A limited time domain unmanned vehicle obstacle avoidance control method and device

Abstract

This application proposes a method and device for obstacle avoidance control of an unmanned vehicle in a limited time domain, including: describing the safety of the control system based on forward invariance, and constructing an obstacle function of the control system based on the safety of the control system and the obstacle; integrating the obstacle function into the cost function of the control system, and converting the collision avoidance control problem of the control system into an optimal control problem based on the cost function; establishing a control system model with unknown parameters based on a neural network function approximation method, and using an integral parallel learning method to design a parameter adaptive law to identify the unknown parameters in the model; based on the control system model, using a quadratic optimization framework, and using an adaptive iterative learning method to solve the optimal control input that meets the requirements of the collision avoidance control task. This application adopting the above scheme does not rely on the precise parameter model of the system, takes into account both the optimization performance and safety of the control system, and can effectively solve the collision avoidance control problem in a limited time domain.

Description

Technical Field

[0001] The present application relates to the fields of control theory and artificial intelligence technology, and in particular to a method and device for controlling obstacle avoidance of an unmanned vehicle in a limited time domain. Background Art

[0002] Intelligent driving technology is seen as an effective means of reducing traffic accidents, improving traffic efficiency, and alleviating driver burdens. It is generally categorized into active safety systems, intelligent driver assistance systems, and autonomous driving systems. Active safety systems play a crucial role in intelligent vehicle control by sensing and monitoring the vehicle's internal and external conditions, identifying potential hazards, and implementing various interventions to help avoid or mitigate collisions. Key approaches to addressing collision avoidance control include model predictive control (MPC) and safety reinforcement learning (RSL). The key advantage of MPC is its ability to dynamically adjust control inputs to adapt to environmental changes and complex obstacle avoidance requirements. This approach typically divides the collision avoidance task into path planning and trajectory tracking, but consumes significant online computing resources and can lead to performance conflicts between the hierarchical subtasks. Safety reinforcement learning optimizes control strategies and numerically solves the Hamilton-Jacobi-Bellman equations, enabling the vehicle to minimize cost while strictly adhering to safety constraints. However, this approach often presents challenges when faced with finite-horizon optimal control problems. Summary of the Invention

[0003] The present application aims to solve one of the technical problems in the related art at least to a certain extent.

[0004] To this end, the first purpose of this application is to propose a limited time domain unmanned vehicle obstacle avoidance control method, which does not rely on the precise parameter model of the system, takes into account the optimization performance and safety of the control system, and can effectively solve the collision avoidance control problem in a limited time domain.

[0005] The second purpose of this application is to propose a limited time domain unmanned vehicle obstacle avoidance control device.

[0006] The third object of this application is to provide a computer device.

[0007] To achieve the above-mentioned purpose, the first embodiment of the present application proposes a finite time domain unmanned vehicle obstacle avoidance control method, including: describing the safety of the control system based on forward invariance, and constructing the obstacle function of the control system based on the safety of the control system and according to the obstacle; integrating the obstacle function into the cost function of the control system, and converting the collision avoidance control problem of the control system into an optimal control problem based on the cost function; establishing a control system model with unknown parameters based on a neural network function approximation method, and using an integral parallel learning method to design a parameter adaptive law to identify the unknown parameters in the model; based on the control system model, using a quadratic optimization framework, an adaptive iterative learning method is adopted to solve the optimal control input that meets the requirements of the collision avoidance control task.

[0008] To achieve the above-mentioned purpose, the second embodiment of the present application proposes a finite time domain unmanned vehicle obstacle avoidance control device, including: a first system establishment module, used to describe the safety of the control system based on forward invariance, and based on the safety of the control system and according to the obstacle, construct the obstacle function of the control system; a second system establishment module, used to integrate the obstacle function into the cost function of the control system, and based on the cost function, convert the collision avoidance control problem of the control system into an optimal control problem; a system identification module, used to establish a control system model with unknown parameters based on a neural network function approximation method, and use an integral parallel learning method to design a parameter adaptive law to identify the unknown parameters in the model; an iterative solution module, used to use a quadratic optimization framework based on the control system model, and adopt an adaptive iterative learning method to solve the optimal control input that meets the requirements of the collision avoidance control task.

