Overview of Fiscal Year 2001 Research Activities

Since the founding of the AHS Research Association in 1996, the overall research has proceeded as shown here. (Figure 1)

Figure 1

During the initial two years, work in the concepts and requirements field involved analysis of vast amounts of traffic accident data. That analysis was used to select user services that would contribute to safety, and these services were organized into a system. The basic AHS service concept of i, c, a was also established during this time. In the area of infrastructure development, fundamental survey studies were carried out and the orientation of research activity was clearly specified.

During the two-year period including 1998 and 1999, AHS requirements aimed to achieve safety were completed. Along the requirements, activity of research and development on infrastructure proceeded.

In 2000, the research results obtained up to that time were subjected to joint tests held jointly with the advanced safety vehicle (ASV) project on a test course in Tsukuba. The effectiveness of cruise assist systems that contribute to safety, and the feasibility of those systems, were confirmed by means of these joint tests. Public demonstrations held at the same time attracted 2,400 participants from Japan and other countries, and these events proved a great success that also had a significant international impact.

Proving tests were implemented, including some on actual roads, in 2001 and 2002. These were positioned, in effect, as the final stage of research activity in preparation for the start of practical application of the systems beginning in 2003.

Cruise assist systems are collaborative systems that work by linking vehicles, radio communications, and roads, so the framework of research extends to ASV, which carries out research and development on the vehicle side, the Association of Radio Industries and Businesses (ARIB), which is in charge of radio communications, and AHS. Research activities have been carried out on the basis of continuing close coordination among these three parties. (Figure 2)

Figure 2

Our objective is to introduce practical application of these systems in fiscal year 2003. The two years from 2001 to 2002, therefore, are an extremely important period for systematically organizing all issues relating to practical application and completing the preparations for practical application. (Figure 3)

Figure 3

In the area of concepts and requirements, joint system definition documentation for ASV/AHS was created for those systems that are slated for early practical application in 2003. At the same time, the conceptual approach to safety and reliability of systems for road sections of uninterrupted flow was established while research results were obtained that included clarified display requirements for stand-alone infrastructure services (for road sections of uninterrupted flow) using message signs.

Merge assistance service is a service intended for practical application in the future. The needs related to this service were systematically organized and preparations were made for testing its effectiveness.

In the systems area of infrastructure development, upgraded versions were created of designs for those systems earmarked for early practical application. The version upgrades were meant to reflect the ASV/AHS joint system definition documentation, apply spot communication, reflect the findings from safety and reliability analysis, provide for use of AHS technology for road management, and so on.

A further matter was the design of systems for proving tests. Test systems were designed for use on test courses and on actual roads throughout Japan.

Work in the element technologies area of infrastructure development included verification of functional performance of road condition assessment sensors and road surface condition assessment sensors conducted in continuation from the previous year in actual traffic environments. Activity centered mainly on the topics of acquisition of safety and reliability data and improvement in performance.

Work in road-to-vehicle communication systems included verification and confirmation of technology both on the test course and on actual roads for the newly introduced spot communication systems. The requirements and specifications for introduction of spot communications were also clarified at the same time.

Work on position detection technology included organization of the results from research on lane markers, which has been maintained at a high level of activity from early on. There has recently been conspicuous progress in the technology and the related needs have also been increasing. Therefore position detection technologies considered hopeful candidates for future practical application were surveyed and studied again from a new perspective. Fundamental experimental data were acquired on the technologies identified as promising, including pseudo-satellite, DOA, and so on.

In the area of system evaluation, proving test plans were first formulated. The items, methods, and locations of measurement were decided for testing on the test course, on actual roads, and on a driving simulator, bearing in mind their respective characteristics and the roles allotted to them. The locations for tests on actual roads were determined, and a total of seven locations were selected throughout Japan, including six road sections of uninterrupted flow and one location selected in terms of effective utilization for road management. (Figure 4)

Figure 4

3. Overview of Results from Public Information Technical Exchange and Intellectual Property Activities

Public information activities increase in importance during the time leading up to practical application. Activities of this kind were conducted in order to obtain the fuller understanding and recognition of members of the public, experts, and road managers. Specific activities include opening a Web site and sending related materials to approximately 400 supporting members, including 100 members overseas. In addition, the public information activities were carried out actively at various events in Japan and other countries.

In the area of surveys, a team was dispatched to Europe and America to visit major institutions and survey the state of position identification technology. Active surveys were also conducted with an emphasis on AHS-related projects overseas.

In fiscal 2001, technical exchanges were carried out with 40 organizations in 15 countries. The results from our own research are also gradually being recognized overseas, and opportunities for exchange are increasing. At the same time, these exchanges are also growing perceptibly more substantial in content with every year.

Activities related to intellectual property include patent application filings, searches for prior patents, and outside presentation of research papers.
In Japan, six patents were registered and 42 were filed. Overseas activity included five foreign patents registered and three filed. Searches are conducted for prior patents that could possibly be infringed by our own research results, and 2,500 such domestic and foreign searches were conducted during the year. During fiscal 2001, nine domestic patents and three foreign patents were discovered that analysis determined would require attention.

A total of 41 research papers were presented at the ITS World Congress, IEEE conferences, major academic conferences, and other such occasions.

4. Conceptual Approach to Cruise Assist Systems Safety & Reliability (Road Sections of Uninterrupted Flow)

4.1 Conceptual Approach to Assurance of Safety & Reliability

The purpose of cruise assist systems is to assist drivers to drive safely. Consequently, the assurance of a high degree of safety and reliability is naturally a crucial requirement. On the other hand, this kind of cooperative vehicle-highway system is a new concept in social systems, so the conceptual approach to safety and reliability must be created from the ground up. In fiscal 2001, this activity was focused on road sections of uninterrupted flow.

