Under the direction of the Council for Science, Technology and Innovation, which acts as a control center,the Cross-Ministerial Strategic Innovation Promotion Program (SIP) of the Cabinet Office, a national project to realize innovation through cross-agency as well as industry–government–academia cooperation, embarked on its second phase in 2018, with 2020 marking the third year milestone of that phase. This document constitutes the Mid-Term Results Report summarizing the initiatives of SIP–Automated Driving for Universal Services, one of the 12 topics covered by the program. The composition and contents of the various chapters are described below.

View full abstract

The Cross-Ministerial Strategic Innovation Promotion Program (SIP) of the Cabinet Office is a national project to realize scientific and technical innovation, with the Council for Science, Technology and Innovation acting as its control center. In the field of automated driving, where results are expected from SIP, initiatives aimed at reducing traffic accidents and congestion, and securing mobility in sparsely populated regions are being pursued to realize a society offering safe and secure transportation for all its citizens. With its sights firmly set on facilitating the realization of Society 5.0, the second phase of SIP-Automated Driving for Universal Services (2018 to 2022) is using the achievements of the first phase (2014 to 2018) as a springboard to carry out research and development, field operational tests (FOTs), and other activities in cooperative areas to build a vehicle-infrastructure cooperative driving automation system that will expand the scope of automated driving commercialization

View full abstract

Vehicles equipped with automated driving systems are required to operate safely and smoothly in various traffic environments while sharing those environments with a wide range of traffic users. To realize this objective, it is necessary to construct a cooperative infrastructure system capable of collecting and generating road traffic environment data at the roadside, as well as distributing this data in a timely fashion to vehicles. At the same time, it will also be important for the future to construct a framework for data circulation that embodies the cyber physical system defined under the so-called Society 5.0 concept, namely the generation, distribution, and re-use by vehicles of new road traffic environment data from commercial probe data collected by vehicles themselves. The SIP-adus program is working to promote practical adoption in these cooperative areas, starting from research and development related to the construction of systems for utilizing road traffic environment data and commercial probe data.

View full abstract

The following three points were identified as requirements for traffic signal information from the standpoints of improving the reliability and availability of automated driving: (1) a maximum margin of error of ±300 msec between the timing of the traffic signal information and the actual traffic signal color, (2) the detection of traffic signal information errors and notification to vehicles when errors occur, and (3) the realization of traffic signal information provision with various traffic signal controls. First, focusing on the vehicle-to-infrastructure (V2I) approach, this project verified the accuracy of traffic signal information and defined failsafe specifications for roadside devices, before then validating the feasibility of functions using prototype units. To improve availability, this project organized a proposal to review the operation of traffic signal controls and carried out validation experiments for traffic signal information with special traffic signal controls on proving grounds and the like. It then expanded the scope of information provision to include emergency vehicle priority (FAST) controls, which are difficult to adopt with traffic signal information provision and pedestrian-operated button controls. Based on these validation results, technical specifications for traffic signal information provision infrastructure using V2I systems were determined. In addition, this project has also begun examining the vehicle-to-network (V2N) approach using mobile circuits as a method of providing traffic signal information without using V2I systems From the 2021 fiscal year onward, validation experiments are planned for V2N systems, with the objective of defining the appropriate provision methods based on the experiment results.

View full abstract

Autonomous and automated vehicles drive based on information from a limited range forward of the vehicle. An accident or congestion in this range requires the vehicle to decelerate suddenly and heavy traffic flows may also prevent smooth lane changes. In these cases, if the forward traffic conditions (look-ahead information) can be identified on a lane-by-lane basis, safer and smoother automated driving may be accomplished by enabling cautionary deceleration or seamless lane changes in advance. As connected vehicles become more widespread, traffic environments that can generate such look-ahead information pertaining to traffic conditions are currently being constructed. However, there are no data formats capable of directly identifying the traffic conditions of separate lanes. This project involves the development of technology related to lane-based road traffic information to help realize safer and smoother automated driving, and the verification of the level of lane-based information that can be generated from currently available probe data. Initiatives in this project began by using data with an immediate potential for use in the future.

View full abstract

This research and development project studied technology capable of identifying changes in roads using dashcam drive recorder images and probe vehicle data. These results were then applied to studies envisioning the rapid deployment of technology for identifying road changes. For the studies into technology capable of identifying changes in roads using dashcam drive recorder images, camera image data of existing sections of roads were obtained after changes were carried out. These studies then confirmed and assessed whether the changes to the road could be identified using existing technology to recognize features from the obtained camera image data, identifying those features after the road changed, and comparing the features with the state before the changes.

