Worm-like robotic systems: Generation, analysis and shift of gaits using adaptive control

Inspired by the biological paradigm earthworm, this paper investigates similar mechanical systems—worm-like locomotion systems (WLLS). Their mechanical model comprises of a chain of interconnected, discrete mass points. These shall be used to mimic the wave of muscle contraction that travels through the earthworm’s body and results in locomotion.

Therefore, actuators are assumed to be positioned between each two adjacent mass points. This type of generation of global movement can be described as a form of undulatory locomotion. According to Ostrowski et al. [9], “Undulatory locomotion is the process of generating net displacements of a robotic mechanism via a coupling of internal deformations to an interaction between the robot and its environment.” Akin to the earthworm, the mechanical model has spikes that serve to fix individual segments on the ground.

Regarding the movement pattern of WLLS, current literature [5, 6, 11, 12] often takes the basic idea from the movement of biological worms. However, the mathematical description of these patterns is based on a purely kinematic view, at times just using two states for the system’s actuating elements – elongated and contracted. The generation of specific gaits is discussed, but they are not analyzed and compared in terms of dynamics.

Such analyzes are the focus of this work. For that, existing theory for gait generation [13] is used; the gaits the authors derive from kinematic and dynamic considerations are now investigated, in simulations of the dynamics of the WLLS. Since all system parameters are assumed to be unknown or, more precisely, uncertain, an adaptive controller is used to achieve the tracking of certain gaits. It does not require the knowledge of system parameters and is able to achieve the control goal via high-gain stabilization.

After the inspection of individual gaits, gait shifting is investigated, with the goal of automatic gait shifting based on the measurement of certain system variables to choose optimal locomotion patterns for given restrictions of, e.g., actuator forces.

For full text: click here

(Author: Silvan Schwebke, Carsten Behn

Published by Sciedu Press)