A Study on Construction of LSTM Driver Models for Overtaking Behavior
LSTMによる追い越し動作を対象としたドライバモデルの構築に関する研究
- Delivery
- Available on this site
- Format
- Price
- Non-members (tax incl.)：￥1,100 Members (tax incl.)：￥880
- Publication code
- 20226247
- Paper/Info type
- Proceedings (Autumn)
No.136-22
- Pages
- 1-5(Total 5 p)
- Date of publication
- Oct 2022
- Publisher
- JSAE
- Language
- Japanese
- Event
- 2022 JSAE Annual Congress (Autumn)
Detailed Information
| Author(J) | 1) 馬場 智大, 2) 及川 昌子, 3) 廣瀬 敏也 |
|---|---|
| Author(E) | 1) Tomohiro Baba, 2) Shoko Oikawa, 3) Toshiya Hirose |
| Affiliation(J) | 1) 芝浦工業大学, 2) 芝浦工業大学, 3) 芝浦工業大学 |
| Affiliation(E) | 1) Shibaura Institute of Technology, 2) Shibaura Institute of Technology, 3) Shibaura Institute of Technology |
| Abstract(J) | ドライビングシミュレータで追い越し動作を対象にLSTM（Long Short-Term Memory）を用いて個人適合型のドライバモデルの構築を行った．モデル精度は縦・横方向の位置座標及び車線変更のタイミングで評価を行った．その結果，LSTMを用いることで追い越し動作の個人特性を反映したドライバモデルを構築することが可能であった． Translation |
| Abstract(E) | This study aimed to construct personalized Long Short-Term Memory (LSTM) driver models for overtaking behavior using driving simulator. LSTM can construct a model for long-term time series data and can simulate a driver behavior with high accuracy. The model accuracy was evaluated for the longitudinal and the lateral position coordinates and the timing of lane changing. As a result, it was possible to construct driver models which were simulated individual characteristics of overtaking behavior. In the future, it will be necessary to investigate whether LSTM model of overtaking behavior can be applied to Advanced Driver Assistance Systems. |
