CN109733396B - Overdrive unmanned automobile input saturation self-adaptive hierarchical control system and method - Google Patents
Overdrive unmanned automobile input saturation self-adaptive hierarchical control system and method Download PDFInfo
- Publication number
- CN109733396B CN109733396B CN201811635903.1A CN201811635903A CN109733396B CN 109733396 B CN109733396 B CN 109733396B CN 201811635903 A CN201811635903 A CN 201811635903A CN 109733396 B CN109733396 B CN 109733396B
- Authority
- CN
- China
- Prior art keywords
- unmanned vehicle
- overdrive
- input
- sliding mode
- module
- 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.)
- Active
Links
Images
Landscapes
- Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)
- Feedback Control In General (AREA)
Abstract
An input saturation self-adaptive hierarchical control system and method for an overdrive unmanned vehicle. The control system is provided with a sensing module, an input saturation approximation module, a self-adaptive terminal neural sliding mode upper-level control module and a lower-level control distribution module; the superior control module comprises a parameter rhythm regulation module, a neural network estimator and a self-adaptive terminal sliding mode controller. The control method comprises the following steps: collecting driving surrounding environment information and vehicle state information, and establishing a nonlinear dynamic model for describing the overdriven unmanned vehicle with saturated input and uncertain parameter characteristics; designing an upper-level control module of an adaptive terminal neural sliding mode of the overdrive unmanned vehicle for overcoming nonlinearity and parameter uncertainty, and dynamically planning generalized force/moment required by the motion of the overdrive unmanned vehicle; and designing an under-level control distributor of the overdrive unmanned vehicle based on tire load rate optimization, and dynamically planning the optimal tire force of each actuating mechanism according to the expected generalized force/moment given by an upper-level controller.
Description
Technical Field
The invention relates to intelligent safety and automatic driving of an automobile, in particular to an input saturation self-adaptive hierarchical control system and method for solving the problem of overdrive unmanned automobile with input saturation limitation.
Background
The unmanned automobile can improve the safety of road traffic, improve the utilization rate of a road network, enhance the road traffic capacity, reduce the fuel consumption of vehicles, realize environmental protection and energy conservation, and is a hot spot of research of governments and scientific research institutions of all countries in the world.
Motion control is one of the core technologies for autonomous driving of an unmanned vehicle, and its task is to cause the unmanned vehicle to travel along a desired route based on a certain control strategy according to the own state and positional deviation information of the vehicle. Because the unmanned vehicle has the characteristics of high nonlinear dynamic characteristics, uncertainty of parameters and the like, how to design the motion control system has profound research significance.
In recent years, the problem of lane keeping control of an unmanned vehicle has received much attention. Document 1(b.mohamed, b.abedlkrim.design of an intelligent optimal neural network-based tracking controller for non-holomatic mobile systems, neuro-computing, 226(2017)46-57) designs a lateral motion controller based on an intelligent optimization neural network for an incomplete constraint unmanned vehicle, and proves the stability of a tracking closed-loop system by using an optimization control theory and lyapunov stability analysis. An active disturbance rejection Control system is proposed in document 2 (Riccamo M, Stefano S, Mariana N.active disturbance rejection Control to automatic steering for lane keeping in automatic vehicles, Control engineering practice,74(2018)13-21), and the robust characteristics of the Control system to speed change and uncertainty of vehicle physical parameters are verified through experiments.
However, due to the physical limitations of the actuators themselves of an overdriven drone vehicle, a saturated control input may produce poor control effects that undermine the stability of the overall closed loop path control system.
Disclosure of Invention
The invention aims to solve the problems in the prior art and provide an input saturation adaptive hierarchical control system of an overdrive unmanned vehicle, which can effectively solve the motion control problem of the unmanned vehicle under the saturation input limit and has the input saturation limit and parameter uncertainty.
Another object of the present invention is to provide a method for input saturation adaptive hierarchical control of an overdriven unmanned vehicle.