For the studies into technology capable of identifying changes in roads using probe vehicle data, the requirements for using probe data to update high precision 3D maps were examined on paper and then validated on a proving ground. This verification process confirmed that driving operations and vehicle behavior accompanying changes to roads can be identified using probe vehicle data in an ideal environment. Based on these results, consultations were held with OEMs about the provision of actual probe vehicle data, and the potential of using the provided real-world probe vehicle data to identify changes to roads was confirmed.

The studies envisioning the rapid deployment of technology for identifying road changes were carried out by (1) carrying out a field operational test (FOT) under a scheme envisioning the rapid deployment of technology after research and development, followed by the organization of requirements such as device specifications and the like. Then, after investigating the latest trends and so on, (2) the required conditions for starting operation on ordinary roads and globally were identified. The gap between the requirements identified at phases (1) and (2) were analyzed and summarized into a procedure for commonization.

View full abstract

The concept of cooperative driving automation has been studied by corporations and research institutes for some time. Field operational tests (FOTs) and standardization activities for the communication protocols that will realize this functionality are also underway. In Japan, ITS wireless communication has been commercialized and is already used to support driving safety. Expanding it to automated driving can be envisioned. However, while the feasibility of its application and the communication protocols to adopt in the future, has been pursued by individual organizations, there have been no discussions for Japan as a whole. Since SIP has established a structure uniting industry, government and academia under one roof to work toward the realization of automated driving, it was decided to hold discussions under that framework. The Task Force (TF) on V2X communication for Cooperative Driving Automation was established in 2019 and initiated a 3-year plan to study future communication protocols. The TF defined use cases for assessing communication, and used them as a basis to clearly define communication requirements and study protocols that satisfy those requirements. The goal is to propose communication protocols for cooperative driving automation and formulate a roadmap that clearly indicates when they will be required.

View full abstract

A vast array of technologies is being deployed to realize advanced automated driving. One of those is technology that enables automated vehicles to use information obtained via V2X communication to recognize objects that are beyond the scope of on-board sensors such as camera, radar, or other sensor. Given concerns over the impact on traffic flow as automated vehicles using only on-board sensors are predicted to stop or slowly move until they can determine it is safe to proceed, that technology can be expected to mitigate such impact.

The research and development presented here focused on core technologies that effectively collect vehicle, pedestrian and other object data obtained from multiple roadside infrastructure units serving as sources of dynamic information, integrate those pieces of object data to streamline the generation of dynamic information, and efficiently transmit the processed data. This project addressed proposing parameters to integrate the object data as well as the communication protocols and common interfaces to collect and transmit the dynamic information from multiple sources to automated vehicles.

View full abstract

Overview: As part of the Japanese government’s “Investments for the Future Strategy” and ahead of the 2020 Tokyo Olympic and Paralympic Games, field operational tests (FOTs) in the Tokyo waterfront area were conceived and carried out as a domestic and international showcase for the latest automated driving technologies, to create a legacy by demonstrating the feasibility of these state-of-the-art technologies in the Haneda Airport and Tokyo waterfront areas, and to facilitate the establishment of systems, frameworks, and the like. Roadside-to-vehicle communication infrastructure was set up for different purposes in three districts in the Tokyo waterfront area with the objectives of demonstrating the effectiveness of and identifying issues with automated driving technologies. In the Haneda district, tests of automated buses were carried out that included precise arrival (docking) control and the utilization of magnetic markers. In the waterfront district, the provision of traffic signal information via intensive infrastructure installation was tested. Finally, on the Metropolitan Expressway that links these two districts, support for merging and passing through electronic toll collection (ETC) gates was studied.

View full abstract

In the FOTs in Tokyo Waterfront City area^{(1), (2), (3)}, we validated data for a total of 29,728 intersection traversals to confirm that traffic signal information provided by ITS Roadside Units installed at 33 intersections within the test area could be used to smoothly and safely traverse intersections with traffic signals on general roads by using vehicle-infrastructure cooperative driving automation (of these intersection traversals, roughly 18% were performed by automated driving using traffic signal information). We confirmed that by acquiring and utilizing current traffic signal color information, we could perform automated driving in road transport environments in which traffic signal colors could not be recognized by on-board cameras due to backlighting, direct lighting, rainfall, concealment or obstruction by preceding vehicles or curves, poor nighttime visibility, traffic signals blending into the background, or other factors. Furthermore, in dilemma zones where there is significant variation in intersection traversal judgement results, we confirmed that traffic signal remaining seconds information could be used to safely stop before reaching intersections without sudden deceleration or to safely traverse intersections without sudden acceleration. We believe that in implementing advanced automated driving on general roads, there is potential for the introduction and expansion of safe, smooth automated driving through the use of vehicle-infrastructure cooperation by defining areas where automated vehicles will be used and preparing ITS Roadside Units that cover entire areas and supply traffic signal information.

