SIP-adus(Automated Driving for Universal Service) started in June 2014 as a national research and development project for innovation. It has a social significance, providing a fundamental solution to suchissues as the reduction of traffic fatalities, the reduction of the environmental burden by easing traffic congestions, travel support for elderly people and other vulnerable road users, and the revitalization of rural areas. The improvement of the competitivenessof the automobile industry and the expansion of related markets are significant from an industrial point of view. Under the Steering Committee, we established the System Implementation WG, the International Cooperation WG, and the Next-Generation TransportWG. In 2016, we identified five important area of focus (1) dynamic maps, (2) human-machine interface(HMI), (3) cyber-security, (4) pedestrian accident reduction, and (5), next-generation transport. We also conducted a large-scale FOTs on these area from 2017and delivered the achievement of the research and development.

View full abstract

Dynamic maps are three-dimensional high precision maps on which various kinds of semi-static (scheduled phenomena), semi-dynamic (current phenomena), and dynamic information (real-time changes), are combined with position data for smart automated driving (AD) and advanced driver assistance systems (ADAS).

Many SIP R&D measures are carried out to achieve the early realization, wide deployment of dynamic maps, as well as reduce their cost.

Dynamic map use will not be limited to vehicles featuring AD/ADAS, but also address social needs such as the prevention or mitigation of the e? ects of natural disasters, advanced agriculture, or infrastructure management and maintenance. These elements are therefore also considered in SIP-adus.

The final stage of SIP R&D involves a large scale Field Operational Test (FOT) carried out by many of the stakeholders involved.

View full abstract

Over a five year period, beginning in 2014, we deliberated the approaches to use in relation to dynamic map static information requirements, data specifications, preparation guidelines, and encoding specifications, as well as the location referencing methods used to link static information and dynamic information in the dynamic maps necessary for automated driving systems. Through the Large-Scale Field Operational Test (Dynamic Map), we conducted static information evaluation with the collaboration of test participants. Based on the evaluation results, we concluded that the essential features that make up static information are sufficient for use by automated driving systems. We also defined features for which there were numerous requests from test participants as quasi-essential features. Through this test, we found that road signs, road markings, and the like change over the course of three to four months afier dynamic maps are created, and we identified a need to organize approaches to updating static information.

View full abstract

Vehicle probe information is getting more important than ever. It is useful to get data on road situations such as traffic flow, objects, as well as weather conditions, in near real-time. It is expected that this kind of information will be shared among industries to realize a more efficient and safer society especially when automated vehicles become a reality.

The purpose of this project is to evaluate the feasibility of the vehicle probe information exchange interface between cloud services through the large-scale field operational test conducted by SIP. The interface speciffications are currently under development by the Dynamic Vehicle Information Sharing Working Group in Japan Automotive Software Platform and Architecture (JASPAR).

View full abstract

For automated driving to be able to use lane-level traffic information, it will be necessary to verify whether or not there are problems with regard to the following issues: 1. errors due to data conversion in the multiple conversion processes conducted when providing traffic information, 2. differences in map accuracy, updating frequency and so on between the maps used for human drivers to provide route navigation and the maps used by systems, including automated driving, and 3. time lag at each stage in the generation, transmission and utilization of traffic regulatory information by the vehicle. A proposed testing plan has been prepared to conduct tests to verify these issues as part of a different project during the next fiscal year. In this study, an equipment configuration to receive the regulatory information distributed by road operators and overlay this information onto the dynamic maps was proposed.

View full abstract

Static high-precision 3D maps, which form the basis of dynamic maps promoted and maintained for use as automatic driving maps on highways, are required to remain up to date and maintain a high level of quality. We proposed an automation technology for map generation to reduce map creation costs and realize quick updates. We confirmed the effectiveness of improving the map generation cost by comparing and verifying the automation technology and existing map generation method. As improvement measures included not only map generation but also data measurement, we implemented real-time measurements and diagrams. We will proceed with a verification of the technology for practical application.

