Control method and device for autonomous driving vehicle

Abstract

The present invention discloses a control method and device for an autonomous driving vehicle, and relates to the field of computer technology. A specific implementation of the method includes: obtaining the state information and road structure information of the controlled vehicle at the current moment; inputting the state information and road structure information of the controlled vehicle into a pre-constructed comprehensive traffic environment model to output the vehicle control instructions at the next moment, wherein the comprehensive traffic environment model is obtained by modeling based on the vehicle position information and road structure information using a virtual potential energy field, and the vehicle dynamics equation is used as a constraint condition of the comprehensive traffic environment model; and controlling the controlled vehicle to travel according to the vehicle control instructions. This implementation does not require designing different driving strategies for different driving scenarios, and does not require hierarchical design of driving behavior decision-making, trajectory planning, and tracking control, making the control system of the autonomous driving vehicle simpler, more compact, and capable of handling different traffic environments.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a control method and device for an autonomous driving vehicle. Background Art

[0002] In the vehicle-based control system, three layers—behavior decision-making, motion planning, and tracking control—play a major role. Many research efforts in this area assume that these layers have no impact on each other, focusing on studying one of them or decomposing the problem into separate layers, hoping to solve a specific problem within a specific traffic environment. Traditional approaches to designing a fully autonomous driving control system require developing distinct processing methods at each layer and designing information exchange interfaces between the layers. In other words, a complete autonomous vehicle control system would require a wide variety of methods within different layers to address diverse driving scenarios, making the control system more complex and designing more challenging. Real-world traffic environments are extremely complex and dynamic, making it difficult to divide the dynamic environment into a finite number of scenarios. Furthermore, switching between scenarios requires designing entry and exit mechanisms, determining when and how to switch from one driving behavior to another. Switching between different driving behavior strategies can result in uneven control command output.

[0003] In the process of implementing the present invention, the inventors discovered that the prior art has at least the following problems:

[0004] An autonomous driving control system designed in a traditional way will become extremely bloated and may still not have sufficient capabilities to handle most traffic scenarios. Summary of the Invention

[0005] In view of this, an embodiment of the present invention provides a control method and device for an autonomous driving vehicle, which eliminates the need to design different driving strategies for different driving scenarios and the need to design driving behavior decisions, trajectory planning, and tracking control in a hierarchical manner, thereby making the control system of the autonomous driving vehicle simpler, more compact, and capable of handling different traffic environments.

[0006] To achieve the above-mentioned purpose, according to one aspect of an embodiment of the present invention, a control method for an autonomous driving vehicle is provided, comprising: obtaining state information and road structure information of a controlled vehicle at the current moment; inputting the state information and road structure information of the controlled vehicle into a pre-constructed comprehensive traffic environment model to output a vehicle control instruction for the next moment, wherein the comprehensive traffic environment model is modeled based on vehicle position information and road structure information using a virtual potential energy field, and a vehicle dynamics equation is used as a constraint condition of the comprehensive traffic environment model; and controlling the controlled vehicle to travel according to the vehicle control instruction.

[0007] Optionally, the road structure information includes road boundary information, lane line information and obstacle information; the comprehensive traffic environment model is constructed in the following manner: using a virtual potential energy field, modeling is performed based on the vehicle position information and the road boundary information to obtain a road boundary potential energy field; using a virtual potential energy field, modeling is performed based on the vehicle position information and the lane line information to obtain a lane line potential energy field; using a virtual potential energy field, modeling is performed based on the vehicle position information and the obstacle information to obtain an obstacle potential energy field; superimposing the road boundary potential energy field, the lane line potential energy field and the obstacle potential energy field to obtain a road structure potential energy field; generating an objective function based on the road structure potential energy field, and constructing the comprehensive traffic environment model with the vehicle dynamics equation as a constraint condition.

[0008] Optionally, modeling is performed based on the vehicle position information and the road boundary information to obtain a road boundary potential energy field, including: obtaining the distance between the vehicle and the left and right boundaries of the road based on the vehicle position information and the road boundary information; establishing a road boundary potential energy field based on the distance between the vehicle and the left and right boundaries of the road, wherein, when the distance between the vehicle and the left and right boundaries of the road decreases, the road boundary potential energy increases; when the vehicle is in the middle of the road, the road boundary potential energy is zero and remains unchanged.

[0009] Optionally, a lane line potential energy field is obtained by modeling based on the vehicle position information and the lane line information, including: judging whether the lane line is a solid line and whether adjacent lanes are lanes traveling in the same direction based on the lane line information; if the lane line is a solid line or the adjacent lane is a lane traveling in the opposite direction, then treating the lane line as a road boundary to establish a lane line potential energy field; if the lane line is a dotted line and the adjacent lane is a lane traveling in the same direction, then obtaining the shortest distance from the vehicle to the lane line based on the vehicle position information and the lane line information, and establishing a lane line potential energy field based on the shortest distance, wherein, when the vehicle crosses the lane line, the lane line potential energy increases.

[0010] Optionally, modeling is performed based on the vehicle position information and the obstacle information to obtain an obstacle potential energy field, including: obtaining a minimum distance between the vehicle and the obstacle based on the vehicle position information and the obstacle information, and establishing an obstacle potential energy field based on the minimum distance, wherein the obstacle potential energy decreases as the minimum distance increases.

[0011] Optionally, the state information of the controlled vehicle includes: the position information and speed of the controlled vehicle; the constraint conditions also include: state variable constraints, control variable constraints, speed constraints, slack variable constraints, vehicle-obstacle safety distance constraints, and initial state value constraints; generating an objective function based on the road structure potential energy field, and constructing the comprehensive traffic environment model with the vehicle dynamics equation as a constraint condition, including: generating an objective function based on the road structure potential energy field, the control variables at the current moment, and the penalty for speed conflict; constructing the comprehensive traffic environment model with the vehicle dynamics equation, the state variable constraints, the control variable constraints, the speed constraints, the slack variable constraints, the vehicle-obstacle safety distance constraints, and the initial state value constraints as constraints.

[0012] Optionally, an objective function is generated based on the road structure potential energy field, the control variables at the current moment and the penalty for speed conflict, including: calculating a first sub-objective function corresponding to the road structure potential energy field based on the product of the road structure potential energy field and a second constant; calculating a second sub-objective function corresponding to the control variables at a certain moment in the prediction domain based on the transposed matrix, the positive definite matrix and the product of the control variables at the certain moment, the prediction domain including multiple sampling periods; calculating a third sub-objective function corresponding to the penalty for speed conflict based on the product of the square of the slack variable and the first constant; calculating a core function of the objective function based on the sum of the first sub-objective function, the second sub-objective function and the third sub-objective function; and generating the objective function by calculating the minimum value of the sum of the core functions of the multiple sampling periods.

