Publication

System Design and Field Testing of the Hominid Robot Charlie

Daniel Kuehn, Moritz Schilling, Tobias Stark, Martin Zenzes, Frank Kirchner

In: Journal of Field Robotics (JFR) 34 Issue 4 Pages 666-703 n.a. 4/2016.

Abstract

Today's and future application scenarios for mobile robots deal with increased requirements regarding autonomy and flexibility in the locomotor system. To cope with these demands, a high sensor quantity and quality allows performing robust locomotion. The authors present a hominid robotic system, which is equipped with multi-point-contact feet, an active artificial spine to incorporate extended sensing, and locomotion capabilities for walking robots. In the proposed robotic system, the front and rear part are connected via an actuated spinal structure with six degrees of freedom. In order to increase the robustness of the system's locomotion in terms of traction and stability, a foot-like structure equipped with various sensors has been developed. Altogether, the robot embodies more than 330 sensors. In terms of distributed local control, the structures feature their own local intelligence and are as autonomous as possible regarding sensing, sensor preprocessing, control, and communication. This allows the robot to respond to external disturbances with minor latency. Within this paper, the proposed robotic system and its distributed and hierarchical control method are presented. For validation purposes of the electro-mechanical and software approach, the authors present results verified experimentally in different environments (in and outdoor) with differing walking speeds on various substrates and in varying inclinations from -20° to 20°. The results show that the presented approach is viable and improves the flexibility of the locomotor system. A hominid design was chosen in order to perform various types of locomotion. To demonstrate the entire functionality of the developed hardware, two different motion modes (quadrupedal and bipedal locomotion) are investigated. This also includes a stable transition from a four-legged posture to an upright posture and vice versa.

iStruct_JFR2.pdf (pdf, 25 MB)

German Research Center for Artificial Intelligence
Deutsches Forschungszentrum für Künstliche Intelligenz