CN-121989989-A - Method for predicting travel track of vehicle and vehicle

CN121989989ACN 121989989 ACN121989989 ACN 121989989ACN-121989989-A

Abstract

The embodiment of the application provides a vehicle running track prediction method and a vehicle, wherein the method comprises the steps of obtaining vehicle state information of the vehicle and state information of at least one road object on a road where the vehicle is currently running, wherein the distance between the road object and the vehicle is smaller than a preset distance threshold; respectively constructing a first kinematic model and a first cost function of the vehicle based on the vehicle state information, and respectively constructing a second kinematic model and a second cost function of the road object based on the state information of the road object; and predicting the running track of the vehicle in the future period based on the first kinematic model, the first cost function, the second kinematic model and the second cost function to obtain a predicted running track. The method solves the technical problem of low accuracy of the prediction of the running track of the vehicle in the related art.

Inventors

ZHOU YIBO
GUO PENGXIN
CHEN YONGYU

Assignees

奇瑞汽车股份有限公司

Dates

Publication Date: 20260508
Application Date: 20260205

Claims (10)

1. A travel track prediction method of a vehicle, characterized by comprising: Acquiring vehicle state information of a vehicle and state information of at least one road object on a road on which the vehicle is currently running, wherein the distance between the road object and the vehicle is smaller than a preset distance threshold; respectively constructing a first kinematic model and a first cost function of the vehicle based on the vehicle state information, and respectively constructing a second kinematic model and a second cost function of the road object based on the state information of the road object, wherein the first kinematic model is used for representing the motion characteristics of the vehicle on the road, the first cost function is used for restraining the running behavior of the vehicle, the second kinematic model is used for representing the motion characteristics of the road object on the road, and the second cost function is used for restraining the motion behavior of the road object; And predicting the running track of the vehicle in a future period based on the first kinematic model, the first cost function, the second kinematic model and the second cost function to obtain a predicted running track.
2. The method of claim 1, wherein constructing a first kinematic model of the vehicle based on the vehicle state information and constructing a second kinematic model of the road object based on the state information of the road object comprises: Identifying a first state variable of the vehicle based on the vehicle state information and a second state variable of the road object based on the state information of the road object, wherein the first state variable is used for representing the current running state of the vehicle and the second state variable is used for representing the current running state of the road object, and Determining a first control variable of the vehicle for driving the first state variable change and a second control variable of the road object for driving the second state variable change; The first control variable and the first state variable are input into an initial kinematic model to obtain the first kinematic model, and the second control variable and the second state variable are input into the initial kinematic model to obtain the second kinematic model, wherein the initial kinematic model is used for describing dynamics of the vehicle and/or the road object through a fixed degree of freedom.
3. The method of claim 1, wherein constructing a first cost function of the vehicle based on the vehicle state information and constructing a second cost function of the road object based on the state information of the road object comprises: Determining a first running performance index of the vehicle based on the vehicle state information, and determining a second running performance index of the road object based on the state information of the road object, wherein the first running performance index is used for representing running performance of the vehicle, and the second running performance index is used for representing running performance of the road object; The first cost function is constructed based on the first running performance index, and the second cost function is constructed based on the second running performance index.
4. The method of claim 1, wherein predicting a travel trajectory of the vehicle over a future period based on the first kinematic model, the first cost function, the second kinematic model, and the second cost function results in a predicted travel trajectory, comprising: Linearizing the first kinematic model to obtain a first linear kinematic model, linearizing the second kinematic model to obtain a second linear kinematic model, and Performing secondary treatment on the first cost function to obtain a treated first cost function, and performing secondary treatment on the second cost function to obtain a treated second cost function; And predicting a running track of the vehicle in a future period based on the first linear kinematic model, the second linear kinematic model, the processed first cost function and the processed second cost function, so as to obtain the predicted running track.
5. The method of claim 4, wherein predicting a travel trajectory of the vehicle over a future period based on the first linear kinematic model, the second linear kinematic model, the processed first cost function, and the processed second cost function, results in the predicted travel, comprising: Splicing the first linear kinematic model and the second linear kinematic model to obtain a linear kinematic model, and splicing the processed first cost function and the processed second cost function to obtain a target cost function, wherein the linear kinematic model is used for representing interaction characteristics between the vehicle and the road object, and the target cost function is used for representing running performance of the vehicle and the road object in a running process; Carrying out inverse solution on the linear kinematic model and the target cost function to obtain an inverse solution result, wherein the inverse solution result is used for representing control parameters of the vehicle and the road object when the vehicle and the road object respectively reach an expected driving target; And predicting the running track of the vehicle in a future period based on the inverse solution result to obtain the predicted running track.
6. The method of claim 5, wherein performing an inverse solution to the linear kinematic model and the objective cost function to obtain an inverse solution result comprises: determining a start time and an end time of the future period; starting from the end time, carrying out inverse solution on the linear kinematic model and the target cost function according to a preset time step, and obtaining first control parameters respectively corresponding to the vehicle at a plurality of times until the start time, and obtaining second control parameters respectively corresponding to the road object at a plurality of times; And determining the first control parameters corresponding to the vehicle at a plurality of moments respectively as the inverse solving result by using the second control parameters corresponding to the road object at a plurality of moments respectively.
7. The method of claim 6, wherein in solving the linear kinematic model and the objective cost function in an inverse manner, the method further comprises: Determining whether the first control parameter predicted at the current moment meets a first preset condition or not, and determining whether the second control parameter predicted at the current moment meets a second preset condition or not, wherein the first preset condition is used for representing a parameter range corresponding to the first control parameter, and the second preset condition is used for representing a parameter range corresponding to the second control parameter; taking the first control parameter at the current moment as an input at a next moment in response to the first control parameter meeting the first preset condition, and taking the second control parameter at the current moment as an input at a next moment in response to the second control parameter meeting the second preset condition, or Updating the first control parameter at the current moment and taking the updated first control parameter as input at a next moment in response to the first control parameter not meeting the first preset condition, and updating the second control parameter at the current moment and taking the updated second control parameter as input at the next moment in response to the second control parameter not meeting the second preset condition.
8. The method of claim 7, wherein predicting the travel trajectory of the vehicle based on the inverse solution results, resulting in the predicted travel trajectory, comprises: and predicting the running track of the vehicle based on the first control parameters respectively corresponding to the vehicle at a plurality of moments in the inverse solution result to obtain the predicted running track.
9. The method according to claim 1, wherein the method further comprises: Determining a first cost function value of the predicted running track based on the first cost function, and acquiring a second cost function value of the historical predicted running track of the vehicle in a previous prediction period, wherein the first cost function value is used for quantifying the running performance of the vehicle when running according to the predicted running track, and the second cost function value is used for quantifying the running performance of the vehicle when running according to the historical predicted running track; Comparing the first cost function value with the second cost function value to obtain a comparison result; Determining that the first cost function value of the predicted travel track converges and determining the predicted travel track as a target travel track of the vehicle in response to the comparison result indicating that the difference between the first cost function value and the second cost function value is less than or equal to a preset threshold value, or And responding to the comparison result to indicate that the difference value between the first cost function value and the second cost function value is larger than the preset threshold, returning to start execution, namely predicting the running track of the vehicle in a future period based on the first kinematic model, the first cost function, the second kinematic model and the second cost function, obtaining a predicted running track until the first cost function value of the predicted running track converges, and determining the predicted running track as a target running track of the vehicle.
10. A vehicle, characterized by comprising: A memory storing an executable program; a processor for executing the program, wherein the program when run performs the method of any of claims 1 to 9.

