Vehicle Lane Travel Control Across Missing Lane Boundaries
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing road shape detection methods for autonomous driving struggle to accurately identify lane center lines and boundaries when they are incompletely drawn, faded, or undetectable by cameras, posing a risk of vehicles leaving their designated lane.
Innovation Solution
A travel control method and device that detect lane boundary lines at the vehicle's widthwise ends, calculate and store prescribed positions when lane boundaries can and cannot be detected, and control the vehicle to travel along a path connecting these positions, ensuring lane adherence even in sections with undetectable lane boundaries.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If lane boundary lines are detected by camera to identify lane center lines and boundaries, then autonomous driving assistance can be executed, but in sections where lane boundary lines are incompletely drawn, have disappeared or faded, accurate identification of lane center lines and boundaries becomes difficult
Solution Approach 1:
The system pre-calculates and stores the travel locus (lane center line) during manual driving before autonomous driving begins. This preliminary data is stored in advance so that when lane boundary lines become undetectable during autonomous driving, the pre-stored travel locus can be immediately used to maintain accurate lane positioning without interruption.
Solution Approach 2:
The patent introduces the travel locus obtained during manual driving as an intermediary element that bridges the gap when lane boundary detection fails. Instead of directly relying on camera detection of lane boundaries, the system uses the pre-recorded travel locus as a mediator to infer and maintain lane center line position, enabling continuous lane keeping even when boundary lines are undetectable.
2Adaptability or versatility
If the vehicle relies on real-time detection of lane boundary lines during autonomous driving, then the system can adapt to current lane conditions, but it cannot maintain prescribed lane position when boundary lines are undetectable
Solution Approach 1:
The system performs preliminary detection and storage of the travel locus during manual driving phase before autonomous driving begins. This advance preparation ensures that when the vehicle enters autonomous driving mode and encounters sections with undetectable lane boundaries, the pre-stored travel locus data is already available to maintain lane position reliability.
Solution Approach 2:
The patent creates a copy of the travel locus data obtained during manual driving and stores it for use during autonomous driving. This copying mechanism allows the system to replicate the accurate lane positioning information from manual driving and use it as a reference during autonomous driving when real-time boundary detection fails, ensuring continuous lane position maintenance.
Data Source
AI summary
A travel control device carries out a travel control for a vehicle based on detected lane boundary lines. A first prescribed position, an absolute vehicle position and an absolute vehicle azimuth angle are stored while changing from a state where the lane boundary lines can be detected to a state where the lane boundary lines cannot be detected. A second prescribed position is stored while changing from a state in which the lane boundary lines cannot be detected to a state in which the lane boundary lines can be detected. The host vehicle is controlled to travel along a travel path connecting the first prescribed position and the second prescribed position where a current absolute position and a current absolute azimuth angle of the host vehicle do not deviate by a prescribed value or more from the stored absolute position and the stored absolute azimuth angle.