CN116243610A - Data-driven vehicle queue fault-tolerant tracking control tracking method and system - Google Patents
Data-driven vehicle queue fault-tolerant tracking control tracking method and system Download PDFInfo
- Publication number
- CN116243610A CN116243610A CN202310530233.1A CN202310530233A CN116243610A CN 116243610 A CN116243610 A CN 116243610A CN 202310530233 A CN202310530233 A CN 202310530233A CN 116243610 A CN116243610 A CN 116243610A
- Authority
- CN
- China
- Prior art keywords
- vehicle
- fault
- queue
- attack
- representing
- 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
Links
- 238000000034 method Methods 0.000 title claims abstract description 47
- 238000013528 artificial neural network Methods 0.000 claims abstract description 19
- 230000001360 synchronised effect Effects 0.000 claims abstract description 11
- 230000006870 function Effects 0.000 claims description 60
- 230000007246 mechanism Effects 0.000 claims description 15
- 238000010276 construction Methods 0.000 claims description 12
- 238000004458 analytical method Methods 0.000 claims description 9
- 241000432812 Pseudobias Species 0.000 claims description 6
- 230000003044 adaptive effect Effects 0.000 claims description 5
- 238000013461 design Methods 0.000 claims description 4
- 210000002569 neuron Anatomy 0.000 claims description 4
- 230000001133 acceleration Effects 0.000 claims description 2
- 230000009347 mechanical transmission Effects 0.000 claims description 2
- 238000005070 sampling Methods 0.000 claims 1
- 238000005516 engineering process Methods 0.000 abstract description 5
- 230000000737 periodic effect Effects 0.000 abstract description 4
- 230000008569 process Effects 0.000 description 8
- 238000010586 diagram Methods 0.000 description 5
- 238000011161 development Methods 0.000 description 4
- 238000012545 processing Methods 0.000 description 3
- 238000011160 research Methods 0.000 description 3
- 230000005540 biological transmission Effects 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 230000007257 malfunction Effects 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 238000013459 approach Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000012938 design process Methods 0.000 description 1
- 238000006073 displacement reaction Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000012886 linear function Methods 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B13/00—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion
- G05B13/02—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric
- G05B13/04—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators
- G05B13/042—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators in which a parameter or coefficient is automatically adjusted to optimise the performance
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Computation (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Feedback Control In General (AREA)
Abstract
The invention belongs to the technical field of vehicle queue control, and discloses a fault-tolerant tracking control tracking method and system for a data-driven vehicle queue. The invention establishes a biased format dynamic linearization model, namely a PFDL model, aiming at an actual nonlinear vehicle queuing system of sensor faults and DoS attacks, converts a complex nonlinear vehicle queuing system into an equivalent PFDL model by adopting a dynamic linearization technology, approximates the sensor faults by adopting a radial basis function neural network method aiming at the sensor faults in the vehicle queues, finally establishes an elastic fault-tolerant model-free self-adaptive queuing safety controller on the basis, and aims at the problem of non-periodic DoS attacks which are likely to exist in network channels. The invention realizes the goal of synchronous tracking of the positions and the speeds of the leader vehicle and the follower vehicle in the vehicle queuing system.
Description
Technical Field
The invention belongs to the technical field of vehicle queue control, and relates to a fault-tolerant tracking control tracking method and a fault-tolerant tracking control tracking system for a data-driven vehicle queue, which are used for solving the problem of safety control of the vehicle queue affected by sensor faults and aperiodic denial of service attacks.
Background
The vehicle queue control is taken as an intelligent traffic method for coordinating the running process of vehicles, has wide prospect in the aspects of improving the safety of vehicles and improving the traffic capacity of roads, and is widely paid attention to. In the operation process of the fleet, many challenges from the inside and the outside, such as network attack risk problems caused by the communication network, possible faults of the physical and electronic components of the vehicle, and model complexity and uncertainty caused by the complex structure of the vehicle, etc., may be faced. These problems have greatly hindered the way of social progress and sustainable development. As an important development direction of the traffic control field, the intelligent traffic system is born for solving the problems, and the application of the intelligent traffic system is increasingly urgent, so that the intelligent traffic system can reduce traffic jams and improve traffic efficiency.
However, the above-described studies on vehicle control are mostly based on a relatively accurate system model. In fact, due to the complex structure of the actual system, with certain time-varying characteristics and nonlinearity, it is almost impossible to obtain an accurate model in the actual system. With the development of information science and technology, actual processes such as chemical industry, machinery, transportation and the like are also changed over the ground. The production technology and equipment in the industries are large in scale, and the production process is more complex. Modeling using first principles of nature or recognition-based processes becomes more difficult. Thus, conventional model-based control theory has been less applicable to such control problems. The development of data-driven control theory and application based on data for these complex processes has become an urgent issue to be solved.
As one of the data-driven control methods, model-free adaptive control technology has recently received attention from many students and has achieved great research results, providing a theoretical framework for control system design and stability analysis.
In summary, the communication network carries a potential risk of network attacks in vehicle queue control, and it is necessary to take network problems into account. Further, considering that the electronic components (sensors) of the vehicle may malfunction during long-term operation, the malfunction of one or more vehicles may have serious consequences during running. Therefore, achieving fault-tolerant control of the vehicle is a necessary condition for smooth operation of the vehicle queuing system. At the same time the complexity and uncertainty of the vehicle model can also hinder the modeling and control process.
