Lane-Keeping Assist Path Adaptation for Stable Curve Interventions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional lane-keeping assist systems intervene without stabilizing vehicle trajectories, leading to over-correction and tracking anomalies, and fail to adapt to road geometry and driving scenarios for optimal driver comfort and occupant safety.
Innovation Solution
A method and system that adapt vehicle trajectories by adjusting the reference path and intervention criteria based on road geometry and driving scenarios, including attributes like road width and curvature, to stabilize vehicle tracking and mitigate over-correction and anomalies, by applying a desired path offset and controlling interventions to ensure consistency and comfort.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional lane-keeping assist systems intervene by steering the vehicle away from the lane marker, then the vehicle is redirected to stay within the lane, but the vehicle trajectory is not stabilized leading to over-correction and tracking anomalies
Solution Approach 1:
The system performs preliminary actions by predicting future lane crossing events and calculating reference paths in advance. The reference path is generated ahead of time to guide the vehicle smoothly to the target path, preventing over-correction by preparing the trajectory adjustment before the actual lane crossing occurs.
Solution Approach 2:
The system continuously monitors trajectory tracking performance and uses this feedback to determine when to exit or abort interventions. The feedback mechanism tracks whether the vehicle has successfully followed the reference path and stabilized on the target path, allowing the system to exit intervention smoothly when stability is achieved, thereby improving both reliability and driver comfort.
2Adaptability or versatility
If lane-keeping assist systems use fixed intervention criteria, then the system operation is simple, but the system fails to adapt to different road geometries and driving scenarios for optimal driver comfort and occupant safety
Solution Approach 1:
The system applies local quality by customizing intervention behavior according to specific road geometry types and driving scenarios. Different reference path generation strategies and intervention criteria are used for different road conditions (e.g., curved roads, straight roads, narrow lanes), allowing the system to adapt locally to each situation while maintaining overall system manageability through structured classification of scenarios.
Solution Approach 2:
The system changes parameters such as path offset, intervention thresholds, and reference path geometry based on detected road characteristics and driving scenarios. By dynamically adjusting these parameters according to the specific situation, the system achieves high adaptability to different road geometries while maintaining a unified underlying control architecture, thus managing complexity.
3Reliability
If intervention exit is based on simple criteria, then the system response is fast, but over-correction and trajectory tracking anomalies occur
Solution Approach 1:
The system uses feedback on trajectory tracking performance to determine intervention exit timing. Instead of using simple fixed-time or fixed-position criteria, the system continuously evaluates whether the vehicle has successfully tracked the reference path and stabilized on the target path. This feedback-based exit criterion ensures accurate trajectory tracking while allowing timely intervention termination, balancing reliability and time loss.
4Adaptability or versatility
If the reference path is aligned with lane center, then the vehicle positioning is straightforward, but the system cannot optimize for different road geometries and driving scenarios
Solution Approach 1:
The system applies local quality by adjusting the reference path offset from the lane center based on specific road geometry types and driving scenarios. For example, on curved roads or in narrow lanes, the reference path may be offset from the center to optimize vehicle positioning and comfort. This scenario-based offset adjustment allows the system to adapt to different conditions while maintaining a unified reference path generation framework.
Data Source
AI summary
In various embodiments, methods, systems, and vehicle apparatuses are provided. A method for implementing a lane-keeping assist unit of a vehicle by receiving information of a plurality of road geometries, and driving scenarios wherein at least one driving scenario is combined with a target path that is parallel and biased from a lane center by a desired path offset, and a reference path for guiding the vehicle to merge with the target path; adapting the reference path with control based on a selected road geometry and driving scenario; adjusting the desired path offset by considering lane markings during an intervention for an inner curve, an outer curve and a straight road; controlling the vehicle trajectory for enabling the vehicle to track the reference path; exiting the intervention once a trajectory tracking performance by the vehicle is confirmed; and aborting once an instability of the trajectory tracking performance is confirmed.