The input saturation self-adaptive hierarchical control system of the overdrive unmanned vehicle is provided with a sensing module, an input saturation approximation module, a self-adaptive terminal neural sliding mode upper-level control module and a lower-level control distribution module; the self-adaptive terminal neural sliding mode upper-level control module comprises a parameter modulation rhythm, a neural network estimator and a self-adaptive terminal sliding mode controller; the lower control distribution module comprises a control distributor based on tire load rate optimization; the output end of the sensing module is connected with the input end of the saturated input approximation module, the output end of the saturated input approximation module is connected with the parameter regulation law of the upper-level control module of the adaptive terminal neural sliding mode, the output end of the parameter regulation law is connected with the input end of the neural network estimator, the output end of the neural network estimator is connected with the input end of the adaptive terminal sliding mode controller, the output end of the adaptive terminal sliding mode controller is connected with the input end of the control distributor based on tire load rate optimization of the lower-level control distribution module, and the output end of the control distributor based on tire load rate optimization is connected with the overdrive unmanned automobile.
The method for the input saturation adaptive hierarchical control of the overdrive unmanned automobile comprises the following steps:
1) collecting running ambient environment information and vehicle state information, and establishing a nonlinear dynamic model for describing the overdriven unmanned vehicle with saturated input and uncertain parameter characteristics;
in step 1), the specific method for acquiring the driving ambient environment information and the vehicle state information and establishing the overdrive unmanned vehicle nonlinear dynamics model with the saturated input and the parameter uncertain characteristic may be:
(1) measuring the transverse deviation, the angle deviation and the longitudinal deviation of the unmanned automobile relative to an expected position by adopting a laser radar and a CCD camera in a sensing module;
(2) establishing an unmanned vehicle kinematic model taking the transverse deviation, the angle deviation and the longitudinal deviation of the overdriven unmanned vehicle relative to an expected position as state variables;
(3) establishing a dynamics model of the overdriven unmanned vehicle by taking the longitudinal speed, the transverse speed and the yaw velocity of the overdriven unmanned vehicle as state quantities and the generalized force/moment of the vehicle as input quantities;
(4) an overdrive unmanned vehicle input saturation approximation module based on a hyperbolic tangent function is designed, and the problem of overdrive unmanned vehicle saturation input is effectively solved.
2) Designing an upper-level control module of an adaptive terminal neural sliding mode of the overdrive unmanned vehicle, which effectively overcomes nonlinearity and parameter uncertainty, and dynamically planning generalized force/torque required by the motion of the overdrive unmanned vehicle;
in step 2), the overdrive unmanned vehicle adaptive terminal neural sliding mode superior control module for effectively overcoming nonlinearity and parameter uncertainty is designed, and a specific method for dynamically planning the generalized force/moment required by the overdrive unmanned vehicle motion can be as follows:
(1) constructing a nonsingular terminal sliding mode switching surface by taking the adjustment of transverse deviation, longitudinal deviation and angle deviation minimization as a control target;
(2) an adaptive terminal sliding mode controller of the overdrive unmanned vehicle is designed to ensure the finite time convergence of the system;
(3) and establishing a neural network estimator, carrying out online estimation on the parameter uncertain items of the unmanned vehicle by adopting the neural network, and designing the parameter modulation rhythm of the neural network estimator.
3) And designing an under-level control distributor of the overdrive unmanned vehicle based on tire load rate optimization, and dynamically planning the optimal tire force of each actuating mechanism according to the expected generalized force/moment given by an upper-level controller.
In step 3), the specific method for designing the overdrive unmanned vehicle subordinate control distributor based on tire load rate optimization to dynamically plan the optimal tire force of each actuator according to the expected generalized force/moment given by the superior controller may be as follows:
(1) the tire force optimization distribution target is formed by minimizing the wheel load rate variance and the mean value in a weighted mode, and an optimization target function is established;
(2) and dynamically planning the tire force of each actuating mechanism of the overdriven unmanned automobile in real time by adopting a quasi-Newton method.