Disclosure of Invention
The invention aims to provide a data-driven vehicle queue fault-tolerant tracking control tracking method so as to realize the self-adaptive queue safety control of an elastic fault-tolerant model-free vehicle queue system under the influence of sensor faults and aperiodic denial of service attacks of the vehicle queue system.
In order to achieve the above purpose, the invention adopts the following technical scheme:
a data-driven vehicle queue fault-tolerant tracking control tracking method comprises the following steps:
step 3, introducing variable PG parameters for auxiliary error analysis into the obtained discrete system equations of the leader vehicle and the follower vehicle, and reconstructing the output of the vehicle queue system by using a biased format dynamic linearization model;
step 4, designing a PG parameter estimator for estimating PG parameters in the partial format dynamic linearization model, and simultaneously introducing an observer and a controller algorithm to control a vehicle queue system;
step 5, approximating the sensor faults by using a radial basis function neural network, re-representing the output with faults by using an approximate fault function, and obtaining a fault-tolerant model-free self-adaptive queue controller;
step 6, on the basis of step 5, considering the aperiodic DoS attack in the network channel, providing an attack index function for indicating whether the attack occurs or not, and providing a corresponding attack compensation mechanism;
and 7, constructing an elastic fault-tolerant model-free self-adaptive queue safety controller by combining the step 5 and the step 6, and realizing synchronous tracking of the positions and the speeds of the leader vehicle and the follower vehicle in the vehicle queue system.
In addition, on the basis of the data-driven vehicle queue fault-tolerant tracking control tracking method, the invention also provides a data-driven vehicle queue fault-tolerant tracking control tracking system which is adaptive to the data-driven vehicle queue fault-tolerant tracking control tracking method, and the data-driven vehicle queue fault-tolerant tracking control tracking system adopts the following technical scheme:
a data-driven vehicle queue fault-tolerant tracking control tracking system, comprising:
the system comprises a dynamics equation construction module, a dynamics equation generation module and a dynamics equation generation module, wherein the dynamics equation construction module is used for establishing dynamics equations of a leader vehicle and a follower vehicle in a vehicle queue system;
the discrete system equation construction module is used for obtaining discrete system equations of the leader vehicle and the follower vehicle;
the system output reconstruction module is used for reconstructing the output of the vehicle queue system by using a partial format dynamic linearization model according to variable PG parameters for introducing auxiliary error analysis in the discrete system equations of the leader vehicle and the follower vehicle;
the system model parameter solving module is used for designing a PG parameter estimator for estimating PG parameters in the partial format dynamic linearization model, and simultaneously introducing an observer and a controller algorithm to control the vehicle queue system;
the fault-tolerant model-free self-adaptive queue control module is used for approximating the sensor faults according to the radial basis function neural network, re-representing the output with the faults by using an approximate fault function, and obtaining a fault-tolerant model-free self-adaptive queue controller;
the attack compensation mechanism construction module is used for providing an attack index function according to the aperiodic DoS attack in the network channel, and is used for indicating whether the attack occurs or not and providing a corresponding attack compensation mechanism;
and the synchronous tracking control module is used for constructing an elastic fault-tolerant model-free self-adaptive queue safety controller and realizing synchronous tracking control of the position and the speed of a leader vehicle by a follower vehicle in the vehicle queue system.
In addition, on the basis of the data-driven vehicle queue fault-tolerant tracking control tracking method, the invention further provides computer equipment which comprises a memory and one or more processors.
The memory stores executable code, and the processor is used for realizing the steps of the data-driven vehicle queue fault-tolerant tracking control tracking method when executing the executable code.
In addition, on the basis of the data-driven vehicle queue fault-tolerant tracking control tracking method, the invention further provides a computer-readable storage medium on which a program is stored. The program, when executed by the processor, is configured to implement the steps of the data-driven vehicle queue fault-tolerant tracking control tracking method described above.
Compared with the prior art, the invention has the following advantages:
as described above, the present invention is directed to a method and system for fault-tolerant tracking control of a data-driven vehicle queue. Aiming at the actual nonlinear vehicle queue system of sensor faults and DoS attacks, a biased format dynamic linearization model, namely a PFDL model, is established according to actual physical significance, a complex nonlinear vehicle queue system is converted into an equivalent PFDL model by adopting a dynamic linearization technology, the sensor faults in a vehicle queue are approximated by adopting a radial basis function neural network method, an elastic fault-tolerant model-free self-adaptive queue safety controller FT-MFAPSC is finally established on the basis, and aiming at the problem of non-periodic DoS attacks possibly existing in network channels, an attack compensation mechanism is provided, so that the problem of vehicle queue tracking control under the non-periodic DoS attacks is solved. The invention realizes the goal of synchronous tracking of the positions and the speeds of the leader vehicle and the follower vehicle in the vehicle queuing system.
Drawings
FIG. 1 is a flow chart of a data driven vehicle queue fault tolerant tracking control tracking method in an embodiment of the invention.
