Curriculum Vitaes
  1. Researcher Search Results
  2. Norifumi Watanabe

Norifumi Watanabe

  (渡邊 紀文)

Profile Information

Affiliation
Musashino University
Degree
政策・メディア(慶應義塾大学)
J-GLOBAL ID
201301073774840492
researchmap Member ID
B000230846
慶應義塾大学大学院政策・メディア研究科後期博士課程修了．玉川大学脳科学研究所嘱託研究員を経て，東京工科大学コンピュータサイエンス学部助教，産業技術大学院大学産業技術研究科情報アーキテクチャ専攻助教，武蔵野大学データサイエンス学部・教育部会准教授．博士(政策・メディア)．
専門は知覚情報処理，神経情報処理，認知科学．視覚情報処理に関係する神経細胞のモデル化と，計算機によるシミュレーション，また近年応用研究として人間の意図を推定し，行動を支援するインタフェースの開発，更に知能を持ったロボットの実現を目指したロボカップへも出場している．

Papers

 46
  • Norifumi Watanabe, Kensuke Miyamoto
    ICA, 37-42, 2023  
  • Kota Itoda, Norifumi Watanabe, Yasushi Kiyoki
    2022 13th International Congress on Advanced Applied Informatics Winter (IIAI-AAI-Winter), Dec, 2022  
  • Kensuke Miyamoto, Norifumi Watanabe, Osamu Nakamura, Yoshiyasu Takefuji
    Applied Sciences, 12(17) 8720-8720, Aug 31, 2022  
    Human cooperative behavior includes passive action strategies based on others and active action strategies that prioritize one’s own objective. Therefore, for cooperation with humans, it is necessary to realize a robot that uses these strategies to communicate as a human would. In this research, we aim to realize robots that evaluate the actions of their opponents in comparison with their own action strategies. In our previous work, we obtained a Meta-Strategy with two action strategies through the simulation of learning between agents. However, humans’ Meta-Strategies may have different characteristics depending on the individual in question. In this study, we conducted a collision avoidance experiment in a grid space with agents with active and passive strategies for giving way. In addition, we analyzed whether a subject’s action changes when the agent’s strategy changes. The results showed that some subjects changed their actions in response to changes in the agent’s strategy, as well as subjects who behaved in a certain way regardless of the agent’s strategy and subjects who did not divide their actions. We considered that these types could be expressed in terms of differences in Meta-Strategies, such as active or passive Meta-Strategies for estimating an opponent’s strategy. Assuming a human Meta-Strategy, we discuss the action strategies of agents who can switch between active and passive strategies.
  • Norifumi Watanabe, Kota Itoda
    Proceedings of the 14th International Conference on Agents and Artificial Intelligence, 299-305, 2022  
  • Norifumi Watanabe, Kensuke Miyamoto
    2022 JOINT 12TH INTERNATIONAL CONFERENCE ON SOFT COMPUTING AND INTELLIGENT SYSTEMS AND 23RD INTERNATIONAL SYMPOSIUM ON ADVANCED INTELLIGENT SYSTEMS (SCIS&ISIS), 1-5, 2022  
    In human cooperative behavior, there are some strategies: a passive behavioral strategy based on others' behaviors and an active behavioral strategy based on the objective-first. However, it is unclear how to acquire a meta-strategy to switch those strategies. In this study, we conduct a collision-avoidance experiment with agents taking multiple strategies in a grid-like corridor to see whether the subject's behavior changes when the agent's strategy changes. Furthermore, we compare the behavior selected by the subjects with the behavior of the agents acquired by reinforcement learning. The experimental results show that subjects can read the change in strategy from the behavior of the opposing agent.
More

Misc.

