| 研究課題/領域番号 |
23K03898
|
|---|
| 研究種目 |
基盤研究(C)
|
| 配分区分 | 基金 |
|---|
| 応募区分 | 一般 |
|---|
| 審査区分 |
小区分21040:制御およびシステム工学関連
|
|---|
| 研究機関 | 群馬大学 |
|---|
研究代表者 |
|
|---|
| 研究分担者 |
山田 功 群馬大学, 大学院理工学府, 教授 (20240012)
|
|---|
| 研究期間 (年度) |
2023-04-01 – 2026-03-31
|
| 研究課題ステータス |
交付 (2023年度)
|
| 配分額 *注記 |
4,550千円 (直接経費: 3,500千円、間接経費: 1,050千円)
2025年度: 1,560千円 (直接経費: 1,200千円、間接経費: 360千円)
2024年度: 1,300千円 (直接経費: 1,000千円、間接経費: 300千円)
2023年度: 1,690千円 (直接経費: 1,300千円、間接経費: 390千円)
|
|---|
| キーワード | Cooperative Driving / Connected vehicle / Edge Computing / Cyber-physical System / Vehicle Control / ITS / Edge computing |
|---|
| 研究開始時の研究の概要 |
次世代の高度道路交通システムの開発に向けて、しばらくこれからの交通システムでは、通信あり・なし車両や自動・手動運転車両の混在が続く．その混在環境で複数の車両の振る舞いを考慮した、個々車両制御による道路交通システム全体を最適にするため、人工知能(AI)と制御理論を複合し、インフラ・クラウド情報や周辺車両の部分的な情報からより正確に交通状態を予測しながら各車両の最適運転を可能にする制御理論を構築することを目的とする．具体的には、サイバーフィジカル交通制御や車両の最適運転技術でエッジインテリジェンス（EI）を導出する理論と、それに基づく各車両の協調制御系の設計理論を構築する．
|
| 研究実績の概要 |
Aiming to develop edge intelligence for cooperative driving in various driving contexts, we have researched several aspects, including vehicle control techniques, cloud-based efficient traffic coordination, driving abnormality detection, and enhancing the perception of autonomous vehicles. The key outcome in the first year is the development of a cooperative lookahead driving technology that can improve traffic flows on multi-lane roads when they are affected by any incidents, which has been published in IEEE Transactions on Intelligent Vehicles. Under this technology, a vehicle unilaterally identifies the need and extends the cooperation to other lane vehicles, significantly improving traffic flows. Besides, cooperative driving at Railway level crossings under infra-based edge intelligence has been developed considering the real driving scenario at Shin-Kiryu Fumikiri. This edge computing-based optimal vehicle coordination techniques will be polished further in the following years.
In addition to the above-mentioned cooperative driving technologies, we have investigated two aspects of other driving systems. We are particularly investigating the detection of lanes or driveways on irregular roads and assessing the driving state (including abnormal maneuvering) based on real driving data. The preliminary studies are ongoing, employing various machine-learning and data-driven techniques. Overall, the progress was smoother, and we had better outcomes in the first year than our original plans.
|
| 現在までの達成度 (区分) |
現在までの達成度 (区分)
1: 当初の計画以上に進展している
理由
We have made significant progress in developing vehicle cooperative control technologies in various scenarios, as international collaborators are also supporting our leading team at Gunma University. The development of other aspects, such as perception of the environment, prediction of traffic, and respective intelligent decision mechanisms, is progressing as initially planned.
|
| 今後の研究の推進方策 |
The edge computing-based perception of the surrounding traffic is crucial for developing cooperative driving technology. Such cooperative techniques are not easily accessible at the low penetration rate of connected vehicles. The key research focus in the next year will be how to use limited information to effectively and precisely predict the surrounding traffic and, based on that, develop a cooperative driving decision scheme. Besides, the achievements in the first year will be further fine-tuned through rigorous examination.
|
