Content area
Full Text
Abstract - The Department of Electrical and Computer Engineering at The University of Alabama in Huntsville (UAH) is developing UAH OnTrack, a system for creating location and velocity aware model trains for teaching about software system safety. UAH OnTrack mimics capabilities of the congressional mandated Positive Train Control system, which will allow centralized analysis and control of US trains in case of imminent danger. The system provides location and velocity information to model trains in GPS challenged environments such as real trains might find in tunnels, urban environments and underground rail yards and stations. It also provides an opportunity to develop advanced scheduling algorithms mimicking the properties of the US airspace Required Navigation Performance capabilities by allowing trains to follow more closely than the traditional block scheduling system. It also provides a platform for development and verification of robust algorithms for monitoring of system safety and security.
Keywords: Positive train control, software safety engineering, DO-178, inertial measurement unit, smart train
1 Introduction
Over the past few years, the Electrical and Computer Engineering Department at The University of Alabama in Huntsville has been developing a Software Safety and Security Laboratory for teaching software safety engineering. Students utilizing the laboratory have developed train scheduling software using an aviation safety standard, DO- 178B to create high reliability software for scheduling trains [1]. Like the current system in use in the US, the system used block scheduling which only requires knowledge of a train's location to the nearest block. This is typical of the system in use in the US on real trains. However, in the Rail Safety Improvement Act of 2008, Congress mandated that a supervisory safety system be created with a centralized authority to detect and manage possibly dangerous conditions such as a halted train, an out of position train, or a train that ran a signal [2]. These systems are primarily based on GPS and as a result can only be used in areas with GPS coverage. GPS-based systems fail in challenged areas such as tunnels and underground yards as found in inner cities. Moreover, a control system that relies only upon GPS data may be spoofed by an attacker. To ameliorate these risks some systems plan on using inertial measurement units (IMU) to...