EP-4737245-A1 - LANE CHANGING CONTROL METHOD AND APPARATUS, AND TERMINAL DEVICE AND STORAGE MEDIUM

EP4737245A1EP 4737245 A1EP4737245 A1EP 4737245A1EP-4737245-A1

Abstract

Disclosed in the present application are a lane changing control method and apparatus, and a terminal device and a storage medium. The method comprises: in response to identifying a lane changing instruction, entering a lane changing execution state, and determining a risk level of the current vehicle; and if the risk level meets a first risk condition, selecting a risk target, and controlling the current vehicle and the risk target to execute interactive lane changing. By means of the solution of the present application, when a risk occurs during lane changing, a flexible response can be made, and interactive lane changing with a risk target is executed, thereby completing a lane changing behavior; thus, the solution of the present application can improve the flexibility of a vehicle while ensuring the safety and comfort of a lane changing behavior, thereby effectively improving the lane changing capability.

Inventors

FAN, Qi
WANG, YIZHOU
XU, Shaolin
QIU, JIE
LI, LIYUN

Assignees

Guangzhou Xiaopeng Autopilot Technology Co., Ltd.

Dates

Publication Date: 20260506
Application Date: 20240627

Claims (14)

A lane change control method, comprising: entering a lane change execution state in response to identification of a lane change instruction, and determining a risk level of a current vehicle; and in response to the risk level meets a first risk condition, selecting a risk target and controlling the current vehicle to perform an interactive lane change with the risk target.
The lane change control method according to claim 1, wherein said selecting a risk target and controlling the current vehicle to perform an interactive lane change with the risk target comprises: selecting the risk target; acquiring driving information of the current vehicle and driving information of the risk target; planning an avoidance path based on the driving information of the current vehicle and the driving information of the risk target; controlling the current vehicle to move to an aligned state following the avoidance path; and maintaining the current vehicle in the aligned state, acquiring lane conditions, and planning a driving path based on the lane conditions.
The lane change control method according to claim 2, wherein: the lane conditions comprise target lane conditions and current lane conditions; and the method, after acquiring the lane conditions, further comprises reselecting a subsequent lane based on the target lane conditions and the current lane conditions.
The lane change control method according to claim 1, after entering the lane change execution state and determining the risk level of the current vehicle, further comprising: changing the lane change execution state to a lane change cancellation state and controlling the current vehicle to return to its original lane, in response to the risk level meets a second risk condition; and controlling the current vehicle to perform a normal lane change, in response to the risk level meets a third risk condition.
The lane change control method according to claim 4, wherein said determining a risk level of a current vehicle comprises: acquiring information about a lane change space, information about a preceding vehicle in a target lane, information about a following vehicle in the target lane, and information about a static obstacle; determining a longitudinal distance between the preceding vehicle in the target lane and the current vehicle based on the information about the preceding vehicle in the target lane, and determining a longitudinal distance between the following vehicle in the target lane and the current vehicle based on the information about the following vehicle in the target lane; determining that the risk level meets the first risk condition, in response to the longitudinal distance between the preceding vehicle in the target lane and the current vehicle is within a first threshold range and the longitudinal distance between the following vehicle in the target lane and the current vehicle included in the information about the following vehicle in the target lane is within a second threshold range; determining that the risk level meets the second risk condition, in response to the longitudinal distance between the preceding vehicle in the target lane and the current vehicle is within a third threshold range, or the longitudinal distance between the following vehicle in the target lane and the current vehicle included in the information about the following vehicle in the target lane is within a fourth threshold range, or a static obstacle is present within a preset range in the target lane; and determining that the risk level meets the third risk condition, in response to no preceding vehicle, no following vehicle, and no static obstacle are present within a preset range in the target lane.