Fig. 2 is a diagram of a vehicle alignment system contemplated by the present invention.
FIG. 3 is a schematic diagram of a partial format dynamic linearization model of a vehicle alignment system contemplated by the present invention.
FIG. 4 is a block diagram of a vehicle queuing system for FT-MFAPSC under an aperiodic DoS attack contemplated by the present invention.
Fig. 5 is a schematic diagram of an aperiodic DoS attack contemplated by the present invention.
Detailed Description
Example 1
In order to solve the problem of safety control of a vehicle queue affected by sensor faults and aperiodic denial of service attacks, the embodiment provides a data-driven vehicle queue fault-tolerant tracking control tracking method.
As shown in fig. 1, the data-driven vehicle queue fault-tolerant tracking control tracking method includes the following steps:
and 1, establishing a dynamics equation of a leader vehicle and a follower vehicle in the vehicle queue system.
wherein ,
、
respectively representing the position and speed of the leader vehicle,
as a time-varying nonlinear function; the kinetic equation for the ith follower vehicle is as follows:
wherein the index i indicates the ith following vehicle,
n represents the number of following vehicles;
、
、
is the position, speed and mass of the ith following vehicle;
a control input for the i-th following vehicle representing a traction/braking force of the i-th following vehicle;
resistance for the ith following vehicle, including throttle, mechanical transmission friction, and aerodynamic resistance;
is about
Is an unknown function of (a).
And 2, obtaining discrete system equations of the leader vehicle and the follower vehicle based on the dynamics equation obtained in the step 1.
The discrete system equations for a follower vehicle are described as:
wherein ,
in order to sample the period of time,
、
respectively represent the first
Vehicle body
The position and speed of the moment in time,
、
、
respectively represent the first
Vehicle body
Position, velocity and acceleration at time.
Represent the first
Vehicle body
Input of time.
Representation of
Is used to determine the degree of freedom of the function,
representation of
Is an unknown function of (a).
The discrete system equations for the leader vehicle are:
wherein ,
、
representing leader vehicles respectively
The position and speed of the moment;
、
representing leader vehicles respectively
The location and speed of the moment.
And 3, introducing variable PG parameters for auxiliary error analysis into the obtained discrete system equations of the leader vehicle and the follower vehicle, and reconstructing the output of the vehicle queue system by using a partial format dynamic linearization model.
Introducing variables for auxiliary error analysis, and designing an output tuning factor
，
The selection of (c) will be explained in detail later.
Reconstructing the output of the follower vehicle in the vehicle queuing system as:
wherein ,
the output at the time of the i-th following vehicle k+1 is indicated.
Reconstructing the output of a leader vehicle in the vehicle queuing system as:
wherein ,
indicating the output of the leader vehicle at time k. Taking into account the limitations of the actual physical structure of the vehicle, the leader increases the vehicle output
It is a matter of course that it is not possible to provide a solution,
。
assuming a constant exists
So that
The following redefined vehicle queue system outputs are obtained:
wherein ：
introducing a parameter L as a control input
When the linear length coefficient of (a)
When the partial format dynamic linearization model is converted into the tight format dynamic linearization modelThe following formula is then defined:
wherein ,
a vector representing the traction/braking force composition from time k to time k-L+1;
the vector of increases in traction/braking force from time k to time k-L + 1 is shown.
Representing the input increment at time k of the ith vehicle,
，
representing the input delta at time k-L +1 for the ith vehicle,
represent the first
Input at time k-L+ 1 of the vehicle.
Assuming a nonlinear function
、
Is about
、
、
The pseudo partial derivative PG is continuous and nonlinear function
Meets the generalized Li Puxi z, namely
、
If (3)
The method comprises the steps of carrying out a first treatment on the surface of the Then:
For nonlinear systems
When the above assumption is satisfied and
find a time-varying PG parameter vector
So that
Converting into the following partial format dynamic linearization model:
wherein 
，
Bounded and of
；
Is that
1 st to L th elements of (b).
As shown in fig. 3, it is known from redefined system output
And (3) with
、
and 
In relation to, among other things,
and
respectively relate to
And
is a non-linear function of (2).
Thus, by selecting an appropriate linearization length factor
Can be used for
Expressed as in the past
Items relating to control inputs at various moments, i.e. associated with
And (5) correlation. While
And
the relation between the two is reflected by a pseudo partial derivative parameter, so the pseudo partial derivative PG parameter is the key of the PFDL-based vehicle queue system.
Based on the differential median theorem, we next further get:
wherein ：
order the
The partial format dynamic linearization model is rewritten as:
wherein ,
an input increment representing the i-th vehicle k-1 time;
and satisfy the following
。
In the above, give
And
a relationship between; based on the above work, the design process of the elastic fault-tolerant model-free self-adaptive queue safety controller can be completed only by the I/O data.
And 4, designing a PG parameter estimator for estimating PG parameters in the partial format dynamic linearization model, and simultaneously introducing an observer and a controller algorithm to control the vehicle queue system.
In analogy to the general PFDL-based control algorithm, the PG parameters of the vehicle are not available, so the present invention designs a PG parameter estimator to estimate the PG parameters, the ith vehicle estimator algorithm is as follows:
wherein ,
、
the estimated values of PG parameters at the k moment and the k-1 moment of the ith vehicle are respectively shown;
the step size coefficient is represented as such,
representing the weight coefficient.