[0009] To achieve the above-mentioned purpose, the third aspect embodiment of the present application proposes a computer device, including: a memory, a processor, and a computer program stored in the memory and runnable on the processor. When the processor executes the computer program, the above-mentioned limited time domain unmanned vehicle obstacle avoidance control method is implemented.

[0010] The finite-time domain unmanned vehicle obstacle avoidance control method of the embodiment of the present application describes the safety of the control system based on forward invariance. On this basis, a corresponding obstacle function is constructed according to the obstacle; the constructed obstacle function is integrated into the cost function, and the collision avoidance control problem is converted into an optimal control problem; a parameter identification model of the unknown system is established based on the neural network function approximation method, and an adaptive law is designed using the integral parallel learning method to identify the unknown parameters in the model online; based on the identification model and the identification parameters, an adaptive iterative learning algorithm is proposed using a quadratic optimization framework to solve the optimal control input that meets the requirements of the collision avoidance control task. This application does not rely on the precise parameter model of the system, takes into account the optimization performance and safety of the control system, and can effectively solve the collision avoidance control problem in a finite time domain.

[0011] Additional aspects and advantages of the present application will be given in part in the description below, and in part will become apparent from the description below, or will be learned through practice of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] The above and / or additional aspects and advantages of the present application will become apparent and easily understood from the following description of the embodiments in conjunction with the accompanying drawings, in which:

[0013] Figure 1 A flowchart of a method for controlling an obstacle avoidance for an unmanned vehicle within a limited time domain provided in Example 1 of the present application is provided;

[0014] Figure 2 A schematic diagram of a "bicycle" model for vehicle steering control according to an embodiment of the present application;

[0015] Figure 3 This is a driving scene design diagram for an embodiment of the present application;

[0016] Figure 4 This is an iteration diagram of the cost function of an embodiment of the present application;

[0017] Figure 5 It is a vehicle trajectory iteration diagram of an embodiment of the present application;

[0018] Figure 6 A comparison chart of the actual value and estimated value of the vehicle status in an embodiment of the present application;

[0019] Figure 7 A schematic diagram of a vehicle path under initial input in an embodiment of the present application;

[0020] Figure 8 This is a schematic diagram of a vehicle path in the 20th iteration of the embodiment of the present application;

[0021] Figure 9 A schematic diagram of a vehicle path when the algorithm of an embodiment of the present application converges;

[0022] Figure 10 A schematic structural diagram of a limited-time-domain unmanned vehicle obstacle avoidance control device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0023] The following describes in detail embodiments of the present application, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present application, and should not be construed as limiting the present application.

[0024] In real-world situations, accurate vehicle models are often difficult to obtain, and driving environments are complex and ever-changing. Leveraging vehicle trajectory data to avoid the reliance of traditional obstacle avoidance control methods on precise system models while simultaneously integrating planning and control and effectively developing control strategies that meet the requirements of the obstacle avoidance task is an effective way to improve driving safety.

[0025] The following describes the obstacle avoidance control method and device for an unmanned vehicle in a limited time domain according to an embodiment of the present application with reference to the accompanying drawings.

[0026] Figure 1 A flow chart of a limited time domain unmanned vehicle obstacle avoidance control method provided in Example 1 of the present application.

[0027] like Figure 1 As shown, the finite time domain unmanned vehicle obstacle avoidance control method includes the following steps:

[0028] Step 101: describing the safety of the control system based on forward invariance, and constructing an obstacle function of the control system based on the safety of the control system and obstacles;

[0029] In this example, the following control system model is considered:

[0030]

[0031] in, is the system state variable, and its initial value is , is the system control input, is the dynamic equation of the unknown system, assuming and its relative status and input The partial derivative of 、 Satisfying the local Lipschitz continuity condition, a common assumption in control theory, guarantees that the trajectory is understood existence and uniqueness.