Description

Method for predicting travel track of vehicle and vehicle Technical Field The embodiment of the application relates to the technical field of vehicles, in particular to a vehicle running track prediction method and a vehicle. Background With the continuous progress of automatic driving technology, users have put higher demands on safety and comfort. In order for an autonomous vehicle to make the right decisions to meet the needs of safe and comfortable driving, accurate perception and prediction of obstacles in the surrounding environment (including other vehicles, pedestrians, etc.) is required. In the related art, a machine learning prediction method is generally adopted to predict obstacles around a vehicle, but most of the machine learning-based methods stay on a prediction layer, and when facing a dynamic interaction scene, it is often difficult to accurately predict the future behavior of the obstacles, so that the running track of the vehicle cannot be accurately predicted. Aiming at the technical problem of low accuracy of the prediction of the running track of the vehicle, no good solution exists at present. Disclosure of Invention The embodiment of the application provides a vehicle running track prediction method and a vehicle, which are used for at least solving the technical problem of low running track prediction accuracy of the vehicle in the related technology. According to one aspect of the embodiment of the application, a driving track prediction method of a vehicle is provided, and comprises the steps of obtaining vehicle state information of the vehicle and state information of at least one road object on a road where the vehicle is currently driving, wherein the distance between the road object and the vehicle is smaller than a preset distance threshold value, respectively constructing a first kinematic model and a first cost function of the vehicle based on the vehicle state information, respectively constructing a second kinematic model and a second cost function of the road object based on the state information of the road object, wherein the first kinematic model is used for representing the motion characteristics of the vehicle on the road, the first cost function is used for restraining the driving behavior of the vehicle, the second kinematic model is used for representing the movement characteristics of the road object on the road, the second cost function is used for restraining the movement behavior of the road object, and predicting the driving track of the vehicle in a future period based on the first kinematic model, the first cost function, the second kinematic model and the second cost function, so as to obtain a predicted driving track. Optionally, constructing a first kinematic model of the vehicle based on the vehicle state information and constructing a second kinematic model of the road object based on the state information of the road object, comprising identifying a first state variable of the vehicle based on the vehicle state information and identifying a second state variable of the road object based on the state information of the road object, wherein the first state variable is used for representing the current running state of the vehicle and the second state variable is used for representing the current running state of the road object, and determining a first control variable of the vehicle and a second control variable of the road object, wherein the first control variable is used for driving the first state variable to change, the second control variable is used for driving the second state variable to change, inputting the first control variable and the first state variable into an initial kinematic model to obtain the first kinematic model, and inputting the second control variable and the second state variable into the initial kinematic model to obtain the second kinematic model, wherein the initial kinematic model is used for describing dynamics of the vehicle and/or the road object through a fixed degree of freedom. Optionally, constructing a first cost function of the vehicle based on the vehicle state information and constructing a second cost function of the road object based on the state information of the road object includes determining a first running performance index of the vehicle based on the vehicle state information and determining a second running performance index of the road object based on the state information of the road object, wherein the first running performance index is used for representing running performance of the vehicle and the second running performance index is used for representing running performance of the road object, constructing the first cost function based on the first running performance index and constructing the second cost function based on the second running performance index. The method comprises the steps of obtaining a predicted running track, wherein the predicted running track comprises the steps of c