Automatic Lane Change Collision Avoidance via Virtual Action Lines

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

During automatic lane changes in autonomous driving, vehicles face challenges in determining effective collision avoidance strategies, leading to increased collision risks due to uncertainties in vehicle states and road conditions, which can result in accidents.

Innovation Solution

A method and apparatus for collision avoidance control that acquire driving parameters of the current vehicle and target vehicles, along with road information, to determine travel and return collision safety detection time periods and virtual action lines, enabling the vehicle to execute a collision avoidance strategy based on positional relations and detected abnormalities during lane changes.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If the vehicle stops changing the lane and returns to the original lane upon detecting a target vehicle, then collision avoidance with the target vehicle is improved, but collision risk with vehicles in the original lane increases

Engineering Contradiction:
Improvecollision avoidanceVSAvoidcollision risk
Core Design Contradiction:
ReliabilityVSObject-affected harmful factors

Solution Approach 1:

The system performs preliminary detection of vehicles in both the target lane and original lane before executing lane change. By pre-identifying vehicles in the original lane and calculating their potential collision risk, the system can make informed decisions about whether to proceed with lane change, abort and hold position, or return to original lane, thereby avoiding the situation where returning to original lane causes new collisions.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The lane change control system dynamically adjusts its behavior based on real-time detection of vehicle positions and movement states. Instead of a fixed response pattern, the system continuously evaluates the collision risk with vehicles in the original lane and modifies the lane change execution accordingly - proceeding when safe, aborting when risk is moderate, or holding position when risk is high, thus adapting to changing traffic conditions.

Inventive Principle:
Principle #15Dynamics

2Productivity

If the vehicle executes a lane change operation without comprehensive safety detection, then lane change efficiency is improved, but collision risk increases

Engineering Contradiction:
Improvelane change efficiencyVSAvoidcollision avoidance
Core Design Contradiction:
ProductivityVSReliability

Solution Approach 1:

The safety detection process is segmented into distinct phases: pre-detection of target vehicles in the target lane, detection of vehicles in the original lane, calculation of collision risk for both scenarios, and conditional execution decisions. This segmented approach allows comprehensive safety checks without creating a single bottleneck, as different detection tasks can be processed in parallel, maintaining efficiency while ensuring thorough collision avoidance assessment.

Inventive Principle:
Principle #1Segmentation

3Device complexity

If the vehicle uses a simple lane change decision-making system, then system complexity is reduced, but ability to handle abnormal working conditions worsens

Engineering Contradiction:
Improvedecision-making system complexityVSAvoidabnormal condition handling
Core Design Contradiction:
Device complexityVSAdaptability or versatility

Solution Approach 1:

The decision-making system dynamically adapts its complexity based on the detected working conditions. During normal lane change conditions, the system uses a streamlined decision process. However, when abnormal conditions are detected (such as vehicles in unexpected positions or rapid changes in traffic patterns), the system automatically activates additional detection and evaluation layers, increasing its analytical depth without requiring a permanently complex system architecture.

Inventive Principle:
Principle #15Dynamics

Data Source

PatentEP4063214B1Method and device for collision avoidance control in automatic lane change, vehicle, and storage medium
Publication Date: 2024.12.18 CHINA FAW CO LTD
  • EP4063214B1 patent drawingFigure 1
  • EP4063214B1 patent drawingFigure 2
  • EP4063214B1 patent drawingFigure 3~4

AI summary

The present invention provides a method for collision avoidance control of automatic lane change. The method includes that: driving parameters of a current vehicle, driving parameters of at least one target vehicle, and road information are acquired in response to the current vehicle executing a lane change operation (S101); a travel collision safety detection time period, a return collision safety detection time period, and lane change virtual action lines are determined according to the driving parameters of the current vehicle, the driving parameters of the at least one target vehicle, and the road information (S102); and in response to a lane change abnormality being detected, a collision avoidance strategy is determined according to a positional relation between the current vehicle and the lane change virtual action lines, the travel collision safety detection time period, and the return collision safety detection time period to control the current vehicle to execute the collision avoidance strategy (S103). The present invention further provides an apparatus for collision avoidance control of automatic lane change, a vehicle, and a storage medium.