In addition, the observer is added to observe output data, so that the controller algorithm is more universal, and the observer algorithm is as follows:
wherein ,
、
output estimates representing times k +1 and k of the vehicle,
representing the observer gain. Vehicle queuing system as shown in fig. 2, the vehicle queuing system of the invention establishes an elastic fault-tolerant model-free adaptive queuing security controller FT-MFAPSC for a vehicle queuing system subject to sensor faults and aperiodic DoS attacks.
To optimize tracking control performance, an optimal performance function for the vehicle queuing system is defined as follows:
wherein ,
representing an optimal performance function;
as the weight coefficient of the light-emitting diode,
is the safe distance between the ith following vehicle and the leader vehicle.
The optimal performance function consists of two parts, namely a second term
In order to enable smooth changes in the control input. First item
The tracking control performance is optimized on the basis.
According to the extremum optimizing condition, the controller algorithm with the self-adaptive structure is obtained as follows:
wherein 
Is a step size coefficient.
The invention aims at the estimation of the pseudo-gradient parameter vector obtained by using an estimator in a low conservation and no-model self-adaptive control (MFAC) algorithm based on an observer provided by a vehicle queue system, and solves the problem that the pseudo-gradient parameter is unavailable. Compared with the traditional pseudo-gradient algorithm, the method has the advantages that the observer is introduced to eliminate the limitation of unchanged pseudo-gradient parameter sign, and the conservation is reduced.
And 5, approximating the sensor faults by using a radial basis function neural network, re-representing the output with faults by using an approximate fault function, and obtaining the fault-tolerant model-free self-adaptive queue controller.
In practical systems, damage to vehicle components often results in failure of the system during operation.
The invention aims at the sensor fault unfolding research in the vehicle queue system, and researches how to continue to ensure the control effect under the condition of the sensor fault, which is also a typical fault problem.
Radial Basis Function Neural Networks (RBFNN) are one of the better approaches to functions. Approximation of sensor faults for an entire vehicle queuing system using a radial basis function neural network, fault approximation errors
Is defined as:
wherein ,
and
respectively isActual fault functions and approximate fault functions of the sensor.
Then, a neural network approximation error function is defined
The following are provided:
basis functions in hidden layers of neural networks
The method comprises the following steps:
wherein ,
the number of nodes is indicated and,
is an output function of the neural network;
the functions in the hidden layers are represented separately, and the specific equations are shown in equation (20).
Specifically, the Gaussian function is selected as
Namely, expressed as:
wherein ,
，
is the first
The center of the individual neurons is referred to as the center,
is the width of the basis function.
Based on formulas (17) to (20), the approximation function given for the sensor fault is as follows:
wherein 
Is provided with a threshold value
Is used for the output layer weight factor of (a),
which respectively represent the positions of 1, …,
output layer weighting factors of individual neurons, and
the update rule of (2) is as follows:
wherein ,
is the learning rate. To facilitate writing, will
Abbreviated as
So the rewrite is:
the convergence of the radial basis function algorithm is demonstrated by means of the Lyapunov function, and the following is obtained:
wherein ,
is a constant.
Based on radial basis function neural network, using
Approximation sensor failure; the output with failure is expressed as:
wherein ,
representing the output of the ith vehicle with the fault.
Then, the following fault-tolerant model-free adaptive queue controllers are obtained according to the formulas (13) to (16) and (24):
wherein ,
indicating that the leader vehicle has a faulty output.
And 6, on the basis of the step 5, considering the aperiodic DoS attack in the network channel, providing an attack index function for indicating whether the attack occurs or not, and providing a corresponding attack compensation mechanism.
Fig. 4 is a system block diagram of the FT-MFAPSC. As can be seen from fig. 4, data packets are transmitted over a network
An attacker blocks data transmission from the sensor to the controller by attacking the network channel.
The aperiodic DoS attack scheme is shown in fig. 5. Wherein:
represent the first
At the beginning of the time of the secondary attack,
is the end time. The gray area indicates that the system is currently being attacked by DoS, while the white area indicates that the system is not currently being attacked by DoS. Will be
And
respectively abbreviated as
And
。
the invention provides the following attack index function
To indicate whether an attack has occurred, the attack compensation mechanism is as follows:
wherein ,
for attack index functionIndicating whether an attack has occurred.
The aggregate of all attack periods, i.e. in the interval
In (a)
All times of (3)
Is set of (a)
Expressed as:
wherein N represents a non-negative integer; obviously, when DoS attacks do not occur, the set of all time intervals is as follows:
wherein ,
is an initial attack parameter for handling the situation where the vehicle queuing system is initially attacked,
is an attack duration coefficient to be determined; therefore, an attack compensation mechanism is proposed as follows:
wherein ,
representing the output after the compensation of the attack at time k of the ith vehicle with the fault,
the output after the compensation is attacked at the moment k-1 of the ith vehicle with the fault.
And 7, constructing an elastic fault-tolerant model-free self-adaptive queue safety controller by combining the step 5 and the step 6, and realizing synchronous tracking of the position and the speed of the leader vehicle by the follower vehicle in the vehicle queue system.