Compared with the prior art, the invention has the following outstanding technical effects and benefits:
the invention establishes a nonlinear dynamics model for describing an overdrive unmanned vehicle with saturated input and uncertain parameter characteristics, and provides a saturated input approximation model based on a hyperbolic tangent function. The adaptive terminal neural sliding mode upper-level motion control module is established, the generalized force and the generalized moment required by the overdrive unmanned vehicle are solved in real time and dynamically, the lower-level control distributor of the overdrive unmanned vehicle based on tire load rate optimization is provided, the dynamic distribution of tire force of each actuating mechanism of the overdrive unmanned vehicle is realized, and therefore the motion control of the overdrive unmanned vehicle under the limitation of saturated input is achieved. The invention effectively solves the problem of motion control of the unmanned automobile under the limitation of uncertainty and saturation input, and obviously improves the performance of the motion control system of the unmanned automobile.
Drawings
Fig. 1 is a block diagram illustrating the structural components of an embodiment of an input saturation adaptive hierarchical control system for an overdriven unmanned vehicle according to the present invention.
Fig. 2 is a kinematic model diagram of an overdriven unmanned vehicle according to an embodiment of the present invention.
Fig. 3 is a dynamic model diagram of an overdriven unmanned vehicle according to an embodiment of the present invention.
Fig. 4 is a diagram of an input saturation approximation model of an overdrive unmanned vehicle according to an embodiment of the invention.
Detailed Description
The following examples will further illustrate the present invention with reference to the accompanying drawings.
As shown in fig. 1, the input saturation adaptive hierarchical control system of the overdrive unmanned vehicle is provided with a sensing module 1, an input saturation approximation module 2, an adaptive terminal neural sliding mode upper control module 3 and a lower control distribution module 4; the adaptive terminal neural sliding mode upper control module 3 comprises a parameter modulation rhythm 31, a neural network estimator 32 and an adaptive terminal sliding mode controller 33; the lower control distribution module 4 comprises a control distributor 41 based on tire load rate optimization; the output end of the sensing module 1 is connected with the input end of the saturated input approximation module 2, the output end of the saturated input approximation module 2 is connected with the parameter regulation law 31 of the adaptive terminal neural sliding mode upper-level control module 3, the output end of the parameter regulation law 31 is connected with the input end of the neural network estimator 32, the output end of the neural network estimator 32 is connected with the input end of the adaptive terminal sliding mode controller 33, the output end of the adaptive terminal sliding mode controller 33 is connected with the input end of the control distributor 41 based on tire load rate optimization of the lower-level control distribution module 4, and the output end of the control distributor 41 based on tire load rate optimization is connected with the overdrive unmanned automobile A.
According to the invention, firstly, the running state information of the unmanned vehicle is collected through a CCD camera and a laser radar, a saturation input approximation model is established, secondly, an upper-level adaptive neural terminal sliding mode controller of the unmanned vehicle is deduced based on intelligent control and nonlinear control theories, and then, the tire force of each actuating mechanism is solved in real time through a lower-level control distributor based on tire load rate optimization, so that the input saturation motion control of the unmanned vehicle is realized. Referring to fig. 2-4, the method for the input saturation adaptive hierarchical control of the overdrive unmanned vehicle comprises the following specific steps:
step 1: and establishing an overdrive unmanned lane keeping nonlinear dynamic model with saturated input and uncertain parameter characteristics based on the fusion information of the laser radar and the CCD camera. The process comprises the following substeps:
step 1.1, measuring the transverse deviation y of the unmanned vehicle relative to the expected position by adopting a laser radar and a CCD cameraeAngle deviation thetaeAnd a longitudinal deviation xe。
Step 1.2, establishing a lateral deviation y of the vehicle from the desired positioneAngle deviation thetaeAnd a longitudinal deviation xeThe unmanned vehicle kinematics model, which is a state variable, is as follows:
wherein v isxAnd vyRepresenting the longitudinal and lateral speed of the vehicle, ω representing the yaw rate of the vehicle, vdAnd ωdRepresenting a desired longitudinal velocity and a desired yaw rate, x, of the vehiclecAnd ycRespectively represents the longitudinal coordinate and the transverse coordinate, x, of the vehicle in an inertial coordinate systemdAnd ydRespectively representing the longitudinal coordinate and the transverse coordinate of the desired preview point in the inertial coordinate system. ThetacIndicating that the vehicle is coastingHeading angle, θ, under the scaledRepresenting a desired heading angle of the vehicle.
