Learning of exploratory behaviors for object recognition using reinforcement learning

被引：0

作者：

Gouko, Manabu ^{[1
]}

Kim, Chyon Hae ^{[2
]}

Kobayashi, Yuichi ^{[3
]}

机构：

[1] Department of Mechanical Engineering and Intelligent Systems, Tohoku Gakuin University

[2] Department of Mechanical Engineering, Graduate School of Engineering, Shizuoka University

来源：

Transactions of the Japanese Society for Artificial Intelligence | 2014年 / 29卷 / 01期

基金：

日本学术振兴会;

关键词：

Active perception; Exploratory behavior; Mobile robot; Reinforcement learning;

D O I：

10.1527/tjsai.29.120

中图分类号：

学科分类号：

摘要：

In this study, we propose a reinforcement learning method for discernment behaviors of robot. Discernment behavior, which is a type of exploratory behaviors that support object feature extraction, is a fundamental tool for a robot to orientate itself, operate objects and establish higher classes of knowledge. In this method, a robot learns the discernment behaviors through the interaction with multiple objects. While the interaction, the robot takes reinforcement signal according to the cluster distance of the observed data. We validated the effectiveness of the model in a mobile robot simulation. Three different shaped objects were placed beside the robot one by one. In this learning, the robot learned different behaviors corresponding to each object. Then, we confirmed the kind of feature that is extracted from an object using learned exploratory behaviors.

引用

页码：120 / 128

页数：8

共 19 条

[1]

Atil I., Dag N., Kalkan S., Sahin E., Affordances and Emergence of Concepts, Proc. of the 10th International Conference On Epigenetic Robotics, pp. 11-18, (2010)

[2]

Gibson J.J., Observations on active touch, Psychological Review, 69, 6, pp. 477-491, (1962)

[3]

Gibson E.J., Exploratory behavior in the development of perceiving, acting, and the acquiring of knowledge, Annual Review of Psychology, 39, 1, pp. 1-42, (1988)

[4]

Griffith S., Sinapov J., Sukhoy V., Stoytchev A., How to Separate Containers From Non-Containers? A Behavior-Grounded Approach to Acoustic Object Categorization, Proc. of 2010 IEEE International Conference On Robotics and Automation (ICRA2010), pp. 1852-1859, (2010)

[5]

Held R., Hein A., Movement-produced stimulation in the development of visually guided behavior, Journal of Comparative and Physiological Psychology, 56, 5, pp. 872-876, (1963)

[6]

Maris M., Boeckhorst R., Exploiting physical constraints: Heap formation through behavioral error in a group of robots, Proc. of the 1996 IEEE/RSJ International Conference On Intelligent Robots and Systems (IROS1996), 3, pp. 1655-1660, (1996)

[7]

Minato T., Asada M., Towards selective attention: Generating image features by learning a visuo-motor map, Robotics and Autonomous Systems, 45, 3-4, pp. 211-221, (2003)

[8]

Morimoto G., Ikegami T., Evolution of Plastic Sensory-motor Coupling and Dynamic Categorization, Artificial Life IX, pp. 188-193, (2004)

[9]

Nakamura T., Nagai T., Iwahashi N., Multimodal Object Categorization by a Robot, Proc. of the 2007 IEEE/RSJ International Conference On Intelligent Robots and Systems (IROS2007), pp. 2415-2420, (2007)

[10]

Nishide S., Ogata T., Tani J., Komatani K., Okuno H.G., Predicting Object Dynamics From Visual Images Through Active Sensing Experiences, Advanced Robotics, 22, 5, pp. 527-546, (2008)

← 1 2 →