Review of the Federal Railroad Administration’s Research and Development Program (2020)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

65 5 Train Control and Communications The Train Control and Communications (TCC) Division consists of two major research programs on train control and communications and tres- passer and highway–rail grade-crossing safety. In conjunction with this work, the division sponsors projects on modeling and simulation, automa- tion, and intelligent transportation systems (ITS). In general, the division’s research focuses on reducing train collisions with other trains and with people and objects at grade crossings and other railroad rights-of-way. The largest TCC Division program, train control and communications, focuses on furthering the ability of train operations to make more effective use of information and communications technology to improve safety and railroad network efficiencies. This work includes investigations of levels of train automation to aid in the identification of future automation require- ments and standards. Another major area of emphasis is on data collection and analysis tools for monitoring the performance of positive train control (PTC) systems. The division also advances automation, communications, and sensor technologies to enhance PTC. As noted in Chapter 3, the TCC Division shares responsibility for grade-crossing and trespasser safety research with the Human Factors Divi- sion. A focus of the TCC Division’s research is on investigating whether and how ITS, including connected vehicle technologies, can help ensure safer interactions of passenger cars, buses, and trucks with trains at grade cross- ings. The program is also studying how machine learning can help identify risky trespasser behaviors and how the use of technologies such as LiDAR (i.e., laser imaging, detection, and ranging) and drones can help detect and prevent trespassing.

66 REVIEW OF FRA’S R&D PROGRAM To review the TCC Division’s work, the train control and communica- tions subcommittee followed a methodology similar to that of the other subcommittees. First, the subcommittee reviewed relevant incident data, their causes and severity, and whether insights gleaned from such safety data seem to inform project programming and budget allocations. The subcommittee then sought information about other methods and criteria used by the division to identify and prioritize candidate research topics and projects. Additional insights into project selection, relevance, and impacts were obtained from consultations with external parties familiar with work of the division. Finally, the subcommittee reviewed three projects in the division’s portfolio for concrete examples of how research sponsored by the division addresses real safety needs and problems. Based on this review of safety data, information obtained from the discussions with division staff and external parties, and insights from the sampled projects, the chapter concludes with two observations and a recommendation. ROLE OF INCIDENT DATA IN THE IDENTIFICATION OF RESEARCH NEEDS To begin assessing the TCC Division’s two research programs, the train control and communications subcommittee reviewed the Federal Railroad Administration (FRA) data on the frequency and severity of derailments in which the cause was reported to involve a train control or communications factor. Of the 1,300 derailments averaged per year from 2015 to 2019 (see Appendix A, Figure A-1), about 25 (or 2 percent) were derailments caused by train control and communications factors (see Table 5-1). The most common causes, and also the causes leading to the reportable damage, were automatic control system failures in classification yards (including software, switch, and retarder failures) and power switch failures.1 These causes account for nearly 80 percent of the damage reported from derail- ments associated with train control and communications factors from 2015 to 2019 (see Figure 5-1). 1 A classification yard is where rolling stock are assembled into freight trains (i.e., switch- ing). A retarder is a braking device adjacent to the rail installed in a classification yard that reduces the speed of freight cars for switching. TABLE 5-1 Number of Derailments Resulting from Train Control and Communications–Related Causes Reported to FRA 2015 2016 2017 2018 2019 Derailments 31 22 22 25 17 SOURCE: Federal Railroad Administration n.d.b.

