Flapping flight is one of the wonders of nature. The state-of-the-art of bioinspired aerial robots still has to bridge a big gap, where energy efficiency, multi-modal operation and complex environments pose challenges due to a lack of functionality and autonomy. By drawing inspiration from distributed sensing and actuation in nature, the shortcomings of current bio-inspired aerial robots can be mitigated by means of embedding flow sensing, proprioception and variable stiffness. This can be achieved by means of embedding functional materials and structures through 3D-printing. With the advent of 3D printing it has become possible to generate new multi-material structures over multiple orders of magnitude in scale, enabling complex geometries, embedded electronics or composites in a single fabrication step.
Especially Fused Filament Fabrication (FFF) is widely preferred due to its simple, low cost, multi-material capabilities. This technique has been used for 3D printing sensors and robotics, where conductive polymer composites can be used as smart material due to their electro-thermo-mechanical properties. Using these materials, FFF introduces anisotropic physical properties in 3D prints due to the line-by-line, layer-per-layer FFF printing process. Therefore, research is still required for embedding 3D printed functionality in aerial robots.
The first goal of this thesis is to investigate how the specific properties of 3D printed functional materials can be exploited, i.e. anisotropic electrical resistivity, that are otherwise hard to obtain and tailor.
Two characterization techniques are presented to investigate the electrical anisotropy of 3D printed conductors: (1) infrared thermography in combination with Joule heating to characterize electrical anisotropy, and (2) voltage contrast scanning electron microscopy method (VCSEM) to determine potential distributions.
The anisotropic electrical conduction is also modeled analytically, where the electrical properties are described as an electrical network with bulk and contact properties in and between neighbouring printed track elements or traxels.
Next it is shown that the anisotropy can be tuned by the FFF print parameters, transforming the anisotropic electrical properties into a design tool for creating DC electric metamaterials. This can be used to steer, e.g. to concentrate or cloak, the electrical current through 3D-prints, e.g. for electronics, sensors, actuators and heaters.
Electrically conductive filament can also be used for monitoring the print process to gather real-time information for nondestructive defect detection of voids and bonding quality. In-situ monitoring of FFF is realized through electrical resistance measurements between the nozzle and single or multi-electrodes on the bed. The anisotropic electrical conduction model is used to interpret the measurements and to detect defects.
The second goal is to study how stiffness modulation, deformation sensing and flow sensing can be implemented in 3D printed flapping flight robots with functional materials.
A novel energy-based control architecture on Euler-Bernoulli beams equipped with a variable stiffness mechanism is presented that can stabilize the beam or induce limit cycles on it. 3D-printed, embedded, piezoresistive strain gauges are used as sensing units for closed-loop control, showing how to deal with the non-ideal sensor response.
Finally flow sensing is presented in the form of a 3D printed, piezoresistive cantilever flow sensor in a quasistatic application in the wind tunnel. However, non-linearities and underdamping of the sensor need to be addressed before becoming applicable. Due to the limitations of 3D printing and the accessory polymers, it might be worthwhile to focus on qualitative sensors that measure the presence of certain phenomena, like stall and flow reversal, instead of performing quantitative measurements.
Overall the electrical anisotropy in 3D prints is clearly modeled, measured and tailored in fabrication, and it is shown how it can be used or reduced in sensor applications. Promising sensing and variable stiffness ideas for implementation in a robotic bird have been presented, where future research can close the gap between the lab and implementation in aerial robots.