Autonomous Vehicle Traffic Light Fusion for Reliable Intersection Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current traffic light detection systems for autonomous vehicles face challenges in accuracy and robustness, particularly in urban environments, due to limitations in sensor-based and vehicle-to-infrastructure (V2I)-based approaches, including slower update rates, reliance on infrastructure communication, and susceptibility to errors or interference.
Innovation Solution
A system that combines sensor-based and V2I-based traffic light detection methods, using processors to perform fusion operations, selecting or generating a final detection output based on confidence levels, and employing learning-based classifiers to enhance accuracy and reliability, thereby improving the vehicle's ability to navigate intersections safely.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If sensor-based traffic light detection is used, then the system can operate independently of infrastructure communication, but the detection accuracy and reliability are reduced due to susceptibility to errors and interference
Solution Approach 1:
The patent combines sensor-based detection and V2I-based detection into a unified system that fuses data from both sources. The sensor-based detector provides independence from infrastructure while the V2I-based detector enhances reliability through direct communication with traffic light controllers. The fusion module integrates both detection results to achieve both independence and reliability simultaneously.
2Reliability
If V2I-based traffic light detection is used, then the detection accuracy is improved through direct infrastructure communication, but the system becomes dependent on infrastructure communication and slower update rates
Solution Approach 1:
The fusion module acts as an intermediary that receives and reconciles data from both V2I communication and sensor-based detection. When V2I data is available, it provides accurate traffic light state information. When V2I communication is unavailable or slow, the sensor-based detector serves as a backup, ensuring continuous operation without being bottlenecked by infrastructure update rates.
3Measurement precision
If multiple detection methods are combined, then the accuracy and robustness of traffic light detection is enhanced, but the system complexity increases due to fusion operations and multiple classifiers
Solution Approach 1:
The system is segmented into distinct functional modules: a sensor-based traffic light detector, a V2I-based traffic light detector, a fusion module, and a control module. Each module has a specific function and can be developed, tested, and maintained independently. This modular architecture reduces overall system complexity while enabling the benefits of multiple detection methods.
4Adaptability or versatility
If sensor-based detection is used, then infrastructure independence is achieved, but the update rate is slower compared to V2I-based detection
Solution Approach 1:
The system dynamically switches between V2I-based detection and sensor-based detection based on availability and performance. When V2I communication is available and providing updates, it is used for its higher update rate. When V2I communication is unavailable or slow, the system transitions to sensor-based detection, maintaining infrastructure independence while adapting to current conditions to optimize update rate.
Data Source
AI summary
Systems and methods for controlling the operation of an autonomous vehicle are disclosed herein. One embodiment performs traffic light detection at an intersection using a sensor-based traffic light detector to produce a sensor-based detection output, the sensor-based detection output having an associated first confidence level; performs traffic light detection at the intersection using a vehicle-to-infrastructure-based (V2I-based) traffic light detector to produce a V2I-based detection output, the V2I-based detection output having an associated second confidence level; performs one of (1) selecting as a final traffic-light-detection output whichever of the sensor-based detection output and the V2I-based detection output has a higher associated confidence level and (2) generating the final traffic-light-detection output by fusing the sensor-based detection output and the V2I-based detection output using a first learning-based classifier; and controls the operation of the autonomous vehicle based, at least in part, on the final traffic-light-detection output.