View full abstract

The field operational tests (FOTs) carried out in the Haneda Airport area used automated driving support infrastructure such as magnetic markers, bus lanes, traffic signal information, public transportation priority systems (PTPS), and the like to verify whether buses equipped with automated driving technology could achieve a stable and regular service on a circular bus route on public roads in the vicinity of Haneda Airport and realize precise arrival (docking) control at bus stops. Automated driving tests consisting of a total of 322 laps of the circular bus route were carried out, and it was confirmed that a stable and regular service by automated driving could be achieved under a mixed traffic environment. In addition, approximately 80% of manual interventions in the tests were to avoid parked vehicles or because the test vehicle came too close to the stop line of the oncoming lane when turning left. It was also confirmed that the use of PTPS helped to realize a more punctual service with stable travel times. The FOTs also confirmed that bus lanes helped to improve the continuity of automated driving. However, it was found that the resulting conflicts with ordinary vehicles would necessitate publicity and educational activities, as well as campaigns to encourage people to follow the rules of bus lanes. At bus stops, a total of 416 automated driving docking control tests were carried out, confirming the feasibility of highly reproducible docking controls to a standard deviation of less than 10 mm through the use of magnetic markers.

View full abstract

In the FOTs on the Metropolitan Expressway^{(1), (2), (3)}, data was validated from a total of 365 traversals of the Airport West Entrance (12 of which (approx. 3%) were performed by automated driving using information provided by infrastructure) in order to verify if ETC gate passing support information and merging lane assistance information provided by ETC2.0 wireless roadside units was effective as support information for vehicle-infrastructure cooperative driving automation and as support information for drivers. The FOTs confirmed that ETC gate passing support information can be used to rapidly and accurately determine the open/close states of ETC gates. In particular, this information appears likely to be particularly effective for toll booths whose operating status cannot be visually confirmed until late and toll booth areas with numerous toll booths. Merging lane assistance information is generated based on information detected by roadside sensors installed in upstream areas of expressway cruising lines. It is effective in enabling automated vehicles and drivers to determine the state of the cruising line in advance. However, it was confirmed that because the information that is provided is based on spot detection of vehicles on cruising lines, the information could not reflect changes in the speeds of vehicles on cruising lines that occurred after the sensor was passed, which presented problems with smoothly merging in critical traffic flow and when there is traffic congestion.

View full abstract

In the impact assessment field operational tests (FOTs), various automated vehicles were driven in mixed traffic environments on general roads, and the impact of these vehicles on ordinary vehicles and pedestrians was evaluated. For these tests, a visualization system was constructed and adopted that combines image data from fixed point cameras installed at multiple intersections, image data from on-board cameras, and vehicle behavior data. The tests confirmed that the addition of automated vehicles did not have a significant negative effect on action times when turning left or right, and that the variance in these action times was smaller for mixed traffic than traffic containing only ordinary vehicles. In contrast, sudden deceleration by automated vehicles on normal stretches of roads, immediately before intersections, and the like, caused various near-miss incidents involving surrounding vehicles. Additionally, the tests did not find a lack of eye contact been drivers and pedestrians on crossings and the like as a result of automated driving. As a result, it was confirmed that the automated vehicles in the tests drove safely in the presence of pedestrians. To help realize advanced automated driving in harmony with ordinary vehicles and pedestrians, similar assessments to those carried out in these FOTs should be continued as automated driving technology becomes more sophisticated and automated vehicles become more widespread.

View full abstract

Automated driving equivalent to level four in urban areas requires both advanced and autonomous recognition and decision making functions using on-board artificial intelligence (AI), as well as supporting infrastructure such as roadside and communication equipment. However, since installing such infrastructure throughout Japan would require a massive budget, the feasibility of using the bare minimum of infrastructure and recognition/decision-making technologies must be examined. Therefore, to stimulate discussions into future cooperative areas, this project is researching and investigating the infrastructure and recognition and decision making technologies that are absolutely necessary for automated driving systems through experiments on public roads by academic researchers who have been allowed a certain degree of disclosure of acquired data and the technologies.

View full abstract

Ensuring safety and reliability are two issues of the highest importance for the practical adoption and popularization of automated vehicles. For this reason, the establishment of safety assurance methods for automated vehicles is an urgent task. At the same time error-free communication must be realized between automated vehicles and other traffic users. SIP-adus is working to construct a safety assurance environment in a virtual space, and to establish sustainable and effective countermeasures against cyberattacks targeting connected vehicles, appropriate methods for communicating the intentions behind automated vehicles to people, and effective educational and awareness-building methods for people using automated vehicles and automated driving services.

