Lane Line Detection for Intelligent Driving Control

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current lane line detection technologies in automatic driving systems are inadequate in accurately determining the estimated distance and time of a vehicle traveling out of its lane, leading to potential safety issues due to limited precision and complexity in various driving scenarios.

Innovation Solution

A lane line-based intelligent driving control method and apparatus that utilizes a neural network for lane line detection, determining the estimated distance and time of lane departure based on the vehicle's state and detection results, and performs intelligent driving control to prevent accidents.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If current lane line detection technologies are used, then the system can detect lane lines, but the accuracy of determining estimated distance and time of lane departure is insufficient

Engineering Contradiction:
Improveaccuracy of lane departure distance and time determinationVSAvoidcomplexity of detection system
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent segments the lane line detection task into multiple components: lane line detection, vehicle state acquisition, and lane departure risk assessment. By dividing the detection process and using multiple sensors (camera, radar, GPS) to capture different aspects separately, the system improves measurement precision for distance and time estimation without overwhelming complexity in a single detection module

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent transitions from two-dimensional image-based lane line detection to three-dimensional spatial analysis by integrating vehicle state data (position, speed, acceleration) with detection results. This dimensional expansion enables accurate calculation of estimated distance and time to lane departure by combining visual information with kinematic parameters

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

2Adaptability or versatility

If lane line detection is performed in complex driving scenarios, then the system can handle various road conditions, but the precision and complexity become unbalanced

Engineering Contradiction:
Improvecapability to handle complex driving scenariosVSAvoidprecision of lane departure estimation
Core Design Contradiction:
Adaptability or versatilityVSMeasurement precision

Solution Approach 1:

The patent creates a universal lane departure warning system that functions across multiple driving scenarios (curved roads, straight roads, intersections, night conditions) by integrating multiple detection methods and adaptive algorithms. The system adjusts its parameters and sensitivity based on the detected scenario type, maintaining measurement precision while achieving broad adaptability

Inventive Principle:
Principle #6Universality (Multi-functionality)

Solution Approach 2:

The patent dynamically adjusts detection parameters (thresholds, sensitivity, weighting factors) based on the complexity and type of driving scenario. For example, it modifies lane line detection thresholds and vehicle state weighting according to road curvature, traffic conditions, and environmental factors, thereby maintaining precision across diverse scenarios without requiring completely separate systems for each condition

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentUS11314973B2Lane line-based intelligent driving control method and apparatus, and electronic device
Publication Date: 2022.04.26 SHANGHAI SENSETIME INTELLIGENT TECH CO LTD
  • US11314973B2 patent drawing
  • US11314973B2 patent drawing
  • US11314973B2 patent drawing

AI summary

Embodiments of the present disclosure disclose a lane line-based intelligent driving control method and apparatus, and an electronic device. The method includes: obtaining a lane line detection result of a vehicle traveling environment; determining, according to a traveling state of the vehicle and the lane line detection result, an estimated distance of traveling out of the lane line by the vehicle and/or estimated time of traveling out of the lane line by the vehicle; and performing intelligent driving control on the vehicle according to the estimated distance and/or the estimated time. The embodiments of the present disclosure implement lane line-based intelligent control on the vehicle traveling state, thereby helping to improve the driving safety.