View full abstract

This study was performed in order to identify the challenges for specifications in dynamic map preparation and clarify the necessary basic data structure requirements and the challenges to be resolved, based on on-site demonstration using specific use cases, for the ultimate purpose of employing the dynamic maps and three-dimensional map common platform data acquired during the dynamic map preparation process in public services.

View full abstract

The concept of Society 5.0, a super smart society aiming to balance economic development and the resolution of social issues, states that intelligent transport systems, one of the 11 types of system identified in the 2015 Comprehensive Strategy, will be developed as core systems. The construction of a platform enabling various systems to utilize dynamic maps is needed to create new values and services. For that reason, it is important to realize the Dynamic Map Service Platform, a base system for the use of dynamic maps that is planned to be constructed in the context of the SIP-adus project and encompass various fields. The object of this project is to develop the concept of the Dynamic Map Service Platform, which will be utilized in various fields in the future, as a part of government measures involving cutting-edge ICT technologies and extensive collaboration between wide-range of parties.

View full abstract

Recently, new satellite positioning methods with multi-GNSS or Quasi-Zenith Satellite System (QZSS) are becoming available. In this survey, we installed several GNSS devices in a measurement vehicle, conducted experiments and evaluated the performance of each satellite positioning method. In addition, the tasks for using satellite positioning in automated driving systems were summarized. From this survey, it was confirmed that the new satellite positioning method makes driving lane recognition possible with no other positioning sensors. However, it has also been shown that there are some restrictions, such as satellite signals shielded by buildings or slopes. To utilize the map information, it is necessary to acquire the position obtained by satellite positioning, and the next step will be to effectively tackle improving the availability and reliability of satellite positioning to achieve practical application.

View full abstract

The effectiveness of introducing ITS wireless roadside units (the complement rate with infrared beacons and improvement in the percentage of drivers who decelerate in reaction to information from the beacons) was evaluated using a model route (a route equipped with ITS wireless roadside units and infrared beacons) installed in Aichi Prefecture.

The results of the evaluation concluded that an optimal combination of introducing infrared beacons and ITS wireless roadside units will have a positive effect on intersections, particularly important intersections whose traffic signal cycle lengths and other parameters change in real time.

In the future, the accuracy of signal information required for controlling automated vehicles will be examined. Cost reductions for roadside units must also be considered.

View full abstract

This project is aiming to develop driving support systems that are effective at preventing traffic accidents at intersections, as well as reducing traffic congestion and environmental burden, while assessing and examining the cost effectiveness and driver acceptance of such systems. The systems covered by this project utilize traffic signal information (a collection of traffic signal phase and timing information from up to 16 downstream intersections) provided by advanced infrared beacons, which are scheduled to be installed in the future. The project will be implemented under the auspices of the Ministry of Economy, Trade and Industry in consideration of commercialization, and involves the participation of the National Police Agency from the standpoint of traffic safety.

View full abstract

To realize automated driving, use of the latest traffic regulation information is an important aspect of vehicle control. This traffic regulation information is managed by the police in each Japanese prefecture. It is necessary to build a structure capable of gathering and providing traffic regulation information from the whole country. To develop this structure, we carried out the following steps: (1) standardization of the traffic regulation information data format, (2) construction of a traffic regulation information management model system.

View full abstract

The development of automated driving systems is mainly based on the use of in-vehicle sensors, but adding vehicle-to-vehicle (V2V) and vehicle-to-infrastructure (V2I) communication has the potential to realize advanced automated driving systems capable of cooperating with other vehicles and the surrounding infrastructure. The purpose of this project is to develop the necessary technologies for realizing such cooperative automated driving systems. We are confirming the feasibility of communication to support automated driving systems, and developing technologies to improve the characteristics and processing effeciency of this communication.

View full abstract

These two research cases were carried out as a three-year project. In the first year, the accuracy of vehicle and pedestrian positions of V2V and V2P communication systems, communication delay, and the like were investigated in tests on public roads. In the second year, driver acceptance of these kinds of safe driving assistance systems that utilize V2V and V2P communication was investigated in tests on a test track. In the final year, driver acceptance of automated driving systems that utilize V2V and V2P communication was investigated in tests using a driving simulator.