View full abstract

As automated driving systems become more complex, it is necessary to ensure a high level of safety in the countless driving environments that occur. However, currently automated vehicle safety can only be validated through costly (in terms of labor, equipment, financial, and time) and comprehensive practical assessments in actual driving environments. In addition, it is also difficult to validate the limits of external sensors such as cameras, radar, LiDAR, and the like that form the interface between the vehicle and the real world, which complicates decisions about the level required upon establishing system safety. Against this background, this research project is developing an assessment platform using simulators to create a virtual models by adopting a series of models consisting of driving environments, spatial propagation, and sensors that are highly consistent with the actual phenomena required to assess the safety of automated driving. The goal of the project is to precisely and efficiently assure the safety of automated driving under as many environmental conditions (scenarios) as possible.

View full abstract

The information on vehicles, people, infrastructures, and other elements projected onto maps and advanced map data that constitute automated driving system platforms is mostly obtained from external networks and sent to control units and information systems in order to controlling the vehicle. However, this situation is also a factor liable to trigger cybersecurity problems not found in traditional vehicles. The agreement on UN R155/R156 reached at the UNECE WP.29 also makes addressing cyberattacks necessary from a regulatory perspective. To solve such issues, this investigative research project focused on intrusion detection systems (IDS) as a technology to combat new cyberattacks after shipment. We formulated IDS evaluation guidelines to serve as a baseline for testing and evaluation when installing an IDS. At the same time, we assessed ways of collecting and storing information on connected car threats and carried out collection experiments using honeypots and other mechanisms in the context of building a system to provide initial response support in the event of an actual incident. The plan for this project extends to fiscal 2022, and the present fiscal year centered around fundamental research and studies. This report summarizes the outcomes of those activities.

View full abstract

This paper describes the research and development on HMI to carry out proper driving takeover in the event of a deviation from the driving environment conditions or a failure of the system, and research and development on the knowledge drivers and pedestrians should acquire, and effective education methods for that purpose initiatives of the second phase of SIP-Automated Driving for Universal Services. For the former, we studied evaluation parameters for driver situation monitoring before the transition from automated to manual driving, as well as the effectiveness of the HMI in fostering driver understanding of the system. We studied evaluation parameters for driver situation monitoring before the transition from automated to manual driving, as well as the effectiveness of the HMI in fostering driver understanding of the system. For the latter, guided by our main objectives, we defined and researched the three themes of (1) proposing education methods that build upon individual characteristics, (2) proposing motivational approaches, and (3) the development of modular educational material taking topic-specific training into account. Using our draft educational material, we also validated the effectiveness of providing advance general knowledge of automated driving in a driving simulator. The outcomes served as the basis for a workshop on automated driving education we held in the context of Japan–Germany cooperation.

View full abstract

The development of technology for low-speed automated transportation and logistics services vehicles, as well as field operational tests (FOTs) using the services vehicles in semi-mountainous regions, notably, are being carried out as part of efforts aimed at ensuring mobility for vulnerable road users as well as addressing the shortage of drivers in the field of transportation and logistics, reducing costs, and solving other social issues. Unlike traditional manually driven vehicles, low-speed automated services vehicles will not have a driver maneuvering the vehicle at all times in future. This creates safety, security, and traffic efficiency issues with respect to communication with other traffic participants such as pedestrians or other drivers. We analyzed the characteristics of communication measured in FOTs and other observations, and conducted experimental studies of communication methods (e.g., vehicle behavior and external HMIs) aimed at communicating the intent or state of the low-speed automated services vehicle to surrounding traffic participants, for the purpose of achieving safe, confident, and smooth communication between those vehicles and participants. Our research and development focus on recommending designs for communication methods to install in low-speed automated services vehicles and on extracting the knowledge surrounding traffic participants should have, and making proposals in that respect.

View full abstract

Japan is facing the social issues of a decreasing population and super-aging society. Aging is particularly pronounced in semi-mountainous regions, where securing the flow of people and goods has become an urgent problem. Since 2017, we have conducted field operational tests (FOTs) for automated driving transportation services using michi-no-eki roadside stations as hubs in 18 locations around the country to secure that flow in such regions where the population is aging. This paper prevents an overview of the outcomes of those tests, as well as of the issues faced by social implementation and planned further initiatives.