TRAIN CONTROL AND COMMUNICATIONS 67 As shown in Table 5A-1 in the chapter’s annex, the TCC Division’s portfolio consisted of about 25 projects during 2018 and 2019. Several projects pertain to automated train operation, largely in the context of PTC assessments. PTC-related projects include track circuit and positive train location research, as well as next generation PTC capabilities such as the advancement of quasi- and full-moving block signaling that would allow trains to run closer together while maintaining required safety margins. Because PTC systems are mandated by Congress to enhance railroad safety, one of the goals of this research is to decrease some of the operational disadvantages of PTC such as reducing message communications failures, overly conservative braking algorithms, and GPS issues that stop or slow trains prematurely or unnecessarily. FIGURE 5-1 FRA-reportable damage from incidents caused by train control and communications–related factors, 2015 to 2019. SOURCE: Federal Railroad Administration 2020b. C la ss ifi ca ti o n y ar d au to m at ic c o n tr o l s ys te m C la ss ifi ca ti o n y ar d a u to co n tr o l s ys te m s w it ch f ai l O th er s ig n al f ai lu re s C la ss ifi ca ti o n y ar d a u to co n tr o l s ys te m r et ar d er f ai lu re R em o te c o n tr o l t ra n sm it te r lo ss o f co m m u n ic at io n s C o m p u te r sy st em co n fi g u ra ti o n /m an ag em en t In te rl o ck s ig n al d is p la ye d f al se p ro ce ed O th er s ig n al f ai lu re s C o m p u te r sy st em d es ig n e rr o r (v en d o r) R ad io c o m m u n ic at io n e q u ip m en t fa ilu re O th er c o m m u n ic at io n eq u ip m en t fa ilu re R ad io c o n tr o lle d s w it ch n o t lo ck ed e ff ec t R ad io c o n tr o lle d s w it ch c o m m u n ic at io n f ai lu re R em o te c o n tr o l tr an sm it te r d ef ec ti ve $4.0 $3.5 $3.0 $2.5 $2.0 $1.5 $1.0 $0.5 $0.0 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% Millions

68 REVIEW OF FRA’S R&D PROGRAM The TCC Division’s research on grade crossings and trespassing is more readily linked to railroad incident data. As discussed in Chapter 3, trespass- ing and grade-crossing incidents caused more than 70 percent of all railroad fatalities from 2015 to 2019 (also see Appendix A, Figure A-2). Table 5-2 provides a more complete count of grade-crossing incidents, including both fatalities and injuries. Human factors, as shown in Figure 5-1, are associated with the most consequential grade-crossing incidents as measured by reportable damage. (High damage incidents tend to involve deaths and injuries.) Driver inat- tentiveness and misjudgment when crossing the track as well as purposeful disregard of crossing warning signs, signals, and gates are the main causes of the most consequential incidents (see Figure 5-2). Grade-crossing research is prominent in the TCC Division’s portfolio, addressing these human factors causes of incidents. For instance, projects on in-pavement grade-crossing lights, enhanced emergency signage, and connected vehicle–crossing warning systems are aimed at reducing incidents caused by driver inattentiveness and misjudgment. Other work in areas such as incursion prevention systems and on gate skirt research also seek to reduce deliberate crossing violations. The division’s research on grade crossings is also accompanied by projects to detect, deter, and reduce tres- passing through means such as elevated grade crossings, law enforcement techniques, and drone surveillance. OTHER CONSIDERATIONS WHEN PROGRAMMING RESEARCH After a train-to-train collision in Chatsworth, California, the National Transportation Safety Board issued a recommendation for railroads to implement the use of PTC to reduce or eliminate a range of potential incidents caused by human error (National Transportation Safety Board 2008). Congress subsequently mandated the use of this technology on all Class I railroad main lines that transport poison- or toxic-by-inhalation hazardous materials and any main lines with regularly scheduled intercity or commuter rail passenger service. The PTC mandate represents a marked TABLE 5-2 Grade-Crossing Fatalities, Injuries, and Reported Incidents, 2015–2019 Incident Type 2015 2016 2017 2018 2019 Fatalities 237 255 271 260 293 Injuries 1,048 853 846 845 805 Incidents 1,784 1,739 1,842 1,883 1,920 SOURCE: Federal Railroad Administration 2020a.