[0032] Control Input The resulting set is defined as a convex set of the following form:

[0033]

[0034] Assume that for closed bounded control input , at the initial value Under this condition, the solution trajectory of the control system is bounded, which is also a common assumption in control theory, avoiding the phenomenon of system escape in a finite time. The formula is described as: , , making ;

[0035] The collision avoidance control problem requires that the state of the control system is always within the given safety set. The definition is as follows:

[0036]

[0037]

[0038]

[0039] in, is a continuously differentiable function, the set for The boundary of the set for inside, and there is as well as ;

[0040] If for any initial state , solution trajectory Always keep in the collection Inside, it is called a set Has forward invariance. If the system state is relative to the set If the forward invariance condition is satisfied, the control system is said to be It is safe;

[0041] According to the properties of the security set, the following barrier function can be defined: :

[0042]

[0043] It can be obtained that the barrier function satisfies the following properties:

[0044]

[0045]

[0046] Consider having To ensure that the system can avoid collisions, define a circular area of ​​the following form. As the first Unsafe areas with obstacles:

[0047]

[0048] in, It is The center of the obstacle, is the radius of the corresponding unsafe area. For a circular area, Obstacle function It can be specifically expressed as:

[0049]

[0050] in, ;

[0051] In actual numerical simulation, in order to avoid the infinite situation, a saturated hyperbolic function can be used to limit the barrier function. The barrier function can be re-expressed as:

[0052]

[0053] In this embodiment, a simplified "bicycle" model of vehicle steering control in a plane rectangular coordinate system is taken as an example, and its schematic diagram is shown as follows: Figure 2 As shown, the vehicle dynamics equation described by the model is:

[0054]

[0055] in, and Respectively represent the horizontal and vertical coordinates of the vehicle in the earth coordinate system. and Denote the lateral velocity and longitudinal velocity in the vehicle coordinate system respectively. In the "bicycle" model, it is assumed that is a fixed value. is the yaw angle, is the yaw rate, Indicates the steering wheel rotation angle, and are the front wheel steering stiffness and the rear wheel steering stiffness, is the vehicle mass, and are the distances from the vehicle's center of mass to the front and rear axles, is the moment of inertia, is the steering gear ratio;

[0056] In this embodiment, taking the above-mentioned "bicycle" model as an example, the model in CarSim is selected as the simulation model, the vehicle model is selected as a Class C hatchback, the tire model is selected as 215 / 55 R17, and some vehicle model parameters provided by CarSim are shown in Table 1.

[0057] Table 1

[0058]

[0059] Select the vehicle status as , the control input is , the allowable control input set is set to , which is described in formula as:

[0060]

[0061] In this embodiment, the obstacle avoidance problem focuses more on the relative positions of the vehicle's horizontal and vertical coordinates and the obstacles, so the obstacle function is related to the vehicle's horizontal and vertical coordinates. ,

[0062] A situation with five obstacles is proposed, wherein the specific positions of the obstacles are shown in Table 2 below. Table 2 is:

[0063] Table 2

[0064]

[0065] The danger zone radius of the obstacle is set to , therefore, the barrier function can be described as:

[0066]

[0067] in, .

[0068] Step 102: Integrate the obstacle function into the cost function of the control system, and transform the collision avoidance control problem of the control system into an optimal control problem based on the cost function;

[0069] In this embodiment, in order to indicate the degree of perception of the unmanned vehicle to obstacles, a circular perception area of ​​the following form is defined for each obstacle: :

[0070]

[0071] in, is the outer diameter of the annular area, representing the minimum distance at which the system can begin to perceive obstacles. is the inner diameter of the annular area, representing the maximum distance at which the system can perceive obstacles. Within the annular area, the system's ability to perceive obstacles decreases as the distance from the center of the obstacle increases. The control system's ability to perceive obstacles decreases as the distance from the center of the obstacle increases. It can be formulated as:

[0072]

[0073] in, is the distance from the center of the obstacle;