 77
  • 2015 2K4OS14a3-2K4OS14a3, 2015  
  • HAGIMOTO Yuki, SUZUKI Toshiaki, WATANABE Norifumi, OMORI Takashi, KAMEDA Hiroyuki
    JSAI Technical Report, Type 2 SIG, 2014(SAI-020) 01, Nov 8, 2014  
    Our study aims at discover and create a new method of how inductively to reason effective attack patterns by using much amounts of log data in RoboCup soccer simulation 2D. The inductive reasoning of those patterns makes it feasible to implement a RoboCup soccer team easier, and also to predict results of soccer games more precisely than before. This paper presents an experiment to reason effective attack patterns from a collection of simple actions of ball-holder soccer players. As a result, some of effective attack patterns can be handled successfully, but sequences of actions which are of importance in soccer game analysis cannot properly. This suggests sampling method should be improved when reasoning inductively.
  • NAGANO Yoshihiro, WATANABE Norifumi, AOYAMA Atsushi
    IEICE technical report. Neurocomputing, 113(500) 47-52, Mar 17, 2014  
    Recent physiological studies have indicated that visual attention modulates neural activity depending on the cell types in V4. Specifically, narrow spiking (NS) neurons and broad spiking (BS) neurons, which are classified according to the spiking waveforms, have both reported to exhibit increased firing rate and less inter-trial variations for attention. We simulated the NS and BS neural activity for visual attention using a neural network model based on spontaneous activity, under the assumption that the visual attention modulation would be shown by the change in variance of input firing rate distributed with a lognormal distribution. Consistent with the physiological studies, an increase in firing rate and a decrease in inter-trial variance was simultaneously obtained in simulations by increasing variance of input firing rate distribution. These results indicate that visual attention forms strong sparse and weak dense input or a ' winner-take-all' state, to improve the signal-to-noise ratio of the target information.
  • 糸田孝太, 渡邊紀文, 武藤佳恭
    知能と情報(Web), 26(3), 2014  
  • 2014 1M5OS05b4-1M5OS05b4, 2014  
  • SUZUKI Toshiaki, HAGIMOTO Yuki, WATANABE Norifumi, KAMEDA Hiroyuki, OMORI Takashi
    JSAI Technical Report, Type 2 SIG, 2013(SAI-018) 04, Nov 10, 2013  
    We analyzed the shared knowledge of one-two-pass in the game of human soccer. From the analysis, we suggested an action algorithm for one-two-pass and have implementation it in RoboCup 2D Soccer Simulation. From an evaluation, we realized success rate of 20 percent in the simulation.
  • TSUKIE Nobuhiro, HATTA Takumi, HARADA Toshinobu, WATANABE Norifumi, KUBOMURA Chiaki, IKEBUCHI Emi, KAMEDA Hiroyuki
    Proceedings of the Society Conference of IEICE, 2013 127-127, Sep 3, 2013  
  • 宮本 賢良, 吉岡 裕彬, 渡邊 紀文
    ファジィシステムシンポジウム講演論文集, 29 505-508, Sep, 2013  
  • 渡邊 紀文, 森 文彦, 大森 隆司
    ファジィシステムシンポジウム講演論文集, 29 536-541, Sep, 2013  
  • MORI Fumihiko, WATANABE Norifumi, OMORI Takashi
    IEICE technical report. Neurocomputing, 112(480) 25-30, Mar 13, 2013  
    Human walking is affected by vision, vestibular, somatic and other various sensations that come through the sensory-motor loop. But detail of the sensory-motor loop is not clear. In this study, we examined a possible affect of self motion sensation by an optical-flow stimulus in peripheral vision with a decayed somato-sensory feeling by a vibration stimulus on leg and foot area. In the experiment, we presented the optical flow for forward direction to the peripheral vision, and then gave the self-motion sensation by changing the flow to left or right direction. We expect the change of walking direction toward opposite of the self-motion sensation because of the decayed somato-sensory feeling. In this paper, we discuss on the unifying mechanism of visual and somatic sensations based on the experimental result.
  • YOSHIOKA Hiroaki, MIYAMOTO Kensuke, WATANABE Norifumi, TAKEFUJI Yoshiyasu, ISHIZAKI Shun
    IEICE technical report. Neurocomputing, 112(480) 107-112, Mar 13, 2013  
    For robots to assist human behaviors, they must check human actions and. estimate the behavioral intentions. This study focuses on the person following behaviors by an autonomous robot to estimate human behavioral intentions. We analyzed the following human smooth behavior by checking pedestrians to keep up with people and estimating the behavioral intentions. And we had to clarify the characteristics on the timing of moving to the left and right and the person following behavioral trajectory. In addition, we have implemented the following behavior by an autonomous robot that appeared characteristics of human behavior by the reinforcement learning.
  • 吉岡裕彬, 宮本賢良, 渡邊紀文, 武藤佳恭
    計測自動制御学会システム・情報部門学術講演会講演論文集(CD-ROM), 2013, 2013  
  • OYAMA Eimei, SHIROMA Naoji, NIWA Masataka, WATANABE Norifumi, SHINODA Shunsuke, OMORI Takashi, SUZUKI Natsuo
    The Proceedings of JSME annual Conference on Robotics and Mechatronics (Robomec), 2013 _1A2-R17_1-_1A2-R17_3, 2013  
    A Head Mounted Display (HMD) is the most popular display device for Virtual Reality (VR), Augmented Reality (AR), Mixed Reality (MR), Telexistence, and remote Behavior Navigation. However, it is difficult to realize both large field of view (FOV) and natural correspondence between the displayed image and the real image with an HMD. In order to improve the performance of the HMD, we propose the simultaneous utilization of both a slenderized HMD for central vision and a projection display for peripheral vision. In this paper, the configuration of the prototype of the HMD/projector hybrid display system, and the experimental evaluation are presented.
  • Miyamoto Kensuke, Yoshioka Hiroaki, Watanabe Norifumi, Takefuji Yoshiyasu
    Proceedings of the Fuzzy System Symposium, 29 116-116, 2013  
    In order for the robots to help people, it is necessary to observe behavior of human and estimate their intention. In this study, use collision avoidance as example of the cooperative action. We measured and analyzed human walking of collision avoidance. And we found that speed of opponent foot probably enables robot to determine human moves which direction.
  • Watanabe Norifumi, Mori Fumihiko, Omori Takashi
    Proceedings of the Fuzzy System Symposium, 29 123-123, 2013  
