Rail Vehicle Speed Sensing Using Dual Trackside Image Matching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for determining the speed of track-guided vehicles suffer from cumulative errors during extended periods of operation, especially in relative positioning, and are costly due to complex implementations.
Innovation Solution
The method involves capturing images of the route and its surroundings using imaging sensors at both ends of the vehicle, identifying characteristic image areas, and calculating speed based on the time difference between image captures, utilizing known sensor distances and feature points, without requiring object recognition.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If complex odometry methods using balises, radar, GPS, and odometers are used, then positioning accuracy is improved, but device complexity and cost increase
Solution Approach 1:
The patent extracts the essential function of speed measurement from complex odometry systems by using only imaging sensors to capture characteristic image areas. This eliminates the need for balises, radar, GPS, and odometers while maintaining measurement accuracy through a simplified system that relies on image processing and feature point tracking.
Solution Approach 2:
The patent uses imaging sensors to create visual copies of characteristic areas along the track. By capturing and comparing sequences of images featuring distinct visual patterns, the system determines speed and position without physical contact with the track, replacing complex physical measurement devices with optical copying and analysis.
2Reliability
If relative positioning is used for extended periods, then positioning is maintained when absolute positioning is unavailable, but cumulative measurement errors increase
Solution Approach 1:
The patent implements feedback by continuously capturing sequences of images and comparing characteristic image areas across multiple frames. This continuous visual feedback loop allows the system to maintain accurate speed and position determination over extended periods without drift, as each new image is compared against previous images to correct any accumulating errors.
Solution Approach 2:
The system maintains continuous speed measurement by constantly capturing image sequences and tracking characteristic areas. This uninterrupted visual monitoring ensures that positioning accuracy is maintained throughout the entire measurement period without the error accumulation that occurs in discrete measurement systems.
3Productivity
If conventional odometry with speedometers is used, then speed measurement is obtained, but measurement accuracy deteriorates due to wheel slippage
Solution Approach 1:
The patent replaces the mechanical speedometer system with an optical imaging system. Instead of measuring wheel rotation mechanically, the system uses imaging sensors to capture and analyze characteristic image areas along the track, eliminating the mechanical connection that is susceptible to wheel slippage and improving measurement accuracy.
Data Source
Figure 1~2
Figure 3~4
Figure 5
AI summary
The invention comprises a method for determining the speed of a track-guided vehicle (FZ) traveling on a track, in which a sequence of images of the track and its surroundings is acquired and stored by imaging sensors (SN), and the respective acquisition times of the images are stored in such a way that they are assigned to the respective images and characteristic image areas (CBB) are recognized in the images. Of the imaging sensors (SN), a first one is mounted at one end of the vehicle (FZ) and a second one at the other end of the vehicle (FZ). A sensor distance (DSTsen) is stored between the first imaging sensor (BS1) and the second imaging sensor (BS2).As soon as a specific characteristic area detected in a first image captured by the first imaging sensor (BS1) is also detected in a second image captured by the second imaging sensor (BS2), the vehicle's speed (FZ) is calculated taking into account the time difference between the capture times of the respective images and the sensor spacing (DSTsen), and a speed signal representing this speed is generated. The system also includes a vehicle, a computer program, and a computer-readable storage medium.