Effectiveness Testing of ADAS in the PTI (Periodical Inspection Test)

Andreas Himmler･Patrizio Agostinelli (dSPACE)･Hiroki Hanaoka (dSPACE Japan)

Mandatory installation of important ADAS functions is being promoted in many countries. However, ADAS are not subjected to regular inspections or functional tests even the legally mandated. On the one hand, there are no legal requirements, and testing methods and equipment are not yet available. In this presentation, the options discussed in the EU for such test methods and the functional principles and developing status of the associated test equipment will be discussed.

RESILIENT AND CONTINUOUS SAFETY ASSURANCE METHODOLOGY FOR CCAM AND ITS HMI COMPONENTS

Oana Moldovan (Applus+Idiada)

The CERTAIN project develops a comprehensive Safety Assurance Framework (SAF) for Cooperative, Connected, and Automated Mobility (CCAM) systems, emphasizing resilience, human-centric design, and continuous assessment. The methodology combines real-world with virtual validations, incorporating stakeholder needs, AI trustworthiness, human-machine interactions and V2X connectivity. The framework ensures ongoing safety through in-service monitoring and reporting, including handling over-the-air updates. Multiple use cases across different vehicle types, from passenger cars to trucks, and automation levels (L2-L4), validate the approach, focusing on safety, trust, user acceptance and comfort. This holistic approach aims to facilitate safe, reliable deployment of CCAM systems with industrial and regulatory consensus.

Automatic Generation of Training Data for Long-Range Nighttime Recognition Using a Long-Baseline Stereo Camera

Shunya Kumano･Shoji Matsuzaka･Nao Ikeda･Yumi Yamada･Yusuke Ueda･Naoki Kawasaki (SOKEN)･Kiichiro Kawakami (Toyota Motor)

Creating training data required for long-range nighttime recognition involves substantial manual annotation work on development-camera images. In this study, we propose an automatic annotation technique that accurately aligns results detected up to approximately 1 km by a 90-cm baseline stereo camera to development-camera images using a calibration-based two-stage processing method, achieving approximately 90% reduction in labor.

Scenario Extraction from Drive Recorder Footage for Traffic Scene Retrieval

Masafumi Tsuyuki (Hitachi)･Tetsuya Nishida･Taminori Tomita･Yoshitaka Atarashi (Astemo)

Enhancing the development efficiency of Advanced Driver Assistance Systems (ADAS) requires robust video retrieval methods. However, comprehensively addressing diverse retrieval requirements remains a significant challenge. This study introduces a scenario-based retrieval approach that leverages the capability of driving simulators to reproduce arbitrary traffic scenes. The proposed method extracts scenario definitions exclusively from monocular camera footage and positional data, enabling structured and searchable representations. By facilitating efficient access to relevant driving situations, this approach supports systematic evaluation and accelerates the development of advanced driver assistance technologies.

A New Method for Forward Parallel Parking Assistance Using Rear-Wheel Steering

Kenta Maeda (Hitachi)･Tatsuya Shiraishi･Daisuke Tsuga･Hiroki Ishimaru･Miki Koso･Atsushi Yokoyama (Astemo)

Parallel parking is one of the driving tasks with which many drivers feel uncomfortable. With conventional front-wheel steering, drivers are forced to reverse when parallel parking; however, introducing rear-wheel steering enables forward parking. This presentation proposes a new method that applies in-phase rear-wheel steering when the parallel parking starts, in contrast to the conventional counter-phase (opposite) rear-wheel steering, aiming to reduce driver anxiety.

A Study on Road Traffic Flow Enhancement with Optimized ACC Strategies

Yoshiaki Irie･Tomoaki Morimoto (Toyota Motor)

Anticipating the widespread adoption of AD/ADAS, this study explores ACC control methods beyond the conventional focus on occupant safety and comfort. By analyzing the detailed behavior of individual vehicles on urban highways, we investigate strategies to optimize ACC operation for smoother traffic flow, contributing not only to individual vehicle performance but also to overall traffic efficiency.

Development of a Non-Contact In-Vehicle Human Presence Detection Sensor Using Spatial Potential Fluctuations

Kenji Kouno (The University of Tokyo)･Hiroyuki Suto (Toyota Motor)･Yusuke Yokota (The University of Tokyo)･Yusuke Umetani (Toyota Motor)･Yoshihiro Suda (Tokyo University of Technology)

This study proposes a non-contact human detection method that captures in-vehicle potential fluctuations caused by body motion and breathing using a passive sensor. Although the sensor can be implemented with a simple configuration based on a field-effect transistor, its noise tolerance has been limited. We developed an improved sensor that achieves higher SNR than conventional sensors and supports wired integration with in-vehicle systems.