    We presented the lateral optical flow for forward direction to the peripheral vision, and then gave the body sway by vibrating subject's foot. We expect the change of walking direction toward opposite of the self-motion sensation because of the decayed somato-sensory feeling. From experimental results, there is a difference between right and left to the guiding effect, the effect appears dependent on right and left foot take in stimulus presentation time.
  • Watanabe Norifumi, Omori Takashi
    Proceedings of the Fuzzy System Symposium, 28 512-515, 2012  
    Our environmentally-based perceptual capacity is often challenged under crowded conditions like emergency evacuations. In order to achieve more effective direction guidance, we considered the use of unconscious reflexes in human walking. In this study, we experimented with vision-guided walking direction control by inducing subjects to shift their gaze direction using optical flow in peripheral visual field. We confirmed that a shift in a subject's walking direction could be induced by a combination of optical flow and vibratory stimulus on his legs. We propose a switching mechanism for visual and somatosensory input based on inducement timing.
  • 渡邊 紀文, 三門 裕明, 大森 隆司
    ファジィシステムシンポジウム講演論文集, 27 291-294, Sep, 2011  
  • WATANABE Norifumi, NISHINO Kenta, OMORI Takashi
    IEICE technical report, 110(462) 7-11, Mar 1, 2011  
    In daily life, our action is induced to us by various media like the signature of the direction board and the gateway etc. However, our sense obtained from the environment is restricted in a crowded event site and the emergency evacuation, etc. Therefore, it is difficult that we pay attention to the inducement medium. In this study, we aimed at the person's action inducement in congestion avoidance, controlled our gaze with the device, and have a pedestrian navigation experiment that induced the body to the line of sight.
  • Watanabe Norifumi, Mikado Hiroaki, Omori Takashi
    Proceedings of the Fuzzy System Symposium, 27 68-68, 2011  
    In daily life, our behavior is induced by various visual media like signatures of the direction board and gateway etc. However, our environment perception is restricted in a crowded event site and the emergency evacuation, etc. Therefore, it is difficult to pay attention to these inducement media. In this study, we have a vision-guided walking experiment by controlling subject's gaze using visual illusion and body sway. In this experiment, we verified that subject's body was induced to eye gaze direction. From the analysis of the inducement timing, we propose a perceptual switching machinery model of visual and somatic sense.
  • 三門 裕明, 渡邊 紀文, 大森 隆司
    人工知能学会第二種研究会資料, 2010(SAI-011) 07, Dec 4, 2010  
    人はある程度混雑した状況でも安全な歩行が可能であるが，イベント会場や災害時などの非常に混雑した状況での安全な歩行は難しい．このようなパニック状況では，非常灯や矢印等の提示・認識による意識的な誘導よりは，無意識的な歩行誘導のほうがより効果的であると考えられる．このような無意識下で歩行者を誘導するデバイスとしてはGVS等があるが，それらの効果を最大限に引き出すには，歩行時の人の行動決定過程を理解したうえで働きかけをデザインする必要があろう．そこで本研究では，人のすれ違い状況における回避判断に必要となる他者の行動についての情報と，自身の運動生成に影響を与える身体部位について調査し，歩行者の行動決定過程についてのモデルを検討した．
  • 渡邊紀文
    日本知能情報ファジィ学会ファジィシステムシンポジウム講演論文集, Sep. 2010, 26 276-279, 2010  
  • 矢作 拓也, 渡邊 紀文
    人工知能学会全国大会論文集, 2010 1C51-1C51, 2010  
    <p>本研究では，人間のサッカープレイデータを分析し，そこで得られた自己と他者の相互作用に必要な視線パラメータを用いたモデルの構築及び，ロボカップサッカーシミュレーションへ実装することで，チームワーク形成過程を解明し，効率的なチームプレーを形成した．このような人間の機能を理解する構成論的手法を用いることで，新たなチームワーク研究手法の構築及び提案を行う．</p>
  • 篠田 孝祐, 渡邊 紀文, 大山 英明
    人工知能学会全国大会論文集, 24 1-3, 2010  
  • Watanabe Norifumi, Mikado Hiroaki, Omori Takashi
    Proceedings of the Fuzzy System Symposium, 26 62-62, 2010  
    We decide and execute our action from many types of environmental information in our daily lives even if we are not conscious of being guided. The human action induced from the opposing person's movement is the collision avoidance of passing each other. In collision avoidance, we chiefly judge the avoidance direction from visual information. Especially, it is important to get the information from oncoming person's body part and avoidance timing in each other. Then, we make an experiment to judge the avoidance direction by watching the masking movie of oncoming person's body part. By evaluating this judgment time, it was clarified oncoming person's body part is leg in collision avoidance. Next, it especially paid attention to oncoming person's leg, and the relation between the walking cycle and leg position in avoidance judgment is evaluated. From this result, the avoidance judgment is possible because the traveling direction can be controlled by the leg when the leg is lifting and landing. It was clarified that oncoming person's walking cycle is important in the action decision in collision avoidance. So we propose the action decision model based at the walking cycle from these results.
  • 渡邊 紀文
    人工知能, 24(1) 123_2-123_2, Jan 1, 2009  
  • Watanabe Norifumi, Toshima Iwaki, Omori Takashi, Maeda Taro
    Proceedings of the Fuzzy System Symposium, 25 26-26, 2009  
    Galvanic Vestibular Stimulation (GVS) is known as a method of giving influence on sense of equilibrium. The vestibular system is stimulated by a weak current through an electrode placed on the mastoid behind the ear. In this paper, we had a motion guidance experiment of walking and the standstill operation by using GVS to evaluate the effect. And we measured the amount of the motion guidance with a motion capture system. We propose a motion guidance model to the change in a fixed fuel spray travel, time, the walking speed and compared it with the experiment result.
  • WATANABE Norifumi, ISHIZAKI Shun
    IEICE technical report. Neurocomputing, 101(737) 1-8, Mar 13, 2002  
    It is believed that the processes of perception, cognition, and reasoning can be ex-pressed with time series of neuron spikes in the brain. We do not have, however, any answer to the question, "What kind of neuron dynamics in the brain works and is able to express the processes?" In order to solve this problem, we need to study methods of neural coding which shows the way to express information on a spike sequence in the brain. It is also necessary to systematize the dynamics on computers. Another important problem is to study a structure of binding information of different media, for example, shapes or colors. In this research, the simulation of the information binding problem adopts theory of "Temporal Coding". Then adopted another model which is extended from the simple model of the abstract coincidence detector which Mr. Fujii proposed in 1996. This model uses also the theory of morephoelectrotonic transform used in the neurophysiology to show dynamical changes of a model. This model was named Binding Simulator Model, and could show plausibility and effectiveness by simulation experiments.