The design of the elastic fault-tolerant model-free self-adaptive queue safety control controller is as follows:
wherein ,
is a normal number, has smaller value,
representing the estimated initial value of the pseudo-bias guide of the ith vehicle.
As shown in fig. 4, the method of the present invention contemplates a vehicle consist of n+1 vehicles, wherein the front of the vehicle consist has a leader vehicle behind it
A vehicle follower vehicle. First of all speed and displacement data measured by sensors of a vehicle queuing system
Because PG parameters in the partial format dynamic linearization are not available, a PG parameter estimator is designed at the sensor end, and therefore, the data packet
And transmitting to the controller through a network. Considering that there is a non-periodic DoS attack in the network channel, an attacker blocks data transmission from the sensor to the controller by attacking the network channel, therefore the invention proposes an attack compensation mechanism to make the data packet after compensation when transmitted to the controller
At the same time consider pairs of
The controller based on the observation period for observation makes the designed observer have more universality, and the controller obtains the output through calculation
The system acts on an actuator to control the vehicle queue system so as to achieve the aim of synchronously tracking the positions and the speeds of the follower vehicle and the leader vehicle.
Example 2
A data-driven vehicle queue fault-tolerant tracking control tracking system, comprising:
the system comprises a dynamics equation construction module, a dynamics equation generation module and a dynamics equation generation module, wherein the dynamics equation construction module is used for establishing dynamics equations of a leader vehicle and a follower vehicle in a vehicle queue system;
the discrete system equation construction module is used for obtaining discrete system equations of the leader vehicle and the follower vehicle;
the system output reconstruction module is used for reconstructing the output of the vehicle queue system by using a partial format dynamic linearization model according to variable PG parameters for introducing auxiliary error analysis in the discrete system equations of the leader vehicle and the follower vehicle;
the system model parameter solving module is used for designing a PG parameter estimator for estimating PG parameters in the partial format dynamic linearization model, and simultaneously introducing an observer and a controller algorithm to control the vehicle queue system;
the fault-tolerant model-free self-adaptive queue control module is used for approximating the sensor faults according to the radial basis function neural network, re-representing the output with the faults by using an approximate fault function, and obtaining a fault-tolerant model-free self-adaptive queue controller;
the attack compensation mechanism construction module is used for providing an attack index function according to the aperiodic DoS attack in the network channel, and is used for indicating whether the attack occurs or not and providing a corresponding attack compensation mechanism;
and the synchronous tracking control module is used for constructing an elastic fault-tolerant model-free self-adaptive queue safety controller and realizing synchronous tracking control of the position and the speed of a leader vehicle by a follower vehicle in the vehicle queue system.
It should be noted that, in the fault-tolerant tracking control tracking system for a data-driven vehicle queue, the implementation process of the functions and roles of each functional module is specifically described in the implementation process of the corresponding steps in the method in the above embodiment 1, and will not be described herein.
Example 3
Embodiment 3 describes a computer apparatus for implementing the steps of the data-driven vehicle queue fault-tolerant tracking control tracking method described in embodiment 1 above.
The computer device includes a memory and one or more processors. Executable code is stored in the memory for implementing the steps of the data driven vehicle queue fault tolerant tracking control tracking method when the executable code is executed by the processor.
In this embodiment, the computer device is any device or apparatus having data processing capability, which is not described herein.
Example 4
Embodiment 4 describes a computer-readable storage medium for implementing the steps of the data-driven vehicle queue fault-tolerant tracking control tracking method described in embodiment 1 above.
The computer-readable storage medium of this embodiment 4 has stored thereon a program for implementing the steps of the data-driven vehicle queue fault-tolerant tracking control tracking method when executed by a processor.
The computer readable storage medium may be an internal storage unit of any device or apparatus having data processing capability, such as a hard disk or a memory, or may be an external storage device of any device having data processing capability, such as a plug-in hard disk, a Smart Media Card (SMC), an SD Card, a Flash memory Card (Flash Card), or the like, which are provided on the device.
The foregoing description is, of course, merely illustrative of preferred embodiments of the present invention, and it should be understood that the present invention is not limited to the above-described embodiments, but is intended to cover all modifications, equivalents and alternatives falling within the spirit and scope of the present invention as defined by the appended claims.
Claims (9)
1. A data-driven vehicle queue fault-tolerant tracking control tracking method is characterized by comprising the following steps:
step 1, establishing a dynamics equation of a leader vehicle and a follower vehicle in a vehicle queue system;
step 2, obtaining discrete system equations of the leader vehicle and the follower vehicle based on the dynamics equation obtained in the step 1;
step 3, introducing variable PG parameters for auxiliary error analysis into the obtained discrete system equations of the leader vehicle and the follower vehicle, and reconstructing the output of the vehicle queue system by using a biased format dynamic linearization model;
step 4, designing a PG parameter estimator for estimating PG parameters in the partial format dynamic linearization model, and simultaneously introducing an observer and a controller algorithm to control a vehicle queue system;
step 5, approximating the sensor faults by using a radial basis function neural network, and re-representing the output with faults by using an approximate fault function to obtain a fault-tolerant model-free self-adaptive queue controller;
step 6, on the basis of step 5, considering the aperiodic DoS attack in the network channel, providing an attack index function for indicating whether the attack occurs or not, and providing a corresponding attack compensation mechanism;
and 7, combining the step 5 and the step 6, and constructing an elastic fault-tolerant model-free self-adaptive queue safety controller so as to realize synchronous tracking of the positions and speeds of the leader vehicle and the follower vehicle in the vehicle queue system.
