Kazuya Yoshida

Connected tracked robot with offset joint mechanism for multiple configurations

This paper describes various configurations of two connected unit crawlers. By changing the relative position of the two connected vehicle units, the overall robot comprising such a mechanism can automatically adapt to surface obstacles on the field, including complicated structures such as disaster-generated debris. In addition, we analyzed the effect of axis arrangements in order to simplify the realization of the switching function so as to achieve the four basic configurations.

Path Planning and Execution for Planetary Exploration Rovers Based on 3D Mapping

The unmanned exploration of the Moon has steadily increased in the past years due to the renewed interest in creating a permanent human settlement on our natural satellite. After detailed remote sensing from orbiters, the focus is now shifting onto autonomous landing and surface locomotion by robotic devices on and around a specific area of highest interest. Especially, robotic exploration is a precursor for future human settlements on the Moon and in-situ resource utilization (ISRU) is one important aspect in the selection process of candidate locations for such settlements. In the early 1960’s, Watson (Watson et al., 1961) estimated that it would be possible to find deposits of ice caps at the bottom of craters located at the Moon’s polar regions. He argued that the shadows produced on these craters due to the very small deviation of the Moon’s equatorial plane position with respect to the Sun create an environment that would present the proper conditions to retain such deposits. Robotic exploration of these regions becomes a vital initial step on the way to build a permanent base of operations for humans to live on extended periods. The navigation of a vehicle on the Moon is presented with a series of issues such as the trafficability over the lunar soil (called “regolith”), the irregularity of the topography and the poor illumination at the Polar Regions, especially at the craters. In this chapter, attention is centered on the irregularity of the topography and the poor illumination issues. Given that we assume an environment with a very low angle of illumination which results in large shadowed areas, we propose the use of Light Detection and Ranging (LIDAR) systems to perceive the surroundings of the vehicle. These systems are not impaired by the lighting conditions assumed and have been successfully utilized in several outdoor mobile robotic applications (Langer et al., 2000),(Morales et al., 2008),(Skrzypcznski, 2008). The features of the topography of the surface of the Moon may translate as obstacles for the navigation of the vehicle and also may limit the visual range of the exteroceptive sensors at a given location. These obstacles cause an “occluding” effect on the readings of the sensors therefore decreasing the size of the perceivable area. Various researchers have approached the occlusion problem presented on indoor and outdoor autonomous navigation of mobile robotic systems. In (Dupuis et al., 2005),(Rekleitis et al., 2009), an irregular triangulated mesh is created based on a LIDAR’s sensory data, and then “filtered” in order to find “shadows” or occluding obstacles and eliminate them from the map. A description of the problems that occluding obstacles may present in teleoperated navigation is presented in (Kunii & Kubota, 2006). In (Heckman et al., 2007), a method to classify different voxels based on its location with respect to an occluding obstacle is given, whereas 13

J192011 Traversal Experiments on rough terrain for "TrackWalker II" : in Mihara-Yama and Nakatajima dune

To explore in extreme environments, such as a lunar surface or volcanic area in the earth, our research group has been developing small-sized tracked robots, called “TrackWalker” and “TrackWalker II”, in a joint research project with JAXA. Each robot mounts two side sub-tracks, and each sub-track has a swing mechanism to enable simple legged locomotion for traversal on a weak a soil. From 2010 to 2011, we have conducted some field tests using the tracked robots and improved its weakness, and executed performance assessments. In 13 th , March, 2012, JAXA organized a debrief session combined with field test in Nakatajima dune, in Hamamatsu, and we performed hill-climbing demonstration. In this paper, we introduce a brief of mechanism of TrackWalker-II, and report some field tests, including Nakatajima experiment.

1P1-E01 Development of High-frequency 3D LIDAR for mobile robots to detect moving objects

近年,家庭や病院など人間が生活する環境で作業を行うサー ビスロボットの研究開発が盛んに行われている.このような人 間と共存するロボットには,環境に応じた動作を実行するため 様々な機能が要求される.その一つとして近年注目されている のが,三次元環境認識である. ロボットの三次元環境認識には,いくつかの手法があるが,主 な手法としてステレオカメラを用いたものと,3D LIDAR(Light Detection And Ranging)を用いたものに大別される.ステレオ カメラによる環境認識は広く用いられ,多くの研究報告がなさ れているが,この手法では,測定距離に応じて計測精度が低下 する,照明などの影響を受けやすいといった欠点がある.そこ で本研究グループでは,これまで3D LIDARを用いた三次元環 境認識に関する研究を進めてきた[1]. LIDARを用いた三次元環境認識では,三次元環境を距離情報 として詳細に取得できるという利点がある.しかしながら,こ の手法では,三次元環境情報をデータ点の集合によって表現す るため,計測には時間が掛かるという欠点がある.特に移動障 害物を検知する場合には,計測に時間が掛かると,計測中に移 動障害物の位置がずれてしまい,その位置を正確に把握できな いといった問題が生じる. そこで本研究では,測域センサを用いて三次元環境計測を行 い,環境中から移動物体を検出するシステムの構築を研究目的 に設定した.まず,移動物体の三次元位置を高速に把握するた め,高速三次元スキャンが可能なスキャナの開発を行った.三 次元スキャナには,水銀接点のロータリーコネクタを使用し, 測域センサを搭載したステージを無限回転可能とした.さらに 周囲環境を高速スキャンするため,ステージ回転速度を最大 180[rpm]に設定し,高速スキャンを可能とした.ただし,測域 センサで得られるデータ量は一定であるため,一回の三次元ス キャンで得られる測距データ数は少なくなる. 次に,この開発したセンサを利用して移動物体を検知するため, 逐次獲得した三次元環境情報を時間方向に比較し,差分処理を 行うことで,環境中から移動物体を検出する手法を開発した. これまで,測域センサを用いた移動障害物検知に関する研究 は,いくつか行われてきた[2][3][4].Carballoらは,測域センサ を用いて床からの高さが異なる二層の水平面を二次元スキャン Fig. 1 3D LIDAR