TRAIN CONTROL AND COMMUNICATIONS 69 difference from the other FRA Office of Research, Development, and Tech- nology (RD&T) divisions in that it establishes a long-term research priority for the TCC Division. Thus, in addition to using incident data to inform project selection, as evident in the grade-crossing portfolio, the TCC Division engages in extensive discussions with the railroad industry on issues associated with implementing and operating PTC systems. Division staff reported that they consult regularly with the Association of American Railroad’s (AAR’s) Train Control, Communications, and Operations (TCCO) Committee on PTC technical and operational issues as well as other issues. The TCCO Committee convenes monthly by conference call and at several in-person meetings annually to discuss implementation of PTC, and the RD&T staff regularly attend those calls and meetings. Through these interactions, the TCC Division obtains industry input on and support for its research. In- deed, the TCCO Committee has created a consensus-based, priority-ranked FIGURE 5-2 FRA-reportable damage from incidents at highway–rail grade cross- ings, 2015 to 2019. SOURCE: Federal Railroad Administration 2020b. H ig h w ay u se r in at te n ti ve n es s H ig h w ay u se r d is re g ar d ed c ro ss in g H ig h w ay u se r m is ju d g m en t H ig h w ay u se r ci te d f o r vi o la ti o n H ig h w ay u se r u n aw ar e, en vi ro n m en ta l f ac to r O th er c au se s (h ig h w ay –r ai l c o lli si o n s) Su ic id e (g ra d e cr o ss in g ) H ig h w ay u se r u n ab le t o s to p , w ea th er H ig h w ay u se r im p ai rm en t M al fu n ct io n o f tr ai n ac ti va te d w ar n in g $40 $45 $40 $35 $30 $25 $20 $15 $10 $5 $0 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% Millions

70 REVIEW OF FRA’S R&D PROGRAM research program list through 2030 that it updates and submits to the TCC Division annually (Newcomb 2020). TCC Division staff report that its grade-crossing research portfolio is informed by the results of a workshop sponsored by the division every 3 or 4 years. The workshop, which includes representatives from Operation Lifesaver, state departments of transportation, municipalities, railroads, engineering and supplier firms, and academia, typically yields dozens of ideas for research projects (Harrison and DaSilva 2019). The committee was not apprised of any formal collaboration with FRA’s Office of Railroad Safety or its regional offices in planning PTC- related case study projects. PROGRAM BUDGET ALLOCATIONS The TCC Division’s budget has maintained a funding level of approxi- mately $8 million per year from 2015 through 2019 (see Figure 5-3). Dur- ing this period, the division has generally allocated the funding between the two major programs such that 80 percent is for work on train control and communications and 20 percent on grade-crossing and trespassing research. EXTERNAL VIEWS ON RESEARCH RELEVANCE AND IMPACT The train control and communications subcommittee consulted members of the AAR TCCO Committee asking them about the relevance and impact of the TCC Division’s work. The nine members consulted included individuals from AAR, the American Public Transportation Association, and each of the seven Class I railroads. The members were unanimous in expressing their satisfaction about the relevance and quality of the division’s research. The general sentiment is that the division collaborates effectively with industry and that the research program continues to produce results having high technical quality and potential to improve safety. The relevance of the work was attributed in part to awareness of the mandate for the train control and communications program to focus on PTC, prompting the industry to submit requests for projects that are needed and that align with this focus. By and large, the TCC staff were described as subject matter experts who are involved, engaged, and effective in communication. None of the individuals consulted expressed concern over the program’s funding levels, expressing satisfaction that the division has been able to leverage its invest- ment by involving the railroads and industry suppliers. Some TCCO mem- bers, however, did express concern about lags in the research publication process that could slow the application of useful results in the field, which relies on quickly evolving technologies. They recommended that report

TRAIN CONTROL AND COMMUNICATIONS 71 summaries be made available in advance of a technical report’s publication and that interim reports could precede the release of a final report, when practical. SAMPLED PROJECTS Three projects from the TCC Division’s programs—one from the train control and communications portfolio (on quasi-moving block train con- trol) and two from the grade-crossing portfolio (using connected vehicle technologies for grade-crossing warnings and LiDAR for identifying cross- ing locations)—were sampled for further insight into how projects were selected and their expected impact. Quasi-Moving Block Train Control Railroads have traditionally used circuits that detect trains to divide the track into signal blocks for ensuring the separation of the leading and FIGURE 5-3 FRA TCC Research Division budget by major program, 2015 to 2019. SOURCE: Federal Railroad Administration Office of Research, Development, and Technology 2019d. 1.7 1.6 1.6 1.6 1.6 6.4 6.5 6.5 6.4 6.4 2015 2016 2017 2018 2019 Grade Crossing and Trespassing Train Control and Communications $7 $6 $5 $4 $3 $2 $1 $0 Millions