View full abstract

In preparation for the commercialization of transportation and logistics services relying on automated driving, local regions with a low volume of other traffic were first considered as locations to introduce automated driving transportation services and find solutions to social implementation issues such as securing driving space on roads and operations management. Guidelines for the adoption of automated driving transportation services in local regions were formulated, and standards for securing road space for automated vehicles were established in anticipation of nationwide deployment. Therefore, examinations focused on building business models allowing sustained operation for automated driving and related services, and the validation of those examinations, are being carried out in cooperation with local governments and other stakeholders. Based on acceptance by the adopting local governments, the field operational tests (FOTs) in local regions are narrowed down to only the areas necessary for social implementation. Consequently, we are working on strengthening coordination between regions and other initiatives needed for social implementation while keeping the financial capacity of the local governments introducing automated driving services in their region in mind.

View full abstract

As part of efforts to bring automated driving services to local regions, a service support system has been developed to help overcome the particular issues of each region and factor in different service launch aims, thereby enabling these services to be introduced, implemented, and run more smoothly. First, an architecture model was arranged and functional layers were established based on short-term requirements, versatility, and so on. The location management, safety management, reservation, as well as boarding management functions were identified. Next, the requirements to realize the identified functions were studied and developed. Actual services were then run and evaluated in two regions, which helped to identify a wide range of issues. Based on these issues, a system package that can be easily deployed by automated driving service managers in local regions was developed. The functions were consolidated and integrated on a cloud server to facilitate both maintenance and service deployment to new regions. At the same time, on-board units were integrated to help minimize costs during introduction and operation. The developed system package was applied to automated driving services in three regions to confirm its effectiveness. In the future, it is planned to deploy the system package to automated driving and mobility services across the whole country.

View full abstract

In addition to technical development and the establishment of systems, the fostering of public acceptance is an important part of encouraging the popularization of logistics and mobility services using automated vehicles and automated driving technology. While focusing on communicating the correct information about automated driving and quantifying its effects, initiatives being carried out by SIP-adus to foster public acceptance also include more broadly targeted activities connected to the field operational tests (FOTs) to achieve this objective more effectively in the long term.

View full abstract

In addition to technical development and the establishment of systems, fostering public acceptance will play an indispensable role in enabling the social implementation of automated driving technologies. Proper understanding and flexible actions by consumers will help to accelerate the early and efficient adoption of these technologies, while also helping to ensure the safety of new road transportation systems. In collaboration with the relevant government ministries, this project has received and carried out commissions for consumer awareness surveys about automated driving over several years. While tracking the changes in these survey results, this project aims to identify what types of information in which areas are necessary to help effectively foster public acceptance of automated driving. Based on some of the results of past questionnaire surveys and the results of qualitative information collected from regions around the country, this article describes an action evaluation checklist to help fostering public acceptance, and underlines the importance of creative systems of collaboration through discussion of the following points: (1) stagnating consumer awareness about automated driving, (2) low user appreciation of the benefits of driving support systems, (3) insufficient understanding of the functions of driving support systems by users, (4) disconnection between the strong need for mobility focused on elderly people and expectations for the technology, and (5) low acceptance of the cost as well as the uniqueness and the technical limits of automated driving.

View full abstract

This project aimed to establish a methodology for the use of driving support systems to help ensure the safety of drivers with visual field loss, and to raise public awareness about this methodology. First, driving simulators (DS) were used at medical institutions to construct a database and identify the particular causes of accidents involving visual field loss. The project then studied conditions for driving support systems to counter these accident risks, focusing on automatic braking and audio guidance. Using these results, information about the utility and effectiveness of driving support systems was communicated widely throughout society as a whole and to the relevant institutions using visual field loss as an example, thereby helping to popularize advanced driving support systems and raise awareness about safety.

In this project, self-organizing map (SOM) analysis of the results obtained from the simple DS was able to identify patterns of visual field loss and situations in which accidents are more likely to occur. It was also found that, using both simple and high-precision DS, accidents involving drivers with visual field loss could be reduced using highly sensitive sensors and automatic braking, as well as by providing specific audio directions rather than simple guidance about the vehicle status. In contrast, the results also showed that automatic braking systems that are unable to completely prevent accidents may actually cause accidents to increase. Additionally, communication of these ideas to society as a whole was started by opening the first safe driving outpatient care programs at two eye clinics using the simple DS results.

As described above, this project underlined the pressing issue of drivers with visual field loss and proposed ways to resolve this issue. It will be necessary to raise the awareness of both drivers with visual field loss and eye doctors, and comprehensive measures to communicate this issue to society as a whole, industry, and government ministries will also be important. It is hoped that this project will help to resolve technical issues, while leading to changes in vehicle markings, rules, and other non-technical aspects.

