International Transportation (68) 1 | 2016 6 Comparison of automated transport modes Synergies for driverless rail transport Transport systems, driverless transport, traic management Automatization takes place in daily life in nearly all areas. Smart watches, smart phones and smart homes-are nowadays quite common. The future of transportation will surely follow this technological advancement. Authors: Viola Klingkusch, Yigit Fidansoy T he increasing demands placed on- transportation, demographic change and new technologies bring new challenges for the future of the transportation systems. Automated transport modes are being developed to reduce the costs and increase the capacity of current systems. Definition of driverless transport Over the past decades, the technologies used in the diferent means of transport have become more and more complex with the aim of enabling driverless transport. An automated transport system can be deined as a system where the driver of the vehicle is partially or fully replaced by an advanced electronic system. Those systems consist of computers, sensors and communication devices in the infrastructure as well as in the vehicles. Driverless vehicle technology has the potential of being a true game changer in traic management. It delivers major beneits in terms of road safety, social inclusion as well as emission and congestion reduction. Transport automation involves the development of the required physical and digital environment, education and acceptance of end users, technological solutions and much more. In this paper the following modes of transport will be discussed: road, air and rail traic (see figure 1). The current situation There are six levels of automation in motor vehicles. Between level zero, in which there is no driver assistance system, and the most highly automated level without a driver, there are four other gradations: assisted Photo: Aurinko | dreamstime.com STRATEGIES Automatization Automatization StrategieS International Transportation (68) 1 | 2016 7 driving, partially, highly and fully automated operation. Driverless operation is also referred to as ‘autonomous operation’. In road transport there are already assisted and partially automated vehicles, in which systems take over longitudinal and transverse guidance for speciic use cases. The car is monitored by the driver. There are currently highly automated vehicles under development that require the driver to intervene only in critical situations. For the automation of the light process, eight diferent stages of development are distinguishable. Between level 1, where manual light guidance is introduced, to level 8 with fully automated autonomous light operation, there are several gradations of assistance systems. It can be assumed that the degree of automation in the cockpit will continue to rise. There are arguments for active intervention of the pilots in the light control in case of a complete system failure. Pilots must be able to intervene in cases of aircraft control failing. Very few maneuvers and trajectories are standardized. Essentially, a pilot can respond more lexibly than a computerized assistance system. According to the International Association of Public Transport (UITP), there are four diferent Grades of Automation (GoA) in rail transport. The irst grade is manual train operation, where a train driver controls the train run. In GoA 2, the driver triggers the movement of the train, which then stops at the next station automatically. This grade is called semi-automatic train operation. In railway systems, some mechanisms of the 2nd and 3rd level of automation have already been implemented. An example for GoA 2 in Germany is the subway in Munich, where the driver only opens and closes the doors. Driverless train operation with automated starting and stopping is called GoA 3, but a train attendant operates the doors and drives the train in case of an emergency. The highest level of automation, where no staf is needed on board, is called ‘unattended train operation’. The train starts and stops automatically, operates the doors and handles emergencies. There are also some metro lines with GoA 3 and GoA 4. An example for GoA 4 in Germany are the metro lines U1, 2 and 3 in Nuremberg. When comparing the number of automation levels in the three diferent transport systems, it becomes clear that there are many more gradations in air traic. The reason may be that with aircraft more parameters need to be automated. There are many more sensors installed and needed as in other modes of transport. Diferences between legal frameworks In the area of driverless transportation, legal liability and registration is essential for a legal framework. For road traic, the Vienna Convention on Road Traic of 1968 was designed to standardize the international traic rules. In Germany, the requirements of this convention have been implemented in the Road Traic Regulations (StVO). The Vienna Convention deines that the assistance systems used must be actively controllable by the driver, which in some cases is limited by the assistance systems themselves. Some of the assistance systems, such as ABS or ESP, become active when the driver is about to lose control of the vehicle. This raises the question of how far the ‘driver’ is able to monitor a driverless car and to what extent he may turn away from the driving task. Article 8 of the Vienna Convention can be interpreted in a way that monitoring of assistance systems is not necessary if the licensing regulations are Figure 1: Status quo of road, air and rail transport automation 05-07 October 2016 Casa Convalescència, Barcelona, Spain The 44th European Transport Conference Booking Discounts - Deadline 30 th June 2016 Delegates are now invited to book their place at the European Transport Conference to beneit from an Early Booking Discount. The Early Booking Discount applies to delegates booking 3-day attendance only, with payment received by 30 th June 2016: AET or ECTRI Members Standard Fee* Early Booking Fee* Individual Member £765 €995 £690 €900 Organisation Member £735 €955 £660 €860 Non-Member £900 €1170 £810 €1055 Single days from £275 may be booked from 1st July 2016. In addition, a 50% discount on booking fees is applicable to attendees from new EU Member States (joined since 2004) and for young professionals under the age of 26 or with less than 5 years’ professional experience. For those involved in transport planning, research and practice, the European Transport Conference is the event to ind in-depth presentations on policy issues, best practice and research indings across a broad spectrum of transport modes. To secure your place, please complete an Early Booking Form online at: www.etcbookings.org *All fees shown are subject to 20% VAT alescència, Barcelona, Spain STRATEGIES Automatization International Transportation (68) 1 | 2016 8 changed. Besides the registration, the liability of driverless cars needs to be