Computer Vision Railway Tie Spike Hole Detection and Injection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing railway maintenance systems require skilled operators to manually detect and fill spike holes, which can lead to operator fatigue and errors, and do not allow for automated detection and filling of other railway features.
Innovation Solution
A computer vision-based railroad tie maintenance vehicle equipped with an imaging device and an injection device that can move vertically and horizontally to automatically detect and fill spike holes, as well as detect other railway features like ties, tie plates, and anchors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If manual detection and filling by skilled operators is used, then flexibility and adaptability are maintained, but operator fatigue and errors occur, reducing reliability
Solution Approach 1:
The system enables automatic detection and filling operations through computer vision algorithms and automated injection mechanisms, eliminating the need for manual operator intervention in detection tasks. The machine autonomously identifies spike holes, determines their locations, and triggers injection operations without continuous human input.
Solution Approach 2:
The patent replaces manual visual detection and manual injection operations with an automated system combining computer vision technology and mechanically-controlled injection devices. The imaging device captures images, software processes them to detect spike holes, and automated mechanisms position and operate injection devices based on detected locations.
2Productivity
If automated injection systems are introduced, then productivity increases, but system complexity increases requiring sophisticated control mechanisms
Solution Approach 1:
The system integrates multiple functions into a unified automated platform: image capture, image processing, spike hole detection, location calculation, and injection control all operate through a single integrated system. The same hardware platform performs both detection and injection tasks, reducing the need for separate specialized systems.
Solution Approach 2:
The patent introduces software as an intermediary layer between the imaging device and injection device. The software processes images to detect spike holes, calculates their locations, and translates this information into control signals for the injection mechanisms, simplifying the overall control architecture.
3Measurement precision
If manual injection methods are used, then equipment simplicity is maintained, but measurement precision and detection accuracy decrease
Solution Approach 1:
The system creates a digital copy of the railway tie structure through image capture and processing. The computer vision software generates a digital representation of the tie, identifies spike hole locations in this digital model, and uses this information to guide physical injection operations, enabling precise location detection without complex mechanical measurement systems.
Solution Approach 2:
The patent replaces manual visual inspection and estimation methods with computer vision-based detection. The imaging device and image processing software provide automated, precise location determination of spike holes, eliminating the imprecision inherent in manual detection while using relatively simple optical hardware.
4Productivity
If a single operator controls multiple guns, then labor efficiency improves, but control precision and response time deteriorate
Solution Approach 1:
The system autonomously performs detection and triggers injection operations automatically based on detected spike hole locations. The automated system responds immediately to detected features without human intervention, eliminating the delay and potential precision loss associated with manual operation of multiple injection devices.
Solution Approach 2:
The system implements a feedback loop where the imaging device continuously monitors the railway infrastructure, the software processes images to detect spike holes, and the injection devices are automatically controlled based on this real-time detection information. This closed-loop system ensures precise positioning and timing of injection operations.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A railroad tie maintenance machine (20) is configured to detect and seal spike holes (22) in a railroad tie (24) formed by the removal of railroad spikes. The vehicle includes a mothership vehicle (32) and a shuttle cart (34) that are connected by a tether (62). The shuttle cart is movable relative to the mothership vehicle. The shuttle cart includes a filler workhead (66) having an imaging device (78), a light source (80), and an injection wand assembly (68). The imaging device is configured to take photographs or video beneath the shuttle cart while it travels along the railway. Using a computer vision system, the location of the ties and the spike holes in the ties are identified, after which the injection wand assembly is moved in lateral and longitudinal direction to overlie a spike hole. Thereafter, the injector can be inserted into the spike hole, after which a chemical solution is used to fill the spike holes.