View full abstract

Assessments of the socioeconomic impacts of automated driving have an extremely important role to play in fostering public acceptance, corporate management, and defining government policies. This article introduces the analysis and methods of the various simulations conducted in the research concerning the impacts of automated driving including traffic accident reduction carried out from 2018 to 2021 based on that understanding. First, we present two models (the dynamic and static models) for the simulations of automated vehicle diffusion that forms the basis of socioeconomic impact assessment. That presentation is then followed by an overview of the impact analyses covering road traffic, the transportation services field, and the industry & society field conducted using the results of the simulations.

View full abstract

Fostering social acceptance is necessary for the smooth introduction and use of automated vehicles and vehicles providing driver assistance. In this project, we reproduced the traffic environment of selected model cities using a multi-agent simulation of the perception, recognition, decision-making, and actions of various traffic participants (developed in a project⁽¹⁾ in the first phase of SIP-Automated Driving for Universal Services). We then integrated driver distraction and other factors that cause accidents in the simulation to recreate real-world accident conditions. We also used the automated driving (driver assistance) system deployment scenarios for every five years from 2015 to 2050 provided by the Socioeconomic Impacts of Reducing Traffic Accidents project to estimate effectiveness at reducing traffic accidents. Since those deployment scenarios differ for each vehicle classification, we further subdivided those classifications and defined rates of automated driving (driver assistance) system market propagation separately for categories such as passenger car, bus, and truck. The nationwide effectiveness at reducing traffic accidents was estimated from traffic accident statistics using the ratio of accident reduction calculated using the simulation applied to model cities.

View full abstract

SIP-adus is involved in building the dynamic maps through generating and distributing the high-precision 3D maps required for automated driving and the traffic environment information linked to those 3D maps. This geographical data is anticipated to prove useful beyond automated driving and advanced driver assistance systems, in areas such as MaaS, support for evacuations during a disaster, and managing the flow of people and goods. As part of its efforts to realize Society 5.0, in which linking and using data across disciplines will be essential, SIP-adus is not only generating and distributing the traffic environment information used in automated driving, but also working to building a portal to that information that promotes the marketing of geographic data for the purpose of creating a data business market.

View full abstract

The traffic environment information consisting of the high-precision map data prepared for automated driving, as well as data collected form road traffic or vehicle probes, also offers promising uses in industries other than the automotive industry. It is therefore important to build a system that enables the safe distribution of such information in an easy to use format. The MD communet™ traffic environment information portal site that consolidates data from the mobility field and provides coordination with other fields was built and launched to realize that system. Our activities to expand the use of the site include raising awareness by attracting corporate members and conducting various promotions, as well as offering matching opportunities both offline and online, in an effort to make the operation of MD communet? sustainable. In addition, we are validating example services effectively use combinations the traffic environment information that plays a key role in social implementation and information from other fields.

View full abstract

A competition for applications and ideas to resolve social issues related to tourism and traffic in Kyoto, a city popular with tourists from around the world, was held based on the concept of identifying use cases that utilize road traffic environment data to help resolve issues concerning traffic in urban areas (an app competition to resolve issues related to tourism and traffic called the KYOTO Raku Mobi Competition). Before holding the competition, the cooperation of the Kyoto Municipal Transportation Bureau and other business operators involved in traffic, logistics, and tourism was obtained in the compilation and provision of data on mass transit bus stops and train stations, routes, schedules, fares, baggage checking in and delivery services, and shops, information on facilities and sightseeing locations in the tourism sector, statistical data about past congestion, as well as future predictions for congestion, map application programming interfaces (APIs), and so on. The competition served to enhance awareness of the traffic environment information portal site, while collecting and utilizing data posted on the portal site through collaboration and consultation with various relevant parties.

View full abstract

Although mobility-related data has various potential applications, issues include how the data is handled and the federation of data between stakeholders. For this reason, it is necessary to prepare an environment to facilitate the participation of stakeholders in data utilization. In this study and research project, first, the issues hampering greater data utilization were identified and analyzed. Then, based on these results, the appropriate approach for preparing rules to encourage data federation and utilization was considered. As the process for promoting and accelerating data federation, the analysis results indicated that beginning data utilization with public data and then gradually incorporating private data would be a potentially effective way of realizing full-scale federation of public and private data. This project also studied the risks of handling data extracted from the analysis results. Countermeasures were studied from the standpoints of the data handling methods and the matters that must be addressed by businesses. The study results were summarized in guidelines related to data handling in the mobility field (referred to below as the “data handling guidelines”) and a proposal for public/ private data federation in the mobility field (referred to below as the “public/private data federation proposal”).

View full abstract

Automated driving technologies have been improved dramatically, and various activities are being addressed throughout the world. In this circumstance, it is crucial for Japan to take the initiative in standardization and regulatory activities, and to contribute to international harmonization, in order to keep the international competitiveness for the automotive manufacturers and related industries.