2. The method for fault-tolerant tracking control of a data-driven vehicle queue of claim 1,
in the step 1, the dynamics of the vehicle are described as follows:
the kinetic equation of the leader vehicle is:
wherein ,
、/>
respectively representing the position, speed, & lt, & gt of the leader vehicle>
As a time-varying nonlinear function; the kinetic equation for the ith follower vehicle is as follows:
wherein the index i indicates the ith following vehicle,
n represents the number of following vehicles;
3. The method for tracking fault-tolerant tracking control of a data-driven vehicle train according to claim 2, wherein,
in the step 2, the discrete system equation of the follower vehicle is described as:
wherein ,
for the sampling period +.>
、/>
Respectively represent +.>
Vehicle->
Position and speed of moment +.>
、/>
、/>
Respectively represent +.>
Vehicle->
Position, velocity and acceleration at time;
the discrete system equations for the leader vehicle are:
wherein ,
、/>
respectively represent leader vehicle->
The position and speed of the moment;
4. The method for fault-tolerant tracking control of a data-driven vehicle queue of claim 3,
in the step 3, a variable for assisting error analysis is introduced, and an output tuning factor is designed
；
Reconstructing the output of the follower vehicle in the vehicle queuing system as:
wherein ,
an output indicating the time k+1 of the i-th following vehicle;
reconstructing the output of a leader vehicle in the vehicle queuing system as:
wherein ,
representing the output of the leader vehicle at time k; considering the limitation of the actual physical structure of the vehicle, the increase of the leader vehicle output +.>
Is bounded, &>
；
Assuming a constant exists
So that->
The following redefined vehicle queue system outputs are obtained: />
wherein ：
introducing a parameter L as a control input
And defines the following formula:
wherein ,
a vector representing the traction/braking force composition from time k to time k-L+1; />
A vector representing the delta composition of traction/braking force from time k to time k-L+1;
assuming a nonlinear function
、/>
Is about->
、/>
、/>
The pseudo partial derivative PG is continuous and a nonlinear function +.>
Satisfy the generalized sense Li Puxi z, i.e. +.>
、/>
If (3)
；/>
Then
, wherein />
；
For nonlinear systems
When the above assumption is satisfied and +.>
Find a time-varying PG parameter vector +.>
Make->
Converting into the following partial format dynamic linearization model:
wherein 
，/>
Bounded and->
；/>
Is->
1 st to L th elements of (b);
based on the differential median theorem, we next further get:
wherein ：
order the
The partial format dynamic linearization model is rewritten as:
wherein ,
an input increment representing the i-th vehicle k-1 time; />
5. The method for fault-tolerant tracking control of a data-driven vehicle queue of claim 4,
in the step 4, a PG parameter estimator is designed to estimate PG parameters, and the ith vehicle estimator algorithm is as follows:
wherein ,
、/>
the estimated values of PG parameters at the k moment and the k-1 moment of the ith vehicle are respectively shown;
wherein ,
、/>
output estimation values representing the times k+1 and k of the vehicle,/>
Representing observer gain; to optimize tracking control performance, an optimal performance function for the vehicle queuing system is defined as follows:
wherein ,
representing an optimal performance function;
according to the extremum optimizing condition, the controller algorithm with the self-adaptive structure is obtained as follows:
wherein 
Is a step size coefficient.
6. The method for fault-tolerant tracking control of a data-driven vehicle queue of claim 5,
in the step 5, a radial basis function neural network is utilized to approximate the sensor faults of the vehicle queue system;
error of fault approximation
Is defined as:
wherein ,
and />
The actual fault function and the approximate fault function of the sensor are respectively;
then, a neural network approximation error function is defined
Is prepared from (1)>
Basis functions in hidden layers of neural networks
The method comprises the following steps:
wherein ,
indicates the number of nodes, ++>
Is an output function of the neural network; />
Respectively representing functions in the hidden layers, wherein the equation is shown as a formula (20); selecting the Gaussian function to be +.>
Expressed as:
wherein ,
，/>
is->
Center of individual neurons, < >>
Is the width of the basis function;
based on formulas (17) to (20), the approximation function given for the sensor fault is as follows:
wherein 
Is with threshold +.>
Is used for the output layer weight factor of (a),
respectively represent 1 st, … th,/->
Output layer weighting factors of individual neurons, and +.>
The update rule of (2) is as follows:
the convergence of the radial basis function algorithm is demonstrated by means of the Lyapunov function, and the following is obtained:
wherein ,
is a constant;
based on radial basis function neural network, using
Approximation sensor failure; the output with failure is expressed as:
wherein ,
representing an output of the ith vehicle with the fault;
then, the following fault-tolerant model-free adaptive queue controllers are obtained according to the formulas (13) to (16) and (24):
wherein ,
indicating that the leader vehicle has a faulty output.