View full abstract

The Human Factors and HMI research project consisted of three tasks: A, B, and C, and started in FY2016 as a three-year project. Task A investigated the effects of system information on driver takeover performance. The provision of well-defined knowledge about the system before driving and experience of using the system were found to have a positive effect. The provision of some dynamic information about the system state was also shown to be effective. Task B investigated the effects of the driver state (readiness) on takeover performance and metrics of readiness. It was found that different driver states influenced takeover performance in different ways. Some metrics for states (readiness) influencing takeover performance were found. Task C investigated effective ways of functionalizing automated vehicles to communicate with nearby road users. An external HMI was found to be an effective additional cue for pedestrians to make a decision to cross a road when an approaching automated vehicle intended to yield but did not decelerate largely enough to clearly signal this intention. It was observed that the external HMI also induced unsafe behavior in some pedestrians. This report was produced in the middle of the third year of the project and does not include all the studies conducted in the project.

View full abstract

While a driver is driving on the highway using a level 2 automated driving system and a system malfunction occurs, control of driving operations are handed over from the system to the driver. This research examines the necessary time margin for the driver to take over manual operation based on experiments conducted with normal drivers using a driving simulator.

View full abstract

Information used as the foundation for automated driving systems is expected to be obtained from external networks, which could cause cybersecurity issues that did not exist in conventional cars. In this project, the activities related to automated driving/connected cars were researched, categorized and organized on a factual basis to derive a common model for automated driving systems and formulate a comprehensive threat model for automated driving systems. In addition, an information security evaluation method was defined and issued as a draft to serve as a guideline for the overall assessment of V shape vehicle development models.

View full abstract

Communication technologies such as V2X are expected to be used in cars to acquire dynamic maps and the surrounding environment information necessary for automated driving. Meanwhile, external connectivity through communications makes cybersecurity as an important issue. To derive the requirements for vehicle cybersecurity, a common model of automated driving systems was constructed and used to carry out threat analysis. Furthermore, evaluations and verifications ranging from the component to the vehicle levels were carried out. Based on this, the testbed requirements for vehicle security studies were developed. In addition, methods of simplifying certificate verification were studied.

View full abstract

Pedestrians and cyclists account for almost half of traffic accident fatalities. To achieve the national goal of reducing traffic fatalities to 2,500 or less by the year 2020, it is critical to establish a vehicle-to-pedestrian (V2P) communication system that can help to prevent collisions by exchanging position data between pedestrians and vehicles. This research establishes the fundamental technologies for a V2P communication system. Furthermore, we have created prototypes of the pedestrian terminal and the in-vehicle terminal for experiments and verified their effectiveness.

View full abstract

To promote the realization of safe driving support and automated driving, it is essential to observe the conditions surrounding the vehicle, including beyond the vehicle's line of sight, and to provide vehicles with traffic information in real time that contributes to the prevention of traffic accidents using roadside sensors. Therefore, research and surveys were conducted into a system that provides vehicles with constantly changing traffic information using radio communication. In fiscal 2014 and 2015, research and surveys were conducted into the enhancement of driving safety support systems (DSSS) promoted by police, and the specifications of low cost versions of the systems, with the aim of implementing the systems across the country. In the following fiscal year of 2016, examinations were carried out with an eye toward the realization of automated driving. These involved the classification of traffic accident types and vehicle behavior at intersections, identification of the types of moving objects (vehicles, pedestrians, bicycles, and the like) that need to be detected, together with the required range of detection to ensure vehicle safety in each accident scenario, followed by an examination of the configuration of roadside sensors required for detection.

View full abstract

This article reports the results of a research initiative to apply high-resolution millimeter-wave radar to roadside sensors for the purpose of reducing pedestrian accidents using cooperative driving support. Millimeter-wave radar technology installed at intersections and the like can realize sensing functions, such as the detection of pedestrians and bicycles with high accuracy, regardless of the weather and time. To facilitate the practical adoption of this technology, detection processing software compatible with high-resolution radar has been developed and verified under severe weather conditions such as hard rain and snowstorms. A prototype 79 GHz band millimeter-wave radar system was installed at actual intersections to evaluate the detection performance of crossing pedestrians and so on. This report mainly describes the development of the radar technology and the progress of the public road experiment.

