- Muñoz Romero, José
- Bijalwan, Ashutosh
Worm-like soft bodies are able to propel by activating bending waves along their bodies. However, this muscle activity is not sufficient, and needs to be accompanied by a non-isotropic frictional conditions, with higher friction orthogonal to the midline than along the tangent direction. We here present a computational model that allows: i) simulating worm locomotion on non-isotropic frictional surfaces, ii) compute an optimal bending strategy for given frictional coefficients, and iii) infer the frictional coefficients and bending from experimental measurements of the midline positions, in our case a C elegans organism. Objective i) is achieved by imposing balance of translational and angular momenta of a 2D beam formulation with active bending and configuration dependent frictional conditions. This model is employed in objective ii) to optimise an objective functional that aims at minimising the positions of the current centre of mass with respect to a target position. Finally, by changing the objective functional we solve an optimal control problem that minimises the error with respect to measured positions at each time-step, and deduce best fitting bending and frictional coefficients. We show that worms exhibit different frictional coefficients in time and space, and are therefore able to adapt their interface with respect to the substrate., This work has been financially supported by the Spanish Ministryof Science and Innovation, under Grants CEX2018-000797-S andPID2020-116141GB-I00, the local government of Generalitat deCatalunya, with Grant SGR 2021 01049, and by the Department ofMathematics of Universitat Politècnica de Catalunya., Peer Reviewed, Postprint (published version)