Skip to main content

Advanced integrated obstacle and track intrusion detection system for smart automation of rail transport

Periodic Reporting for period 1 - SMART2 (Advanced integrated obstacle and track intrusion detection system for smart automation of rail transport)

Reporting period: 2019-12-01 to 2021-05-31

SMART2 project will build on the results achieved in previous Shift2Rail project SMART by advancement, innovation and implementation of SMART2 on-board long-range all-weather obstacle detection (OD) and track intrusion detection (TID) system. The SMART2 on-board system will be complemented with two new systems to build a holistic OD&TID system. These new systems are advanced SMART2 trackside (TS) and airborne (drone-based) OD&TID systems. All three systems will be integrated into a holistic OD&TID system via interfaces to central Decision Support System (DSS). A holistic approach to autonomous obstacle detection for railways would enable increased detection area including areas behind a curve, slope, tunnels and other elements blocking the train’s view on the rail tracks, in addition to a long-range straight rail-tracks OD. The data recorded by the sensors of each of three OD&TID systems, on-board and TS/airborne will be processed to inform DSS about possible obstacles and track intrusions in their fields of view. DSS will integrate information coming from three OD&TID sub-systems and will make final decision on OD&TID and will suggest possible actions for the train control. SMART2 platform will be flexible and open for interfacing additional OD&TID modules based on future technologies. SMART2 project aims at developing a working prototype of the foreseen holistic OD&TID that will be evaluated in different real-world railway use-case scenarios representing benefit of three OD&TID systems, on-board and TS/airborne. By development of advanced innovative solution for obstacle and track intrusion detection system, SMART2 will contribute to competitiveness, efficiency and operational reliability of railway traffic through the OD&TID automation necessary for GoA 3/4 operation. In order to achieve its main goal of contributing to competitiveness, efficiency and operational reliability of railway traffic, SMART2 will deliver the following measurable objectives:

• Definition of requirements and specifications for the long-range all-weather conditions on-board OD&TID system, as well as for its interfaces to other possible OD&TID systems such as airborne-based systems, in line with the achievement of SIL 4 for the entire GoA 3/4 system;
• Development of TRL 6/7 prototypes of on-board OD&TID system and trackside (TS)/airborne systems;
• Development of TRL6/7 prototype of holistic OD&TID for the railway consisting of on-board and TS/airborne OD&TID subsystems together with decision support and selfdiagnostics subsystems to detect potentially dangerous objects in the train’s path at ranges of up to 2000 m and on potential collision paths with the train
The main results achieved in the first reporting period involve definition of requirements (technical and RAMS-related) for OD&TID in railways, and the definition and analysis of potential use cases for OD&TID systems in railway operations. Also, the definition of high-level specifications of OD&TID sub-systems and interfaces, with respect to identified key requirements, was completed in the first reporting period. Starting from the defined requirements and specifications, the sensors of OD&TID sub-systems were selected/developed. All sensors of the on-board OD&TID sub-system were integrated into the custom made sensor housing to enable easy mount and dismount onto/from different test vehicles (locomotives). Several static field tests were conducted on railtrack site in Niš, Serbia, involving on-board sensors and drone camera and the initial data recording for the purpose of dataset generation was performed. Development of software algorithms for reliable OD for on-board system has progressed, building on the dataset and software of predecessor project SMART and some new results have been published. Development of software algorithms for reliable obstacle detection for drone-based and TS sub-systems has started. The HW infrastructure is fully completed and ready for further field tests to be performed in the next project period to support further development of OD&TID software for autonomous recognition of obstacles on the rail tracks ahead of the locomotive and for the calculation of distances to those obstacles.
SMART2 will progress beyond the state of the art in several aspects.
-The SMART2 approach will be to review and assess the relevant requirements for OD in railways, including existing ones (in regulation or related to current technologies), as well as those relating to emerging technologies (as being defined in past and ongoing R&D projects).
-The SMART2 will research, develop and implement the first holistic approach to autonomous obstacle detection in railways including novel use of drones in obstacle detection. The holistic approach of SMART2 and using of knowledge and expertise from previous project SMART will enable the increased range and detection area. Extension of the dataset generated in SMART through the extensive field tests (static and dynamic) to be performed in SMART2 will provide the basis for knowhow of ideal set-up of holistic OD&TID system. The testing of SMART2 prototype in real environment on challenging track sections (gorges, small radius curves, existing vegetation, etc.) will further generate knowledge and experiences of how ideal system setup should be obtained. The dataset will also contain the map of the track with position and type of all infrastructure elements which will be used to improve the results generated by algorithms for OD&TID.
-The acquisition of data from multiple sources (stationary TS/airborne and on-board systems) will enable the detection of possible threats for the whole track regardless of existence/non-existence of direct visibility of the obstacle/intrusion by the train. Inclusions of multiple sources of data will also lead to less fault positives. The openness of the system and possibility to connect additional trackside systems via the specified interface will allow further reduction of the number of fault positives.
-The addition of active change of cameras’ directions for rail track tracking will allow the on-board OD&TID system to increase detection range and address the curves. The SWIR technology in combination with thermal vision, to be developed in SMART2 WP2, will enable the system to increase
detection in challenging weather conditions.

The SMART2 project aims to deliver an integrated solution for OD&TID in railways that would respond to specific requirement of freight use cases (Shift2Rail IP5), as well as to general requirements identified by Technology Demonstrators in Shift2Rail IP2, which apply to all railway segments. The potential impact of SMART2 on railways is hugely significant, in so far as it forms a key contribution to the route to implementation of autonomously operated trains through the implementation of the outcomes of the Shift2Rail on the European rail network in the future. The social impact of SMART2 is as a key element in the autonomous operation of trains which will increase rail transport capacity with minimal environmental impact; this will increase the competitiveness of rail freight, producing economic and societal benefits, and environmental benefits through modal shift from road.
Vision of SMART2 holistic system for OD&TID with on-board, trackside and drone-based sub-system