View full abstract

In Japan, the authors analyzed accident data using automated driving systems under the auspices of the Cross-ministerial Strategic Innovation Promotion Program (SIP) of the Cabinet Office of the Government of Japan. Based on ITARDA macro data from 2013, 255 SIP accident patterns with three or more fatalities were established and studied for four years. This article describes the background, accident patterning methodology, a pattern sheet sample, analysis of sheets, and some study results using the 255 SIP accident patterns. It also introduces the future plans for these SIP accident patterns in a national accident database.

View full abstract

The aim of this project is to develop novel multi-agent traffic simulation software capable of predicting the impact of automated driving systems on traffic safety. The software is designed to incorporate variables for human perception and recognition, decision-making, and errors, dependent on age and gender. In this study, the software was applied to a 6 x 3 km² area in Tsukuba city. Over a simulated period of time that included more than 500 agents (vehicles, drivers, and pedestrians), the software simulated and identified at least five types of accidents. Different automated driving technology penetration scenarios were applied to conduct preliminary estimations of the potential impact of these technologies on safety.

View full abstract

The safety of automated driving systems, which are regarded as a promising means of helping to reduce traffic accidents, is greatly dependent on functions for recognizing pedestrians and objects, and on making early and accurate judgments of potential collision risks. The collection of large volumes of driving data and the development of technologies and tools for using this data efficiently are recognized as common issues for strengthening competitiveness in the development of such recognition and judgment functions. This article describes efforts undertaken to research and evaluate the utility of a driving video database and simulation evaluation technology needed for developing camera image recognition technology.

View full abstract

We have developed a tool for visualizing the CO₂ emissions of vehicular traffic, which can quantitatively evaluate how the introduction of automated driving systems might affect traffic flow and, consequently, changes in the amount of CO₂ emissions from vehicular traffic. To develop this evaluation tool, we combined traffic simulation with CO₂ emissions models and performed necessary verifications in accordance with an international joint report issued as part of NEDO's “Development of Energy-saving ITS Technologies” project. In addition, we used the tool to evaluate certain automated driving systems in specific scenarios, where the tool proved applicable to the evaluation of those systems.

View full abstract

The following are regarded as preconditions for encouraging the development and widespread use of more advanced automated driving systems (ADS): Clarification of the impacts on society and industry inside and outside Japan, as well as the risks that will result from such changes, the implementation of measures to mitigate these impacts and risks, the formulation of scenarios from a long-term perspective, and greater understanding of ADS on the part of the general public. This study clarified the evolution of ADS, proposes a future vision for these systems and scenarios for issue resolution, and defined specific actions to prepare for the implementation of ADS through a study team composed of academic specialists from various fields.

View full abstract

To create social acceptance of automated driving systems, public events called “citizens' dialogues” have been held since FY2016. These events provide information about automated driving to the public to facilitate understanding of these systems. In addition, we have been sorting and analyzing new awarenesses and visions of these systems based on opinions and responses, including the concerns and expectations of the public, to provide input for guiding future research and development activities, as well as to foster social acceptance. At the dialogue sessions held over the past two years, opinions were exchanged among a variety of stakeholders, accurate information on automated driving was provided, and expectations and concerns about automated driving were extracted. The positive outcomes of the initial events and acceptance of the activities are illustrated by dissemination of the activities to provincial areas during FY 2018.

View full abstract

As Japan's society ages, the percentage of traffic accident fatalities involving elderly people is increasing. For this reason, it is necessary to develop more flexible, efficient, and useful public transport systems. With this background, the SIP-adus Next Generation Transport Working Group decided to develop a partially automated driving system, in addition to the related services and advanced infrastructure applications for public transport. These services and applications include a precise docking system, an advanced public transportation priority system (PTPS) application, an Advanced Rapid Transit (ART) information center, and a walking assistance service system for vulnerable road users. We set the 2020 Tokyo Olympic and Paralympic Games as the milestone for the initial implementation of these systems.

