U.S. Department of Transportation
Federal Highway Administration
1200 New Jersey Avenue, SE
Washington, DC 20590
202-366-4000
Federal Highway Administration Research and Technology
Coordinating, Developing, and Delivering Highway Transportation Innovations
REPORT |
This report is an archived publication and may contain dated technical, contact, and link information |
|
Publication Number: FHWA-HRT-12-033 Date: December 2012 |
Publication Number: FHWA-HRT-12-033 Date: December 2012 |
This report covers current international activities for developing, testing, and deploying CVHAS, augmenting the literature review that was prepared separately for this project. The focus of this report is on activities outside of the United States, because U.S. CVHAS activities are already familiar. In this report, there will be occasional references to activities in the United States to help set the context for the overseas activities and to offer points of comparison. Cooperative vehicle–highway automation means that the systems involve some form of vehicle–to–vehicle (V2V), vehicle–to–infrastructure (V2I), or infrastructure–to–vehicle (I2V) cooperation (and/or interactions with the driver in partially automated systems), but this report generally avoids addressing the fully autonomous systems that do not involve active cooperation. The term automation covers multiple degrees of automation of the driving function, ranging from driver warning and control assistance to partial automation and full automation.
The dominant international activities in CVHAS, as indeed in all of ITS, are sponsored by the Japanese government and the EC. Although they have the largest budgets and are most inclined to publicize their work in international technical forums, they are not the only sponsors active in this research area. There are also substantial national programs of research and development (R&D) in individual European countries and in other Asian countries, particularly Korea and China.
There has been considerable heritage of prior work on cooperative automation systems in several countries, although none has previously had as much activity as the United States; however, several of the current international automation activities are substantially larger and more ambitious than any current U.S. activities.
Japan has the most extensive heritage of relevant prior work. The pioneering work on vehicle automation by Dr. Sadayuki Tsugawa, working for the Ministry of International Trade and Industry beginning in the 1970s, was based on autonomous vehicle concepts.(1) The same ministry, however, also supported the development of one of the earliest and most ambitious automated personal rapid transit (PRT) system concepts—the Computer–Controlled Vehicle System (CVS)—in the 1970s.(2) This included special-purpose passenger and freight vehicles captive to a guideway, as well as “dual mode” vehicles that could operate automatically on the guideway and manually on normal roads. CVS had an extensive test track, with multiple test vehicles, and performed elaborate simulation studies of how large networks of automated vehicles would operate, but it never advanced beyond the test track stage.
When the ITS program became active throughout the world, Japan was one of the most ambitious participants. As soon as the United States formed the National Automated Highway Systems Consortium (NAHSC) in 1994, Japan’s Minister of Land, Infrastructure and Transport visited to learn more about it and immediately formed a counterpart government–industry, public–private partnership organization in Japan, known as AHSRA. In the first few years, AHSRA concentrated on fully automated vehicle–infrastructure cooperative systems, as the NAHSC had been doing; however, by 1997 it had renamed itself the Advanced Cruise–Assist Highway Systems Research Association, and changed its emphasis to much nearer term systems for driver warning and control assistance. AHSRA continued with this emphasis through its dissolution in 2010, when its research mission was declared accomplished and the results of the research were handed off for deployment by the Highway Industry Development Organization (HIDO).
Vehicle automation work shifted in different directions in Japan in the meantime. Toyota concentrated its attention on fully automated driving of passenger cars and small buses by using magnetic guidance based on the Partners for Advanced Transit and Highways (PATH) technology from the United States and very strong infrastructure-based intelligence for controlling vehicles longitudinally (moving block control analogous to rail and automated guideway transit technology). Toyota conducted public demonstrations of its automated cars at its MegaWeb amusement complex in Tokyo and of its three-bus Intelligent Multimode Transit System (IMTS) at several amusement parks and the 2005 Aiichi Expo (World’s Fair) near Nagoya, obtaining valuable practical experience with automated vehicle operations.
Japan’s Ministry of Economy, Trade and Industry (METI; the new name for the former Ministry of International Trade and Industry) initiated the Energy ITS Program in 2008, with a 5-year plan to develop an automated truck platoon to reduce aerodynamic drag, thereby saving energy and CO2 emissions. This important project will be described in more depth in the section on Japan in the next chapter of this report.
