Field testing of an intelligent train monitor

Developments such as LEADER and TrainStar have pioneered the concept of on-board train state estimation. Research at the Centre for Railway Engineering has been progressing toward a system that exceeds these capabilities and provides estimates of in-train forces in future time. The result is an in-cabin device that can predict the train speed and longitudinal coupler force at selected train positions for a future time period. The device is known as the Intelligent Train Monitor. The device has been developed with the capability of providing 50 seconds of future predicted data. Inputs to the system include present and future control sequences, Global Positioning System (GPS) location and a track database. The train type, locomotive type and wagon loadings are user selectable. Algorithms within the system fine tune the monitor to match the characteristics of each individual train. The driver can view past and future in-train forces and train velocities along with maps displaying curves and grades. Energy usage is also reported on the interface. The software and Linux operating system of the in-cabin system has been designed to take maximum advantage of hardware advancements. The monitor software is built upon a scalable message-passing framework, providing high performance and robustness for future development. This paper presents details of the in-cabin device and early results from in-field testing.