Back to Search
Start Over
Morphologically induced stability on an underwater legged robot with a deformable body.
- Source :
- International Journal of Robotics Research; Jan2021, Vol. 40 Issue 1, p435-448, 14p
- Publication Year :
- 2021
-
Abstract
- For robots to navigate successfully in the real world, unstructured environment adaptability is a prerequisite. Although this is typically implemented within the control layer, there have been recent proposals of adaptation through a morphing of the body. However, the successful demonstration of this approach has mostly been theoretical and in simulations thus far. In this work we present an underwater hopping robot that features a deformable body implemented as a deployable structure that is covered by a soft skin for which it is possible to manually change the body size without altering any other property (e.g. buoyancy or weight). For such a system, we show that it is possible to induce a stable hopping behavior instead of a fall, by just increasing the body size. We provide a mathematical model that describes the hopping behavior of the robot under the influence of shape-dependent underwater contributions (drag, buoyancy, and added mass) in order to analyze and compare the results obtained. Moreover, we show that for certain conditions, a stable hopping behavior can only be obtained through changing the morphology of the robot as the controller (i.e. actuator) would already be working at maximum capacity. The presented work demonstrates that, through the exploitation of shape-dependent forces, the dynamics of a system can be modified through altering the morphology of the body to induce a desirable behavior and, thus, a morphological change can be an effective alternative to the classic control. [ABSTRACT FROM AUTHOR]
- Subjects :
- REMOTE submersibles
BODY size
BUOYANCY
SYSTEM dynamics
SOFT robotics
Subjects
Details
- Language :
- English
- ISSN :
- 02783649
- Volume :
- 40
- Issue :
- 1
- Database :
- Complementary Index
- Journal :
- International Journal of Robotics Research
- Publication Type :
- Academic Journal
- Accession number :
- 149575970
- Full Text :
- https://doi.org/10.1177/0278364919840426