Refine
Year of publication
- 2020 (52) (remove)
Document Type
- Article (40)
- Part of a Book (5)
- Conference Proceeding (4)
- Doctoral Thesis (2)
- Book (1)
Keywords
- Adaptive control (1)
- Brownian Pillow (1)
- Choleratoxin B (1)
- Crámer–von-Mises distance (1)
- Hoeffding-Blum-Kiefer-Rosenblatt independence test (1)
- Incomplete data (1)
- Iterative learning control (1)
- Marginal homogeneity test (1)
- Paired sample (1)
- Resampling test (1)
- Robotic rehabilitation (1)
- Trinkwassersicherheit (1)
- agility (1)
- cholera toxin B (1)
- community dwelling (1)
- drinking water safety (1)
- healthy aging (1)
- magnetic frequency mixing technique (1)
- not identically distributed (1)
- prevention (1)
Institute
- Fachbereich Medizintechnik und Technomathematik (52) (remove)
Three-dimensional (3D) full-field measurements provide a comprehensive and accurate validation of finite element (FE) models. For the validation, the result of the model and measurements are compared based on two respective point-sets and this requires the point-sets to be registered in one coordinate system. Point-set registration is a non-convex optimization problem that has widely been solved by the ordinary iterative closest point algorithm. However, this approach necessitates a good initialization without which it easily returns a local optimum, i.e. an erroneous registration. The globally optimal iterative closest point (Go-ICP) algorithm has overcome this drawback and forms the basis for the presented open-source tool that can be used for the validation of FE models using 3D full-field measurements. The capability of the tool is demonstrated using an application example from the field of biomechanics. Methodological problems that arise in real-world data and the respective implemented solution approaches are discussed.