Books and Other Publications

 2

Presentations

 74
  • 糸田孝太, 渡邊紀文, 清木康
    第40回日本ロボット学会学術講演会（RSJ2022）, Sep, 2022
  • MIYAMOTO Kensuke, WATANABE Norifumi, TAKEFUJI Yoshiyasu, NAKAMURA Osamu
    Proceedings of the Annual Conference of JSAI, 2022, The Japanese Society for Artificial Intelligence
    In humans' cooperative behavior, there are two types of behavioral strategies: passive behavioral strategies based on the others, and active behavioral strategies based on the objective-first. In order to realize a robot that can use different strategies and communicate like a person, we created an agent that can switch between active and passive strategies. However, it is not clear whether people change their own behavioral strategies according to each strategy. In this study, we conducted an experiment in which agents with multiple strategies of actively giving way and passively giving way passed each other in a grid-like space, and analyzed whether people's behavior changed when the agents' strategies changed. The results show that, in addition to subjects who change their own behavior in response to changes in the agent's strategy, there are also subjects who behave in a certain way regardless of the agent's strategy and subjects whose behavior is not clearly divided.
  • 糸田孝太, 渡邊紀文
    日本知能情報ファジィ学会 ファジィ システム シンポジウム 講演論文集, Sep, 2021
  • MIYAMOTO Kensuke, WATANABE Norifumi, TAKEFUJI Yoshiyasu
    Proceedings of the Annual Conference of JSAI, 2021, The Japanese Society for Artificial Intelligence
    In human's cooperative behavior, there are some strategies: a passive behavioral strategy based on others’behaviors and an active behavioral strategy based on the objective-first. However, it is not clear how to acquire a meta-strategy to switch those strategies. In this study, we conduct a collision avoidance experiment with agents taking multiple strategies in a grid-like corridor to see whether subject's behavior changes when agent's strategy changes. We compare the behavior selected by the subjects with the behavior of the agents acquired by reinforcement learning. The experimental results show that subjects can read the change in strategy from the behavior of the oncoming agent.
  • Dec 12, 2020
More

Research Projects

 4

Social Activities

 1

Media Coverage

 4
To the list screen