Modeling multiagent cooperative behavior on soccer games and its simulation evaluation
Project/Area Number |
25350798
|
Research Category |
Grant-in-Aid for Scientific Research (C)
|
Allocation Type | Multi-year Fund |
Section | 一般 |
Research Field |
Sports science
|
Research Institution | Fukuoka University |
Principal Investigator |
|
Co-Investigator(Kenkyū-buntansha) |
OMORI TAKASHI 玉川大学, 工学部, 教授 (50143384)
YAMADA NOBUYUKI 玉川大学, 教育学部, 准教授 (70245968)
WATANABE NORIFUMI 東京工科大学, コンピュータサイエンス学部, 助教 (30534721)
|
Co-Investigator(Renkei-kenkyūsha) |
SHINODA KOUSUKE 独立行政法人理化学研究所, 光グリーンテクノロジー特別研究ユニット, 協力研究員 (90533191)
KAWARABAYASHI TOMOMI 福井工業高等専門学校, 電気電子工学科, 講師 (10455131)
|
Project Period (FY) |
2013-04-01 – 2016-03-31
|
Project Status |
Completed (Fiscal Year 2015)
|
Budget Amount *help |
¥4,420,000 (Direct Cost: ¥3,400,000、Indirect Cost: ¥1,020,000)
Fiscal Year 2015: ¥1,170,000 (Direct Cost: ¥900,000、Indirect Cost: ¥270,000)
Fiscal Year 2014: ¥1,170,000 (Direct Cost: ¥900,000、Indirect Cost: ¥270,000)
Fiscal Year 2013: ¥2,080,000 (Direct Cost: ¥1,600,000、Indirect Cost: ¥480,000)
|
Keywords | サッカー / シミュレーション / 行動モデル |
Outline of Final Research Achievements |
In this research, we examined the computational model of decision making for cooperative behavior, including the intention estimation of others in a group situation. In the tactical pass model problem, we examined the model of pass receiver prediction acquired by logistic regression and the dominant region model, using the real data of human soccer games. We also used the movement of the player's line of sight as a key feature to extract cooperative action pattern form the gaze behavior in real ball games. In addition, we proposed a cooperative action planning technique using a tree search method and a learning method to acquire the evaluation function for the search.
|
Report
(4 results)
Research Products
(18 results)