Autonomous Vehicle Drive Control Using Oncoming Traffic Light Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing autonomous driving technologies struggle to assess the color of an illuminated traffic light when it is not visible from the host vehicle, due to curved roads or other obstructions, leading to incomplete vehicle control.
Innovation Solution
A vehicle control method that estimates the color of the illuminated traffic light based on oncoming vehicle information when the traffic light is not visible, and adjusts the drive source output accordingly to implement eco-driving control.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If the traffic light color is determined using only onboard camera, then the measurement system is simple, but the traffic light color cannot be assessed when the traffic light is not visible from the host vehicle
Solution Approach 1:
The patent uses oncoming vehicles as intermediary carriers to transmit traffic light color information. When the host vehicle's camera cannot directly observe the traffic light, the system captures images of oncoming vehicles that have passed the traffic light, extracts the traffic light color from those images, and uses it for control decisions. This mediator approach resolves the contradiction by maintaining simple onboard equipment while improving assessment reliability through external information sources.
2Reliability
If the traffic light color is estimated based on oncoming vehicle information, then the traffic light color can be assessed even when not visible, but the information processing complexity increases
Solution Approach 1:
The system performs preliminary actions by continuously capturing and storing images of oncoming vehicles before the host vehicle reaches the traffic light. When the traffic light becomes invisible, the pre-captured images containing traffic light color information are immediately available for processing, eliminating the need for complex real-time estimation algorithms and reducing processing complexity while maintaining reliability.
Solution Approach 2:
The system establishes a feedback mechanism where traffic light color information is continuously updated based on oncoming vehicle images. The control unit monitors the visibility of the traffic light and switches between direct observation mode and indirect estimation mode, creating a closed-loop feedback system that automatically selects the most reliable information source without requiring complex processing.
3Loss of time
If the vehicle maintains current speed, then the travel time is short, but the fuel efficiency is reduced when approaching a red traffic light
Solution Approach 1:
The system performs preliminary deceleration before the host vehicle reaches the traffic light when a red light is detected or estimated. By reducing speed in advance rather than braking abruptly at the stop line, the system recovers kinetic energy more efficiently and reduces overall energy consumption while still maintaining acceptable travel time. This preliminary action resolves the contradiction between travel time and fuel efficiency.
Data Source
Figure 1
Figure 2~3
Figure 4
AI summary
A vehicle control method for controlling an output of a drive source based on a color of an illuminated signal of a traffic light in a vehicle-advancing direction during travel in autonomous driving, wherein when the color of the illuminated signal of the traffic light in the vehicle-advancing direction cannot be acquired by an onboard camera, the color of the illuminated signal of the traffic light in the vehicle-advancing direction is estimated based on oncoming vehicle information, and the output of the drive source is controlled based on an estimation result.