Masaki Fukuchi

3D Perception and Environment Map Generation for Humanoid Robot Navigation

A humanoid robot that can go up and down stairs, crawl underneath obstacles or simply walk around requires reliable perceptual capabilities for obtaining accurate and useful information about its surroundings. In this work we present a system for generating three-dimensional (3D) environment maps from data taken by stereo vision. At the core is a method for precise segmentation of range data into planar segments based on the algorithm of scan-line grouping extended to cope with the noise dynamics of stereo vision. In off-line experiments we demonstrate that our extensions achieve a more precise segmentation. When compared to a previously developed patch-let method, we obtain a richer segmentation with a higher accuracy while also requiring far less computations. From the obtained segmentation we then build a 3D environment map using occupancy grid and floor height maps. The resulting representation classifies areas into one of six different types while also providing object height information. We apply our perception method for the navigation of the humanoid robot QRIO and present experiments of the robot stepping through narrow space, walking up and down stairs and crawling underneath a table.

Stair climbing for humanoid robots using stereo vision

For the fully autonomous navigation in a 3 dimensional world, a humanoid robot must be capable of stepping up and down staircases and other obstacle where a sufficient large flat surface can support the robots feet. This paper presents methods for the recognition of stairs and a control architecture that enable the humanoid robot QRIO to safely climb up and down in its environment The approach is based on data captured by a stereo vision system and segmented into planar surfaces. From the segmented planes, stairs that can be climbed by the robot are extracted and fed to a control system which decides the action to be taken next. Experimental results on a staircase as well as climbing up and down a sill are presented.

A Floor and Obstacle Height Map for 3D Navigation of a Humanoid Robot

With the development of biped robots, systems became able to navigate in a 3 dimensional world, walking up and down stairs, or climbing over small obstacles. We present a method for obtaining a labeled 2.5D grid map of the robots surroundings. Each cell is marked either as floor or obstacle and contains a value telling the height of the floor or obstacle. Such height maps are useful for path planning and collision avoidance. The method uses a novel combination of a 3D occupancy grid for robust sensor data interpretation and a 2.5D height map for fine resolution floor values. We evaluate our approach using stereo vision on the humanoid robot QRIO and show the advantages over previous methods. Experimental results from navigation runs on an obstacle course demonstrate the ability of the method to generate detailed maps for autonomous navigation.

A modular architecture for humanoid robot navigation

We present a three level architecture for humanoid robot navigation. The perception layer creates a 2D environment model from range data and foot step information and classifies areas into one of six different types. The control layer is organized into behavior modules each representing different motion capabilities of the robot. Based on the environment classification, a suitable sequence of actions is selected for navigating the robot to a goal location. On the planning layer, collision-free paths are searched on the environment map and provided to the control layer. The system is modular as behaviors can be added and removed. We verify our approach on Sonys QRIO robot and present experimental results in an environment containing stairs, a table, and obstacles of different height

Robot device with changing dialogue and control method therefor and storage medium

Sentences corresponding to internal statuses of a robot device or the like are created and uttered, thereby expressing the internal statuses. The robot device or the like comprise means for recognizing an external status, and means for generating an emotion based on the internals status, whereby a change in the emotion is reflected upon a dialogue. The internal status is not associated with a sentence, but it exists independently of the system and is always varied depending on various external inputs and internal changes of the system. Accordingly, even when the same question is made on the robot device or the like, the contents of a reply are changed depending on the internal status at that time, and a manner of providing a reply also differs depending on the internal status.

Environment identification by comparing maps of landmarks

This paper describes a method for identifying an environment a robot is operating in by comparing the geometry of landmarks of a map the robot is currently building with a set of previously created maps. Landmark maps are created using the stochastic map approach originally presented by Smith, Self and Cheeseman [1990]. The paper provides a method for measuring the similarity of such maps and presents a closed form solution for the special case that covariances are constant. Experiments carried out on a developer version of the AlBO robot and on a prototype of the humanoid SDR-4X robot show that the approach is applicable even for robots with poor odometry.

The focus of expansion attracts attention and eye gaze

s S59 O2-I1-6 Early postnatal stress alters fear extinction-associated prefrontal synaptic plasticity of adult rats Machiko Matsumoto, Daisuke Yamazaki, Sachiko Hiraide, Chika Jyudo, Hiroko Togashi Dept. Phyamacol., Health Sci. Univ. of Hokkaido, Japan The present study elucidated whether synaptic transmission in the medial prefrontal cortex (mPFC) underlying contextual fear extinction was altered in freely moving adult rats exposed to footshock (FS) during the third (3WFS) postnatal week. Evoked potential in the mPFC exerted a sustained increase after extinction training in Non-FS control, whereas this LTP-like response was not observed in 3WFS group, which impaired the extinction processes. The partial NMDA receptor agonist D-cycloserine administered before and after extinction training, ameliorated impairment of extinction accompanied with synaptic facilitation, in 3WFS group. These results suggest that early postnatal stress participates in long-lasting alterations in the synaptic plasticity in the mPFC underlying fear extinction. Considering the similarities between exposure psychotherapy and fear extinction training in animals, D-cycloserine may have clinical significance in the treatment of anxiety-related disorders, such as posttraumatic stress disorder. doi:10.1016/j.neures.2009.09.157 O2-I2-1 Approach to neural processing of calculation based on cerebral oscillation: comparison between abacus experts and nonexperts Masayuki Hirata1,2, Kanako Kiso2, Tetsu Goto1, Toshiki Yoshimine1, Shiro Yorifuji 2 1 Dept. Neurosurg., Osaka Univ. Med. School, Japan; 2 Div. of Func. Diag. Sci., Osaka Univ. Grad. School of Med., Japan Abacus experts exert marvelous mental calculation capability. They say that they perform metal calculation using ‘an imaginary abacus’. The aim of this study is to elucidate neural processing of calculation and neural mechanism of special brain ability. Using beamforming MEG, we investigated spatiotemporal distribution of cerebral oscillatory changes during complex mental calculation, and compared between abacus experts and non-experts. In non-experts, sustained desynchronization was observed in intraparietal sulcus. In contrast, in experts, desynchronization in parietooccipital junction was transient, while desynchronizations in right inferior frontal gyrus and in left precentral gyrus were sustained. These results suggested that non-experts use mainly parietal numerical processor as expected, while experts, after transient numeral processing in pareitooccipital areas, utilize prefrontal and motor working memory exactly as ‘an imaginary abacus’. doi:10.1016/j.neures.2009.09.158 O2-I2-2 Granger causality defines neural networks of orienting and holding of attention Takashi J. Ozaki1, Seiji Ogawa2 1 BSI, RIKEN, Japan; 2 Kansei. Fukushi. Res. Cent., Tohoku Fukushi Univ.,

Multiple plane segmentation from stereo with the application of stair climbing for humanoid robots

Abstract Segmenting range images into planar regions is an important technique for many applications including mobile robot navigation. We present an efficient method for precise segmentation of stereo range images into multiple planar regions. The method is adaptive to noise contained in the data, and thus can produce superior results than other standard techniques. We apply the method on the humanoid robot QRIO for staircase recognition and develop a behavior for climbing stairs. Experimental results show the real-time capability and accuracy of the proposed method.