72 REVIEW OF FRA’S R&D PROGRAM succeeding trains. However, because the blocks’ sizes are usually fixed and determined by worst-case scenarios and safe braking margins, they can reduce track throughput. The quasi-moving block is a concept that breaks the dependency on track circuits for train detection and replaces it with virtual end of train location. Long fixed blocks are replaced by detection of the actual position of the rear of the train, reducing excessive train spacing and thus increasing capacity. This project, with a first phase funded at $400,000, seeks to develop a concept of operations, with industry input and review, describing the technical approach for implementing quasi-moving block train control and to analyze communications, hardware, and software needs and identify potential technology gaps. The project is being performed in collaboration with the AAR Interoperable Train Control Committee as a way to increase capacity on rail lines equipped with Interoperable Electronic Train Manage- ment System–type PTC. Industry has interest in this work as a means of increasing capacity over the current implementation of PTC and ultimately to yield savings from the future elimination of wayside signals. These sav- ings could be substantial for future train control systems. At the time of this review, the first phase report was awaiting publication. Grade-Crossing Violation Warning This project is investigating the use of an interface to connected vehicles that alerts highway users of imminent violation of a grade-crossing pro- tection system. Once a crossing activation has occurred, messages will be transmitted to vehicles in the vicinity of the crossing to alert the drivers, and, if necessary, cause the motor vehicles to stop short of the intersection. The goals of the project are to propose a reference system for in-vehicle driver warning, evaluate the technical feasibility of implementing the pro- posed system, and make recommendations on future data and analysis needs. Research has applied a systems engineering methodology to design, develop, test, and evaluate a prototype application, and demonstrated the potential for leveraging real-time connected vehicle concepts and services to enhance and transform rail crossing safety. Initially funded by the Federal Highway Administration (FHWA), the project is an example of an effort to leverage the latest developments in con- nected vehicle components and technologies developed by past U.S. Depart- ment of Transportation (U.S. DOT) connected vehicle deployment projects. The project has involved coordination and collaboration with Honda R&D Americas; the Transportation Technology Center, Inc.; FHWA; the Federal Transit Administration; and the Federal Motor Carrier Safety Administra- tion (FMCSA).

TRAIN CONTROL AND COMMUNICATIONS 73 Enhanced Humped Crossing Database Using LiDAR Highway–rail crossings are continually changing owing to roadway and railway improvements and because the surface approaching the railway crossing will change over time. Of particular interest is the profile of the roadway out of concern that highway vehicles may become struck on the track. The American Association of State Highway and Transportation Officials recommends that the surface of the highway be not more than 3 inches higher or lower within 30 feet of the top of the nearest rail. Apart from this guidance, however, there is currently no formula or threshold for determining if a highway–rail grade crossing presents a risk for low-ground clearance vehicles. Because the actual risk for vehicles to become stuck on tracks has more to do with the rate of change in the roadway grade than its magnitude, accurate three-dimensional models of crossings are needed to properly assess the risk. This project is investigating a LiDAR-based solution to measure all ap- proach profiles to grade crossings. LiDAR equipment installed on a railcar, such as a track geometry car, collects surface profile data at and around the grade crossing for post-processing. Should the system identify a profile that could cause a vehicle to become stuck while traversing the crossing, an alarm would be generated to alert the railroad and approaching vehicles to protect the crossing until repairs are made. The data collected will also enhance FRA’s inventory of highway–rail crossings. To date, nearly 8,000 crossings have been surveyed by the system. Moreover, because the system runs on a continuous basis, it collects data for the entire railway surveyed, which may have other valuable applications such as for asset management, object recognition, and signal sighting. OBSERVATIONS The TCC Division’s Work to Improve the Operational Capabilities and Performance of PTC Aligns with Its Safety Mission The TCC Division’s major emphasis on PTC is responsive to the public interest of furthering the use of this safety technology and railroad interests in expanding the operational capacity-enhancing benefits of PTC while reducing deployment costs and any operational disadvantages. In recog- nizing the importance of efficient, precise, and reliable PTC operations to achieving the safety promise of these systems, the division has forged strong connections with the railroad industry to identify and articulate critical PTC-related issues requiring research and industry involvement in projects. While the legislative mandate for PTC deployment underpins this