The lane change control method according to claim 5, wherein the information about the preceding vehicle in the target lane comprises a relative position, speed, acceleration, time headway, and risk coefficient of the preceding vehicle in the target lane with respect to the current vehicle; and the information about the following vehicle in the target lane comprises a relative position, speed, acceleration, time headway, and risk coefficient of the following vehicle in the target lane with respect to the current vehicle; and said determining a longitudinal distance between the preceding vehicle in the target lane and the current vehicle based on the information about the preceding vehicle in the target lane comprises: calculating a minimum forward following distance based on the relative position, speed, acceleration, and time headway of the preceding vehicle in the target lane with respect to the current vehicle; and calculating the longitudinal distance between the preceding vehicle in the target lane and the current vehicle based on the minimum forward following distance and the risk coefficient of the preceding vehicle; and said determining a longitudinal distance between the following vehicle in the target lane and the current vehicle based on the information about the following vehicle in the target lane comprises: calculating a minimum rearward following distance based on the relative position, speed, accelerations, and time headway of the following vehicle in the target lane with respect to the current vehicle; and calculating the longitudinal distance between the following vehicle in the target lane and the current vehicle based on the minimum rearward following distance and the risk coefficient of the following vehicle.
The lane change control method according to claim 1, before entering the lane change execution state in response to identification of a lane change instruction, further comprising: determining a lane change need based on at least one of a navigation need, a need to overtake a slow-moving vehicle, and a need to bypass an obstacle; determining whether a lane change space is available based on the lane change need; and generating the lane change instruction in response to a lane change space is available.
The lane change control method according to claim 2, where said planning an avoidance path based on the driving information of the current vehicle and the driving information of the risk target comprises: determining a position of a front axle center of the current vehicle based on the driving information of the current vehicle; determining a target lane guidance line of a lane where the risk target is located based on the driving information of the risk target; calculating a lateral offset based on the position of the front axle center of the current vehicle and the target lane guidance line; and determining the avoidance path based on the lateral offset.
A lane change control apparatus, comprising: a determining module, configured to enter a lane change execution state in response to identification of a lane change instruction, and determine a risk level of a current vehicle; and an executing module, configured to select a risk target in response to the risk level meets a first risk condition, and control the current vehicle to perform an interactive lane change with the risk target.
The lane change control apparatus according to claim 9, wherein the executing module is configured to: select the risk target; acquire driving information of the current vehicle and driving information of the risk target; plan an avoidance path based on the driving information of the current vehicle and the driving information of the risk target; control the current vehicle to move to an aligned state following the avoidance path; and maintain the current vehicle in the aligned state, acquire lane conditions, and plan a driving path based on the lane conditions.
The lane change control apparatus according to claim 9, wherein the executing module is further configured to: change the lane change execution state to a lane change cancellation state and control the current vehicle to return to its original lane, in response to the risk level meets a second risk condition; and control the current vehicle to perform a normal lane change, in response to the risk level meets a third risk condition.
The lane change control apparatus according to claim 11, wherein the determining module is configured to: acquire information about a lane change space, information about a preceding vehicle in a target lane, information about a following vehicle in the target lane, and information about a static obstacle; determine a longitudinal distance between the preceding vehicle in the target lane and the current vehicle based on the information about the preceding vehicle in the target lane, and determine a longitudinal distance between the following vehicle in the target lane and the current vehicle based on the information about the following vehicle in the target lane; determine that the risk level meets the first risk condition, in response to the longitudinal distance between the preceding vehicle in the target lane and the current vehicle is within a first threshold range, and the longitudinal distance between the following vehicle in the target lane and the current vehicle included in the information about the following vehicle in the target lane is within a second threshold range; determine that the risk level meets the second risk condition, in response to the longitudinal distance between the preceding vehicle in the target lane and the current vehicle is within a third threshold range, or the longitudinal distance between the following vehicle in the target lane and the current vehicle the information about the following vehicle in the target lane is within a fourth threshold range, or a static obstacle is present within a preset range in the target lane; and determine that the risk level meets the third risk condition, in response to no preceding vehicle, no following vehicle, and no static obstacle are present within a preset range in the target lane.
A terminal device, comprising a memory, a processor, and a lane change control program stored in the memory and executable on the processor, wherein the lane change control program, when executed by the processor, implements the steps of the lane change control method according to any one of claims 1 to 8.
A computer-readable storage medium having a lane change control program stored thereon, wherein the lane change control program, when executed by a processor, implements the steps of the lane change control method according to any one of claims 1 to 8.

