Onboard Vision and LiDAR for Train Localization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current Positive Train Control (PTC) systems rely heavily on wayside signaling infrastructure, which is costly, complex, and limited by environmental conditions, making it inefficient for accurate and real-time train localization and control, especially in developing countries and the United States.
Innovation Solution
Implementing a system that uses onboard machine vision systems, such as LiDAR, combined with GPS and remote databases to process data for precise vehicle localization and control, reducing the need for extensive wayside signaling equipment by utilizing local environmental sensors to identify tracks, obstructions, and track conditions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If wayside signaling infrastructure is deployed for PTC, then train localization accuracy is improved, but system cost and complexity increase
Solution Approach 1:
The patent replaces the mechanical/physical wayside signaling infrastructure with an optical/electronic vision-based system. Instead of relying on physical transponders and signaling equipment along the tracks, the system uses cameras and image processing algorithms mounted on the train to detect track features, signals, and environmental markers, thereby eliminating the need for extensive wayside infrastructure while maintaining localization accuracy
Solution Approach 2:
The system creates a digital copy or representation of the physical track environment through computer vision. By capturing images of track features, signals, and surroundings, the system generates a virtual model that can be processed to determine train position and track status without requiring physical wayside equipment at every location
2Reliability
If wayside signaling equipment is deployed throughout the railway network, then train control reliability is improved, but deployment cost and maintenance requirements increase
Solution Approach 1:
The train itself performs the sensing and detection functions that previously required external wayside equipment. The vision system on the train autonomously captures and processes images to determine track status, signal states, and position, making the train self-sufficient and eliminating the need for expensive wayside infrastructure deployment and maintenance
3Device complexity
If GPS is used for train positioning, then system simplicity is improved, but positioning accuracy is insufficient to distinguish between tracks
Solution Approach 1:
The patent merges multiple sensing modalities by combining GPS (for coarse positioning) with computer vision (for fine-grained track identification). The vision system processes images of track features, signals, and environmental markers to provide precise track distinction and localization, while GPS provides overall position context, creating a complementary hybrid positioning system
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
This solution enables accurate and reliable train localization and control without relying on extensive wayside infrastructure, improving safety and reducing costs by using onboard sensors to process data for real-time tracking and operation.
Implementation Method 1
Local environment sensors, which may include a machine vision system such as LiDAR
Data Source
AI summary
Methods and apparatus for real time machine vision and point-cloud data analysis are provided, for remote sensing and vehicle control. Point cloud data can be analyzed via scalable, centralized, cloud computing systems for extraction of asset information and generation of semantic maps. Machine learning components can optimize data analysis mechanisms to improve asset and feature extraction from sensor data. Optimized data analysis mechanisms can be downloaded to vehicles for use in on-board systems analyzing vehicle sensor data. Semantic map data can be used locally in vehicles, along with onboard sensors, to derive precise vehicle localization and provide input to vehicle to control systems.