7. The method for fault-tolerant tracking control of a data-driven vehicle queue of claim 6,
in the step 6, the attack compensation mechanism is as follows:
wherein ,
an attack index function is used for indicating whether an attack occurs;
the aggregate of all attack periods, i.e. in the interval
Middle->
Is +.>
Is>
Expressed as:
wherein N represents a non-negative integer; obviously, when DoS attacks do not occur, the set of all time intervals is as follows:
wherein ,
is an initial attack parameter for handling the situation where the vehicle queuing system is initially attacked,
is an attack duration coefficient to be determined; therefore, an attack compensation mechanism is proposed as follows:
wherein ,
representing the output after compensation of the i-th vehicle with fault attack at time k,/>
Representing the output after attack compensation at the time of the ith vehicle k-1 with the fault;
8. The method for fault-tolerant tracking control of a data-driven vehicle queue of claim 7,
in the step 7, the design of the elastic fault-tolerant model-free self-adaptive queue safety controller is as follows:
wherein ,
is a normal number,/->
Representing the estimated initial value of the pseudo-bias guide of the ith vehicle.
9. A data-driven vehicle queue fault-tolerant tracking control tracking system, comprising:
the system comprises a dynamics equation construction module, a dynamics equation generation module and a dynamics equation generation module, wherein the dynamics equation construction module is used for establishing dynamics equations of a leader vehicle and a follower vehicle in a vehicle queue system;
the discrete system equation construction module is used for obtaining discrete system equations of the leader vehicle and the follower vehicle;
the system output reconstruction module is used for reconstructing the output of the vehicle queue system by using a partial format dynamic linearization model according to variable PG parameters for introducing auxiliary error analysis in the discrete system equations of the leader vehicle and the follower vehicle;
the system model parameter solving module is used for designing a PG parameter estimator for estimating PG parameters in the partial format dynamic linearization model, and simultaneously introducing an observer and a controller algorithm to control the vehicle queue system;
the fault-tolerant model-free self-adaptive queue control module is used for approximating the sensor faults according to the radial basis function neural network, re-representing the output with the faults by using an approximate fault function, and obtaining a fault-tolerant model-free self-adaptive queue controller;
the attack compensation mechanism construction module is used for providing an attack index function according to the aperiodic DoS attack in the network channel, and is used for indicating whether the attack occurs or not and providing a corresponding attack compensation mechanism;
and the synchronous tracking control module is used for constructing an elastic fault-tolerant model-free self-adaptive queue safety controller and realizing synchronous tracking control of the position and the speed of a leader vehicle by a follower vehicle in the vehicle queue system.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310530233.1A CN116243610B (en) | 2023-05-12 | 2023-05-12 | Data-driven vehicle queue fault-tolerant tracking control tracking method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310530233.1A CN116243610B (en) | 2023-05-12 | 2023-05-12 | Data-driven vehicle queue fault-tolerant tracking control tracking method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116243610A true CN116243610A (en) | 2023-06-09 |
| CN116243610B CN116243610B (en) | 2023-08-01 |
Family
ID=86631716
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310530233.1A Active CN116243610B (en) | 2023-05-12 | 2023-05-12 | Data-driven vehicle queue fault-tolerant tracking control tracking method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116243610B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116488167A (en) * | 2023-06-20 | 2023-07-25 | 青岛大学 | Attack compensation data driving voltage safety control method, system, equipment and medium |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105242544A (en) * | 2015-10-30 | 2016-01-13 | 山东科技大学 | Non-linear multi-unmanned-aerial-vehicle-system fault-tolerance formation control method with consideration of random disturbance |
| CN109373999A (en) * | 2018-10-23 | 2019-02-22 | 哈尔滨工程大学 | Combinated navigation method based on failure tolerant Kalman filtering |
| CN110244747A (en) * | 2019-08-02 | 2019-09-17 | 大连海事大学 | Heterogeneous fleet fault-tolerant control method based on actuator fault and saturation |
| CN114428460A (en) * | 2022-01-28 | 2022-05-03 | 青岛科技大学 | Data-driven dynamic internal model control technology |
| CN114594784A (en) * | 2022-01-13 | 2022-06-07 | 南京航空航天大学 | Neural network-based finite time fault-tolerant control method for formation of fixed-wing unmanned aerial vehicles |
| CN114924488A (en) * | 2022-06-07 | 2022-08-19 | 清华大学 | Distributed finite time self-adaptive fault-tolerant control method and system based on vehicle queue |
| US20220276620A1 (en) * | 2020-08-25 | 2022-09-01 | Dalian University Of Technology | Improved model-free adaptive control method |
| CN115022031A (en) * | 2022-05-31 | 2022-09-06 | 重庆邮电大学 | Security consistency control method for solving influence of FDI attack on multi-agent system |
