Traffic Light Cycle Prediction for Autonomous Vehicle Navigation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles face challenges in navigating due to the sheer volume of data from sensors and traditional mapping technologies, which can limit navigation accuracy and efficiency.
Innovation Solution
The use of cameras to analyze visual information for lane marking, directional arrows, traffic lights, and free space mapping, combined with a server-based navigation model, to enhance autonomous vehicle navigation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional mapping technology is used for autonomous vehicle navigation, then navigation coverage is provided, but data storage volume and processing complexity increase significantly
Solution Approach 1:
The patent extracts only the essential navigation elements (lane marks, directional arrows, traffic lights, free spaces) from the complete traditional map data. By selectively extracting and mapping only these critical features rather than storing entire high-resolution maps, the system reduces data volume while maintaining navigation accuracy.
Solution Approach 2:
The navigation model is segmented into discrete, independently mappable elements such as lane marks, directional arrows, traffic lights, and free spaces. Each element can be detected, mapped, and updated separately by individual vehicles, allowing distributed data collection without requiring centralized storage of complete map datasets.
2Reliability
If complete map data is stored and updated continuously, then navigation accuracy is maintained, but processing time and computational resources increase
Solution Approach 1:
The system performs partial mapping by focusing only on essential navigation elements rather than processing complete map data. Vehicles detect and map only lane marks, directional arrows, traffic lights, and free spaces that are immediately relevant to navigation, avoiding unnecessary processing of unrelated geographic information.
Solution Approach 2:
Each autonomous vehicle independently detects, maps, and updates navigation elements in its own environment using its sensors and processors. The vehicle serves its own navigation needs by creating and maintaining its own navigation model, eliminating the need for centralized map processing and distribution.
3Adaptability or versatility
If real-time environmental data is processed for navigation decisions, then navigation adaptability improves, but data processing complexity increases
Solution Approach 1:
The system processes environmental data locally at each vehicle, creating a personalized navigation model based on the specific road segment being traversed. Each vehicle adapts its navigation model to local conditions (specific lane marks, arrows, traffic lights, and free spaces) rather than processing global environmental data, reducing processing complexity while maintaining adaptability.
Data Source
AI summary
A system for autonomous vehicle navigation. The system includes at least one processor programmed to: receive one or more images representative of an environment of a host vehicle; identify a representation or indicator of a traffic light in the images; and determine a state of the traffic light based on the analysis of the images. The one more process may also be programmed to receive an autonomous vehicle road navigation model including associations between traffic lights and trajectories collected from multiple vehicles The processor is also configured to make determinations of traffic lights state, relevancy of traffic lights to host vehicle trajectory, and navigation actions (in response to the determination of a traffic light being relevant). The system is also configured to cause one or more actuator systems associated with the host vehicle to implement determined navigational actions.