[0013] Optionally, the state variable constraint is implemented based on the maximum and minimum values ​​of the state variable under the system mechanical constraint conditions; the control variable constraint is implemented based on the maximum and minimum values ​​of the control variable under the system mechanical constraint conditions; the speed constraint is implemented based on the maximum speed, minimum speed and slack variables; the safety distance constraint between the vehicle and the obstacle is implemented based on the constraint relationship between the position of the vehicle and each obstacle within a set range and a set threshold; the initial state value constraint is implemented based on the state value of the controlled vehicle at the sampling moment of each sampling period.

[0014] According to another aspect of an embodiment of the present invention, a control device for an autonomous driving vehicle is provided, comprising: an information acquisition module for acquiring status information and road structure information of a controlled vehicle at a current moment; an instruction generation module for inputting the status information and road structure information of the controlled vehicle into a pre-constructed comprehensive traffic environment model to output a vehicle control instruction for the next moment, wherein the comprehensive traffic environment model is modeled based on vehicle position information and road structure information using a virtual potential energy field, and vehicle dynamics equations are used as constraints of the comprehensive traffic environment model; and a vehicle control module for controlling the driving of the controlled vehicle according to the vehicle control instruction.

[0015] According to another aspect of an embodiment of the present invention, an electronic device is provided, comprising: one or more processors; a storage device for storing one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors implement the control method of the autonomous driving vehicle provided by an embodiment of the present invention.

[0016] According to another aspect of an embodiment of the present invention, a computer-readable medium is provided, on which a computer program is stored. When the program is executed by a processor, the control method of the autonomous driving vehicle provided by an embodiment of the present invention is implemented.

[0017] One embodiment of the above invention has the following advantages or beneficial effects: by obtaining the state information and road structure information of the controlled vehicle at the current moment; inputting the state information and road structure information of the controlled vehicle into a pre-built integrated traffic environment model to output vehicle control instructions for the next moment, the integrated traffic environment model is modeled based on vehicle position information and road structure information using a virtual potential field, and the vehicle dynamics equation is used as a constraint condition of the integrated traffic environment model; and controlling the driving of the controlled vehicle according to the vehicle control instructions. The technical solution realizes the control of the controlled vehicle based on the model predictive control theory. The virtual potential field is used to model road facilities such as road boundaries, lane lines, and surrounding obstacles or obstructing vehicles, transforming the control problem of the autonomous vehicle on structured roads into a model predictive control problem, and using the vehicle dynamics equation as a constraint of the model predictive control problem. This eliminates the need to design different driving strategies for different driving scenarios and eliminates the need to hierarchically design driving behavior decision-making, trajectory planning, and tracking control, making the control system of the autonomous vehicle simpler, more compact, and capable of handling different traffic environments. At the same time, other constraint functions can be introduced to constrain the safe distance between the controlled vehicle and obstacles to add collision avoidance constraints.

[0018] The further effects of the above-mentioned non-conventional optional manner will be described below in conjunction with specific embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The accompanying drawings are provided for a better understanding of the present invention and are not intended to limit the present invention.

[0020] Figure 1 is a schematic diagram of the main steps of a method for controlling an autonomous driving vehicle according to an embodiment of the present invention;

[0021] Figure 2 is a schematic diagram of a dynamic traffic environment model according to an embodiment of the present invention;

[0022] Figure 3 is a schematic cross-sectional diagram of a road boundary potential energy field according to an embodiment of the present invention;

[0023] Figure 4 is a schematic cross-sectional diagram of the lane line potential energy field of a dotted lane line according to an embodiment of the present invention;

[0024] Figure 5 1 is a schematic cross-sectional diagram of the potential energy field of lane lines of a three-lane road divided by dashed lines according to an embodiment of the present invention;

[0025] Figure 6 Schematic diagram of the obstacle potential energy field of a rectangular obstacle according to an embodiment of the present invention;

[0026] Figure 7 is a schematic diagram of a nonlinear vehicle dynamics model according to an embodiment of the present invention;

[0027] Figure 8 is a schematic diagram of main modules of a control device for an autonomous driving vehicle according to an embodiment of the present invention;

[0028] Figure 9 is an exemplary system architecture diagram in which embodiments of the present invention may be applied;

[0029] Figure 10 It is a schematic diagram of the structure of a computer system of a terminal device or server suitable for implementing an embodiment of the present invention. DETAILED DESCRIPTION

[0030] The following description of exemplary embodiments of the present invention is made in conjunction with the accompanying drawings, in which various details of the embodiments of the present invention are included to facilitate understanding. These details should be considered as merely exemplary. Therefore, it should be appreciated by those skilled in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present invention. Similarly, for the sake of clarity and conciseness, descriptions of well-known functions and structures are omitted in the following description.

[0031] The acquisition, storage, use, and processing of data in the technical solution of the present invention are in compliance with the relevant provisions of national laws and regulations.

[0032] An autonomous driving control system designed in a traditional way will become extremely bloated, but may still not have sufficient capabilities to handle most traffic scenarios. In order to solve these problems, the present invention proposes a control method for autonomous driving vehicles based on model predictive control theory. A virtual potential energy field is used to model road facilities, such as road boundaries, lane lines, and surrounding obstacles or obstacle vehicles, and the control problem of autonomous driving vehicles on structured roads is converted into a model predictive control problem, with the vehicle dynamics equations used as part of the constraints of the model predictive control problem. This method does not require the design of different driving strategies for different driving scenarios, and does not require the hierarchical design of driving behavior decisions, trajectory planning, and tracking control, making the control system of autonomous driving vehicles simpler, more compact, and capable of handling different traffic environments.

[0033] Figure 1 FIG. 1 is a schematic diagram of the main steps of the control method of the automatic driving vehicle according to an embodiment of the present invention. Figure 1 As shown, the control method of the autonomous driving vehicle according to an embodiment of the present invention mainly includes the following steps S101 to S103.

[0034] Step S101: Acquire the state information of the controlled vehicle and the road structure information at the current moment.

[0035] In an embodiment of the present invention, in order to control an autonomous vehicle, it is necessary to predict the control instructions for the controlled vehicle at the next moment based on the current state information of the controlled vehicle and road structure information. The current state information of the controlled vehicle includes, for example, the position, speed, acceleration, and size of the controlled vehicle; the road structure information includes, for example, the position information of the left and right road boundaries, lane marking information (e.g., whether the lane marking is solid or dashed, the position of the lane marking, etc.), lane direction information, and obstacle information (e.g., information about vehicles traveling in each lane and roadblocks in the lane).

[0036] Step S102: Input the state information and road structure information of the controlled vehicle into a pre-built comprehensive traffic environment model to output the vehicle control instructions at the next moment. The comprehensive traffic environment model is obtained by modeling based on the vehicle position information and road structure information using a virtual potential field, and the vehicle dynamics equation is used as the constraint condition of the comprehensive traffic environment model.