Step 1.3, solving the state quantity x of the formula (1)e，yeAnd thetaeThe second derivative of the sum, expressed as follows:
step 1.4, longitudinal speed v of overdriven unmanned vehiclexTransverse velocity vyThe sum yaw rate omega is a state quantity and takes the longitudinal generalized force FxTransverse generalized force FyAnd a generalized moment MzFor the input quantities, an overdrive driverless vehicle dynamics model is established as follows:
wherein, IzRepresenting the moment of inertia of the vehicle, m representing the mass of the vehicle, cwxAnd cwyExpressing longitudinal and lateral air resistance coefficients, Λ1,Λ2And Λ3Representing unmodeled dynamics and external disturbances. Longitudinal generalized force FxTransverse generalized force FyAnd a generalized moment MzMay be the following expression:
wherein lfAnd lrRepresenting the distance from the front and rear axles to the center of mass of the vehicle, dwIndicates the track width, FxiAnd FyiRepresenting the longitudinal and lateral forces of each tire.
Step 1.5, substituting the formula (3) into the formula (2) to obtain the following expression:
step 1.6, let Y1＝X1＝[xe,ye,θe]TAnd
equation (5) can be written as the following expression:
wherein X ═ X1X2]；u＝[FxFyMz]TRepresented as a control input; d (t) ═ d1(t) d3(t) d3(t)]TAn external perturbation represented as bounded;
is the upper bound of d (t), Δ f (x) and Δ g (x) are expressed as uncertainty: f (X) and g (X) may be the following expressions:
and 2, designing an overdrive unmanned vehicle saturated input approximation module based on the hyperbolic tangent function, and effectively processing the problem of overdrive unmanned vehicle saturated input. The process comprises the following substeps:
step 2.1, writing the control input to a saturated input form, considering that u (t) is a control input limited by a nonlinear saturation constraint, as follows:
u(t)＝[sat(v1) sat(v2) sat(v3)]T(9)
wherein:
wherein v isi(t) denotes the control input to be designed, uiMIs a control input ui(t) boundary.
Step 2.2, the saturation input function (9) is approximately represented by a smooth hyperbolic tangent function, as follows:
step 2.3, let sat (v) in formula (10)i(t)) is rewritten as the following expression:
sat(vi(t))＝κ(vi)+Δ(vi) (12)
wherein, Δ (v)i)＝sat(vi(t))-κ(vi) Is a bounded function, namely:
step 2.4, the following expression can be obtained by the mean theorem:
wherein the content of the first and second substances,
wherein λ isiExpressed as one satisfying the inequality 0 < lambdaiA constant of < 1.
Step 2.5, order
Then the nonlinear system equation (6) containing the uncertainty term can be written as:
wherein the content of the first and second substances,
p(X,v)＝[p1(X,v) p2(X,v) p3(X,v)]T
where p (X, v) represents an uncertainty term, which can be expressed as:
p(X,v)＝Δf(X)+g(X)v+Δg(X)sat(v) (18)
and step 3: a self-adaptive neural sliding mode terminal superior controller which effectively overcomes nonlinearity and parameter uncertainty is designed, a neural network estimator is established to estimate the uncertainty of the unmanned vehicle on line, and the generalized force/moment required by the lane keeping of the unmanned vehicle is solved.