View full abstract

To make it easier for bus users, including wheelchair users, to get on and off the bus, a survey was conducted to help develop an automated control technology that stops the bus so that the gap between the bus entrance and the bus stop is minimized. Section 1 reports the results of an investigation about the permissible gap and step between the bus entrance and the bus stop. Sections 2 and 3 describe the performance of recognition technology used for controlling the accurate arrival of the bus at a bus stop. Section 4 describes the performance of the arrival control using the recognition technology described in Section 3.

View full abstract

This research is part of the Strategic Innovation Promotion Program(SIP)/Autonomous driving system project and examines the issues of the so-called Advanced Rapid Transit (ART) automated driving system, which is considered to be a promising next-generation transport system for urban areas. This article describes the docking control project, which aims to clarify basic control issues related to sensing, steering linkage, integrated control with the brakes, and so on. These items are being researched and development to realize docking maneuvering technology with excellent functionality.

View full abstract

Precise docking at bus stops is a key technical requirement for achieving the practical implementation of the next-generation Advanced Rapid Transit (ART) transportation system. Among the control methods for precise docking that are being studied in this research and development project, the road surface guidance lines described in this article are regarded as an effective means of reducing control error. However, some concerns remain, such as their use on roads not used by buses, on which they may cause needless driver confusion. This study investigated guidance lines for precise docking that are readable by cameras mounted on ART vehicles, and conducted several verification experiments to confirm their safety. The effect of this precise docking system on the driving behavior of ordinary vehicles and the recognition rate of these guidance lines were investigated in experiments held at a test course.

View full abstract

Next-generation urban transport systems are being developed in order to ensure safe and smooth traffic around the venues of the 2020 Tokyo Olympic and Paralympic Games, and to secure means of transportation for the elderly in Japan's super-aging society. The UTMS Society of Japan developed a system that performs bus priority control using vehicle-to-infrastructure communication via the 700 MHz radio frequency and conducted verification through field operational tests. The equipment specifications and communications standards established in this development will be adopted as standard specifications to promote future the introduction and dissemination of the system.

View full abstract

This study on the speediness and safety of the Advanced Rapid Transit (ART) system involved the building of on-board units for a Public Transportation Priority System (PTPS), and the evaluation of the effectiveness of the resulting Advanced PTPS. Specifically, the basic requirements were designed for the advanced PTPS on-board units, which communicate with roadside units that control traffic signal timings (i.e., to extend the green light interval or truncate the red, based on bus locations sent by the on-board units). These units also have a priority arbitration function, which grants priority over other buses when a bus meets the criteria for priority, as well as an HMI function, which notifies the priority status information and supports passage through the intersection. The extent to which Advanced PTPS improves timeliness and speediness, and its effects on general traffic were evaluated. It was clarified that Advanced PTPS and priority arbitration can improve both speediness and timeliness.

View full abstract

The functions of an Advanced Rapid Transit (ART) information center were investigated to help realize services using ART-related information and to associate various public traffic information. Experimental receiving/publishing data APIs were produced to realize five ART information center functions. A bus information-provision service for bus users was developed to exemplify the function of utilizing the information collected and stored in ART information center platform. We plan to investigate the usefulness of the information and the service provided by ART information center function in a large scale field operational test from October to November 2018.

View full abstract

The purpose of this study is to identify behavioral and psychological characteristics by analyzing a large quantity of data about major events held in Tokyo in close cooperation with relevant entities, including the Olympic Organizing Committee, the Tokyo Metropolitan Government, and railway/bus operators. We are conducting social experiments regarding behavior modification and guidance targeted at relatively major events to form and verify hypotheses of models to be used for congestion prediction. Based on these verification results, we intend to suggest a new scheme that can provide congestion predictions and traffic information appropriate for the 2020 Olympic and Paralympic Games to minimize impacts on individuals attending or affected by the games, mainly by targeting people who will visit the Olympic and Paralympic Games and citizens that live or work in Tokyo and use public transportation including Advanced Rapid Transit (ART).