[0074] In order to transform the collision avoidance control problem into an optimal control problem, we first define the cost function in the following form:

[0075]

[0076] in, is the instantaneous cost, including system performance and barrier function, which can be described as:

[0077]

[0078] Among them, the matrix and matrix is a real symmetric positive definite matrix; in actual numerical simulation, a saturated hyperbolic function is used to restrict the barrier function, and the cost function can be re-expressed as:

[0079]

[0080]

[0081]

[0082] For the purpose of collision avoidance problem, the control goal is to obtain an optimal control input , so that for , the following inequality holds:

[0083]

[0084] According to the Pontryagin minimum principle, the optimal control input The following necessary conditions must be met:

[0085]

[0086] in, is the Hamiltonian function, is the Hamiltonian function for the state The partial derivative of is the co-state quantity.

[0087] In this embodiment, taking the above-mentioned "bicycle" model as an example, the inner and outer diameters of the sensing area are set to , ; Driving scene design such as Figure 3 As shown, the initial position of the vehicle is at the origin of the coordinate system, and the control goal is to make the vehicle track the given trajectory. , while avoiding hitting obstacles, where the position information of the obstacles is the same as Table 2. The target trajectory tracked by the vehicle is a cosine function path in the coordinate axis plane, which is specifically described as:

[0088]

[0089] in, is a constant, the horizontal coordinate of the target trajectory at any time Value It can be solved by the following formula:

[0090]

[0091] in, is an elliptic integral of the second kind, and It can be calculated with the following boundary conditions:

[0092]

[0093] Based on the proposed control purpose, the obstacle function is integrated into the cost function through the perception function, and the finite time domain is selected as , the cost function is formulated as:

[0094]

[0095] Among them, the matrix and matrix is a real symmetric positive definite matrix, choose , ;

[0096] Therefore, the optimal control problem is described as optimizing the control inputs within a given set of permissible control inputs based on a given vehicle control system so that the vehicle minimizes the proposed cost function within a given time domain.

[0097] Step 103: establishing a control system model with unknown parameters based on a neural network function approximation method, and designing a parameter adaptive law using an integral parallel learning method to identify the unknown parameters in the model;

[0098] In this embodiment, the control system with unknown parameters is expressed as:

[0099]

[0100] in, is the unknown coefficient matrix, is the designed activation function, is the approximation error, when the number of neurons When is large enough, the approximate error can converge to a small enough range. Therefore, the unknown control system can be approximately expressed as:

[0101]

[0102] in, is the estimated value of the unknown coefficient matrix;

[0103] The integral parallel learning method uses the numerical integration of historical data over a period of time to identify the unknown parameters in the model online, which gets rid of the need for state derivative data in traditional parallel learning methods. The historical trajectory data within is recorded as , represents a set of W sets of system trajectories and corresponding control inputs at different times. For each set of data, calculate:

[0104]

[0105] in, , define the matrix , when collecting historical trajectory data, the matrix is ​​required satisfy ;

[0106] definition For the The estimated value of the system state during the iteration is designed as follows:

[0107]

[0108] in, is the learning rate, It is The estimation error of the system state during the iteration;

[0109] definition For the The estimated value of the coefficient matrix in the iteration process is designed as follows using the historical data set:

[0110]

[0111] in is the learning rate, It is The identification error of the coefficient matrix during the iteration process.

[0112] In this embodiment, referring to the “bicycle model” of vehicle steering control, a single hidden layer neural network is selected, which is expressed as: ;

[0113] The control inputs of the human driver and the autonomous driving system are randomly given as ,in is a Gaussian distributed random number, randomly selecting the time interval A total of 200 sets of trajectory data are used as historical trajectory data.

[0114] Step 104 : Based on the control system model, using a quadratic optimization framework, and employing an adaptive iterative learning method, an optimal control input that satisfies the requirements of the collision avoidance control task is solved.

