Neuro-Based Position and Force Hybrid Control of Six-Legged Walking Robot
Qing-jiu Huang* and Kenzo Nonami**
*Artificial System Science, Graduate School of Science and Technology, Chiba University, 1-33 Yayoi-cho, Inage-ku, Chiba, 263-8522 Japan
**Department of Electronics and Mechanical Engineers, Faculty of Engineers, Chiba University, 1-33 Yayoi-cho, Inage-ku, Chiba, 263-8522 Japan
We propose a six-legged walking robot having two manipulators which offers added stability, mobility, and functionality. We studied neuro-based position and force hybrid motion control for walking on irregular terrain. Comparison with conventional position and force hybrid control demonstrates the high efficiency of the proposed neuro-based position and force hybrid control. The neuro-based position and force hybrid control includes six-axis force sensors in each leg, which provide control vertically, i.e., in the direction of gravity, and in the walking direction. This platform has proven to be very useful on irregular terrain including obstacles of random height and random position. Consequently, autonomous stable walking in an unknown environment has been realized through experiments.
Copyright© 2002 by Fuji Technology Press Ltd. and Japan Society of Mechanical Engineers. All right reserved.