Multiple Object Motion Prediction Using Deep Convolutional Neural Networks

Deep Convolutional Neural Network を用いた複数障害物の行動予測

This paper presents multiple object motion prediction using Deep Convolutional Neural Network (DCNN). Specifically the authors' approach generates a potential map from the location data of multiple objects, and the DCNN learns to predict the future potential from the previous time frame. Advantages of this model are enabling the behavior prediction without using tracking by acquiring each object’s successive state. In addition, it was confirmed in the authors' experiment that their model enables the multiple object detection at about 6 msec. per frame using GeForce GTX TITAN X of NVIDIA’s GPU. Deep Convolutional Neural Network(DCNN)を用いた行動予測手法を提案する．本手法は，複数の障害物の位置情報からリスクを表現するポテンシャルマップを作成し，過去から現在のポテンシャルマップから，未来のポテンシャルマップの予測を訓練する．本手法の特徴は，対象の連続的な変化をトラッキングすることなくDCNNで解釈し，予測を実現することである．

• Record URL:
  https://doi.org/10.11351/jsaeronbun.49.522
• Record URL:
  http://scholar.google.com/scholar_lookup?title=Multiple%20Object%20Motion%20Prediction%20Using%20Deep%20Convolutional%20Neural%20Networks&author=R.%20Saiin&author=T.%20Daimon&author=T.%20Yamashita&author=M.%20Nakamura&publication_year=2018
• Availability:
  • Find a library where document is available. Order URL: http://worldcat.org/issn/02878321
• Authors:
  • Saiin, Ryuji
  • Daimon, Tomohiro
  • Yamashita, Takayoshi
  • Nakamura, Masahiko
• Publication Date: 2018-3

Language

• English
• Japanese

Media Info

• Media Type: Digital/other
• Features: Figures; Photos; References; Tables;
• Pagination: pp 522-527
• Serial:
  • Transactions of Society of Automotive Engineers of Japan
  • Volume: 49
  • Issue Number: 2
  • Publisher: Society of Automotive Engineers of Japan
  • ISSN: 0287-8321
  • EISSN: 1883-0811
  • Serial URL: https://www.jstage.jst.go.jp/browse/jsaeronbun
• Publication flags:

  Open Access (libre)

Subject/Index Terms

• TRT Terms: Advanced vehicle control systems; Automatic vehicle detection and identification systems; Control systems; Field studies; Intelligent vehicles; Machine learning; Mapping; Mathematical prediction; Neural networks; Passing
• Subject Areas: Data and Information Technology; Highways; Vehicles and Equipment;

Filing Info

• Accession Number: 01670663
• Record Type: Publication
• Source Agency: Japan Science and Technology Agency (JST)
• Files: TRIS, JSTAGE
• Created Date: May 29 2018 4:03PM