Step 3.1, using the longitudinal distance deviation x of the unmanned vehicleeAnd the minimization is a control target, and an adaptive neural terminal sliding mode control law for adjusting the longitudinal distance deviation is designed.
Step 3.1.1, defining the 1 st nonsingular terminal sliding mode switching surface s1As follows:
wherein, a1∈R+,β1∈R+,q1∈N+,p1∈N+,q1And p1Is to satisfy inequality p1＞q1And is an odd integer. To terminal sliding form surface s1Taking the derivative, the following expression is obtained:
step 3.1.2, designing a sliding mode index approach law to realize finite time convergence, as follows:
wherein k is1,r1For control coefficients larger than zero, sgn represents the switching function.
Step 3.1.3, combining expressions (20) and (21) to obtain the control input v required for adjusting the longitudinal deviation1The expression of (c) is as follows:
step 3.1.4, establishing a neural network estimator and adopting a neural network model
Uncertainty p for unmanned vehicle1(X, v) on-line estimation, neural network model
The following expression is designed:
wherein the content of the first and second substances,
representing estimates of neural network weights, ξ1Is a smooth basis function vector.
Step 3.1.5, establish estimate
And the actual value p1The error expression between is as follows:
wherein the content of the first and second substances,
representing the ideal neural network weight, εξ1Representing the net approximation error.
Step 3.1.6, design weights
The adaptive control law of (2) is as follows:
where ρ is1Representing real numbers greater than zero.
Step 3.2, transverse distance deviation y of the unmanned vehicle is overdriveneAnd the minimization is a control target, and an adaptive neural terminal sliding mode upper-level control law for adjusting the transverse distance deviation is designed.
Step 3.2.1, defining the 2 nd nonsingular terminal sliding mode switching surface s2As follows:
wherein, a2∈R+,β2∈R+,q2∈N+,p2∈N+，q2And p2Is to satisfy inequality p2＞q2And is an odd integer. To terminal sliding form surface s2Taking the derivative, the following expression is obtained:
step 3.2.2, designing an index approach law to realize finite time convergence, as follows:
wherein k is2,r2For control coefficients larger than zero, sgn represents the switching function.
Step (ii) of3.2.3, combining expressions (27) and (28) to obtain the control input v required to adjust the lateral deviation2The expression of (c) is as follows:
step 3.2.4, establishing a neural network estimator and adopting a neural network model
Uncertainty p for unmanned vehicle2(X, v) on-line estimation, neural network model
The following expression is designed:
wherein the content of the first and second substances,
representing estimates of neural network weights, ξ2Is a smooth basis function vector.
Step 3.2.5, establish an estimate
And the actual value p2The error expression between is as follows:
wherein
Representing the ideal neural network weight, εξ2Representing the net approximation error.
Step 3.2.6, designing neural network weight
The adaptive control law of (2) is as follows:
where ρ is2Representing real numbers greater than zero.
Step 3.3, the angle deviation theta of the unmanned vehicle is overdriveneMinimization as a control target, designed to accommodate angular deviation θeThe self-adaptive neural terminal sliding mode superior control law.
Step 3.3.1, defining the 3 rd nonsingular terminal sliding mode switching surface s3As follows:
wherein, a3∈R+,β3∈R+,q3∈N+And p3∈N+；q3And p3Is to satisfy inequality p3＞q3And is an odd integer. To terminal sliding form surface s3Taking the derivative, the following expression is obtained:
step 3.3.2, designing an exponential approximation law to realize finite time convergence, as follows:
wherein k is3,r3For control coefficients larger than zero, sgn represents the switching function.
Step 3.3.3, combining the expressions (34) and (35) to obtain the control input v required for adjusting the lateral deviation3The expression (c) of (a),as follows:
step 3.3.4, establishing a neural network estimator and adopting a neural network model
Uncertainty p for unmanned vehicle3(X, v) on-line estimation, neural network model
The following expression is designed:
wherein the content of the first and second substances,
representing estimates of neural network weights, ξ3Is a smooth basis function vector.