[0037] According to the technical solution of this invention, a virtual potential field is used to model road infrastructure, such as road boundaries, lane markings, and surrounding obstacles or vehicles. This transforms the control problem of autonomous vehicles on structured roads into a model predictive control problem, incorporating the vehicle dynamics equations as part of the constraints of the model predictive control problem. This approach eliminates the need to design different driving strategies for different driving scenarios, and eliminates the need for a hierarchical design of driving behavior decision-making, trajectory planning, and tracking control. This makes the control system of autonomous vehicles simpler, more compact, and more capable of handling diverse traffic environments.

[0038] According to one embodiment of the present invention, road structure information includes road boundary information, lane line information, and obstacle information. Furthermore, the comprehensive traffic environment model is constructed in the following manner: using a virtual potential energy field to model based on the vehicle position information and the road boundary information to obtain a road boundary potential energy field; using a virtual potential energy field to model based on the vehicle position information and the lane line information to obtain a lane line potential energy field; using a virtual potential energy field to model based on the vehicle position information and the obstacle information to obtain an obstacle potential energy field; superimposing the road boundary potential energy field, the lane line potential energy field, and the obstacle potential field to obtain a road structure potential field; generating an objective function based on the road structure potential field, and using the vehicle dynamics equation as a constraint condition to construct the comprehensive traffic environment model. The construction process of the comprehensive traffic environment model of the present invention will be described below in conjunction with specific embodiments.

[0039] Figure 2 It is a schematic diagram of the dynamic traffic environment model of an embodiment of the present invention. Assuming that a high-fidelity model of the controlled vehicle has been established and the various variable parameters therein have been identified, and the size, position, speed, acceleration and other state information of the surrounding vehicles traveling on the same road are known, this information of the surrounding vehicles can be obtained through sensors such as cameras and lidars, using a variety of environmental perception technologies such as deep learning and SLAM (simultaneous localization and mapping, real-time positioning and map construction), and can also be obtained through V2X (vehicle to vehicle, that is, vehicle-to-vehicle information exchange) communication technology, such as vehicle-road collaborative technology. Figure 2 As shown in the figure, the vehicle pattern represents the controlled autonomous vehicle, and the gray rectangle represents the surrounding vehicles, which are referred to as "obstacles" or "obstacle vehicles" in the subsequent embodiments. The black solid line represents the road boundary, and the black dotted line represents the lane line that can be crossed. The width of each lane is D, the position coordinates of the controlled vehicle are (X, Y), and the distance from the controlled vehicle to the left and right boundaries of the road are d respectively. l and d r , the distance from the controlled vehicle to the i-th lane line is recorded as d iOnly the obstacle vehicles within the distance L from the rear of the controlled vehicle in the road coordinate system are considered, and L is called the "influence range". (X j ,Y j ) is the position of the obstacle vehicle with the label j, (v jx ,v jy ) is its horizontal and vertical speed. The shortest distance between the controlled vehicle and the jth obstacle vehicle is d j .

[0040] To establish a unified traffic environment model and consider the combined impact of all traffic participants on the controlled vehicle's travel, this paper introduces the concept of potential energy fields from physics. By superimposing the potential energy fields of different obstacles, we can rationally and vividly represent their combined impact on the controlled vehicle. The following describes the potential energy fields of road boundaries, lane markings, and obstructing vehicles (obstacles).

[0041] According to one embodiment of the present invention, modeling is performed based on the vehicle position information and the road boundary information to obtain a road boundary potential energy field, which can specifically include: obtaining the distance between the vehicle and the left and right boundaries of the road based on the vehicle position information and the road boundary information; establishing a road boundary potential energy field based on the distance between the vehicle and the left and right boundaries of the road, wherein the road boundary potential energy increases as the distance between the vehicle and the left and right boundaries of the road decreases; and when the vehicle is in the middle of the road, the road boundary potential energy is zero and remains unchanged.

[0042] Figure 3 : is a schematic cross-sectional view of the road boundary potential energy field in an embodiment of the present invention. In order to take into account the constraints of the road boundary, a road boundary potential energy function is designed. When the potential energy is close to the road boundary, it is very large, and when it is far away from the boundary, it is almost zero. Figure 3 The cross-section of the potential energy field at the left and right boundaries of the road is shown. The superimposed potential energy field at the road boundary is as follows:

[0043]

[0044] Among them, k R is a constant term, (X, Y) is the coordinate of the controlled vehicle in the world coordinate system, d l (X,Y) and d r (X, Y) are the distances from the controlled vehicle to the left and right edges of the road, respectively. The closer the controlled vehicle gets to a road edge, the more dramatically the integrated potential energy field at that edge increases. When the controlled vehicle is in the middle of the road, its value is zero and remains constant.

[0045] According to another embodiment of the present invention, when modeling a lane line potential energy field based on the vehicle position information and the lane line information, the following steps may be specifically included: determining whether the lane line is a solid line and whether the adjacent lanes are lanes traveling in the same direction based on the lane line information; if the lane line is a solid line or the adjacent lanes are lanes traveling in the opposite direction, the lane line potential energy field is established using the lane line as a road boundary; if the lane line is a dashed line and the adjacent lanes are lanes traveling in the same direction, the shortest distance from the vehicle to the lane line is determined based on the vehicle position information and the lane line information, and a lane line potential energy field is established based on this shortest distance. When the vehicle crosses a lane line, the lane line potential energy field increases. In other words, when establishing the lane line potential energy field, it is necessary to differentiate and construct it based on the specific attributes of the lane line. In this embodiment of the present invention, if the lane line is a solid line or the adjacent lanes are lanes traveling in the opposite direction, the lane line can be treated as a road boundary to establish its virtual potential energy field; if the lane line is a dashed line and the adjacent lanes are lanes traveling in the same direction, the lane line potential energy field function can be designed as a Gaussian-like distribution function. When a vehicle crosses the dashed lane line, the potential energy is large, but there is no sudden change.

[0046] Figure 4 : is a schematic cross-sectional diagram of the lane potential field of the dotted lane line according to an embodiment of the present invention. Figure 4 As shown, a potential energy field cross-section diagram of two dashed lane lines of three lanes traveling in the same direction is shown, and the virtual potential energy function of the dashed lane lines is expressed as follows:

[0047]

[0048] Among them, α and σ are constants, (X, Y) is the coordinate of the controlled vehicle in the world coordinate system, d i (X, Y) is the shortest distance from the controlled vehicle to the i-th lane line. These lane lines form small bumps in the virtual potential energy field, which prevent the vehicle from staying on the lane line for a long time while allowing the vehicle to cross the lane line to another lane without consuming too much energy.

[0049] Figure 5 : is a schematic cross-sectional diagram of the potential energy field of lane lines of a three-lane road divided by dashed lines according to an embodiment of the present invention. Figure 5 As shown, a cross-sectional diagram of a virtual potential energy field that integrates two dotted lane lines and the road boundary is shown. The curve in the figure has three local minimum points, which vividly and intuitively shows that when the vehicle is traveling in the middle of each lane, its potential energy value is small.

