Autonomous Stopped-Vehicle Avoidance Using Taillight Recognition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles face challenges in navigating around stopped vehicles within their driving lane without manually switching to manual driving mode, leading to potential traffic congestion and safety risks.
Innovation Solution
An autonomous driving method that utilizes taillight recognition information to determine whether to avoid a stopped vehicle, sets an appropriate avoidance method (either returning to the original lane or performing lane changing), and generates an avoidance path based on a predetermined time point to ensure safe navigation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If autonomous vehicle maintains appropriate distance from stopped vehicle using distance maintenance function, then collision avoidance is achieved, but traffic congestion occurs and productivity decreases
Solution Approach 1:
The system dynamically switches between distance maintenance mode and avoidance path execution mode based on real-time conditions. When a stopped vehicle is detected, the system transitions from passive distance maintenance to active avoidance navigation, allowing the vehicle to adapt its behavior dynamically rather than remaining static in one mode.
Solution Approach 2:
The system performs preliminary detection of stopped vehicles using taillight recognition before executing avoidance maneuvers. By identifying stopped vehicles early through taillight detection and determining avoidance necessity in advance, the system can plan and execute avoidance paths proactively rather than reactively, improving traffic flow while maintaining safety.
2Ease of operation
If driver manually avoids stopped vehicle by switching from autonomous to manual driving mode, then avoidance capability is improved, but system complexity and ease of operation deteriorate
Solution Approach 1:
The autonomous driving system performs avoidance maneuvers autonomously without requiring driver intervention. The system detects stopped vehicles, determines avoidance necessity, generates avoidance paths, and executes maneuvers all through self-service automation, eliminating the need for drivers to manually switch modes and reducing operational complexity.
Solution Approach 2:
The autonomous driving system integrates multiple functions including stopped vehicle detection, avoidance necessity determination, path generation, and maneuver execution into a single unified system. This multi-functionality allows the system to handle various driving scenarios (distance maintenance, avoidance, lane changes) through one integrated control apparatus rather than requiring separate manual interventions.
3Difficulty of detecting and measuring
If autonomous vehicle recognizes stopped vehicle using basic detection, then detection capability is achieved, but avoidance determination accuracy deteriorates
Solution Approach 1:
The system uses taillight recognition to detect stopped vehicles by identifying the characteristic red color and pattern of taillights. This color-based detection method provides more accurate identification of stopped vehicles compared to basic obstacle detection, enabling better avoidance determination by distinguishing actual stopped vehicles from other objects.
Data Source
AI summary
Disclosed herein are an autonomous driving method for avoiding a stopped vehicle and an apparatus for the same. The autonomous driving method for avoiding a stopped vehicle is performed by an autonomous driving control apparatus provided in an autonomous vehicle, and includes obtaining taillight recognition information for a stopped vehicle identified ahead of the autonomous vehicle, determining whether the stopped vehicle is to be avoided in consideration of the taillight recognition information, when it is determined that the stopped vehicle is to be avoided, setting an avoidance method in consideration of whether lane returning is to be performed, which is determined based on an autonomous driving task, and setting an avoidance time point corresponding to the avoidance method and controlling the autonomous vehicle to avoid the stopped vehicle by traveling along an avoidance path generated in conformity with the avoidance time point.