Virtual Lane Generation for Collision Avoidance

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current autonomous driving systems lack the capability to prevent collisions when a lane or forward vehicle is invisible due to deteriorating weather conditions or external factors, as they do not implement a virtual lane or vehicle to avoid accidents.

Innovation Solution

A collision avoidance method and apparatus using GPS, navigation systems, and vehicle information to generate virtual lanes and vehicles, allowing the autonomous vehicle to perform control operations and avoid collisions by fusing data from multiple vehicles and considering road curvature.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If sensors are used to detect lanes and forward vehicles, then collision avoidance capability is improved under normal conditions, but the system fails when lanes or vehicles become invisible due to weather deterioration or external factors

Engineering Contradiction:
Improvecollision avoidance capabilityVSAvoidvisibility loss due to weather conditions
Core Design Contradiction:
ReliabilityVSObject-affected harmful factors

Solution Approach 1:

The patent introduces GPS information and vehicle specification information as intermediary data sources. When sensor detection fails due to poor visibility, the system uses GPS coordinates and vehicle specs from multiple vehicles as mediators to construct virtual lane and vehicle models, enabling collision avoidance without direct visual detection

Inventive Principle:
Principle #24Intermediary (Mediator)

Solution Approach 2:

The patent creates virtual copies of lanes and vehicles using GPS data and vehicle specifications. Instead of relying on direct sensor detection of physical lanes and vehicles, the system generates virtual representations (virtual lane, virtual forward vehicle) that replicate the essential spatial and dimensional characteristics needed for collision avoidance

Inventive Principle:
Principle #26Copying

2Reliability

If virtual lanes are generated using GPS and vehicle specification information, then collision avoidance is enabled under invisible conditions, but system complexity increases

Engineering Contradiction:
Improvecollision avoidance capabilityVSAvoidsystem architecture
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent makes the collision avoidance system multi-functional by enabling it to operate in both normal visibility conditions (using sensor detection) and poor visibility conditions (using virtual construction). The same system architecture handles both detection modes, eliminating the need for separate systems and reducing overall complexity despite the added virtual construction capability

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

Solution Approach 2:

The system uses publicly available GPS information and vehicle specification data that vehicles already possess and transmit. Instead of requiring additional specialized sensors or infrastructure, the system serves itself by utilizing existing data sources that are already part of the vehicle's normal operation

Inventive Principle:
Principle #25Self-service

Data Source

PatentUS20230182722A1Collision avoidance method and apparatus
Publication Date: 2023.06.15 HYUNDAI MOBIS CO LTD
  • US20230182722A1 patent drawing
  • US20230182722A1 patent drawing
  • US20230182722A1 patent drawing

AI summary

A collision avoidance method and apparatus are provided. The collision avoidance method includes sensing a forward vehicle and a lane of a front road, receiving global positioning system (GPS) information and vehicle specification information from the forward vehicle, generating a virtual lane corresponding to the forward vehicle upon failing to the lane of the front road, and performing a control operation to avoid collision with the forward vehicle based on the generated virtual lane.