[0050] According to another embodiment of the present invention, when modeling the obstacle potential energy field based on the vehicle position information and the obstacle information, specifically: a minimum distance between the vehicle and the obstacle is determined based on the vehicle position information and the obstacle information, and the obstacle potential energy field is established based on the minimum distance, wherein the obstacle potential energy decreases as the minimum distance increases. In this embodiment of the present invention, the obstacle is described using an obstructing vehicle as an example; however, the obstacle can also be a roadblock in the middle of the road, a roadblock separating lanes in different directions of travel, and so on. The process of a controlled vehicle traveling through the potential energy field of these obstructing vehicles can be compared to traveling through a valley. Driving through the valley floor can be smooth, but the driving distance may be long, potentially taking a long time to reach the destination. However, traveling over mountains and ridges causes significant vehicle jolting, reducing driving comfort. Therefore, the present invention requires an indicator that can describe the impact of an obstructing vehicle on the controlled vehicle. This indicator can decrease with increasing distance until it approaches zero; when the two are very close, the indicator increases sharply. In an embodiment of the present invention, the obstacle potential field is described using the Yukawa potential. The Yukawa potential is a theory used to describe short-range interactions between nuclei. The obstacle potential field of an embodiment of the present invention has the following form:

[0051]

[0052] Among them, d j (X, Y) represents the minimum distance between the controlled vehicle and the obstacle vehicle numbered j; the coefficient α j Determines the speed of change of potential energy, α j The larger the value, the faster the potential energy increases when the vehicle approaches the obstacle; j It is a proportional factor. The larger its value is, the larger the influence range of the obstacle potential energy field of the obstacle vehicle is.

[0053] Figure 6 Schematic diagram of the obstacle potential field of a rectangular obstacle in an embodiment of the present invention. Figure 6 , which gives a schematic diagram of the obstacle potential field of a 2x4 rectangular size, where α j =0.001, k j =1.

[0054] The above describes the creation of a virtual potential energy field based on road structure information, including road boundaries, lane markings, and obstacle information. Similarly, virtual potential energy fields can also be created based on the geometric and functional characteristics of other traffic elements in the traffic environment. Subsequently, by overlaying the virtual potential energy fields of road boundaries, lane markings, and obstacles, we can construct a comprehensive function of the complex traffic environment's impact on the controlled vehicle, i.e., a comprehensive traffic environment model. The resulting road structure potential energy field function is as follows:

[0055]

[0056] After constructing the road structure potential energy field function, the objective function of the comprehensive traffic environment model can be generated based on the road structure potential energy field function. In order to simultaneously generate lateral control and longitudinal control instructions, the present invention uses a nonlinear vehicle dynamics equation as a constraint condition for the comprehensive traffic environment model.

[0057] Figure 7 Schematic diagram of the nonlinear vehicle dynamics model of an embodiment of the present invention. In extreme motion situations such as high-speed driving and sharp turns, lateral motion and longitudinal motion are strongly coupled, and a slight deviation in control may lead to danger. In this case, it is particularly important to use more realistic nonlinear vehicle and tire dynamics models in simulation research and control research. Methods based on nonlinear vehicle dynamics models can improve control accuracy, driving safety, and ride comfort. Taking an autonomous driving vehicle traveling on a horizontal ground as the controlled object, considering that its complete vehicle dynamics model is very complex, in order to simplify the analysis, the present invention adopts a simplified two-degree-of-freedom nonlinear vehicle dynamics model on the horizontal plane, and replaces the two front wheels and two rear wheels of a real vehicle with one front wheel and one rear wheel respectively.

[0058] like Figure 7 As shown. Figure 7 In the figure, XY represents the world coordinate system, X B -Y B represents the vehicle body coordinate system, A and B are the centers of the front and rear wheels respectively, CG is the center of gravity of the vehicle, and the meanings of other variables and symbols are shown in Table 1.

[0059] Table 1

[0060]

[0061]

[0062] The differential equation of the vehicle dynamics model used in the embodiment of the present invention is shown below:

[0063]

[0064]

[0065]

[0066]

[0067]

[0068]

[0069] The lateral forces of the front and rear wheels can be approximated using a linear tire model. The specific calculation formula is as follows:

[0070]

[0071]

[0072] Among them, C f and C r Represent the steering stiffness coefficients of the front and rear tires respectively.

[0073] Taking into account the different types of vehicles, their different power sources, energy conversion methods and conversion efficiencies, as well as the differences in braking methods and steering assistance systems, in order to enhance the portability of the control system and the universality of the control method, the present invention sets the front wheel steering angle δ f and longitudinal force F xT As the control input variable of the original system. Then, it is normalized to obtain:

[0074] [δ f ,F xT ] T =[C δ u δ ,C F u F ] T ,

[0075] Among them, u δ ∈[-1,1] and u F ∈[-1,1] is a dimensionless variable, C δ and C F is a constant that takes into account the constraints of steering, engine, transmission and other mechanisms. Finally, the control input variables of the system can be obtained as:

[0076] u=[u δ ,u F ] T ,

[0077] Then u is the variable that the autonomous vehicle control algorithm needs to solve. The state variable of the system is:

[0078] x=[u x ,u y ,γ,ψ,X,Y] T .

[0079] Finally, the vehicle dynamics model can be discretized and simplified as:

[0080] x k+1 =f(x k ,u k ).

[0081] After obtaining the road structure potential energy field and vehicle dynamics equation according to the above steps, the objective function can be generated according to the road structure potential energy field, and the vehicle dynamics equation can be used as a constraint condition to construct a comprehensive traffic environment model.

[0082] In an embodiment of the present invention, control is performed based on a comprehensive traffic environment model, which is based on model predictive control. The potential energy field function is introduced into the objective function of the optimization problem, and the objective function in the prediction domain is minimized under the constraints of vehicle dynamics. This method integrates the driving behavior decision layer, the motion planning layer, and the tracking control layer. More precisely, steps such as behavior decision-making and trajectory planning are deleted, thereby simplifying the model and making the autonomous driving control system simpler. The control method will complete driving behaviors such as lane changing, following a vehicle, and overtaking within a certain length of time, and can complete the movement process without making any decisions or path planning in advance. This method does not require the system to be decomposed, and will simplify the control system of the entire autonomous driving vehicle, avoiding the layered instruction transmission from the decision-making system to the motion planning system and then to the control execution system.

[0083] According to an embodiment of the present invention, the state information of the controlled vehicle includes: the position information and speed of the controlled vehicle; and the constraint conditions also include: state variable constraints, control variable constraints, speed constraints, slack variable constraints, vehicle-obstacle safety distance constraints, and initial state value constraints. Specifically, generating an objective function based on the road structure potential energy field and constructing the comprehensive traffic environment model using the vehicle dynamics equation as a constraint condition includes: generating an objective function based on the road structure potential energy field, the current control variables, and the penalty for speed conflicts; and constructing the comprehensive traffic environment model using the vehicle dynamics equation, the state variable constraints, the control variable constraints, the speed constraints, the slack variable constraints, the vehicle-obstacle safety distance constraints, and the initial state value constraints as constraints.