Step 3.3.5, establish estimate
And the actual value p2The error expression between is as follows:
wherein the content of the first and second substances,
representing the ideal neural network weight, εξ3Representing the net approximation error.
Step 3.3.6, designing neural network weights
The adaptive control law of (2) is as follows:
where ρ is3Representing real numbers greater than zero.
Step 3.4, in order to attenuate the dithering phenomenon caused by the sign function in equations (22), (29) and (36), a function sat(s) is introducedi/Δi) (i-1, 2,3) instead of the sign function sgn(s) in equations (22), (29) and (36)i)(i＝1,2,3)。
And 4, designing a lower-level control distributor of the overdrive unmanned vehicle based on tire load rate optimization, and dynamically planning the optimal longitudinal force and the optimal transverse force of each tire of the overdrive unmanned vehicle according to the expected generalized force/moment.
Step 4.1, the tire force optimization distribution target is formed by minimizing the wheel load rate variance and the mean value in a weighting mode, and an optimization objective function is established as follows:
wherein epsilonvLoad rate variance and mean weight coefficient. Var (. cndot.) denotes the deviation, E (. cndot.) denotes the mean, χiThe tire load factor is expressed as follows:
wherein, FxiAnd FyiRepresents the longitudinal and transverse forces, mu, of the tireiRepresenting the road adhesion coefficient.
Step 4.2, establishing an equality constraint condition of the optimal distribution of the tire force of the unmanned vehicle, as follows:
in the step 4.3, the step of the method,substituting the equation constraint condition formula (42) into the optimization objective function formula (40) to establish a model containing 8 variables (F)x1,Fx2,Fx3,Fx4,Fy1,Fy2,Fy3,Fy4) To the unconstrained optimization problem of (1). The method is solved by adopting a quasi-Newton method, and an iterative formula is designed as follows:
wherein:
θ＝[Fx1Fx2Fx3Fx4Fy1Fy2Fy3Fy4]T(44)
where k is the iteration step.
The method solves the difficult problems in the prior art, firstly establishes an overdrive unmanned automobile nonlinear dynamics model with saturated input and uncertain parameter characteristics, secondly designs a saturation input approximation module based on a hyperbolic tangent function, then designs an adaptive neural terminal sliding mode upper-level control module to dynamically plan expected generalized moment and generalized force in real time, and distributes the expected generalized moment and the generalized force to each actuating mechanism through a lower-level control distributor, thereby realizing the motion control of the overdrive unmanned automobile under the limitation of uncertainty and input saturation.
Claims (5)
1. The input saturation self-adaptive hierarchical control system of the overdrive unmanned vehicle is characterized by being provided with a sensing module, an input saturation approximation module, a self-adaptive terminal neural sliding mode upper-level control module and a lower-level control distribution module; the self-adaptive terminal neural sliding mode upper-level control module comprises a parameter modulation rhythm, a neural network estimator and a self-adaptive terminal sliding mode controller; the lower control distribution module comprises a control distributor based on tire load rate optimization; the output end of the sensing module is connected with the input end of the saturated input approximation module, the output end of the saturated input approximation module is connected with the parameter regulation law of the upper-level control module of the adaptive terminal neural sliding mode, the output end of the parameter regulation law is connected with the input end of the neural network estimator, the output end of the neural network estimator is connected with the input end of the adaptive terminal sliding mode controller, the output end of the adaptive terminal sliding mode controller is connected with the input end of the control distributor based on tire load rate optimization of the lower-level control distribution module, and the output end of the control distributor based on tire load rate optimization is connected with the overdrive unmanned automobile.
