Intersection Lane Passability Detection for Self-Driving Vehicles
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles face challenges in determining the passage state of traffic lights at intersections due to the need to process multiple traffic lights and their logical relationships, which can lead to misjudgments from illegal vehicle or pedestrian movements.
Innovation Solution
A method involving acquiring intersection information, including lane data and traffic lights, and matching the historical trajectory of obstacles with lane center lines to determine the lane with the smallest matching error, thereby identifying the passable state of the lane based on the traffic light status.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If a self-driving vehicle uses sensor arrays to detect objects and determine right of way, then the vehicle can autonomously navigate intersections, but the system complexity and computational requirements increase significantly
Solution Approach 1:
The system divides the intersection navigation task into distinct phases: detection phase (identifying objects and their positions), classification phase (determining object types and right-of-way status), and execution phase (controlling vehicle movement). This segmentation allows each subsystem to be optimized independently while maintaining overall automation.
Solution Approach 2:
The system performs preliminary detection and classification of objects at the intersection before the self-driving vehicle makes its navigation decision. By pre-identifying objects and determining right-of-way status in advance, the system reduces real-time computational complexity while maintaining autonomous navigation capability.
2Reliability
If the vehicle stops at every intersection to determine right of way, then safety is improved, but travel time increases significantly
Solution Approach 1:
The system applies partial stopping action by only stopping when necessary to yield to objects with right of way. Instead of stopping at every intersection regardless of conditions, the system selectively stops based on real-time detection of other objects and their right-of-way status, thereby maintaining safety while reducing unnecessary travel time loss.
Solution Approach 2:
The system continuously monitors the intersection environment using sensor arrays, providing real-time feedback on detected objects and their movements. This feedback loop allows the vehicle to adjust its stopping decision dynamically based on current conditions, ensuring safety while minimizing travel time through intelligent decision-making.
3Loss of time
If the vehicle proceeds without stopping to determine right of way, then travel time is reduced, but the risk of collision increases
Solution Approach 1:
The system performs preliminary detection and assessment of the intersection environment before proceeding. By identifying objects and determining right-of-way status in advance, the vehicle can make informed decisions about whether to proceed or stop, reducing collision risk while maintaining efficient travel time through intersections.
Solution Approach 2:
The system uses continuous sensor feedback to monitor objects in the intersection and adjust its proceeding decision in real-time. This feedback mechanism allows the vehicle to proceed efficiently when safe while automatically stopping when detection indicates potential collision risks, thereby balancing travel time reduction with collision prevention.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A method and apparatus for determining passing by a self-driving vehicle, a computer readable storage medium and an electronic device. The method comprises: acquiring intersection information along a driving route of a self-driving vehicle, the intersection information comprising lane data; acquiring the historical track of an obstacle in an intersection during a specific time period in the intersection; matching the historical track of the obstacle with the center line of each lane in the lane data so as to acquire the lane having the smallest matching error; and determining that the lane having the smallest matching error is of a passable status; the lane having the smallest matching error is the lane in which the obstacle is located. The described method enables the determination of passing status at an intersection.