A swimming robot controlled by a dual-muscle work loop rig


Meeting Abstract

28.6  Monday, Jan. 5 09:15  A swimming robot controlled by a dual-muscle work loop rig RICHARDS, C.T.*; RIVERA, A.R.V.; The Royal Veterinary College; West Chester University ctrichards@rvc.ac.uk http://www.rvc.ac.uk/Staff/ctrichards.cfm

The work loop (WL) method measures muscular mechanical power underlying cyclical behaviours. A WL rig cyclically shortens and lengthens an in vitro muscle while electrically stimulating the muscle. By manipulating activation timing parameters (phase), one maps the relationship between neural input and muscle power output. However, such a map may be limited in that 1) it cannot predict locomotor performance and 2) differences in the mechanical load (e.g. water vs. land) distort the relationship between neural input and limb motion. Can shifts in WL phase modulate locomotor speed? We built a robotic Xenopus laevis frog foot ‘remote controlled’ by a modified WL rig housing the Plantaris longus (PL) and Tibalis anticus longus (TA) muscles. Extending a previous technique, PL extensor and TA flexor torques were transmitted to the foot to drive oscillating propulsion along a 2 m tank. Preliminary data suggest that shifting the PL phase earlier causes an apparent force enhancement due to activation during lengthening. However, this force enhancement did not result in faster robotic swimming speed. We speculate that speed modulation arises not by activation timing, but rather by other mechanisms such as the activation intensity and/or dynamic control of foot morphology.

the Society for
Integrative &
Comparative
Biology