[0084] According to one embodiment of the present invention, when generating an objective function based on the road structure potential energy field, the control variables at the current moment and the penalty for speed conflict, the following steps may be included: calculating the first sub-objective function corresponding to the road structure potential energy field according to the product of the road structure potential energy field and a second constant; calculating the second sub-objective function corresponding to the control variable at a certain moment in the prediction domain according to the transposed matrix, the positive definite matrix and the product of the control variable at the certain moment, the prediction domain includes multiple sampling periods; calculating the third sub-objective function corresponding to the penalty for speed conflict according to the product of the square of the slack variable and the first constant; calculating the core function of the objective function according to the sum of the first sub-objective function, the second sub-objective function and the third sub-objective function; generating the objective function by calculating the minimum value of the sum of the core functions of the multiple sampling periods.

[0085] Moreover, in an embodiment of the present invention, the state variable constraint is implemented based on the maximum and minimum values ​​of the state variable under the system mechanical constraint conditions; the control variable constraint is implemented based on the maximum and minimum values ​​of the control variable under the system mechanical constraint conditions; the speed constraint is implemented based on the maximum speed, the minimum speed and the slack variable; the safety distance constraint between the vehicle and the obstacle is implemented based on the constraint relationship between the position of the vehicle and each obstacle within the set range and the set threshold; the initial state value constraint is implemented based on the state value of the controlled vehicle at the sampling moment of each sampling period.

[0086] The present invention is based on model predictive control, and the comprehensive traffic environment model constructed is as follows:

[0087]

[0088] stx k+1 =f(x k ,u k ),k=0,1,…,(N p -1),

[0089] x min ≤x k ≤x max ,k=1,2,…,N p ,

[0090] u min ≤u k ≤u max ,k=0,1,…,(N p -1),

[0091] Δu min ≤u k+1 -u k ≤Δumax ,k=0,1,…,(N p -2),

[0092] v min -ξ k ≤v x,k ≤v max +ξ k ,k=1,2,…,N p ,

[0093] ξ k ≥0,k=0,1,…,(N p -1),

[0094]

[0095] P j >Q j >0,

[0096] x0=x(t).

[0097] Among them, the formula is the objective function, in which N p is the total number of prediction steps, that is, the number of sampling periods, assuming H p is the time length of the prediction domain, Δt is the sampling time interval, that is, the time interval between every two sampling cycles, then H p =N p Δt, U k is the road structure potential energy field, is the control variable at the kth moment in the prediction domain, K p and K ξ is a constant, is a positive definite matrix; k is the slack variable, the quadratic term of the slack variable in the objective function It is the penalty for speed conflicts. As can be seen, there is no reference trajectory in the objective function, which simplifies the model and makes the autonomous driving control system simpler.

[0098] x k+1 =f(x k ,u k ),k=0,1,…,(N p -1) is the discrete form of the vehicle dynamics equation, which serves as one of the important constraints here.

[0099] x min ≤x k ≤x max ,k=1,2,…,N p 、u min ≤uk ≤u max ,k=0,1,…,(N p -1)、Δu min ≤u k+1 -u k ≤Δu max ,k=0,1,…,(N p -2) It is assumed that the constraints of the state variables and control variables take into account the mechanical constraints of the system.

[0100] v min -ξ k ≤v x,k ≤v max +ξ k ,k=1,2,…,N p is the velocity constraint, ξ k is a slack variable.

[0101] The safe distance constraint between the controlled vehicle and the obstacle is expressed in the formula and j=1,2,…,N0, where (X j,k ,Y j,k ) is the position coordinate of the jth obstacle, N0 is the total number of obstacles or obstacle vehicles within the considered range, P j >Q j >0, indicating that the longitudinal safety distance is greater than the lateral safety distance, because the longitudinal speed is usually greater than the lateral speed. j and Q j Adjustments should be made accordingly based on the speed of the obstacle.

[0102] x0=x(t) indicates that the system should capture the state of the controlled vehicle at the current sampling moment in real time during each sampling period.

[0103] In step S102, by inputting the state information of the controlled vehicle and the road structure information into the aforementioned integrated traffic environment model, the vehicle control instructions for the next moment can be output. This process is the process of solving the integrated traffic environment model. The integrated traffic environment model is solved online in real time. The first variable of the obtained control sequence (i.e., the vehicle control instructions for the next moment) or the first part of the sequence is input into the integrated traffic environment model. The values ​​of each state variable (i.e., the state information of the controlled vehicle) are then obtained. The above solution problem is then repeated, and this process continues until the control process terminates. By periodically solving the integrated traffic environment model, the controlled vehicle can automatically handle various complex traffic scenarios, as long as the influence of participants in the traffic environment on the controlled vehicle is converted into a potential energy field function. If there are no obstacles near the controlled vehicle or the relative distance between them is almost unchanged, the controlled vehicle will maintain a relatively constant speed in its lane. If there is an obstacle in front of the controlled vehicle or the speed of the vehicle ahead is very low, the controlled vehicle will change lanes to an adjacent lane and drive normally in the new lane. All of these complex driving behaviors are completed step by step, bottom-up, by the control system, without relying on external instructions. The autonomous driving control system, constructed from this comprehensive traffic environment model, can also readily respond to driving requests input via the human-machine interface. For example, when the control system receives a lane change instruction from the operator, a virtual obstacle is placed at the minimum safe distance in front of the controlled vehicle, preventing it from continuing on the current road. The controlled vehicle will then spontaneously change lanes to the adjacent lane without following any reference trajectory pre-set by other planning methods.

[0104] The solution of the comprehensive traffic environment model can be transformed into a nonlinear optimization problem. However, the solution of the comprehensive traffic environment model is actually a nonlinear non-convex optimization problem. Its online solution on the vehicle system will inevitably consume a lot of resources and take a long time to calculate. In order to reduce the computational complexity, the general practice is to linearize the nonlinear dynamic system near the operating point. At time t, u x ≠0,u y =γ=ψ=0, the linearized system can be expressed as:

[0105]

[0106] in,

[0107]

[0108]

[0109] Then, the state space system is discretized using the zero-order hold method to obtain:

[0110] xk+1 =A k x k +B k u k ,

[0111] Among them, A k =e A(t)Δt , Δt is the sampling time interval of the digital system. In this way, the integrated predictive control problem (i.e., the problem of solving the integrated traffic environment model) can be transformed into an optimization problem with linear vehicle dynamics constraints, which greatly reduces the complexity of the problem and the solution time.

[0112] Step S103: Controlling the controlled vehicle according to the vehicle control command. By inputting the controlled vehicle's state information and road structure information into a pre-built integrated traffic environment model, the integrated traffic environment model is solved and the vehicle control command for the next moment is output. The controlled vehicle is then controlled according to the vehicle control command.

[0113] Based on steps S101 to S103 described above, a control method for autonomous vehicles in dynamic traffic environments is proposed based on model predictive control theory. A virtual potential field is used to model the road structure, including road boundaries, lane markings, and obstacles. By synthesizing the potential fields of road boundaries, lane markings, and obstacles, a road structure potential field function for the entire traffic environment is derived. This function is a key component of the objective function of the model predictive controller. A nonlinear vehicle dynamics model is used to simultaneously generate lateral and longitudinal control commands. Vehicle dynamic constraints are also one of the primary constraints in the model predictive control problem. Furthermore, other constraint functions can be introduced to constrain the safe distance between the controlled vehicle and obstacles, thereby adding collision avoidance constraints. This invention establishes a comprehensive traffic environment model for autonomous vehicle motion control and proposes a control method for integrated vehicle lateral and longitudinal motion control based on model predictive control theory. This method integrates the three key aspects of driving behavior decision-making, motion planning, and tracking control in traditional autonomous vehicle control systems without switching between different control modes for different traffic environments.

[0114] Figure 8 FIG. 1 is a schematic diagram of the main modules of the control device for an autonomous driving vehicle according to an embodiment of the present invention. Figure 8 As shown, the control device 800 of the autonomous driving vehicle according to an embodiment of the present invention mainly includes an information acquisition module 801, an instruction generation module 802 and a vehicle control module 803.

[0115] The information acquisition module 801 is used to obtain the status information of the controlled vehicle and the road structure information at the current moment;

[0116] The command generation module 802 is configured to input the state information of the controlled vehicle and the road structure information into a pre-built integrated traffic environment model to output the vehicle control command for the next moment. The integrated traffic environment model is modeled using a virtual potential energy field based on the vehicle position information and the road structure information, and the vehicle dynamics equation is used as a constraint condition of the integrated traffic environment model.

[0117] The vehicle control module 803 is used to control the driving of the controlled vehicle according to the vehicle control instruction.

[0118] According to one embodiment of the present invention, the road structure information includes road boundary information, lane line information and obstacle information; and the comprehensive traffic environment model is constructed in the following manner: using a virtual potential energy field, modeling is performed based on the vehicle position information and the road boundary information to obtain a road boundary potential energy field; using a virtual potential energy field, modeling is performed based on the vehicle position information and the lane line information to obtain a lane line potential energy field; using a virtual potential energy field, modeling is performed based on the vehicle position information and the obstacle information to obtain an obstacle potential energy field; the road boundary potential energy field, the lane line potential energy field and the obstacle potential energy field are superimposed to obtain a road structure potential energy field; generating an objective function according to the road structure potential energy field, and constructing the comprehensive traffic environment model with the vehicle dynamics equation as a constraint condition.

[0119] According to another embodiment of the present invention, modeling is performed based on the vehicle position information and the road boundary information to obtain a road boundary potential energy field, which may specifically include: obtaining the distance between the vehicle and the left and right boundaries of the road based on the vehicle position information and the road boundary information; establishing a road boundary potential energy field based on the distance between the vehicle and the left and right boundaries of the road, wherein, when the distance between the vehicle and the left and right boundaries of the road decreases, the road boundary potential energy increases; when the vehicle is in the middle of the road, the road boundary potential energy is zero and remains unchanged.

[0120] According to another embodiment of the present invention, a lane line potential energy field is obtained by modeling based on the vehicle position information and the lane line information, which may specifically include: judging whether the lane line is a solid line and whether adjacent lanes are lanes traveling in the same direction based on the lane line information; if the lane line is a solid line or the adjacent lane is a lane traveling in the opposite direction, then using the lane line as a road boundary to establish a lane line potential energy field; if the lane line is a dotted line and the adjacent lane is a lane traveling in the same direction, then obtaining the shortest distance from the vehicle to the lane line based on the vehicle position information and the lane line information, and establishing a lane line potential energy field based on the shortest distance, wherein the lane line potential energy increases when the vehicle crosses the lane line.

[0121] According to another embodiment of the present invention, modeling is performed based on the vehicle position information and the obstacle information to obtain an obstacle potential energy field, which may specifically include: obtaining a minimum distance between the vehicle and the obstacle based on the vehicle position information and the obstacle information, and establishing an obstacle potential energy field based on the minimum distance, wherein the obstacle potential energy decreases as the minimum distance increases.

[0122] According to another embodiment of the present invention, the state information of the controlled vehicle includes: the position information and speed of the controlled vehicle; the constraint conditions also include: state variable constraints, control variable constraints, speed constraints, slack variable constraints, vehicle-obstacle safety distance constraints, and initial state value constraints; and, generating an objective function based on the road structure potential energy field, and constructing the comprehensive traffic environment model with the vehicle dynamics equation as a constraint condition, which may specifically include: generating an objective function based on the road structure potential energy field, the control variables at the current moment, and the penalty for speed conflict; and constructing the comprehensive traffic environment model with the vehicle dynamics equation, the state variable constraints, the control variable constraints, the speed constraints, the slack variable constraints, the vehicle-obstacle safety distance constraints, and the initial state value constraints as constraints.

[0123] According to another embodiment of the present invention, an objective function is generated based on the road structure potential energy field, the control variables at the current moment and the penalty for speed conflict, including: calculating a first sub-objective function corresponding to the road structure potential energy field based on the product of the road structure potential energy field and a second constant; calculating a second sub-objective function corresponding to the control variables at a certain moment in a prediction domain based on the transposed matrix, the positive definite matrix and the product of the control variables at the certain moment, the prediction domain including multiple sampling periods; calculating a third sub-objective function corresponding to the penalty for speed conflict based on the product of the square of the slack variable and the first constant; calculating a core function of the objective function based on the sum of the first sub-objective function, the second sub-objective function and the third sub-objective function; and generating the objective function by calculating the minimum value of the sum of the core functions of the multiple sampling periods.

[0124] According to another embodiment of the present invention, the state variable constraint is implemented based on the maximum and minimum values ​​of the state variable under the system mechanical constraint conditions; the control variable constraint is implemented based on the maximum and minimum values ​​of the control variable under the system mechanical constraint conditions; the speed constraint is implemented based on the maximum speed, the minimum speed and the slack variable; the safe distance constraint between the vehicle and the obstacle is implemented based on the constraint relationship between the position of the vehicle and each obstacle within a set range and a set threshold; the initial state value constraint is implemented based on the state value of the controlled vehicle at the sampling moment of each sampling period.

[0125] According to the technical solution of an embodiment of the present invention, by obtaining the state information and road structure information of the controlled vehicle at the current moment; inputting the state information and road structure information of the controlled vehicle into a pre-built integrated traffic environment model to output the vehicle control command for the next moment, the integrated traffic environment model is modeled based on the vehicle position information and road structure information using a virtual potential field, and the vehicle dynamics equation is used as the constraint condition of the integrated traffic environment model; and controlling the driving of the controlled vehicle according to the vehicle control command. The technical solution realizes the control of the controlled vehicle based on the model predictive control theory. The virtual potential field is used to model road facilities such as road boundaries, lane lines, and surrounding obstacles or obstructing vehicles, transforming the control problem of the autonomous vehicle on structured roads into a model predictive control problem. The vehicle dynamics equation is used as a constraint in the model predictive control problem. This eliminates the need to design different driving strategies for different driving scenarios and eliminates the need to design driving behavior decisions, trajectory planning, and tracking control in a hierarchical manner. The control system of the autonomous vehicle is simpler, more compact, and capable of handling different traffic environments. At the same time, other constraint functions can be introduced to constrain the safe distance between the controlled vehicle and obstacles to add collision avoidance constraints.

[0126] Figure 9 An exemplary system architecture 900 is shown to which the control method for an autonomous driving vehicle or the control device for an autonomous driving vehicle according to an embodiment of the present invention can be applied.

[0127] like Figure 9 As shown, system architecture 900 may include terminal devices 901, 902, 903, a network 904, and a server 905. Network 904 is used to provide a medium for communication links between terminal devices 901, 902, 903 and server 905. Network 904 may include various connection types, such as wired or wireless communication links or fiber optic cables.

[0128] Users can use terminal devices 901, 902, and 903 to interact with server 905 via network 904 to receive or send messages, etc. Various communication client applications can be installed on terminal devices 901, 902, and 903, such as vehicle navigation applications, map navigation applications, location search applications, etc. (only as examples).

[0129] The terminal devices 901 , 902 , and 903 may be various electronic devices having a display screen and supporting web browsing, including but not limited to smart phones, tablet computers, laptop computers, and desktop computers.

[0130] Server 905 may be a server that provides various services, such as a background management server (for example only) that supports the vehicle navigation application browsed by users using terminal devices 901, 902, and 903. The background management server may obtain the current state information and road structure information of the controlled vehicle from the received vehicle navigation request and other data; input the state information and road structure information of the controlled vehicle into a pre-built comprehensive traffic environment model to output vehicle control instructions for the next moment, and feed back the processing results (for example, vehicle control instructions - for example only) to the terminal device.

[0131] It should be noted that the control method of the autonomous driving vehicle provided in the embodiment of the present invention is generally executed by the server 905. Accordingly, the control device of the autonomous driving vehicle is generally set in the server 905.

[0132] It should be understood that Figure 9 The number of terminal devices, networks and servers in the embodiment is merely illustrative. Any number of terminal devices, networks and servers may be provided as required.

[0133] Reference below Figure 10 , which shows a schematic structural diagram of a computer system 1000 of a terminal device or server suitable for implementing an embodiment of the present invention. Figure 10 The terminal device or server shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.

[0134] like Figure 10 As shown, the computer system 1000 includes a central processing unit (CPU) 1001, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 1002 or a program loaded from a storage unit 1008 into a random access memory (RAM) 1003. Various programs and data required for the operation of the system 1000 are also stored in the RAM 1003. The CPU 1001, the ROM 1002, and the RAM 1003 are connected to each other via a bus 1004. An input / output (I / O) interface 1005 is also connected to the bus 1004.

[0135] The following components are connected to the I / O interface 1005: an input section 1006 including a keyboard, a mouse, and the like; an output section 1007 including devices such as a cathode ray tube (CRT), a liquid crystal display (LCD), and a speaker; a storage section 1008 including a hard disk; and a communication section 1009 including a network interface card such as a LAN card or a modem. The communication section 1009 performs communication processing via a network such as the Internet. A drive 1010 is also connected to the I / O interface 1005 as needed. A removable medium 1011, such as a magnetic disk, an optical disk, a magneto-optical disk, or a semiconductor memory, is installed in the drive 1010 as needed, so that computer programs read therefrom can be installed into the storage section 1008 as needed.

[0136] In particular, according to the embodiments disclosed herein, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, the embodiments disclosed herein include a computer program product comprising a computer program carried on a computer-readable medium, the computer program comprising program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network via the communication section 1009, and / or installed from a removable medium 1011. When the computer program is executed by the central processing unit (CPU) 1001, the above-mentioned functions defined in the system of the present invention are performed.

[0137] It should be noted that the computer-readable medium described in the present invention can be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. A computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of computer-readable storage media can include, but are not limited to, an electrical connection having one or more conductors, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In the present invention, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In the present invention, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. This propagated data signal can take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium that can transmit, propagate, or transport a program for use by or in conjunction with an instruction execution system, apparatus, or device. Program code embodied on a computer-readable medium may be transmitted using any suitable medium, including but not limited to wireless, wireline, optical fiber cable, RF, or any suitable combination thereof.

[0138] The flowcharts and block diagrams in the accompanying drawings illustrate the possible implementation architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present invention. In this regard, each box in the flowchart or block diagram can represent a module, program segment, or a part of code, and the above-mentioned module, program segment, or a part of code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in an order different from that marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flowchart, and the combination of boxes in the block diagram or flowchart, can be implemented with a dedicated hardware-based system that performs the specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0139] The units or modules involved in the embodiments of the present invention may be implemented in software or hardware. The units or modules described may also be provided in a processor. For example, they may be described as: a processor including an information acquisition module, an instruction generation module, and a vehicle control module. The names of these units or modules do not, in certain circumstances, constitute limitations on the units or modules themselves. For example, a vehicle control module may also be described as a "module for controlling the movement of the controlled vehicle according to the vehicle control instructions."

[0140] As another aspect, the present invention further provides a computer-readable medium, which may be included in the device described in the above embodiment; or may exist independently and not be incorporated into the device. The computer-readable medium carries one or more programs, and when the one or more programs are executed by the device, the device is configured to: obtain state information and road structure information of a controlled vehicle at the current moment; input the state information and road structure information of the controlled vehicle into a pre-built integrated traffic environment model to output a vehicle control instruction for the next moment, wherein the integrated traffic environment model is modeled using a virtual potential field based on vehicle position information and road structure information, and uses vehicle dynamics equations as constraints of the integrated traffic environment model; and control the controlled vehicle according to the vehicle control instruction.

[0141] According to the technical solution of an embodiment of the present invention, by obtaining the state information and road structure information of the controlled vehicle at the current moment; inputting the state information and road structure information of the controlled vehicle into a pre-built integrated traffic environment model to output the vehicle control command for the next moment, the integrated traffic environment model is modeled based on the vehicle position information and road structure information using a virtual potential field, and the vehicle dynamics equation is used as the constraint condition of the integrated traffic environment model; and controlling the driving of the controlled vehicle according to the vehicle control command. The technical solution realizes the control of the controlled vehicle based on the model predictive control theory. The virtual potential field is used to model road facilities such as road boundaries, lane lines, and surrounding obstacles or obstructing vehicles, transforming the control problem of the autonomous vehicle on structured roads into a model predictive control problem. The vehicle dynamics equation is used as a constraint in the model predictive control problem. This eliminates the need to design different driving strategies for different driving scenarios and eliminates the need to design driving behavior decisions, trajectory planning, and tracking control in a hierarchical manner. The control system of the autonomous vehicle is simpler, more compact, and capable of handling different traffic environments. At the same time, other constraint functions can be introduced to constrain the safe distance between the controlled vehicle and obstacles to add collision avoidance constraints.

[0142] The above specific embodiments do not limit the scope of protection of the present invention. Those skilled in the art will appreciate that various modifications, combinations, sub-combinations, and substitutions may occur depending on design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention are intended to be included within the scope of protection of the present invention.

Claims

1. A control method for an autonomous driving vehicle, characterized in that: include: Obtain the status information of the controlled vehicle and road structure information at the current moment; Inputting the state information and road structure information of the controlled vehicle into a pre-built integrated traffic environment model to output vehicle control instructions at the next moment, wherein the integrated traffic environment model is modeled based on the vehicle position information and road structure information using a virtual potential energy field, and the vehicle dynamics equation is used as a constraint condition of the integrated traffic environment model; Controlling the controlled vehicle to travel according to the vehicle control instruction; The comprehensive traffic environment model is constructed in the following way: calculating a first sub-objective function corresponding to the road structure potential energy field according to the product of the road structure potential energy field and a second constant; the road structure potential energy field is constructed based on road structure information; Calculating a second sub-objective function corresponding to the control variable at a current moment according to a product of a transposed matrix of the control variable at a certain moment in a prediction domain, a positive definite matrix, and the control variable at the certain moment, wherein the prediction domain includes multiple sampling periods; Calculate the third sub-objective function corresponding to the penalty for speed conflict according to the product of the square of the slack variable and the first constant; Calculate the core function of the objective function according to the sum of the first sub-objective function, the second sub-objective function and the third sub-objective function; Generating the objective function by calculating the minimum value of the sum of the core functions of the plurality of sampling periods; According to the objective function, a comprehensive traffic environment model is constructed with vehicle dynamics equations as constraints.

2. The method according to claim 1, characterized in that The road structure information includes road boundary information, lane line information and obstacle information; The road structure potential energy field is constructed in the following way: Using a virtual potential energy field, modeling is performed based on the vehicle position information and the road boundary information to obtain a road boundary potential energy field; Using a virtual potential energy field, modeling is performed based on the vehicle position information and the lane line information to obtain a lane line potential energy field; Using a virtual potential energy field, modeling is performed based on the vehicle position information and the obstacle information to obtain an obstacle potential energy field; The road boundary potential energy field, the lane line potential energy field and the obstacle potential energy field are superimposed to obtain the road structure potential energy field.

3. The method according to claim 2, characterized in that Modeling based on the vehicle position information and the road boundary information to obtain a road boundary potential energy field includes: Determining the distance between the vehicle and the left and right boundaries of the road based on the vehicle position information and the road boundary information; A road boundary potential energy field is established based on the distance between the vehicle and the left and right boundaries of the road, wherein the road boundary potential energy increases as the distance between the vehicle and the left and right boundaries of the road decreases; and when the vehicle is in the middle of the road, the road boundary potential energy is zero and remains unchanged.

4. The method according to claim 2, characterized in that Modeling is performed based on the vehicle position information and the lane line information to obtain a lane line potential energy field, including: Determining whether the lane line is a solid line and whether adjacent lanes are lanes traveling in the same direction according to the lane line information; If the lane line is a solid line or the adjacent lane is a lane for driving in the opposite direction, the lane line is regarded as a road boundary to establish a lane line potential energy field; If the lane line is a dotted line and the adjacent lane is a lane traveling in the same direction, the shortest distance from the vehicle to the lane line is obtained based on the vehicle position information and the lane line information, and a lane line potential energy field is established based on the shortest distance, wherein the lane line potential energy increases when the vehicle crosses the lane line.

5. The method according to claim 2, characterized in that Modeling is performed based on the vehicle position information and the obstacle information to obtain an obstacle potential energy field, including: A minimum distance between the vehicle and the obstacle is obtained based on the vehicle position information and the obstacle information, and an obstacle potential energy field is established based on the minimum distance, wherein the obstacle potential energy decreases as the minimum distance increases.

6. The method according to claim 2, characterized in that The state information of the controlled vehicle includes: the position information and speed of the controlled vehicle; The constraints also include: state variable constraints, control variable constraints, speed constraints, slack variable constraints, safety distance constraints between the vehicle and the obstacle, and initial state value constraints.

7. The method according to claim 6, characterized in that The state variable constraint is implemented based on the maximum and minimum values ​​of the state variable under the mechanical constraint conditions of the system; The control variable constraint is realized based on the maximum and minimum values ​​of the control variable under the mechanical constraint conditions of the system; The speed constraint is implemented based on the maximum speed, the minimum speed and the slack variable; The safety distance constraint between the vehicle and the obstacle is implemented based on the constraint relationship between the position of the vehicle and each obstacle within the set range and the set threshold; The initial state value constraint is implemented based on the state value of the controlled vehicle at the sampling moment of each sampling period.

8. A control device for an autonomous vehicle, characterized in that: include: The information acquisition module is used to obtain the status information of the controlled vehicle and the road structure information at the current moment; an instruction generation module, configured to input the state information of the controlled vehicle and the road structure information into a pre-built integrated traffic environment model to output the vehicle control instruction for the next moment, wherein the integrated traffic environment model is modeled based on the vehicle position information and the road structure information using a virtual potential energy field, and the vehicle dynamics equation is used as a constraint condition of the integrated traffic environment model; A vehicle control module, configured to control the movement of the controlled vehicle according to the vehicle control instruction; The comprehensive traffic environment model is constructed in the following way: calculating a first sub-objective function corresponding to the road structure potential energy field according to the product of the road structure potential energy field and a second constant; the road structure potential energy field is constructed based on road structure information; Calculating a second sub-objective function corresponding to the control variable at a current moment according to a product of a transposed matrix of the control variable at a certain moment in a prediction domain, a positive definite matrix, and the control variable at the certain moment, wherein the prediction domain includes multiple sampling periods; Calculate the third sub-objective function corresponding to the penalty for speed conflict according to the product of the square of the slack variable and the first constant; Calculate the core function of the objective function according to the sum of the first sub-objective function, the second sub-objective function and the third sub-objective function; Generating the objective function by calculating the minimum value of the sum of the core functions of the plurality of sampling periods; According to the objective function, a comprehensive traffic environment model is constructed with vehicle dynamics equations as constraints.

9. An electronic device, characterized in that: include: one or more processors; a storage device for storing one or more programs, When the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1 to 7.

10. A computer-readable medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images