Autonomous snake robot locomotion in rough terrain depends on the robot's ability to rise from a horizontal position to a vertical, for its ability to climb obstacles. At first sight, lifting N body segments seems to require O(N/sup 2/) dynamic torque. However, in this article we describe a practical algorithm that requires only O(1) dynamic torque. The algorithm is applicable to a wide range range of snake robot morphologies. The algorithm requires little space, and control is simple, since motion occurs only in a plane. Analysis of the algorithm reveals a strong relation between the maximum dynamic torque required and the joint pitch range. We discuss some consequences for the design of snake robots. For instance, an adjustment of the mass center of the robot's end segment can reduce the maximum dynamic torque. Implications are studied for some snake robot joint designs, including Dragon, an autonomous snake robot constructed for the study of unconventional locomotion.
[1]
Timothy Ohm,et al.
The JPL Serpentine Robot: a 12-DOF system for inspection
,
1995,
Proceedings of 1995 IEEE International Conference on Robotics and Automation.
[2]
Hisato Kobayashi,et al.
Shape control of hyper redundant manipulator
,
1995,
Proceedings of 1995 IEEE International Conference on Robotics and Automation.
[3]
Gregory S. Chirikjian,et al.
A hyper-redundant manipulator
,
1994,
IEEE Robotics Autom. Mag..
[4]
Yoram Koren,et al.
Design and motion planning of a mechanical snake
,
1993,
IEEE Trans. Syst. Man Cybern..
[5]
Masahiko Hiraki,et al.
Development of a Snake-like Robot
,
1996
.
[6]
Bernhard Klaassen,et al.
GMD-Snake: A Semi-Autonomous Snake-like Robot
,
1996,
DARS.