In the second phase of SIP-adus, enhancing international cooperation is one of the four major activity pillars. Seven focus international cooperation themes have been identified, and an international cooperation theme leader has been assigned in each research theme. Addition to that, an international cooperation framework featuring collaborative research coordinator who promotes international cooperation activities, has been established. Currently, joint research activities with the German Federal Ministry of Education and Research, and cooperation activities with research projects in the EU Framework Programme for Research and Innovation etc., have been progressing. We are also making the SIP-adus research outcomes known to other countries through events such as SIP-adus Workshops, as well as carrying out the investigative research activities. At the same time, we are coordinating with the relevant standardization bodies about international standardization activities and contributing standardization from both the “de jure” and “de facto” standards perspectives.

View full abstract

The SIP-adus Workshop is an international conference held by SIP-adus since the first phase of SIP was started in 2014. Counting the seventh in 2020, this series of conferences hosted by Japan is gaining recognition as one of the world’s major international events in the field of automated driving. The SIP-adus Workshop is more than just an event. This is a forum to enrich international networks by gathering many automated driving specialists in both fields of the research and development and the implementation of businesses from around the world and exchanging their knowledge and expertise. It has always evolved in gaining momentum. The Eighth edition, SIP-adus Workshop 2021, will be held in November of that same year. Focusing on the results of SIP-adus research and development, as well as social implementation efforts, it will feature various discussions and exchanges of views and offer an opportunity to communicate information with one another on automated driving at an international level, and therefore contribute to the SIP-adus international cooperation activities. This report also gives an overview of the SIP-adus Workshop carried out during the first phase of SIP of which the report did not cover it.

View full abstract

As inter-governmental international cooperation activities in the second phase of SIP-adus, collaborative activities with the German Federal Ministry of Education and Research (BMBF) and the European Commission Directorate-General for Research and Innovation (DG-RTD) have been progressing by receiving offers of joint research through the frameworks of Japan-Germany cooperation and of EU funded project. With respect to Japan–Germany cooperation, joint research plans in the fields of “Human Factors” and “Impact Assessment” related to automated driving technologies were approved in January 2019, with the subsequent approval of new joint research plans in the fields of “Safety Assurance” and “Cybersecurity” in 2020. Now cooperative research covers four areas of research field. With respect to Japan?EU cooperation, cooperation activities with three projects under the Horizon 2020 Framework Programme for Research and Innovation funded by the European Commission are currently ongoing.

View full abstract

International cooperation activities on dynamic maps are based on the SIP-adus research and development results. The main objectives of these activities are to make them, along with the SIP-adus research results, known outside Japan, and to obtain feedback particularly in the context of dynamic maps. The specific activities can be broadly classified as presentations and participation in international conferences, international standardization activities, industry standardization activities, and Japan?U.S.?EU trilateral cooperation. Achievements from activities of the second phase of SIP to date have included presentations at the ITS World Congress and several other international conferences, and the establishment of three international standards—ISO 17572-4 (Precise relative location references), and ISO 20524 (Geographic Data Files) Parts 1 and 2. In addition, SIP-adus has officially become a member of the steering committee of the Open AutoDrive Forum (OADF), an organization involved in industry standardization of digital maps, and has concluded a memorandum of understanding with the Advanced Driver Assistance Systems Interface Specifications (ADASIS) Forum, another member of the OADF, to cooperate in the next SIP-adus field operational tests (FOTs).

View full abstract

Human factors in automated driving constitute a critical aspect of the safety and social acceptance of that technology. In addition, understanding people can be defined as fundamental research, and encompasses a large portion of the cooperative area. Through the first and second phases of SIP, we have relied on proactive cooperation with countries outside Japan both to define issues, as well as validate research methods and appropriateness, for projects concerning human factors and to convey our results internationally. We have also actively worked to incorporate those results in international standards. This article introduces the actual activities.

View full abstract

The development and deployment of practical automated vehicles is regarded as an important way of realizing an even safer, more efficient mobility society that ensures freedom of mobility for all. In this situation, the definition of safety assurance methods is an urgent task, specifically how to judge safety to ensure the public acceptance of this technology, and how to comprehensively assess safety in various traffic conditions. Japan has established the Driving Intelligence Validation Platform (DIVP) with the support of the Cabinet Office to promote the construction of a virtual safety assurance environment and launched the SAKURA project with the support of the Ministry of Economy, Trade and Industry (METI) to promote the construction of a database of safety assurance scenarios. The Japan Automobile Manufacturers Association (JAMA) is responsible for the overall strategy for technical development projects across these two government-led programs and is also liaising with industry, government, and academia to actively support the establishment of international standards. As various safety assurance research projects have been started up around the world, it is important to coordinate closely between projects both inside and outside Japan, and to construct and operate cooperative systems capable of creating leverage toward the establishment of international standards and common basic technology.