2. The method for the input saturation adaptive hierarchical control of the overdrive unmanned automobile is characterized by comprising the following steps of:
1) collecting running ambient environment information and vehicle state information, and establishing a nonlinear dynamic model for describing the overdriven unmanned vehicle with saturated input and uncertain parameter characteristics;
2) designing an upper-level control module of an adaptive terminal neural sliding mode of the overdrive unmanned vehicle, which effectively overcomes nonlinearity and parameter uncertainty, and dynamically planning generalized force/torque required by the motion of the overdrive unmanned vehicle;
3) and designing an under-level control distributor of the overdrive unmanned vehicle based on tire load rate optimization, and dynamically planning the optimal tire force of each actuating mechanism according to the expected generalized force/moment given by an upper-level controller.
3. The method for the input saturation adaptive hierarchical control of the overdriven unmanned vehicle according to claim 2, wherein in the step 1), the method for collecting the driving ambient information and the vehicle state information and establishing the nonlinear dynamics model of the overdriven unmanned vehicle with the characteristics of saturated input and uncertain parameters comprises the following specific steps:
(1) measuring the transverse deviation, the angle deviation and the longitudinal deviation of the unmanned automobile relative to an expected position by adopting a laser radar and a CCD camera in a sensing module;
(2) establishing an unmanned vehicle kinematic model taking the transverse deviation, the angle deviation and the longitudinal deviation of the overdriven unmanned vehicle relative to an expected position as state variables;
(3) establishing a dynamics model of the overdriven unmanned vehicle by taking the longitudinal speed, the transverse speed and the yaw velocity of the overdriven unmanned vehicle as state quantities and the generalized force/moment of the vehicle as input quantities;
(4) an overdrive unmanned vehicle input saturation approximation module based on a hyperbolic tangent function is designed, and the problem of overdrive unmanned vehicle saturation input is effectively solved.
4. The method for input saturation adaptive hierarchical control of the overdrive unmanned vehicle as claimed in claim 2, wherein in the step 2), the overdrive unmanned vehicle adaptive terminal neural sliding mode superior control module for effectively overcoming nonlinearity and parameter uncertainty is designed, and the specific method for dynamically planning the generalized force/moment required by the motion of the overdrive unmanned vehicle is as follows:
(1) constructing a nonsingular terminal sliding mode switching surface by taking the adjustment of transverse deviation, longitudinal deviation and angle deviation minimization as a control target;
(2) an adaptive terminal sliding mode controller of the overdrive unmanned vehicle is designed to ensure the finite time convergence of the system;
(3) and establishing a neural network estimator, carrying out online estimation on the parameter uncertain items of the unmanned vehicle by adopting the neural network, and designing the parameter modulation rhythm of the neural network estimator.
5. The method for inputting saturation adaptive hierarchical control to the unmanned vehicle according to claim 2, wherein in step 3), the method for designing the lower-level control distributor of the unmanned vehicle based on tire load rate optimization dynamically planning the optimal tire force of each actuator according to the expected generalized force/moment given by the upper-level controller is as follows:
(1) the tire force optimization distribution target is formed by minimizing the wheel load rate variance and the mean value in a weighted mode, and an optimization target function is established;
(2) and dynamically planning the tire force of each actuating mechanism of the overdriven unmanned automobile in real time by adopting a quasi-Newton method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811635903.1A CN109733396B (en) | 2018-12-29 | 2018-12-29 | Overdrive unmanned automobile input saturation self-adaptive hierarchical control system and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811635903.1A CN109733396B (en) | 2018-12-29 | 2018-12-29 | Overdrive unmanned automobile input saturation self-adaptive hierarchical control system and method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109733396A CN109733396A (en) | 2019-05-10 |
