vehicle transverse and longitudinals coupling cooperative control system and control method
CN110333728A (en) * 2019-08-02 2019-10-15 大连海事大学 A kind of isomery fleet fault tolerant control method based on change time interval strategy
CN110703761A (en) * 2019-10-30 2020-01-17 华南理工大学 Networked autonomous fleet scheduling and cooperative control method based on event triggering

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP0828071A2 (en) * 1996-09-05 1998-03-11 Toyota Jidosha Kabushiki Kaisha Acceleration control apparatus for a diesel engine provided with supercharger
CN108973998A (en) * 2018-07-11 2018-12-11 清华大学 A kind of heterogeneous vehicle platoon distribution energy-saving control method based on MPC
CN109144076A (en) * 2018-10-31 2019-01-04 吉林大学 A kind of more vehicle transverse and longitudinals coupling cooperative control system and control method
CN110333728A (en) * 2019-08-02 2019-10-15 大连海事大学 A kind of isomery fleet fault tolerant control method based on change time interval strategy
CN110703761A (en) * 2019-10-30 2020-01-17 华南理工大学 Networked autonomous fleet scheduling and cooperative control method based on event triggering

Non-Patent Citations (6)

* Cited by examiner, † Cited by third party
Title
LIGANG WU,等: "Cooperative Adaptive Cruise Control with Communication Constraints", 《PROCEEDINGS OF THE 34TH CHINESE CONTROL CONFERENCE》 *
YANG ZHOU,等: "Rolling horizon stochastic optimal control strategy for ACC and CACC under uncertainty", 《TRANSPORTATION RESEARCH PART C》 *
YU BAI，等: "Optimal control based CACC: problem formulation, solution, and stability analysis", 《2019 IEEE INTELLIGENT VEHICLES SYMPOSIUM》 *
于晓海，郭戈: "车队控制中的一种通用可变时距策略", 《自动化学报》 *
华雪东，等: "车辆协同巡航控制***设计改进与试验评价", 《交通运输***工程与信息》 *
高振宇，郭戈: "基于扰动观测器的AUVs固定时间编队控制", 《自动化学报》 *

Cited By (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113034911A (en) * 2020-12-14 2021-06-25 湖南大学 Vehicle queue control method and system with parameter and structure heterogeneity
CN113034911B (en) * 2020-12-14 2022-06-14 湖南大学 Vehicle queue control method and system with parameter and structure heterogeneity
CN113147764A (en) * 2021-06-01 2021-07-23 吉林大学 Vehicle control method based on hybrid potential function of cooperative adaptive cruise system
CN113391553A (en) * 2021-06-15 2021-09-14 东北大学秦皇岛分校 Self-adaptive optimal control method of heterogeneous CACC system with actuator saturation
CN113391553B (en) * 2021-06-15 2022-04-01 东北大学秦皇岛分校 Self-adaptive optimal control method of heterogeneous CACC system with actuator saturation
CN114253274A (en) * 2021-12-24 2022-03-29 吉林大学 Data-driven-based online hybrid vehicle formation rolling optimization control method
CN114253274B (en) * 2021-12-24 2024-03-26 吉林大学 Data-driven-based networked hybrid vehicle formation rolling optimization control method

Also Published As

Publication number Publication date
CN112034859B (en) 2022-08-02

Similar Documents

Publication Publication Date Title
CN112034859B (en) Self-adaptive dynamic planning method of anti-interference CACC system
CN110333728B (en) Heterogeneous fleet fault tolerance control method based on variable time interval strategy
Wang et al. Robust output-feedback based vehicle lateral motion control considering network-induced delay and tire force saturation
CN103381826B (en) Based on the self-adapting cruise control method of approximate Policy iteration
CN110703761B (en) Networked autonomous fleet scheduling and cooperative control method based on event triggering
CN111694366B (en) Motorcade cooperative braking control method based on sliding mode control theory
CN103324085A (en) Optimal control method based on supervised reinforcement learning
CN112083719B (en) Finite time fleet control method based on preset performance function
CN112150808B (en) Urban traffic system scheduling strategy generation method based on deep learning
Kuutti et al. End-to-end reinforcement learning for autonomous longitudinal control using advantage actor critic with temporal context
Hu et al. Distributed coordinated brake control for longitudinal collision avoidance of multiple connected automated vehicles
CN103246200B (en) A kind of motor train unit synchronization and tracking control method based on distributed model
CN116339155B (en) High-speed motor train unit data driving integral sliding mode control method, system and equipment
CN113391553B (en) Self-adaptive optimal control method of heterogeneous CACC system with actuator saturation
CN105160870A (en) Bidirectional autonomous fleet control method
He et al. Gaussian learning‐based fuzzy predictive cruise control for improving safety and economy of connected vehicles
CN113359483A (en) Vehicle cooperative control method based on nonsingular rapid terminal sliding mode control
Chen et al. Hybrid physics-data-driven online modelling: Framework, methodology and application to electric vehicles
CN116224802B (en) Vehicle team longitudinal composite control method based on interference observer and pipeline model prediction
CN116088530A (en) Reinforced learning automatic driving fleet control method based on model predictive control guidance
Ghasemi Directional control of a platoon of vehicles for comfort specification by considering parasitic time delays and lags
CN113759701A (en) High-speed train speed control method and system
Juárez et al. Assisted cooperative adaptive cruise control with human memory effects
Wang et al. Cooperative adaptive cruise control using delay-based spacing policy: a robust adaptive non-singular terminal sliding mode approach
Meng et al. Stability analysis for car following model based on control theory

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant