An automatic driving vehicle speed decision control method and device and a storage medium

Abstract

The application relates to an automatic driving vehicle speed decision control method and device and a storage medium, wherein the method comprises the following steps: online estimation of execution delays of a power system and a braking system, and then, based on a prediction model, considering speed and acceleration information of a preceding vehicle, real-time optimization of speed decision control instructions (acceleration and deceleration) of the vehicle under the constraint of obstacle avoidance, and transmission of the acceleration and deceleration to the power system and the braking system of the controlled vehicle to realize speed control. Compared with other methods, the application can online estimate different execution delays of the power system and the braking system, has higher control precision, can give higher weight to the braking system in the cost function through decoupling modeling of the power system and the braking system, can make the vehicle less likely to decelerate and reduce braking, effectively improves the comfort and ecology of the vehicle, and has high calculation efficiency and can reduce operation load.

Description

Technical Field

[0001] This invention relates to the field of autonomous driving technology control, and in particular to a method, device and storage medium for speed decision control of autonomous vehicles. Background Technology

[0002] Since the beginning of the 21st century, the rapid development of computer technology has brought revolutionary changes to the automotive manufacturing industry. Intelligent vehicles have become the general trend of transformation in the current automotive manufacturing industry, and autonomous driving is the current focus and hot topic of research in intelligent vehicles.

[0003] Autonomous vehicle decision-making and control technology consists of two parts: lateral and longitudinal. The lateral part is mainly responsible for lane keeping and lane changing behaviors of autonomous vehicles, while the longitudinal part is mainly responsible for car-following behaviors. However, due to the significant differences in control delays between the vehicle's transmission and braking systems, and the lack of separate modeling of the transmission and braking systems in current longitudinal decision-making and control methods, longitudinal decision-making and control suffers from inaccurate control, frequent intermittent braking, and poor comfort and ecological performance. Summary of the Invention

[0004] The purpose of this invention is to provide a method, device, and storage medium for speed decision control of autonomous vehicles, thereby improving control accuracy and ensuring better comfort and environmental friendliness.

[0005] The objective of this invention can be achieved through the following technical solutions:

[0006] A method for speed decision control of an autonomous vehicle includes the following steps:

[0007] S1. Real-time acquisition of historical acceleration and deceleration control quantities and actual quantities of autonomous vehicles. Among them, the control quantity is the acceleration and deceleration value of the control command, and the actual quantity is the acceleration and deceleration value exhibited by the autonomous vehicle after the power system and braking system execute the control command.

[0008] S2. Based on the historical acceleration and deceleration control quantities and actual quantities, construct optimization problems respectively, and estimate the execution delay of the vehicle power system and braking system in real time online.

[0009] S3. Obtain current information about vehicles ahead from sensors located on the autonomous vehicle;

[0010] S4. Calculate the status of the vehicle ahead in the future time period based on the information of the vehicle ahead;

[0011] S5. Obtain the status of the autonomous vehicle, combine it with the status of the vehicle in front in the future time period, build a predictive model to calculate control commands, and send the control commands to the power system and braking system of the autonomous vehicle.

[0012] S6, the powertrain and braking systems execute control commands to achieve real-time speed control of the vehicle;

[0013] S7, Return to S1.

[0014] In the control command, at least one of the control values ​​for acceleration and deceleration is 0.

[0015] Obtain the intersection of the historical acceleration and deceleration control curves with the X-axis. Using the control curve intersection as the center point, shift the pre-configuration time forward to obtain the acceleration control command curve, and shift the pre-configuration time backward to obtain the deceleration control command curve.

[0016] Obtain the intersection of the actual acceleration and deceleration curves with the X-axis. Using the intersection of the actual acceleration and deceleration curves as the center point, shift the pre-configured time forward to obtain the actual acceleration curve. Shift the pre-configured time backward to obtain the actual deceleration curve.

[0017] The control command curves and actual performance curves for acceleration and deceleration are used to estimate execution delay.

[0018] Estimated vehicle powertrain execution delay (acc) τ The optimization problem is expressed as:

[0019]

[0020] Among them, acc real (t) represents the actual acceleration curve, acc com (t) represents the acceleration control command curve.

[0021] Estimate vehicle braking system execution delay dec τ The optimization problem is expressed as:

[0022]

[0023] Among them, dec real (t) represents the actual deceleration curve, dec com (t) represents the deceleration control command curve.

[0024] Step S4 includes the following steps:

[0025] S41. Set the prediction step size Δt and the prediction time domain N to represent the prediction of the state of the vehicle in front within the next Δt*N time period;

[0026] S42. Obtain the current longitudinal position P′0, velocity v′0, and vehicle acceleration a′0 of the vehicle in front.

[0027] S43. Predict the longitudinal road position P′ of the preceding vehicle in the k-th prediction time domain. k :

[0028]

[0029] Step S5 includes the following steps:

[0030] S51. Obtain the current longitudinal position P0 of the autonomous vehicle on the road (in meters); velocity v0 (in meters per second); and actual acceleration. Actual amount of deceleration Unit m / s 2 The initial state vector is in, and At least one of them is 0;

[0031] S52, Let the speed limit on the road be v. max If the unit is m / s, then the expected value of the state vector is N represents the prediction time domain;

[0032] S53. Construct a prediction model;

[0033] S54. Considering driving comfort and efficiency, the objective function of the prediction model is set as follows:

[0034]

[0035]

[0036]

[0037]

[0038] Among them, u k Let u be the control state vector in the k-th prediction time domain. N To predict the control state vector in the time domain N, the state cost matrix Q is a 4*4 diagonal matrix, and the diagonal elements q1, q2, q3, q4 are non-negative; the control cost matrix R1 and the final state cost matrix R2 are both 2*2 diagonal matrices, and the diagonal elements r1, r2, r3, r4 are non-negative.

[0039] S55. Determine the state update equation for the prediction model:

[0040] P k+1 =P k +v k ×Δt

[0041]

[0042]

[0043]

[0044] S56. Given a safety distance of l, determine the obstacle avoidance constraints for the prediction model:

[0045]

[0046] S57. Based on the objective function, obstacle avoidance constraints, and state update equation, construct a prediction model for solving a quadratic programming problem to obtain the predictive control state vector u. k , take u k The first element but These are control commands for the power system and braking system, respectively.

[0047] S58. Send control commands to the power system and braking system of the autonomous vehicle.

[0048] The process of sending control commands to the power system and braking system of the autonomous vehicle involves sending acceleration from the control commands to the power system and deceleration from the control commands to the braking system.

[0049] An autonomous vehicle speed decision control device includes a memory, a processor, and a program stored in the memory, wherein the processor executes the program to implement the method described above.

[0050] A storage medium having a program stored thereon, which, when executed, implements the method described above.

[0051] Compared with the prior art, the present invention has the following beneficial effects:

[0052] (1) The present invention can estimate the different execution delays of the transmission system and the braking system online, and has higher control accuracy.

[0053] (2) This invention separates the transmission and braking systems and performs decoupling modeling. By giving the braking system a higher weight in the objective function, the vehicle is less likely to decelerate and the number of braking points can be reduced, which can effectively improve the comfort and ecological nature of the vehicle.

[0054] (3) The method described in this invention has computational efficiency at the engineering application level and can reduce computational load. Attached Figure Description

[0055] Figure 1 This is a flowchart of the method of the present invention;

[0056] Figure 2 This is a schematic diagram of the control command curves and actual performance curves for acceleration and deceleration. Detailed Implementation

[0057] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments. These embodiments are based on the technical solution of the present invention and provide detailed implementation methods and specific operating procedures. However, the scope of protection of the present invention is not limited to the following embodiments.

[0058] A speed decision control method for autonomous vehicles, such as Figure 1 As shown, it includes the following steps:

[0059] S1. Real-time acquisition of historical acceleration and deceleration control quantities and actual quantities of autonomous vehicles. Among them, the control quantity is the acceleration and deceleration value of the control command, and the actual quantity is the acceleration and deceleration value exhibited by the autonomous vehicle after the power system and braking system execute the control command.

[0060] In the control command, at least one of the control values ​​for acceleration and deceleration is 0.

[0061] S2. Based on the historical acceleration and deceleration control quantities and actual quantities, construct optimization problems respectively, and estimate the execution delay of the vehicle power system and braking system in real time online.

[0062] Obtain the intersection points of the historical acceleration and deceleration control curves with the X-axis. Using the control curve intersection points as the center point, shift forward by a fixed time interval and take the control curve within that time interval to obtain the acceleration control command curve. Shift backward by a fixed time interval and take the control curve within that time interval to obtain the deceleration control command curve.

[0063] Obtain the intersection point of the actual acceleration and deceleration curves with the X-axis. Using the intersection point as the center point, move forward a fixed time interval and take the actual curve within that time interval to obtain the actual acceleration curve. Move backward a fixed time interval and take the actual curve within that time interval to obtain the deceleration curve.

[0064] A schematic diagram showing the control command curves and actual performance curves for acceleration and deceleration is shown below. Figure 2 As shown, the curve is used to estimate the execution delay after it is determined.

[0065] Estimated vehicle powertrain execution delay (acc) τ The optimization problem is expressed as:

[0066]

[0067] Among them, acc real (t) represents the actual acceleration curve, acc com (t) represents the acceleration control command curve.

[0068] Estimate vehicle braking system execution delay dec τ The optimization problem is expressed as:

[0069]

[0070] Among them, dec real (t) represents the actual deceleration curve, dec com (t) represents the deceleration control command curve.

[0071] Application refers to using iterative methods to solve two optimization problems, which yields the execution delay.

[0072] S3. Obtain current information about vehicles ahead from sensors located on the autonomous vehicle.

[0073] In order to collect this information, the autonomous vehicle used in this invention should have a vehicle information collection device.

[0074] S4. Based on the information of the vehicles ahead, calculate the status of the vehicles ahead in the future time period.

[0075] S41. Set the prediction step size Δt = 0.1S and the prediction time domain N = 100 to indicate the prediction of the state of the vehicle in front within the next 10 seconds;

[0076] S42. Obtain the current longitudinal position P′0, velocity v′0, and vehicle acceleration a′0 of the vehicle in front.

[0077] S43. Predict the longitudinal road position P′ of the preceding vehicle in the k-th prediction time domain. k :

[0078]

[0079] S5. Obtain the status of the autonomous vehicle, combine it with the status of the vehicle in front in the future time period, build a predictive model to calculate control commands, and send the control commands to the power system and braking system of the autonomous vehicle.

[0080] S51. Obtain the current longitudinal position P0 of the autonomous vehicle on the road (in meters); velocity v0 (in meters per second); and actual acceleration. Actual amount of deceleration Unit m / s 2 The initial state vector is in, and At least one of them is 0;

[0081] S52, Let the speed limit on the road be v. max If the unit is m / s, then the expected value of the state vector is N represents the prediction time domain;

[0082] S53. Construct a prediction model;

[0083] S54. Considering driving comfort and efficiency, the objective function of the prediction model is set as follows:

[0084]

[0085]

[0086]

[0087]

[0088] Among them, u k Let u be the control state vector in the k-th prediction time domain. N To predict the control state vector in the time domain N, the state cost matrix Q is a 4*4 diagonal matrix, and the diagonal elements q1, q2, q3, q4 are non-negative; the control cost matrix R1 and the final state cost matrix R2 are both 2*2 diagonal matrices, and the diagonal elements r1, r2, r3, r4 are non-negative.

[0089] All values ​​in the cost matrix are actively set according to actual needs. Increasing the size of q4 helps reduce the need for intermittent braking, making the vehicle more stable and ecologically sound.

[0090] S55. Determine the state update equation for the prediction model:

[0091] P k+1 =P k +v k ×Δt

[0092]

[0093]

[0094]

[0095] S56. Given a safety distance of l, determine the obstacle avoidance constraints for the prediction model:

[0096]

[0097] S57. Based on the objective function, obstacle avoidance constraints, and state update equation, construct a prediction model for solving a quadratic programming problem to obtain the predictive control state vector u. k , take u k The first element but These are control commands for the power system and braking system, respectively.

[0098] S58. Send the acceleration in the control command to the power system and the deceleration to the braking system.

[0099] S6, the powertrain and braking systems execute control commands to achieve real-time speed control of the vehicle.

[0100] S7, return to S1 to update the control quantity and actual quantity in real time, so as to achieve continuous control of vehicle speed.

[0101] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0102] The preferred embodiments of the present invention have been described in detail above. It should be understood that those skilled in the art can make numerous modifications and variations based on the concept of the present invention without creative effort. Therefore, all technical solutions that can be obtained by those skilled in the art based on the concept of the present invention through logical analysis, reasoning, or limited experimentation on the basis of existing technology should be within the scope of protection defined by the claims.

Claims

1. A speed decision control method for autonomous vehicles, characterized in that, Includes the following steps: S1. Real-time acquisition of historical acceleration and deceleration control quantities and actual quantities of autonomous vehicles. Among them, the control quantity is the acceleration and deceleration value of the control command, and the actual quantity is the acceleration and deceleration value exhibited by the autonomous vehicle after the power system and braking system execute the control command. S2. Based on the historical acceleration and deceleration control quantities and actual quantities, construct optimization problems respectively, and estimate the execution delay of the vehicle power system and braking system in real time online. S3. Obtain current information about vehicles ahead from sensors located on the autonomous vehicle; S4. Calculate the status of the vehicle ahead in the future time period based on the information of the vehicle ahead; S5. Obtain the status of the autonomous vehicle, combine it with the status of the vehicle in front in the future time period, build a predictive model to calculate control commands, and send the control commands to the power system and braking system of the autonomous vehicle. S6, the powertrain and braking systems execute control commands to achieve real-time speed control of the vehicle; S7, Return to S1; Obtain historical acceleration and deceleration control curves and X The control command curve for acceleration is obtained by shifting the control quantity intersection point forward by the length of the pre-configuration time, and the control command curve for deceleration is obtained by shifting the pre-configuration time backward. Obtain the actual quantity curves of historical acceleration and deceleration and X The intersection of the axes, with the actual intersection point as the center point, is shifted forward by the length of the pre-configured time to obtain the actual performance curve of acceleration, and shifted backward by the length of the pre-configured time to obtain the actual performance curve of deceleration; The control command curves and actual performance curves for acceleration and deceleration are used to estimate execution delay; Estimate vehicle powertrain execution delay The optimization problem is expressed as: in, The actual acceleration performance curve, For acceleration control command curves; Estimate the vehicle braking system execution delay The optimization problem is expressed as: in, The actual deceleration performance curve is shown. For deceleration control command curves; Step S5 includes the following steps: S51. Obtain the current longitudinal position of the autonomous vehicle on the road. Unit: meters; speed ,unit Actual acceleration quantity Actual deceleration ,unit The initial state vector is ,in, and At least one of them is 0; S52, The speed limit for the road is set as follows: ,unit Then the expected value of the state vector is , For prediction in the time domain; S53. Construct a prediction model; S54. Considering driving comfort and efficiency, the objective function of the prediction model is set as follows: in, For the first k A predictive time-domain control state vector To predict the control state vector in the time domain N, the state cost matrix is... It is a 4x4 diagonal matrix, and the diagonal elements It is a non-negative value and increases. Size; Control cost matrix Sum of final state cost matrix Both are 2x2 diagonal matrices, and the diagonal elements It is a non-negative value; S55. Determine the state update equation for the prediction model: S56. Let the safe distance be... Determine the obstacle avoidance constraints for the prediction model: S57. Based on the objective function, obstacle avoidance constraints, and state update equation, construct a prediction model for solving a quadratic programming problem to obtain the predictive control state vector. ,Pick The first element ,but , These are control commands for the power system and braking system, respectively. S58. Send control commands to the power system and braking system of the autonomous vehicle.

2. The method for speed decision control of an autonomous vehicle according to claim 1, characterized in that, In the control command, at least one of the control values ​​for acceleration and deceleration is 0.

3. The method for speed decision control of an autonomous vehicle according to claim 1, characterized in that, Step S4 includes the following steps: S41. Set prediction step size and prediction time domain N , indicating prediction of the future The status of the vehicle ahead within a given time period; S42. Obtain the current longitudinal position of the vehicle ahead on the road. ,speed Vehicle acceleration ; S43, Predicting the vehicle ahead is in... The longitudinal location of the road in the prediction time domain : = 。 4. The method for speed decision control of an autonomous vehicle according to claim 1, characterized in that, The process of sending control commands to the power system and braking system of the autonomous vehicle involves sending acceleration from the control commands to the power system and deceleration from the control commands to the braking system.

5. An autonomous vehicle speed decision control device, comprising a memory, a processor, and a program stored in the memory, characterized in that, When the processor executes the program, it implements the method as described in any one of claims 1-4.

6. A storage medium having a program stored thereon, characterized in that, When the program is executed, it implements the method as described in any one of claims 1-4.

Images