[0115] In this embodiment, the specific process of the adaptive iterative algorithm is as follows:

[0116] S1: Initialize algorithm parameters 、 、 And the algorithm convergence accuracy 、 , select the initial control input , by solving the following equations to obtain Estimates of the initial system trajectory and estimates of the initial coefficient matrix:

[0117]

[0118] in, and and are the initial values ​​of the system state estimation error and the identification error of the coefficient matrix, Calculate the initial cost function as the initial value of the system's true state trajectory And order ;

[0119] S2: Solve by calculating the following formula :

[0120]

[0121] in, is the estimated value of the Hamiltonian function, For its state variables The partial derivative of

[0122] S3: Solved through a quadratic optimization framework :

[0123]

[0124] in, is the estimated value of the Hamiltonian function for the variable The partial derivative of ,make ;

[0125] S4: Get online real trajectory data based on new control input , update the estimated value of the system trajectory and the estimated value of the coefficient matrix by solving the following equation:

[0126]

[0127] in, and are the updated system state estimation error and the identification error of the coefficient matrix;

[0128] S5: Calculate the cost function using the updated control input and online trajectory data , determine whether the updated control input reduces the cost function: if , then let , , is the iteration step length, and returns to S3, otherwise, enters S6;

[0129] S6: If , then enter S7, otherwise make , , And return to S2;

[0130] S7: If , stop the iterative process and get the corresponding optimal control input, otherwise let , , , , And return to S2.

[0131] In this embodiment, referring to the vehicle steering control “bicycle model”, the initial state value of the vehicle system is selected as: , the initial control input is selected as: , the initial parameters of the adaptive iterative algorithm are set as: , the initial coefficient matrix is ​​randomly selected as , as shown below:

[0132]

[0133] According to the above adaptive iterative algorithm process, the algorithm converges after 35 iterations. The cost function iteration diagram is as follows: Figure 4 As shown in Figure 2, we can see that after 33 iterations, the cost function has fully decreased, and subsequent iterations cannot further reduce the cost value, indicating that the algorithm has converged to the optimal control input;

[0134] Figure 5 This is a diagram of the vehicle's path during some iterations. It shows how, given the initial control input, the vehicle gradually adjusts its path during the iterations, approaching the target trajectory while avoiding obstacles. It can be seen that when the iterations stop, the vehicle achieves good obstacle avoidance.

[0135] Figure 6This is a comparison chart of the actual value of the vehicle state and its estimated value. It can be observed that the error between the identification model and the actual model in CarSim is very close;

[0136] Figure 7 、 Figure 8 、 Figure 9 The diagrams of the vehicle paths in CarSim under the initial control input, the control input after the 20th iteration, and the converged control input are shown respectively. It can be seen that the initial trajectory is very different from the target trajectory and cannot effectively avoid obstacles. In addition, there is a large error between the estimated trajectory and the true value. After 20 iterations, the vehicle tracking performance and obstacle avoidance effect are significantly improved, and the identification error is also significantly reduced. When the algorithm converges, it can be seen that the identification error has been reduced to a sufficiently small range. At the same time, the vehicle has achieved good results in tracking performance and obstacle avoidance effect in a limited time domain.

[0137] The finite-time domain unmanned vehicle obstacle avoidance control method of the embodiment of the present application describes the safety of the control system based on forward invariance. On this basis, a corresponding obstacle function is constructed according to the obstacle; the constructed obstacle function is integrated into the cost function, and the collision avoidance control problem is converted into an optimal control problem; a parameter identification model of the unknown system is established based on the neural network function approximation method, and an adaptive law is designed using the integral parallel learning method to identify the unknown parameters in the model online; based on the identification model and the identification parameters, an adaptive iterative learning algorithm is proposed using a quadratic optimization framework to solve the optimal control input that meets the requirements of the collision avoidance control task. This application does not rely on the precise parameter model of the system, takes into account the optimization performance and safety of the control system, and can effectively solve the collision avoidance control problem in a finite time domain.

[0138] In order to implement the above embodiments, the present application also proposes a limited time domain unmanned vehicle obstacle avoidance control device.