2A1-E25 Development of a Small-sized Leg-track Robot to traverse on loose slopes and irregular terrains

Track mechanism has high movability on irregular terrains. Therefore, it is typically used for locomotion mechanism of all-terrain robots. However, the track mechanism sometimes slips while it traverses on loose slopes. Therefore, we developed a new locomotion mechanism, called “Surface-contact-type locomotion”, which has high movability on weak soils. It uses simple leg mechanism that has wide contact area with the ground in order not to corrupt the contact surface. However, it has a disadvantage of low movability on irregular terrains. To solve the above trade-off, we developed “Leg-Track Hybrid Robot” by fusing the both locomotion mechanisms. In this paper, we explain details of the developed locomotion mechanism and report some initial experiments.

2A2-D30 Robust Positioning Device with Optical Sensor and Dual Laser Sources for Mobile Robots Traversing Slippery Terrains

This paper describes the development of a sensing device that can be used to estimate the position of mobile robots on slippery terrains. The device consists of an optical sensor designed for a computer mouse and dual laser light sources for generating a laser speckle pattern. It detects the motion of a moving surface at a large distance from the surface, from 80 mm to 300 mm, by tracking the laser speckle pattern. The use of dual laser light sources makes the tracking robust for large distances from the ground. Some fundamental experiments validated the performance of the device, which tracked surfaces with high accuracy under various height conditions. Finally, the device was mounted on our mobile robot, and simple experiments were conducted on a slippery sandy terrain to evaluate the usefulness of the device as a noncontact odometry system.

2P2-C18 Three-Dimensional Odometry for Tracked Vehicles Based on Multiple Internal Sensor Fusion

Gyro-based odometry is a robust and easy-to-use localization method for mobile robots. However, the Gyrobased odometry for tracked vehicles has difficulties to estimate its exact localization because of track-terrain slippage and gyroscopes’ bias-drifts. To solve these problems, we propose an extended 3-dimensional odometry method for tracked vehicles based on multiple internal sensor fusion. The proposed method consists of slippagecompensation using encoders and gyroscopes, attitude correction using an acceleration sensor and gyroscopes’ bias value update. Finally, a performance test was carried out in real environment to confirm a validity of our approach. In this paper, we introduce the three-dimensional localization method by extended gyro-based odometry and report the experimental results.

2P1-L01 Trajectory control of crawler type mobile robot with consideration of a slip

近年,相次ぐ震災・テロをきっかけに,人による救助に先立っ て被災現場の情報収集や要救助者の発見等を行うロボットシス テムの開発が,ロボティクス分野の各研究機関において進められ ており,本研究室でもクローラタイプの遠隔操作型レスキューロ ボットの研究開発を行っている [1]. 本プロジェクトでは,災害時に地上の通信設備が被害を受けて いることを想定し,このレスキューロボットの遠隔操作に衛星通 信を用いる.このとき問題となるのが,衛星通信に伴う数秒の時 間遅れによるロボットの操作性の低下である.このためロボット 周辺環境の情報がリアルタイムに手に入らず,ラジコンのような 速度指令による直接的な遠隔操作は非常に困難となる.この問題 に対処するため,本研究グループでは軌跡追従指令による操作の 導入を検討している.これは,オペレータより与えられた目標軌 跡に対してロボットが追従走行を行うという操作方式であり,ロ ボットに正確な軌跡追従走行性能があれば,時間遅れによる操作 性の低下を最小限に抑えた円滑な遠隔操作の実現が期待できる. 一般に,ロボットの軌跡追従走行は,自己位置推定とそれに基 づいた軌跡追従制御を行うことによって実現される.しかし,ク ローラロボットにおいては,クローラのスリップによって累積す るオドメトリの誤差のため,自己位置推定をオドメトリに依存す る正確な軌跡追従走行は困難とされてきた. そこで,本研究では,エンコーダおよびジャイロセンサより 得られた情報からクローラのスリップの大きさを推定する自己位 置推定手法 [2]を利用し,クローラロボットの軌跡追従制御を実 現した.また,実クローラロボットを用いて,追従性能の検証を 行った.