74 REVIEW OF FRA’S R&D PROGRAM collaboration, it appears to have promoted collaboration across the divi- sion’s train control and communications portfolio. The TCC Division’s Focus on Automation, ITS, Communications, and Sensor Technology Recognizes the Importance of Human Factors to Train Operations and Grade-Crossing Safety Having such a wide breadth of responsibility—from furthering PTC to grade-crossing safety—the TCC Division has demonstrated creativity in harnessing advanced communications, sensors, ITS, and automation tech- nologies to address human performance and behavior issues that can affect safety. Collaborations with other U.S. DOT agencies, such as FHWA and FMCSA, on using connected vehicle technologies to alert drivers to grade crossings exemplify this creativity. While the PTC mandate drives much of the division’s research and accounts for a large portion of its budget, the technical capacity and knowledge gained from this work is being leveraged to address other safety concerns and challenges. RECOMMENDATION While all of the RD&T divisions face obstacles to getting the results of their research published and disseminated, overcoming these obstacles is espe- cially important for ensuring the timely application of the TCC Division’s technology-oriented work. To this end, the TCC Division should make a concerted effort to ensure that research results are made available to indus- try and other users as quickly as possible to contribute to the advancement of PTC and other systems being deployed in the field.

TRAIN CONTROL AND COMMUNICATIONS 75 ANNEX TABLE 5A-1 TCC Division Projects, 2019 TCC Division Project Title Funding Project Duration Rail Software-Defined Radio (SDR) $518,188 August 2016–December 2018 Quasi-Moving Block Train Control $400,013 September 2017–September 2019 Flexible Operator Location Feasibility Analysis $1,105,414 September 2015–December 2020 Rail Crossing Violation Warning $853,156 September 2015–March 2020 Positive Train Control (PTC) Critical Asset Track Data Auditing System $946,641 (total) September 2015–January 2019 Higher Reliability and Capacity Train Control $757,951 September 2017–September 2019 Restricted Speed Enforcement for Positive Train Control (PTC) Systems $400,550 February 2017–May 2020 Pilot Grant Program—Law Enforcement Strategies for Reducing Trespassing $196,357 October 2018–December 2019 Enhanced Humped Crossing Database Using LiDAR $299,804 October 2018–April 2020 Cybersecurity Risk Management for Connected Railroads $799,713 September 2017–November 2019 Trespass Detection and Warning– Drone System $150,000 January 2018–December 2019 Photo Enforcement-Based Education at Crossings $250,000 March 2016–June 2019 Gate Skirts Research $50,000 May 2017–September 2019 Vehicle Railroad Right-of-Way (ROW) Incursion Prevention $165,000 October 2016–March 2019 Vehicle Blocked Crossing Research $150,000 March 2017–June 2019 Rail Right-of-Way (ROW) Trespass Mitigation Treatments $300,000 October 2014–September 2019 LED-Enhanced “Do Not Stop on Tracks” Sign Research $185,000 June 2017–June 2019 Emergency Notification System (ENS) Sign Study $150,000 June 2018–November 2019 National Trespass Workshop $150,000 January 2019–December 2019 In-Pavement Grade-Crossing Lights $190,000 September 2016–September 2018 continued

76 REVIEW OF FRA’S R&D PROGRAM TCC Division Project Title Funding Project Duration Effect of Grade Separation on Trespassing $175,000 October 2014–September 2018 Grade-Crossing and Trespass Research Program Support $210,000 November 2016–March 2019 Positive Train Control (PTC) Interoperability $886,000 (total) March 2017–March 2020 Monitoring and Analysis of the Integrated Network (MAIN) $1,489,300 (total) August 2016–May 2020 Employee in Charge Portable Remote Terminal (EICPRT) $3,200,400 (total) September 2012–March 2019 NOTE: Funding level is for FY 2019 unless otherwise specified. SOURCE: Federal Railroad Administration Office of Research, Development, and Technol- ogy 2019a. TABLE 5A-1 Continued