discussed. The legal basis in Germany for air transportation is created by the Civil Aviation Act (LuftVG), the Air Traic Regulations (LuftVO) and the Certiication Speciications for Large Aeroplanes (CS 25). The CS 25 states that the pilot should always be able to take control of the aircraft. An assistance system provides additional safety, because ideally both, the pilot and the assistance system, have an efect on the aircraft. The socalled type certiicate is the oicial approval of a new type of aircraft for air traic. The aviation authority of the state in which the aircraft is registered issues the license. In Europe the European Aviation Safety Agency (EASA) is responsible of the type certiicates. The EASA regulatory framework diferentiates between development, production, maintenance, and airworthiness of aircrafts. All changes, additions or repair procedures on the aircraft must be approved separately. In case of rail traic, the rules for driverless transport are diferent for trams and railways. Trams and some light-rail systems in Germany are planned and operated according to the Ordinance on the Construction and Operation of Street Railways (BoStrab). The Technical Rules for Trams for driving without driver (TRStrab FoF) are derived from BoStrab and speciied for autonomous driving. The other important German regulation is the Ordinance on the Construction and Operation of Railways (EBO), which speciies rules and regulations for railways. Currently there is no speciication for driverless transport in the EBO, which is an obstacle for the topic. Trains running under EBO are approved by the Federal Railway Authority (EBA). Tram licenses are issued by the Technical Regulation Authorities (TAB). Driver assistance systems Driver assistance systems in cars are additional electronic devices integrated in the vehicles to assist the driver in driving situations. They exert semi-autonomous or autonomous inluence on the control, drive or signaling devices of the vehicle. Autonomous vehicles work without any intervention of a human driver. Assistance systems actively inform and warn drivers, increase comfort by stabilizing the automobile, thereby reducing the driver’s workload. The development of aircraft systems in the past has been inluenced and shaped by the steady increase in air traic, the demand for higher safety, greater eiciency in light operations and the use of new technologies. No human pilot is beyond making mistakes during a light and therefore has limits. The objective of assistance systems in aircraft is therefore to reduce the workload of the pilot. The special feature of air traic is that it involves three-dimensional control variables, which an assistance system must be able to control. For larger machines, such as passenger aircraft, there are three-axis autopilots where in addition the rudder can be controlled. More complex systems also adjust engine power, communication and navigation for aircraft control. The assistance systems that are planned for railways can be divided into driver assistance systems and network-related assistance systems. Driver assistance systems give speed recommendations to save energy and reduce wear. Network-related assistance systems focus on a speciic network segment in order to optimize train movement in this area. This means that the position of the trains relative to each other should be optimized to save energy. These systems work basically with a ‘green wave’ principle like in road transport. In addition, the disposition, punctuality and capacity are in the foreground of network-related assistance systems. Challenges for driverless rail transport For the implementation of Automatic Train Operation there have to be far-reaching changes to the railway tracks, the stations, the signal boxes, the vehicles and the operation processes. There is a need for further research. A limitation of the spatial extent seems necessary for autonomous operation. Reasons for this are mainly the high cost in planning and retroitting the infrastructure, the rail vehicles and the operating centers. A spatial limitation of the tracks in form of a fence is considered too complex and costly. The cost of reconstruction on the tracks and in the stations is rather high. In Automatic Train Operation there is a change in safetyrelated technologies. Achieving safety of the railway tracks can be considered diicult. Through the use of assistance systems not only driverless driving is possible, but there also is a high potential of energy savings. However, the potential savings are dependent on the degree of automation, i.e. how many assistance systems are used. The realization of an automatic railway track requires a long planning period. Additionally, the duration of rebuilding the tracks and railway stations is of great importance. An important aspect in the implementation of an Autonomous Train Operation is the acceptance of diferent groups of people who may be afected by such an introduction. The operator may be in favor of driverless operation after conducting a positive economic study. Opposition will probably come from the train drivers and, to a certain extent, the passengers. Train drivers are afraid of losing their jobs and being patronized by technical means like assistance systems. Tasks that were previously handled by a human must now be carried out by technology. Thus the risk of human error can indeed be reduced and at the same time safety can be increased by a redundant installation. The passengers may have fears of highly automated technologies because they do not see them as reliable. Summary In recent years a lot of research in the ield of driverless transportation has been done. In the means of transport, be it airplanes, cars, or trains, there are diferent approaches to achieve driverless transportation. Yet the diferences and similarities between the means of transport and experiences from present automated operations can be used to advance the automation grade. The challenges involved are not only of a technological, but also of a political and social nature. ■ LITERATURE [1] Lutz, Lennart S.; Tang, Tito; Lienkamp, Markus: Die rechtliche Situation von teleoperierten und autonomen Fahrzeugen. Neue Zeitschrift für Verkehrsrecht NZV 26 (2), 2013, 57-63 [2] Mensen, Heinrich: Handbuch der Luftfahrt. Springer-Verlag, 2013 [3] Verband der Automobilindustrie: Fortschritt Automatisiertes Fahren; http: / / www.internationalesverkehrswesen.de/ english/ special-edition-22015.html? L=%252Fproc%252Fself%252Fenviron (abgerufen am 13.04.2016) [4] Winner, Hermann; Stephan Hakuli and Gabriele Wolf: Handbuch Fahrerassistenzsysteme. Wiesbaden 2012 [5] Brockhaus, Rudolf; Alles, Wolfgang; Luckner, Robert: Flugregelung. Berlin/ Heidelberg 2011 Viola Klingkusch Technische Universität Darmstadt-(DE) viola.klingkusch@yahoo.de Yigit Fidansoy Research Assistant, Department of Railway Engineering, Technische Universität Darmstadt (DE) fidansoy@verkehr.tu-darmstadt.de