RNN-Based Path Prediction of Obstacle Vehicles With Deep Ensemble

Min, Kyushik; Kim, Dongchan; Park, Jongwon; Huh, Kunsoo

doi:10.1109/TVT.2019.2933232

상세 보기

RNN-Based Path Prediction of Obstacle Vehicles With Deep Ensemble

Min, Kyushik;
Kim, Dongchan;
Park, Jongwon;
Huh, Kunsoo

Citations

WEB OF SCIENCE

63

Citations

SCOPUS

82

초록

In this paper, a new approach for obstacle vehicle path prediction, which is important for advanced driver assistance systems (ADAS) and autonomous vehicles, is proposed based on a deep neural network. In order to analyze sequential sensor data, a recurrent neural network (RNN) is used and the input data for RNN is drawn from three sensors: LIDAR, camera and GPS. These sensor data are obtained experimentally with real vehicles. In addition, deep ensemble is used for robustness of the estimation and acquisition of the uncertainty. The predicted path of the proposed method is continuous and it predicts both short-term and long-term path with a single algorithm. The size of the network model is small, but it shows good performance in predicting future trajectory of obstacle vehicles.

키워드

Path prediction; deep learning; ADAS; ensemble; Advanced driver assistance systems; Automobile drivers; Deep learning; Forecasting; Recurrent neural networks; Vehicles; ADAS; ensemble; Network modeling; New approaches; Path prediction; Real vehicles; Recurrent neural network (RNN); Three sensors; Deep neural networks

제목: RNN-Based Path Prediction of Obstacle Vehicles With Deep Ensemble

저자: Min, Kyushik; Kim, Dongchan; Park, Jongwon; Huh, Kunsoo

DOI: 10.1109/TVT.2019.2933232

발행일: 2019-10

유형: Article

저널명: IEEE Transactions on Vehicular Technology

권: 68

호: 10

페이지: 10252 ~ 10256