- Principal investigators
- Staff
- Begin
-
01/07/2017
- End
-
30/06/2017
-
The main goal of the project is to develop and realize efficient model reduction schemes for simulating complex soft tissue models, in particular, structural and functional aspects of the musculoskeletal (human) system in the context of control problems.
Modeling of the soft tissue involves many parameters and nonlinearities depending on the different attributes (internal or external), such as elasticity, viscosity, plasticity, shape etc. In the path of our research, we shall consider a hierarchy of muscle models by addressing different levels of complexity, namely a linearized pure mechanical muscle model, a full nonlinearmechanical muscle model and a model containing electrophysiology. Simulation for such bodies, in the context of a robot with hard gripping facility interacting with the body and transfer to a target position, is expensive, which demands a lot of computational time and storage capacity. In this regard, MOR is the state of art to build a reduced model for obtaining the solution in the case of multiquery and real-time situations. Besides simulation of the body, we concern about the path, along which the soft body moves, the velocity of the motion, the force of the gripper, possibly subject to constraints, especially if there are obstacles along the path of the movement. In the framework of control theory, we formulate suitable optimal control problems. We shall investigate both open loop problems and closed loop or feedback control problems.
To be more precise, in this project we shall aim to implement Model predictive control (MPC) and reduced basis technique to solve the problem.
This project is funded by German Research Foundation (DFG) within IRTG 2198/1