Lane Fusion System Using Forward and Rear Cameras

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing vehicle lane position systems rely heavily on forward-view cameras, which can be obstructed by leading vehicles or obscured by weather conditions, leading to unreliable operation in adverse situations.

Innovation Solution

A method and system that combines image data from both forward-view and rear-view cameras with vehicle dynamics sensors to compute lane curvature and vehicle position relative to lane boundaries, using mathematical models like Kalman and particle filters for robust lane tracking.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If a forward-view camera is used for lane detection, then the system can provide lane position information under normal conditions, but the system becomes unreliable when the camera is obstructed by leading vehicles or obscured by weather conditions

Engineering Contradiction:
Improvelane detection reliabilityVSAvoidobstruction and obscuration effects
Core Design Contradiction:
ReliabilityVSObject-affected harmful factors

Solution Approach 1:

The patent combines image data from both forward-view and rear-view cameras to create a fused lane detection system. The rear-view camera captures images of lane markings behind the vehicle, and through image processing and coordinate transformation, this data is merged with forward-view camera data to provide redundant lane position information that remains reliable even when one camera is obstructed

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The rear-view camera, traditionally used only for backup assistance, is repurposed to also provide lane detection functionality. This multi-functional use of the rear-view camera allows the system to maintain lane detection capability under conditions where the forward-view camera fails, improving overall system reliability without adding dedicated hardware

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

2Reliability

If only forward-view camera data is used for lane sensing, then the system structure remains simple, but the system cannot operate when forward-view imaging conditions are unfavorable

Engineering Contradiction:
Improvesystem operabilityVSAvoidcamera system complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The rear-view camera serves dual purposes: its traditional backup assistance function and a new lane detection function. This allows the system to improve reliability by utilizing existing hardware for multiple functions rather than adding dedicated lane detection equipment

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

Solution Approach 2:

The patent introduces an image fusion module that acts as an intermediary between the forward-view and rear-view camera systems. This module processes and combines data from both cameras, enabling the system to leverage multiple data sources while maintaining a unified lane detection output

Inventive Principle:
Principle #24Intermediary (Mediator)

Data Source

PatentUS9090263B2Lane fusion system using forward-view and rear-view cameras
Publication Date: 2015.07.28 GM GLOBAL TECHNOLOGY OPERATIONS LLC
  • US9090263B2 patent drawing
  • US9090263B2 patent drawing
  • US9090263B2 patent drawing

AI summary

A method and system for computing lane curvature and a host vehicle's position and orientation relative to lane boundaries, using image data from forward-view and rear-view cameras and vehicle dynamics sensors as input. A host vehicle includes cameras at the front and rear, which can be used to detect lane boundaries such as curbs and lane stripes, among other purposes. The host vehicle also includes vehicle dynamics sensors including vehicle speed and yaw rate. A method is developed which computes lane curvature and the host vehicle's position relative to a lane reference path, where the lane reference path is derived from the lane boundaries extracted from a fusion of the front and rear camera images. Mathematical models provided in the disclosure include a Kalman filter tracking routine and a particle filter tracking routine.