Proposal of a Steering Assist System Considering Individual Driver Input Constraints
-Third Report: Actual Vehicle Evaluation with Drivers Simulating Physical Constraints-

Daisuke Nagasaka (J-QuAD Dynamics)･Akira Ito (Aichi Institute of Technology)･Hiroyuki Okuda･Hirofumi Aoki (Nagoya University)･Shigenori Ichinose (J-QuAD Dynamics)

The steering assist system proposed in previous reports, which considers individual driver input constraints, was evaluated in an actual vehicle with drivers simulating physical constraints. This report presents the evaluation results, including the effects on steering workload reduction and driving assistance.

Vehicle Attitude and Disturbance Suppression Control Based on Optimal Torque Allocation to Front and Rear Wheels

Ryo Watanabe･Nobutaka Wada (Hiroshima University)･Minoru Miyakoshi･Yasuhide Yano･Tomohiko Adachi･Seiji Fukui (Mazda)

This paper proposes a control method to stabilize vehicle attitude during acceleration and deceleration and minimize the impact of road irregularities on body motion by optimizing the distribution of braking and driving forces between the front and rear wheels. The proposed control law is derived based on model predictive control. The effectiveness of the proposed method is verified through numerical simulations.

Robust Performance Improvement of Automotive Powertrain Control Systems Using Deep Reinforcement Learning Considering Parameter Variations

Hiroki Shibata･Heisei Yonezawa (Hokkaido University)･Shota Sato･Hiroya Kikuchi･Takashi Hatano･Shuichi Kondo･Shigeki Hiramatsu (Mazda)･Itsuro Kajiwara (Hokkaido University)

In automotive powertrain systems, reducing low-frequency vibration during acceleration remains a challenging issue. Since model-based control systems depend on system models, their control performance may degrade due to parameter variations in actual plants. This study obtains a control policy using deep reinforcement learning that accounts for parameter variations, thereby improving vibration suppression and robust performance under varying conditions.

Development of Method for Deriving Comprehensive Use Cases on Public Roads for Functional Safety Analysis

Yoshihiro Miyata･Shuhei Kashihara･Takayuki Nakano (Nissan Motor)

In hazard analysis for systems operating on public roads, countless use cases must be considered, and defining them requires a tremendous amount of time. Demonstrating coverage is also difficult. This issue is especially relevant for autonomous driving systems that perform diverse vehicle control functions. We have developed a method for deriving use cases that addresses these issues. In this presentation, we report the developed method and its application examples.