| CN109733396B true CN109733396B (en) | 2020-06-23 |
Family
ID=66362400
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811635903.1A Active CN109733396B (en) | 2018-12-29 | 2018-12-29 | Overdrive unmanned automobile input saturation self-adaptive hierarchical control system and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109733396B (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US4222289A (en) * | 1977-04-27 | 1980-09-16 | Toyota Jidosha Kogyo Kabushiki Kaisha | Automatic transmission with overdrive |
| CN106671982A (en) * | 2017-01-09 | 2017-05-17 | 厦门大学 | Multi-intelligent agent based unmanned electric car automatic overtaking system and method |
| CN107139917A (en) * | 2017-04-27 | 2017-09-08 | 江苏大学 | It is a kind of based on mix theory pilotless automobile crosswise joint system and method |
| CN206870883U (en) * | 2017-05-25 | 2018-01-12 | 厦门大学嘉庚学院 | A kind of roadside solid charging pile based on internet |
-
2018
- 2018-12-29 CN CN201811635903.1A patent/CN109733396B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US4222289A (en) * | 1977-04-27 | 1980-09-16 | Toyota Jidosha Kogyo Kabushiki Kaisha | Automatic transmission with overdrive |
| CN106671982A (en) * | 2017-01-09 | 2017-05-17 | 厦门大学 | Multi-intelligent agent based unmanned electric car automatic overtaking system and method |
| CN107139917A (en) * | 2017-04-27 | 2017-09-08 | 江苏大学 | It is a kind of based on mix theory pilotless automobile crosswise joint system and method |
| CN206870883U (en) * | 2017-05-25 | 2018-01-12 | 厦门大学嘉庚学院 | A kind of roadside solid charging pile based on internet |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109733396A (en) | 2019-05-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108227491B (en) | Intelligent vehicle track tracking control method based on sliding mode neural network | |
| Hu et al. | Robust H∞ output-feedback control for path following of autonomous ground vehicles | |
| CN107831761B (en) | Path tracking control method of intelligent vehicle | |
| CN108454623B (en) | A kind of unmanned electric vehicle Trajectory Tracking Control method of four motorized wheels | |
| CN109017778B (en) | Active steering control method for expected path of four-wheel independent drive vehicle | |
| CN109017760B (en) | Vehicle expected trajectory tracking method and device and rolling time domain optimization algorithm | |
| CN110989597B (en) | Adaptive path tracking method of integrated fuzzy neural network | |
| CN107561942A (en) | Intelligent vehicle track following model predictive control method based on model compensation | |
| Boukens et al. | Design of an intelligent optimal neural network-based tracking controller for nonholonomic mobile robot systems | |
| Zha et al. | A survey of intelligent driving vehicle trajectory tracking based on vehicle dynamics | |
| Awad et al. | Model predictive control with fuzzy logic switching for path tracking of autonomous vehicles | |
| CN109017759B (en) | Desired path vehicle yaw control method | |
| Pauca et al. | Predictive control for the lateral and longitudinal dynamics in automated vehicles | |
| Cui et al. | Adaptive tracking control of wheeled mobile robots with unknown longitudinal and lateral slipping parameters | |
| CN112859596B (en) | Nonlinear teleoperation multilateral control method considering formation obstacle avoidance | |
| Hang et al. | Path tracking control of a four-wheel-independent-steering electric vehicle based on model predictive control | |
| Zuo et al. | Adaptive robust control strategy for rhombus-type lunar exploration wheeled mobile robot using wavelet transform and probabilistic neural network | |
| CN109017446B (en) | Expected path vehicle longitudinal speed tracking control method and device | |
| Jiang et al. | Anti-disturbance direct yaw moment control of a four-wheeled autonomous mobile robot | |
| CN109017804A (en) | Torque distribution controller is the method that each hub motor of vehicle distributes driving moment | |
| CN113885548B (en) | Multi-quad-rotor unmanned helicopter output constraint state consistent game controller | |
| CN109017447B (en) | Method for outputting total driving torque of unmanned vehicle | |
| CN107450311A (en) | Inversion model modeling method and device and adaptive inverse control and device | |
| CN109606364A (en) | A kind of variable speed lower leaf formula self study extension neural network lane holding control method of intelligent automobile | |
| CN109733396B (en) | Overdrive unmanned automobile input saturation self-adaptive hierarchical control system and method |
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 |