LIDAR Traffic Light Prediction Using Lane-Based Virtual Boxes
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing traffic infrastructure systems lack the ability to accurately and quickly acquire and utilize traffic data from fixedly installed LIDAR systems, particularly at intersections, for predicting traffic light information.
Innovation Solution
A method and server using fixedly installed LIDAR to detect vehicles and recognize lane lines, allocate metadata to virtual boxes, and predict traffic light information by analyzing start and stop timing information of vehicles within these boxes, with prioritization based on reliability and noise reduction techniques.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If fixedly installed LIDAR is used to detect vehicles and acquire traffic data, then traffic data acquisition capability is improved, but the ability to accurately predict traffic light information deteriorates due to lack of processing methods
Solution Approach 1:
The detection region is divided into multiple virtual boxes along lane lines, with each box tracking vehicle start and stop timing independently. This segmentation allows the system to process traffic data from different locations and directions separately, improving prediction accuracy for each specific lane while maintaining overall system capability.
Solution Approach 2:
The system pre-establishes virtual boxes and metadata for each lane line before vehicles arrive. Start timing information is recorded when vehicles enter virtual boxes and stop timing when they exit, preparing the data structure in advance for efficient traffic light prediction processing.
2Reliability
If start timing and stop timing information of multiple vehicles is collected, then prediction accuracy is improved, but system complexity increases due to noise and reliability variations
Solution Approach 1:
The system calculates reliability for each vehicle's timing information based on trajectory quality and detection consistency. This reliability metric feeds into the prediction process, allowing the system to weight different vehicle data appropriately and filter out noisy measurements, thereby improving accuracy without requiring overly complex processing.
Solution Approach 2:
The system transforms raw timing data into reliability scores and uses these as weighting parameters in the prediction algorithm. By changing the parameter representation from raw timing values to reliability-weighted values, the system simplifies the handling of noisy multi-vehicle data while maintaining prediction accuracy.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Enables accurate prediction of traffic light information by recognizing lane lines and vehicle movements, determining changing orders and durations of traffic lights, and filtering noise to enhance prediction reliability.
Implementation Method 1
LIDAR is a device which precisely illustrates the appearance of a surrounding target object by projecting a laser pulse into the surrounding target object, receiving the light reflected back from the surrounding target object, and thereby measuring the distance, etc. to the surrounding target object
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A method for predicting traffic light information by using a LIDAR is provided. The method includes steps of: (a) on condition that each of metadata has been allocated for each of virtual boxes included in a region covered by the LIDAR, obtaining, by a server, at least part of start timing information and stop timing information of a plurality of vehicles for each of the virtual boxes; and (b) predicting, by the server, each of pieces of the traffic light information respectively corresponding to each of the virtual boxes by referring to at least part of the start timing information and the stop timing information of the vehicles for each of the virtual boxes.