[0139] Figure 10 A schematic structural diagram of a limited-time-domain unmanned vehicle obstacle avoidance control device provided in an embodiment of the present application.

[0140] like Figure 10 As shown, the limited time domain unmanned vehicle obstacle avoidance control device includes:

[0141] A first system establishment module is used to describe the safety of the control system based on forward invariance, and to construct an obstacle function of the control system based on the safety of the control system and obstacles;

[0142] The second system establishment module is used to integrate the obstacle function into the cost function of the control system and transform the collision avoidance control problem of the control system into an optimal control problem based on the cost function;

[0143] The system identification module is used to establish a control system model with unknown parameters based on the neural network function approximation method, and to design a parameter adaptive law using the integral parallel learning method to identify the unknown parameters in the model;

[0144] The iterative solution module is used to solve the optimal control input that meets the requirements of the collision avoidance control task based on the control system model, using the quadratic optimization framework and the adaptive iterative learning method.

[0145] It should be noted that the above explanation of the embodiment of the obstacle avoidance control method for an unmanned vehicle in a limited time domain is also applicable to the obstacle avoidance control device for an unmanned vehicle in a limited time domain of this embodiment, and will not be repeated here.

[0146] In order to implement the above embodiments, the present application also proposes a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the method described in the above embodiments is implemented.

[0147] In the description of this specification, the reference terms "one embodiment", "some embodiments", "example", "specific example" or "some examples" mean that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic expressions of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art can combine and combine different embodiments or examples described in this specification and features of different embodiments or examples without contradiction.

[0148] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of the technical features being referred to. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of such features. Throughout the description of this application, "plurality" means at least two, for example, two, three, etc., unless otherwise specifically defined.

[0149] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, segment or portion of code comprising one or more executable instructions for implementing the steps of a custom logical function or process, and the scope of the preferred embodiments of the present application includes alternative implementations in which functions may be performed out of the order shown or discussed, including performing functions in a substantially simultaneous manner or in the reverse order depending on the functions involved, which should be understood by those skilled in the art to which the embodiments of the present application belong.

[0150] The logic and / or steps represented in a flowchart or otherwise described herein, for example, can be considered a sequenced list of executable instructions for implementing the logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (e.g., a computer-based system, a system including a processor, or other system that can fetch and execute instructions from an instruction execution system, apparatus, or device). For purposes of this specification, a "computer-readable medium" is any device that can contain, store, communicate, propagate, or transport a program for use by, or in conjunction with, an instruction execution system, apparatus, or device. More specific examples (not exhaustive) of computer-readable media include: an electrical connection with one or more wires (electronic devices), a portable computer disk cartridge (magnetic device), random access memory (RAM), read-only memory (ROM), erasable and programmable read-only memory (EPROM or flash memory), fiber optic devices, and a portable compact disc read-only memory (CDROM). Furthermore, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium and then editing, interpreting or processing it in another suitable manner if necessary, and then storing it in a computer memory.

[0151] It should be understood that various parts of this application can be implemented using hardware, software, firmware, or a combination thereof. In the above-described embodiments, multiple steps or methods can be implemented using software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented using hardware, as in another embodiment, any one of the following technologies known in the art or a combination thereof can be used: a discrete logic circuit having logic gate circuits for implementing logical functions on data signals, an application-specific integrated circuit having suitable combinational logic gate circuits, a programmable gate array (PGA), a field-programmable gate array (FPGA), etc.

[0152] Those skilled in the art will understand that all or part of the steps in the method of the above embodiment can be completed by instructing related hardware through a program, and the program can be stored in a computer-readable storage medium. When the program is executed, it includes one or a combination of the steps of the method embodiment.

[0153] In addition, the functional units in the various embodiments of the present application may be integrated into a processing module, or each unit may exist physically separately, or two or more units may be integrated into a module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of a software functional module. If the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it may also be stored in a computer-readable storage medium.

[0154] The storage medium mentioned above may be a read-only memory, a magnetic disk, or an optical disk, etc. Although the embodiments of the present application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present application. Persons skilled in the art may make changes, modifications, substitutions, and variations to the above embodiments within the scope of the present application.

