Wolf-Homeyer, Sabine: Object Localization in Fluids based on a Bioinspired Electroreceptor System. 2019
Inhalt
- Title
- Acknowledgments
- Abstract
- Contents
- List of Figures
- List of Tables
- Publications
- 1 Introduction to biological and bioinspired electrolocation
- 1.1 Sensory modalities and electrosensory system of the weakly electric fish
- 1.2 Scanning behavior of weakly electric fish
- 1.3 Abstraction of the biological electrosensory system for the development of a technical electroreceptor system
- 1.4 Outline of the thesis
- 2 The electric field and its perturbation by spherical objects
- 2.1 Analytical description of the perturbed electric field
- 2.1.1 Macroscopic Maxwell's equations
- 2.1.2 Interface conditions at a surface of different media
- 2.1.3 Perturbation due to a spherical object in a uniform electric field
- 2.1.4 Solution of boundary-value problem by solving Laplace's equation
- 2.2 Numerical description of the perturbed electric field by means of FEM
- 2.2.1 Physical background for using COMSOL's AC/DC module
- 2.2.2 Modeling process by using COMSOL
- 2.2.3 Solution of the Poisson's/Laplace's equation in 2D based on FEM
- 2.3 Dipole field perturbation and its representation with electrical viewpoints
- 3 Optimal concatenation of active sensor movements based on numerically extracted EEVs
- 3.1 Active sensor movements for disambiguation of object positions
- 3.2 Object localization by means of rotated and linearly shifted EEVs
- 3.3 Numerical extraction and superposition of EEV contour-rings
- 3.3.1 Extraction of EEV contour-rings by means of discretization strategies
- 3.3.2 Superposition of EEV contour-rings based on localization matrices D
- 3.3.3 Combination of rotational EEV contour-rings
- 3.3.4 MUP for combination of rotational EEV contour-rings
- 3.3.5 Contrast function of rotational EEV contour-rings
- 3.3.6 Combination of linearly shifted EEV contour-rings
- 3.3.7 MUP for combination of linearly shifted EEV contour-rings
- 3.3.8 Contrast function of linearly shifted EEV contour-rings
- 3.4 Estimation of unique object positions for permutations of rotational and linearly shifted movements
- 3.5 Summary
- 4 Search area partitioning and contour-ring fragmentation as a strategy in an application for active electrolocation
- 4.1 Short range sensor application based on a biomimetic abstraction for active electrolocation
- 4.2 Localization algorithm based on a basic movement sequence
- 4.2.1 Active electrolocation by means of a basic algorithm
- 4.2.2 Extraction of EEV CRPs and modification according to the respective movement sequence
- 4.2.3 Search area partitioning
- 4.2.4 Active electrolocation strategy based on a nearness metric
- 4.2.5 Results by using the basic movement sequence
- 4.3 Active electrolocation by means of a majority voting approach
- 4.3.1 Process sequence of the majority voting approach
- 4.3.2 Method of separate discretization for the superposition process
- 4.3.3 Results of majority voting approach compared to the basic movement sequence
- 4.4 Results of majority voting approach using real measurements
- 4.4.1 Implementation of a real measurement system and first results
- 4.4.2 Optimization of the measurement results
- 4.5 Summary
- 5 Evaluation of a fitted histogram representation of EEVs for the estimation of an optimal composition of movement sequences
- 5.1 Consideration of the general shape of an EEV by means of a histogram representation and density fit curve
- 5.2 Estimation of an optimal composition of movement sequences based on a fitted histogram representation of EEVs
- 5.3 Results of the fitted histogram representation of EEVs
- 5.4 Summary
- 6 Discussion and future research
- A Appendix A
- Bibliography