Camera-Based Track Availability Detection for Guided Vehicles
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for determining lane availability for guided vehicles, such as track circuits and axle counters, are either unreliable, costly, or prone to false readings, and do not meet safety standards for SIL2 to SIL4 levels, which require a probability of dangerous failure below 10^-6 to 10^-9 per hour.
Innovation Solution
A real-time lane availability determination system using cameras to capture images of the track downstream of the vehicle, with an image processing and analysis device that recognizes track features and calculates a safety distance, employing invariant dimensional characteristics to ensure accurate availability assessment, and optionally utilizing radar for confirmation and ground signaling for communication.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If track circuit is used to detect guided vehicle presence, then the detection function is provided, but the system requires complex electrical isolation and multiple rails, increasing device complexity and cost
Solution Approach 1:
The patent replaces the electrical track circuit system with an optical camera-based detection system. Instead of using electrical connections through wheels and rails to detect vehicle presence, the invention uses cameras to capture images and process visual information to determine track occupancy, thereby eliminating the need for complex electrical isolation and multiple rails.
Solution Approach 2:
The patent introduces an image processing device as an intermediary between the camera and the track availability determination. This intermediary processes the captured images to extract relevant information about vehicle presence and track conditions, simplifying the overall system architecture compared to direct electrical detection methods.
2Device complexity
If axle counter is used to count axles, then the track availability can be determined without electrical isolation, but the production cost is higher and false counts may occur
Solution Approach 1:
The patent replaces the mechanical axle counting system with an optical image recognition system. Instead of counting physical axles that may generate false signals, the system uses cameras to capture visual information and image processing algorithms to determine vehicle presence and track availability, eliminating false counting issues.
Solution Approach 2:
The patent creates a visual copy of the track scene through camera imaging. By processing this optical copy of the reality, the system can accurately determine vehicle presence without directly interacting with the physical axles or vehicles, thereby avoiding false counts while maintaining reliability.
3Reliability
If video camera system is used to detect track state, then the availability determination is provided, but the system complexity and processing requirements increase
Solution Approach 1:
The patent applies local quality by focusing the image processing on specific regions of interest within the captured images. Instead of processing the entire image uniformly, the system concentrates computational resources on detecting vehicles and obstacles in critical areas of the track, thereby reducing overall processing complexity while maintaining detection reliability.
Data Source
Figure 1~2
AI summary
The present invention relates to a method and a system for determining the availability of a lane, wherein said system is to be provided on a guided vehicle (1) capable of moving on a lane (2) in an upstream-to-downstream direction, and said system includes: at least one camera (31) capable of acquiring at least one image (311) of a downstream section of said lane (2), said downstream section being a portion of the lane extending from a first point (P1) of said lane (2) located downstream from the guided vehicle (1) to a second point (P2) of said lane (2) located downstream from said guided vehicle (1), the distance between the camera (31) and the second point (P2) being greater than the distance between the camera and the first point (P1); at least one image processing and analysis device (32) capable of analyzing each image (311) acquired by each camera (31), locating said downstream section therein, determining a safety downstream distance as well as the status of the availability of the lane (2) between said first point (P1) and said second point (P2).