Claims

1. A method for avoiding collisions with obstacles of an unmanned vehicle in a limited time domain, characterized in that: include: The safety of the control system is described based on forward invariance, and based on the safety of the control system and the obstacles, the obstacle function of the control system is constructed; The obstacle function is integrated into the cost function of the control system, and based on the cost function, the collision avoidance control problem of the control system is transformed into an optimal control problem; A control system model with unknown parameters is established based on the neural network function approximation method, and an integral parallel learning method is used to design a parameter adaptation law to identify the unknown parameters in the model. Based on the control system model, an adaptive iterative learning method is used to solve the optimal control input that meets the requirements of the collision avoidance control task using a quadratic optimization framework; Among them, the barrier function is integrated into the cost function of the control system, including: Define a circular sensing area for each obstacle , expressed as: in, is the inner diameter of the annular area, representing the minimum distance at which the control system begins to sense obstacles. is the outer diameter of the annular area, representing the maximum distance at which the system can sense an obstacle; Defines the control system's awareness of obstacles for: in, , For distance The distance between the centers of the obstacles; The cost function is defined as: in, is the instantaneous cost, including system performance and barrier function, expressed as: Among them, the matrix and matrix is a real symmetric positive definite matrix; In actual numerical simulation, a hyperbolic function is used to restrict the barrier function so that the cost function Updated to: in, is the instantaneous cost, for: 。 2. The obstacle avoidance control method for an unmanned vehicle in a limited time domain according to claim 1, characterized in that: The security of the control system described based on forward invariance includes: The control system is formulated as: in, is the system state variable, is the initial state variable, is the system control input, is the dynamic equation of the control system; Assumptions and its relative status and input The partial derivative of 、 Satisfy local Lipschitz continuity; Define system control inputs The set formed is a convex set , expressed as: in, is the constraint matrix, is the constraint vector; Assume that for closed bounded control input , at the initial value Under this condition, the solution trajectory of the control system Bounded, the formulation of this assumption is described as: for , , making in, For a finite time domain, Status The set composed of Status The upper bound of the norm of ; The collision avoidance control problem requires that the state of the control system is always within a given safety set. for: in, is a continuously differentiable function, the set for The boundary of the set for inside, , ; If for any initial state , solution trajectory Always keep in the collection In, judge the set It has forward invariance, if the system state is relative to the set Satisfy the forward invariance condition and judge the control system relative to the set Safety.

3. The obstacle avoidance control method for an unmanned vehicle in a limited time domain as claimed in claim 2, characterized in that: The obstacle function of the control system is constructed based on the safety of the control system and according to the obstacles, including: The barrier function is defined based on the safety set as: The barrier function satisfies: Taking into account obstacles, defining a circular area As the first The unsafe area of ​​obstacles is expressed as: in, It is The center of the obstacle, is the radius of the unsafe zone; For circular areas , No. The obstacle function of an obstacle is expressed as: in, ; In the actual data simulation, the barrier function is calibrated using a saturated hyperbolic function. Limit the barrier function Updated to: 。 4. The obstacle avoidance control method for an unmanned vehicle in a limited time domain as claimed in claim 1, characterized in that: The method of converting the collision avoidance control problem of the control system into an optimal control problem based on the cost function includes: Based on the collision avoidance control problem of the control system, the control objective is to obtain the optimal control input , so that for ,satisfy: ; According to the Pontryagin minimum principle, the optimal control input is determined The conditions for meeting the requirements are: in, , is the Hamiltonian function pair The partial derivative of is the co-state quantity.

5. The obstacle avoidance control method for an unmanned vehicle in a limited time domain as claimed in claim 4, characterized in that: The control system model with unknown parameters is established based on the neural network function approximation method, including: The control system with unknown parameters is expressed as: in, is the unknown coefficient matrix, is the activation function, is the approximate error; When the number of neurons When is large enough and the approximate error rate is small enough, the control system of unknown parameters can be approximately expressed as: in, is the estimated value of the unknown parameter matrix.

