Experimental enriched beam model for robust simulation of soft material robotic systems
A recent subject of research are soft material robots that are able to perform nearly unrestricted, continuous deformations and facilitates new opportunities of actuation, control
and sensing. Their ability to adapt to obstacles, safely handle fragile objects and easily change their motion behavior makes them very promising for applications in healthcare,
human-robot interactions and rescue operations.
However, to describe their characteristics and deformation behavior, enriched models are indispensable, which account for the elastic structure, the complex material behavior and the continuous and large deformations of the robot. Hence, the objective of this research project is the development of a robust mathematical model that is valid for arbitrary soft material robots and covers interactions with the surrounding and dynamical effects. For computational efficiency, the promising Cosserat beam approach will be adopted, for which the structure of the robot is reduced to a flexible material curve in space. With regard to the models applicability and validity, the following research questions will be addressed within the research project:
1. Which effects of complex soft material robots in operation needs to be captured and how can they appropriately be addressed by the Cosserat beam approach?
2. What is the correlation between the actuation of the soft material robot and the related parameter in the model, and is a superposition valid?
3. Due to the simplified robot structure, how do contact forces and external loads need to be included in the Cosserat beam model to cover local interaction phenomena?
4. How significant is the error of the Cosserat beam model compared to models which represent the robot structure in more detail?
To answer these research questions, numerical investigations and experimental analyses on fabricated pneumatic actuated soft material robots will be conducted. The fabricated robots differ in their deformation behavior (bending, twist, elongation) and in the complexity of their structure. Basic tests help to identify the correlation between the actuation induced deformation and the related model parameters. Furthermore, additional analyzes of the robot’s behavior under applied loads and contacts help to establish the Cosserat beam approach. Throughout the project, the model’s validity will be examined and results compared to corresponding experiments. To clarify the accuracy and limits of the model, the Cosserat beam approach will be compared to geometrical more detailed finite element models.