View full abstract

As Japan faces a super-aging society, movement support for people with mobility constraints has become an important social issue that needs to be addressed. To cope with this situation, technology development, as well as research and surveys are in progress to help realize safe, secure, and smooth movement support for people with mobility constraints that also satisfies convenience and economic rationality requirements. The UTMS Society of Japan is working on the enhancement of the existing Pedestrian Information Communication Systems (PICS) and proceeding with studies to help realize services that provide intersection crossing support using smartphones.

View full abstract

To improve pedestrian accessibility through route guidance adjusted to the individual characteristics of each pedestrian, the following studies, investigations, and developments were performed. First, the information necessary for transfers and the method of collecting this information were studied and evaluated through field verifications. As a result, data collection applications on smartphones were adopted as a low-cost and continuous method for information collection. Second, demonstration experiments to develop walking networks using the data collection application were implemented. In addition, the serviceability of common platform information was evaluated using a prototype personal navigation application on smartphones, and a field verification of route guidance corresponding to the personal attributes of pedestrians. Furthermore, cooperation between personal navigation systems and advanced PICS was studied. A large-scale demonstration experiment regarding route guidance will be performed in October 2018.

View full abstract

Through the large-scale field operational test (dynamic maps), we designed static information update data for dynamic maps necessary for automated driving systems and location referencing methods to link static information with dynamic information, and collaborated with test participants to evaluate these static information update data and location referencing methods. Based on the static information evaluation results, we created map updating guidelines. We also determined that information usage intent varied by automated driving level. Our findings indicate that linking static information and other information can be effective for automated driving system vehicle control, and we verified dynamic map theory through actual field operational testing.

View full abstract

In the HMI large-scale experimental demonstration, the results of research and development (knowledge and systems) are verified by corporate participation (in actual use). In FY 2017, a baseline for readiness was derived through public road experiments in which (1) the feasibility of driver monitoring systems (DMS) in the real world was verified, and (2) data with variability for use as a baseline in the readiness index was acquired. In FY 2018, the relationship between how and what information on system functions is conveyed to the driver and handover performance are being verified experimentally on a test course. Also, various controlled driver conditions, DMS output, and handover performance will also be veri? ed experimentally on the test course.

View full abstract

A field operational was conducted based on the information security evaluation method for vehicle systems established in FY 2017. In FY 2018, a larger scale information security field operational test will be conducted with an increased number of participants. The evaluation and implementation approach were reviewed and will be finalized as Information Security Evaluation Guidelines for vehicle systems.

View full abstract

The goal of this project is the reduction of pedestrian collisions. For this purpose, various demonstration experiments with attention evaluation tools were conducted using communication technology (V2P). In the demonstration experiments, functionality and effectiveness were verified under the actual traffic conditions, and the impact of promoting the social value of the provided service was examined. The service transmits alerts, conveys information, and provides notifications of the presence of an intersection between cars and pedestrians, and was assigned a certain evaluation. This report presents the overall process and the results of the project for the two years of FY 2017 and FY 2018.

View full abstract

A large-scale field operational test on the introduction of advanced rapid transit (ART) will be conducted with the aim of developing a public transportation system that is convenient and easy to use for various users, including the elderly and people with disabilities. Specifically, there are plans to: verify the effectiveness of advanced public transportation priority systems (PTPS) to reduce the time required for ART, verify the precise docking control system at bus stops to provide safe and quick boarding and alighting for all passengers, verify the effectiveness of the route map and guide information created by aggregating and integrating barrier-free information collected from people with restricted access to transportation and other people, as well as the provision of congestion forecasts by station and time based on the results of analyzing congestion around stations near a venue when events are held. A field operational test on the ART Information Center function as the foundation for realizing services utilizing ART-related information as well as organic and flexible cooperation of ART systems by collecting, managing, and providing various traffic information is also included in the plan.

View full abstract