Autonomous Train Sensing System for Active Obstacle Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current rail transit systems lack an autonomous obstacle perception ability for trains, leading to increased accident rates due to the inability to detect obstacles in time, especially in driverless modes.
Innovation Solution
A comprehensive autonomous sensing system for trains, comprising an on-board subsystem, a trackside subsystem, a center server subsystem, a dispatch terminal subsystem, and a handheld mobile terminal subsystem, which collaboratively detect obstacles, extend detection ranges, and trigger appropriate braking actions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If manual observation and peripheral systems are used to protect against obstacles, then train operation safety is maintained to some extent, but the train itself lacks autonomous obstacle detection ability, leading to inability to detect obstacles in time in driverless modes
Solution Approach 1:
The patent replaces manual observation and peripheral detection systems with an autonomous obstacle detection system that uses sensors (such as cameras, radar, or laser detectors) mounted on the train to automatically detect obstacles. The detection results are directly transmitted to the train control system, enabling autonomous response without human intervention or reliance on external monitoring systems.
Solution Approach 2:
The train is equipped with its own obstacle detection and control systems, making it self-sufficient in detecting and responding to obstacles. The train autonomously monitors its environment, processes detection data, and executes braking or other protective actions without requiring external assistance from manual observers or peripheral systems.
2Reliability
If traditional contact or non-contact obstacle perception systems are used, then accidents that have occurred can be detected, but the systems cannot predict accidents before they happen
Solution Approach 1:
The patent implements preliminary detection by continuously monitoring the track environment before trains arrive at potential hazard zones. The system identifies obstacles ahead of time and transmits warning information to approaching trains, enabling them to take preventive actions (such as slowing down or preparing to brake) before the obstacle becomes an immediate threat, thus preventing accidents rather than merely detecting them after occurrence.
Solution Approach 2:
The system establishes a real-time feedback loop where detection results from sensors are continuously transmitted to the train control system, which then adjusts train operations accordingly. This closed-loop control enables the train to respond dynamically to detected obstacles, with the control system receiving ongoing information about the obstacle's position and the train's distance to it, allowing for timely and appropriate protective actions.
3Reliability
If drivers visually observe the train operation process to ensure safety, then train operation safety is maintained, but the high labor intensity causes drivers to get tired, reducing response time and concentration ability
Solution Approach 1:
The patent replaces the driver's visual observation function with automated sensor systems that continuously monitor the track environment. These sensors (cameras, radar, or other detection devices) take over the task of detecting obstacles, eliminating the need for drivers to maintain constant visual attention, thereby reducing labor intensity and the risks associated with driver fatigue while maintaining or improving safety.
Solution Approach 2:
The train's obstacle detection and safety monitoring functions are performed autonomously by onboard systems without requiring continuous human intervention. The train essentially monitors itself and its environment, making safety decisions based on sensor data, which eliminates the physical and mental burden on drivers while ensuring consistent and reliable safety monitoring.
4Measurement precision
If video processing and machine learning-based obstacle recognition techniques are used, then obstacle detection accuracy is improved, but the systems rarely describe operations to be performed after obstacles are detected
Solution Approach 1:
The patent integrates the obstacle detection system with the train control system, merging the functions of detection, analysis, and control into a unified system. The detection results from video processing and machine learning algorithms are directly fed into the control system, which automatically executes appropriate responses (such as braking or alerting the driver). This integration ensures that the system is not only capable of accurate detection but also of executing comprehensive operations after obstacle detection, eliminating the gap between detection and action.
Data Source
AI summary
The invention relates to a autonomous sensing system for train, comprising: an on-board subsystem used for recognizing an obstacle in a rail operation region; a trackside subsystem used for making up limitations in detection distance of sensors of the on-board subsystem and enhancing detection in key regions out of sight; a handheld mobile terminal subsystem carried and operated by a train attendant and used for checking, displaying and confirmation; a center server subsystem used for acquiring operating states and warning information of other subsystems and performing combinational logic processing based on global state data; and a dispatch terminal subsystem communicating with the center server subsystem and used for acquiring and displaying operating states of lineside subsystems and devices, displaying the position of a train and the installation position of the trackside subsystem, checking alarm information of the subsystems and managing the devices of the subsystems. Compared with the prior art, the autonomous sensing system for train allows trains to find obstacles actively and takes measures according to the condition of the obstacles.