A lightweight simulation model for soft robot's locomotion and its application to trajectory optimization
This letter extends the piecewise constant curvature modeling approach by energy minimization techniques in order to simulate contact with the environment. The new method enables a fast and sufficiently good estimation of the forward kinematics for soft structures whose behavior is mainly determined by boundary conditions. This enables to predict the configuration of such structures for a given reference input, which is demonstrated on a gecko-inspired soft robot. The newly gained simulation capability is then used to find new patterns for straight as well as curved trotting gait. The existing gait pattern for straight gait could be improved by a factor of 1.4, and a new gait pattern for the rotation on the spot could be discovered.
and learning for soft robots
soft robot applications