Here I would like to describe the configuration of cruise assist systems, taking the case of the support system for prevention of collisions with forward obstacles as an example. The circumstances are that an obstacle (such as a stalled vehicle) exists ahead on a curve where it is not visible from the vehicle receiving the service. The system is set up so that sensors detect the obstacle, and the information is passed by means of beacons to the vehicle receiving the service, where the driver is notified using a display or other such device. (Figure 5)

Figure 5

Infrastructure-based systems are basically intended to support and extend driver and ASV functions. Consequently, users expect 100% safety and reliability. For technical and economic reasons, however, it is difficult for infrastructure-based systems to assure safety and reliability approaching 100%. Therefore, the scope of consideration was expanded to include human systems, i.e., from sensors to the human interface inside the vehicle (display device, etc.), and corresponding measures to avoid danger were proposed. These are scheduled to be verified by proving tests beginning this fiscal year.

Before discussing the details, I would like to present the definitions of some terms related to safety and reliability. (Figure 6)

Figure 6

4.2 Categories of Cruise Assist Systems Operational States and Clarification of Failure to Danger<

Next let us consider which of the various cruise assist systems states may interfere with safety.

First of all, system states can be divided into the general category either of passable to vehicles, meaning a period when services should be provided, or closed to vehicles, meaning a period when services are not needed. The periods when services should be provided can be further divided into four types. When categorized according to system operations with respect to a vehicle, these can be evaluated by their combination of states posing no actual danger and states posing danger, so that in terms of safety they are either safe (), dangerous (), or not dangerous (). (Figure 7)

Figure 7

The symbol indicates normal operation, when the system notifies the vehicle that there is danger, or notifies it that there is no danger. Also, when the vehicle is notified that service will be interrupted or that the system is malfunctioning, or that the system is down for maintenance, then the vehicle can respond accordingly. Therefore, these states do not interfere with safety. The symbol indicates failure to safety. For example, the sensor may operate incorrectly so the vehicle is notified that a stalled vehicle is ahead or the sensor may happen to be unable to notify the vehicle of this in cases when there actually is no stalled vehicle. In these cases, there is no danger. The symbol indicates failure to danger. These are cases when the vehicle is not notified even though there is danger (a stalled vehicle), and the system is not able to properly inform the vehicle within the prescribed time about whether there is danger. Analysis of the circumstances of failure to danger show that they correspond to cases when danger was not detected by road sensors, cases of radio wave blockage, and cases of mechanical failure in infrastructure systems.

Here, mistaken operation is also considered to be failure. Working out countermeasures for these three cases will contribute to improved safety.

4.3 Basic Conceptual Approach to Probability of Failure to Danger and Its Countermeasures

The first step here is to gauge the general probability that failure to danger will occur. The rate of non-detection by road surface sensors was derived using sample data from the Ashigara tests, yielding a rate of 1% - 4%. Based on simulation test results, the probability of radio wave blockage (shadowing) came out to be about 0.12%. The probability of mechanical failure, based on hypothetical calculations, came out as 0.00052%. Consequently, this means that for technical and economic reasons, safety approaching 100% cannot be assured by infrastructure-based systems. (Figure 8)

Figure 8

In order to proceed with the discussion, however, it is still necessary to set safety target figures for infrastructure-based systems. Therefore we considered target values that would be realistic and appropriate in light of generally accepted notions, taking the above failure to danger rates and JIS safety standards into consideration, and decided to aim for 95% or higher. In order to achieve this degree of safety, we must establish countermeasures that include human systems.

4.4 Countermeasures that Include Human Systems

We have studied proposals for specific countermeasures to the three cases described above, and have come up with proposals, which I will introduce in order.

First is the case of non-detection by road condition sensors. In this case, information may not be provided to drivers even though danger exists. There is fear that, when drivers are dependent on the system, the danger may actually end up greater than it would have been before introduction of the system. (Figures 9 and 10)

Figure 9

Figure 10

For example, if the message, "There is a dangerous stalled vehicle ahead, so please be careful," is transmitted to a vehicle before it enters a curve, then the driver will decelerate. When drivers are dependent on the system, however, then in the absence of any information, there is a possibility that they will enter the curve at a somewhat higher speed than they would have before the system was introduced, so that their situation could actually become more dangerous than otherwise.

On the other hand, the system can be assumed to be introduced in accident black spots, so we propose that information to encourage careful driving should be provided regardless of whether there is danger. We can expect that doing this will encourage drivers to drive carefully and curb the excessive speeds that could be a negative effect of the system in this case. This kind of effect was observed in tests conducted using a driving simulator, and upcoming proving tests are scheduled to verify it.

In the event of radio wave blockage, the driver has no way to know whether the system is operating normally. It is necessary, therefore, to make the driver aware of the presence of danger. This proposal seeks to provide a message to encourage careful driving even when there is no obstacle ahead, and in this way to avoid danger. (Figure 11)

Figure 11

The intention here is to cause the display to go blank when radio wave blockage actually occurs. Then the driver will be aware that there is an equipment error due to radio wave blockage or some other such reason, and will take action to avoid danger.

The approach in case of mechanical systems failure is much the same as for radio wave blockage. A blank display in this case will also make the driver aware that the system is not able to provide normal service. Use of a blank display will not differentiate between mechanical failure and radio wave blockage. If drivers are made aware that the system is malfunctioning, however, then they will drive with greater care, and this will improve the safety factor. (Figure 12)

Figure 12

4.5 Issues for the Future

The issues that remain to be addressed in the future include intersection systems. Compared to road sections of uninterrupted flow, intersection systems experience much greater congestion. The safety and reliability measures they require are therefore more complex than those for road sections of uninterrupted flow. Work on this issue is presently underway.