Germany’s history of cooperative vehicle automation system work is not quite as long and deep as Japan’s history, but it is nevertheless substantial. In the 1970s, a German consortium developed an advanced PRT concept called the Cabinentaxi under the sponsorship of the Ministry of Research and Technology and tested it extensively on a test track; however, like the Japanese CVS, it never advanced beyond the test track to public deployment.(3)
The German auto industry, with its emphasis on high-end vehicles, has had a significant involvement with advanced technology for automotive vehicles. The emphasis on automated driving has been somewhat muted because of the prevailing ideology about the “fun” of driving oneself, particularly at BMW and Mercedes; however, Mercedes was the founder of the PROMETHEUS project in 1986, the seminal European program that began ITS with a strong theme of automated driving.(4) Mercedes developed an extremely sophisticated and advanced test car, called VITA II, under PROMETHEUS, which they demonstrated performing fully automated driving in 1994.(5) The philosophy behind this vehicle development was to make the car fully autonomous, basing its driving decisions entirely on data about the driving environment collected by its sensors (mainly video cameras), without any cooperation with the infrastructure or other vehicles. This was influenced to a considerable extent by the pioneering work of Ernst Dickmanns of the University of the Bundeswehr, who demonstrated a vision–guided car in 1987.(6) German car makers do not trust their infrastructure agencies to be able to provide any cooperative infrastructure, so they have tended to insist that their vehicles be able to operate without infrastructure assistance.
Subsequent to PROMETHEUS, Mercedes was the leader of the CHAUFFEUR projects, in which platoons of two and three trucks were developed and tested. The first truck in the platoon would be driven manually, but the following truck(s) would use automatic steering and speed control to follow the trajectory of the first. This system was highly cooperative: There was close communication between the leading and following trucks and a distinctive pattern of infrared lamps mounted on the rear of the truck trailers that could be recognized by sensors on the trucks that followed behind them.
Volkswagen has been more receptive to vehicle automation than its German competitors have been. It demonstrated an advanced automated platoon system in the late 1980s, in which a heterogeneous platoon of vehicles was driven at short gaps on a test track with no drivers in the vehicles. V2V cooperation and sensing systems supported effective platoon control. More recently, Volkswagen supported the Stanford team’s testing of autonomous automated vehicles for the Defense Advanced Research Projects Agency (DARPA) Challenges, and they are now key participants in the HAVEit (Highly Automated Vehicles for Intelligent Transport) project, which is described in more detail in the chapter, “Current European Projects in Cooperative Vehicle–Highway Automation” within this report.
In France, the interest in automated vehicles has originated within the research community rather than within the vehicle industry. Following the NAHSC work in the United States, a group of French researchers from several national research institutes developed a plan for an analogous program in France that is based on a very similar operating concept with cooperative automated vehicles driving in protected lanes. These researchers wrote a book about this work (La Route Automatisée) and formed a research consortium, LARA, to take it forward. The funding support for this work has been somewhat limited, focused on nearer term applications and narrower developments of enabling technologies.
The two French automobile companies, PSA and Renault, have been relatively negative about vehicle automation until very recently, in large part because of some adverse experiences with early versions of partially automated vehicle control assist systems that were brought to market prematurely; however, as explained in the section on France in the next chapter, this situation is now changing significantly.
The more substantial research activity in France has been in the area of “CyberCars,” small automated vehicles designed for use at low speed in urban areas. Several generations of these vehicles have been developed and tested under a variety of operating conditions, either physically segregated from other vehicles and pedestrians or else operating in locations with a very low density of other vehicles and pedestrians.
In many ways, the United States pioneered the concepts and technologies of cooperative vehicle–highway automation and had the field to itself until the 1970s.(7) The original concepts were defined at General Motors (GM) in the 1930s and were presented to the public in the Futurama exhibits of the 1939–40 and 1964–65 World’s Fairs in New York, NY.(8) GM and RCA pursued development of automated highway technologies throughout the 1950s and into the 1960s.(9) In the mid-1960s, this line of research was picked up at Ohio State University, where it continued until 1980.(10) In the 1970s, there was considerable parallel research on cooperative automation systems for PRT, with heavy funding support from the Urban Mass Transportation Administration, the precursor to today’s Federal Transit Administration.(11) This led to the implementation of the Morgantown, WV, people–mover system and several other urban people movers, as well as many of the airport people movers now in operation around the world. These systems are captive to their specialized guideways, and their longitudinal control is based on moving block point–follower control systems derived from railroad technology, rather than performing active vehicle–following.
There was a hiatus in vehicle automation research in the United States for most of the 1980s,(12) even while this type of research became active in other countries, until the California Department of Transportation (Caltrans) began to sponsor a new generation of this kind of research via the California PATH Program in the late 1980s.(13) The PATH research on automated highways was State-funded until it started to receive Federal Highway Administration (FHWA) funding in 1993 and was then integrated into the NAHSC research from 1995 onward. Following the demise of the NAHSC in 1998, Caltrans continued to support PATH research on automated trucks and buses through 2003. The Federal Transit Administration provided some support for automatic steering control of buses through its Vehicle Assist and Automation Program, and FHWA’s Exploratory Advanced Research Program supported additional research on automated truck platoons and CACC beginning in 2007. For the past decade, these efforts have been funded at much lower levels than the analogous research overseas, which will be described in subsequent sections of this report.