We had two motives for beginning biomechanical research on the movement of snakes. The first motive was that, up until that time, the fundamental problem of "How is it that a snake can go forward without legs?" largely remained unanswered, and this required an engineering analysis. The second motive grew from the expectation that a "snake-like robot", which would be modeled on a snake would have a particularly broad functionality while maintaining a simple shape. The future possibilities of serpent robots can be anticipated from the fact, as indicated in Photo. 1, that the body of a snake, which has the simple form of a rope, functions as "legs" when moving, as "arms" when traversing branches, and as "fingers" when grasping something.

 When beginning this research, in order to explain the dynamics of the creeping propulsion movement of snakes on level ground, a basic motion equation for this was derived, and numerous running experiments were conducted using striped snakes. Photo. 2, is one example of this. The conditions for moving on level ground were investigated by rigging an electro-muscular meter and a normal force meter on the torso of the snake. From these experiments, it was found that: i) the waveform that the snake assumes during creeping movement is a curve which changes sinusoidally along the curvature of the body, and we made a formula for this, calling it a serpenoid curve; ii) the action by which one part of the body floats up during advancement, called sinus-lifting in Photo. 3, , can be interpreted as an action which concentrates the body weight on the part that can most easily slip, and this functions to prevent slippage, as indicated in Fig. 1, ; and iii) as in Fig. 2, ; , a variety of positions can be considered for the propulsion motion in the corner part within the labyrinth, but the most appropriate body form is (d), and this was also experimentally verified as in Photo. 4.