View full abstract

International cooperation activities on connected vehicles involve conveying information on cooperative automated driving-related measures that make use of the communication initiatives carried out by SIP via the SIP-adus Workshop, ITS World Congress or other international conferences. They also include gathering information on related activities outside Japan, and building a network with experts in other countries. The SIP-adus Workshop, in particular, provides an opportunity to directly introduce SIP activities to expert from outside Japan during the event, while also providing an opportunity to hold frank exchanges of opinion. Conversely, international conferences held in Europe or the U.S. countries make it possible to listen to the initiatives of other countries and ideas from various corporations, making it possible to obtain a broad range of information. This article presents an overview of those activities.

View full abstract

The information on vehicles, people, infrastructure and other elements projected onto maps, along with the advanced map data that constitutes automated driving system platforms, is expected to be primarily retrieved from external networks.

The information retrieved is then sent to units in the control and information systems of the vehicle for the purpose of controlling it. However, this situation is also a factor liable to trigger cybersecurity problems not found in traditional vehicles.

The agreement on UN R155/R156 reached at the UNECE WP.29 also makes addressing cyberattacks necessary from a regulatory perspective.

To solve such issues, our Surveys of New Cyberattack Techniques and Countermeasure Technologies project from the second phase of SIP-Automated Driving for Universal Services focused on intrusion detection systems (IDS) as a technology to combat new cyberattacks after shipment. We formulated IDS evaluation guidelines to serve as a baseline for testing and evaluation when installing an IDS.

At the same time, we assessed the technical requirements for collecting and storing information on connected vehicle threats and carried out collection tests using honeypots and other mechanisms in the context of building a system to provide first response support in the event of an actual incident.

With respect to the theme of formulating IDS evaluation guidelines, we investigated the new cyberattacks disclosed in 2020 and used questionnaires and interviews to examine the specifications of products offered by three vendors to feed that information back into evaluation items for IDS functions. We are also assessing IDS performance evaluation items in actual environments such as test beds and vehicle benches, primarily for the detection functionality of network IDS (NIDS).

With respect to the theme of surveys and research on connected car threat information and first response support, we theorized that sharing information on threats within the industry would contribute to first response support. We therefore analyzed threat intelligence activities and the threat information collection and storage methods from the IT industry, the pioneer in that area, and are planning to conduct threat information collection testing using a honeypot to simulate an after sales product (an external unit connected through OBD, for example), as well as monitoring tests.

View full abstract

International cooperation related to the socioeconomic impacts of automated driving is centered on a collaborative framework established between Japan and Germany. More specifically, this refers to collaboration between the Connected and Automated Driving: Socioeconomic Impact Assessment (CADIA) research project in Germany, which principally features researchers from the Karlsruhe Institute of Technology (KIT) and the German Aerospace Center (DLR), and the SIP-adus phase 2 (fiscal years 2018 to 2021) project in Japan being undertaken by the University of Tokyo and Doshisha University to research the impact of automated driving including traffic accident reduction. This was recognized as a joint Japan-Germany research project by a meeting of the dual Japan-Germany Collaborative Program Steering Committee held in January 2019 and featuring representatives from the German Federal Ministry of Education and Research (BMBF) and SIP-adus in Japan, which operates under the auspices of the Cabinet Office. As part of this project, a gathering of experts was convened in Germany in October 2019, followed by a session at the SIP-adus Workshop 2020 in November 2020 that focused on German-Japanese cooperation. Opinions were exchanged about common international awareness as well as differences due to national and cultural factors with respect to models of automated driving popularization, topics related to the public acceptance of new automated driving transportation services, and so on. At the same time, trilateral meetings featuring representatives from Japan, the U.S., and Europe have also been held regularly to exchange information about impact assessments.

View full abstract

Studies of use cases, business models, and the like for services using automated driving are an important part of realizing the practical application of automated vehicles. Although field operational tests (FOTs) of automated driving services have been held both inside and outside Japan, various studies and discussions remain ongoing about the issues involved in stepping up from FOTs to actual implementation, as well as the countermeasures for these issues. Consequently, international cooperation initiatives have included information exchanges through SIP-adus workshops and the like from the standpoint of promoting service implementation. This article describes the status of these initiatives.

View full abstract

This final chapter introduces a number of SIP-adus projects expected to prove useful in upcoming discussions of automated driving that were not touched upon elsewhere. It also describes already announced research results, the provision of research themes for field operational tests (FOTs) in the Tokyo waterfront area, and the conveying of achievements.

View full abstract