6. The obstacle avoidance control method for an unmanned vehicle in a limited time domain as claimed in claim 5, characterized in that: The control system model with unknown parameters is established based on the neural network function approximation method, and the parameter adaptive law is designed using the integral parallel learning method to identify the unknown parameters in the model, including: Collection W group The historical trajectory data in is recorded as ,express A set of system trajectories and corresponding control inputs at different times, and for each set of data, calculate: in, ; Define the matrix , the collected historical trajectory data meets ; The parameter adaptive law for updating is designed using the integral parallel learning method, and the unknown parameters in the control system are updated using the collected historical trajectory dual data. The designed parameter adaptive law is: definition For the The estimated value of the system state in the iteration process, its adaptive law is: in, , is the learning rate, It is The estimation error of the system state during the iteration; definition For the The estimated value of the coefficient matrix during the iteration, It is The identification error of the coefficient matrix during the iteration process is, , The adaptive law is: in, , is the learning rate.

7. The obstacle avoidance control method for an unmanned vehicle in a limited time domain as claimed in claim 6, characterized in that: The method of solving the optimal control input that meets the requirements of the collision avoidance control task based on the control system model and utilizing a quadratic optimization framework and an adaptive iterative learning method includes: Step S1: Initialize iteration parameters 、 、 and convergence accuracy 、 , select the initial control input , by solving the following equations we get Estimates of the initial system trajectory and estimates of the initial coefficient matrix: in, and are the initial values ​​of the system state estimation error and the coefficient matrix identification error, is the initial value of the system’s true state trajectory; Calculate the initial cost function , is the integral of the instantaneous cost, are the upper and lower limits of the integral, is the integral variable; make ; Step S2: Calculate by the following formula : in, , represents the estimated value of the Hamiltonian function, For its state variables The partial derivative of Step S3: Solve the control input increment by quadratic optimization method : in, , is the estimated value of the Hamiltonian function for the variable The partial derivative of ; make ; Step S4: Get the online true trajectory through the updated control input , update the estimated value of the system trajectory and the estimated value of the coefficient matrix by solving the following equation: in, and are the updated system state estimation error and coefficient matrix identification error respectively; Step S5: Calculate the cost function using the updated control input and online trajectory data , judge whether the updated control input reduces the cost function, if , then let , , If it is the iteration step, return to step S3, otherwise, go to step S6; Step S6: If , then go to step S7, otherwise, let , , , and return to step S2; Step S7: If , stop the iterative process and get the corresponding optimal control input, otherwise, let , , , and return to step S2.

8. A limited time domain unmanned vehicle obstacle avoidance control device, characterized in that: include: A first system establishment module is used to describe the safety of the control system based on forward invariance, and to construct an obstacle function of the control system based on the safety of the control system and obstacles; The second system establishment module is used to integrate the obstacle function into the cost function of the control system and transform the collision avoidance control problem of the control system into an optimal control problem based on the cost function; The system identification module is used to establish a control system model with unknown parameters based on the neural network function approximation method, and to design a parameter adaptive law using the integral parallel learning method to identify the unknown parameters in the model; An iterative solution module is used to solve the optimal control input that meets the requirements of the collision avoidance control task based on the control system model using a quadratic optimization framework and an adaptive iterative learning method; Among them, the barrier function is integrated into the cost function of the control system, including: Define a circular sensing area for each obstacle , expressed as: in, is the inner diameter of the annular area, representing the minimum distance at which the control system begins to sense obstacles. is the outer diameter of the annular area, representing the maximum distance at which the system can sense an obstacle; Defines the control system's awareness of obstacles for: in, , For distance The distance between the centers of the obstacles; The cost function is defined as: in, is the instantaneous cost, including system performance and barrier function, expressed as: Among them, the matrix and matrix is a real symmetric positive definite matrix; In actual numerical simulation, a hyperbolic function is used to restrict the barrier function so that the cost function Updated to: in, is the instantaneous cost, for: 。

Images