| CN116047496A (en) * | 2023-01-10 | 2023-05-02 | 浙江省计量科学研究院 | Road target tracking method based on vehicle-mounted Lei Shidong state weighting fusion structure module |
-
2023
- 2023-05-12 CN CN202310530233.1A patent/CN116243610B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105242544A (en) * | 2015-10-30 | 2016-01-13 | 山东科技大学 | Non-linear multi-unmanned-aerial-vehicle-system fault-tolerance formation control method with consideration of random disturbance |
| CN109373999A (en) * | 2018-10-23 | 2019-02-22 | 哈尔滨工程大学 | Combinated navigation method based on failure tolerant Kalman filtering |
| CN110244747A (en) * | 2019-08-02 | 2019-09-17 | 大连海事大学 | Heterogeneous fleet fault-tolerant control method based on actuator fault and saturation |
| US20220276620A1 (en) * | 2020-08-25 | 2022-09-01 | Dalian University Of Technology | Improved model-free adaptive control method |
| CN114594784A (en) * | 2022-01-13 | 2022-06-07 | 南京航空航天大学 | Neural network-based finite time fault-tolerant control method for formation of fixed-wing unmanned aerial vehicles |
| CN114428460A (en) * | 2022-01-28 | 2022-05-03 | 青岛科技大学 | Data-driven dynamic internal model control technology |
| CN115022031A (en) * | 2022-05-31 | 2022-09-06 | 重庆邮电大学 | Security consistency control method for solving influence of FDI attack on multi-agent system |
| CN114924488A (en) * | 2022-06-07 | 2022-08-19 | 清华大学 | Distributed finite time self-adaptive fault-tolerant control method and system based on vehicle queue |
| CN116047496A (en) * | 2023-01-10 | 2023-05-02 | 浙江省计量科学研究院 | Road target tracking method based on vehicle-mounted Lei Shidong state weighting fusion structure module |
Non-Patent Citations (3)
| Title |
|---|
| LIN ZHU等: "Tracking Control for Constrained Nonlinear Systems", 《2022 IEEE 11TH DATA DRIVEN CONTROL AND LEARNING SYSTEMS CONFERENCE》 * |
| ZHU LIN: "Tracking Control for Constrained Nonlinear Systems", 《2022 IEEE 11TH DATA DRIVEN CONTROL AND LEARNING SYSTEMS CONFERENCE (DDCLS)》 * |
| 余世明;吴赛男;何德峰;: "传感器量程和行驶约束车辆队列模型预测控制", 信息与控制, no. 01 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116488167A (en) * | 2023-06-20 | 2023-07-25 | 青岛大学 | Attack compensation data driving voltage safety control method, system, equipment and medium |
| CN116488167B (en) * | 2023-06-20 | 2023-09-05 | 青岛大学 | Attack compensation data driving voltage safety control method, system, equipment and medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116243610B (en) | 2023-08-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Shi et al. | Fault diagnosis of an autonomous vehicle with an improved SVM algorithm subject to unbalanced datasets | |
| Alcala et al. | Gain‐scheduling LPV control for autonomous vehicles including friction force estimation and compensation mechanism | |
| WO2021212728A1 (en) | Unmanned vehicle lane changing decision-making method and system based on adversarial imitation learning | |
| Li et al. | Velocity-based robust fault tolerant automatic steering control of autonomous ground vehicles via adaptive event triggered network communication | |
| CN116243610B (en) | Data-driven vehicle queue fault-tolerant tracking control tracking method and system | |
| Chen et al. | Resilient filter of nonlinear network systems with dynamic event-triggered mechanism and hybrid cyber attack | |
| CN111159642B (en) | Online track prediction method based on particle filtering | |
| US12017690B1 (en) | Data-driven integral sliding mode control method, system and device for high-speed electric multiple unit | |
| CN113841157A (en) | Training a safer neural network by using local linearity regularization | |
| Zhao et al. | Distributed and self-adaptive vehicle speed estimation in the composite braking case for four-wheel drive hybrid electric car | |
| Xu et al. | Estimation of road friction coefficient and vehicle states by 3‐DOF dynamic model and HSRI model based on information fusion | |
| Wang et al. | Resilient path‐following control of autonomous vehicles subject to intermittent denial‐of‐service attacks | |
| Li et al. | Super‐twisting sliding mode trajectory tracking adaptive control of wheeled mobile robots with disturbance observer | |
| Guo et al. | Enhanced cyber-physical security of steering stability control system for four-wheel independent drive electric vehicles | |
| Jin et al. | High‐Speed Train Emergency Brake Modeling and Online Identification of Time‐Varying Parameters | |
| Guo et al. | Adaptive fuzzy sliding mode control for high‐speed train using multi‐body dynamics model | |
| Zhang et al. | Path tracking control for autonomous vehicles with saturated input: A fuzzy fixed‐time learning control approach | |
| CN117719519A (en) | Vehicle running state estimation method | |
| Guo et al. | Adaptive dynamic surface longitudinal tracking control of autonomous vehicles | |
| CN116736842A (en) | H infinity and passivity control method for vehicle queue under half Markov switching topology | |
| CN113848876B (en) | Low-communication and low-calculation-consumption fault detection method for multi-AGV distance cruise system | |
| CN116039715A (en) | Train virtual marshalling operation control method and device | |
| CN112783133B (en) | AGV running state prediction method | |
| Kang et al. | Attack detection and mitigation for sensor and can bus attacks in vehicle anti-lock braking systems | |
| Jia et al. | Nonlinear hybrid multipoint model of high‐speed train with traction/braking dynamic and speed estimation law |
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 |