Description

CROSS-REFERENCE TO RELATED APPLICATIONS This application claims priority to Chinese Patent Application No. 2023107720680, filed at the China National Intellectual Property Administration on June 27, 2023 and entitled "Lane change control method and apparatus, terminal device, and storage medium", which is incorporated herein by reference in its entirety. TECHNICAL FIELD The present application relates to the technical field of intelligent driving, and in particular to a lane change control method, a lane change control apparatus, a terminal device, and a storage medium. BACKGROUND Lane changing is one of the core capabilities of advanced driver assistance systems and autonomous driving systems. Lane change maneuver planning mainly involves two stages, namely lane change initiation and lane change execution. To ensure safety and comfort of lane change maneuvers, relevant technical solutions generally adopt a strategy of "emphasizing initiation more than execution". Specifically, emphasis is placed on evaluation of the initiation timing of a lane change, and execution of the lane change is allowed only when the longitudinal space is sufficiently safe. Moreover, a lane change is immediately cancelled once a risk appears during the lane changing. Such a strategy exhibits obvious shortcomings in complex scenarios such as urban environments, including conservative initiation and weak lane change capability, which results in low flexibility of vehicles during lane change execution. It is therefore necessary to provide a solution for improving the flexibility of vehicles during lane change execution. SUMMARY To solve all or some of the problems existing in the related technologies, the present application provides a lane change control method, a lane change control apparatus, a terminal device, and a storage medium, aiming to improve the flexibility of vehicles during lane change execution. To achieve the above objective, the present application provides a lane change control method. The lane change control method includes: entering a lane change execution state in response to identification of a lane change instruction, and determining a risk level of a current vehicle; andin response to the risk level meets a first risk condition, selecting a risk target and controlling the current vehicle to perform an interactive lane change with the risk target. In an embodiment, the step of selecting a risk target and controlling the current vehicle to perform an interactive lane change with the risk target includes: selecting the risk target;acquiring driving information of the current vehicle and driving information of the risk target;planning an avoidance path based on the driving information of the current vehicle and the driving information of the risk target;controlling the current vehicle to move to an aligned state following the avoidance path; andmaintaining the current vehicle in the aligned state, acquiring lane conditions, and planning a driving path based on the lane conditions. In an embodiment, the lane conditions include target lane conditions and current lane conditions; and the method, after acquiring the lane conditions, further includes: reselecting a subsequent lane based on the target lane conditions and the current lane conditions. In an embodiment, the method, after entering the lane change execution state and determining the risk level of the current vehicle, further includes: changing the lane change execution state to a lane change cancellation state and controlling the current vehicle to return to its original lane, in response to the risk level meets a second risk condition; andcontrolling the current vehicle to perform a normal lane change, in response to the risk level meets a third risk condition. In an embodiment, the step of determining a risk level of a current vehicle includes: acquiring information about a lane change space, information about a preceding vehicle in a target lane, information about a following vehicle in the target lane, and information about a static obstacle;determining a longitudinal distance between the preceding vehicle in the target lane and the current vehicle based on the information about the preceding vehicle in the target lane, and determining a longitudinal distance between the following vehicle in the target lane and the current vehicle based on the information about the following vehicle in the target lane;determining that the risk level meets the first risk condition, in response to the longitudinal distance between the preceding vehicle in the target lane and the current vehicle is within a first threshold range and the longitudinal distance between the following vehicle in the target lane and the current vehicle included in the information about the following vehicle in the target lane is within a second threshold range;determining that the risk level meets the second risk condition, in response to the longitudinal distance between the preceding vehicle in