Fancy Control With Crappy Sensors

Finished: 2020-10-14

MSc assignment

While micro-electromechanical system (MEMS) based sensors are widely used and can be highly accurate, additive manufacturing (AM) can provide a unique and novel perspective on how to design and manufacture a variety of different sensors. Recent advances in additive manufacturing provide the ability to print with several novel materials including: Flexible, carbon-fiber reinforced filaments as well as filaments integrated with conductive ink. Combining these materials, it is possible to exploit the advantages of 3-D printing (rapid prototyping, in-house manufacturing, print-at-once functional models etc.) in order to develop a system where the sensors will be embodied in the structure.

Current possibilities involve capacitive, biopotential, resistive, piezoelectric and magnetic sensing, as well as hybrid approaches where sensors manufactured by different fabrication techniques are utilized in a common structure. State-of-the-art research on sensors based on thermoplastic materials such as polyactic acid (PLA), thermoplastic polyurethane (TPU) and conductive inks, illustrates that due to the printing procedure (Fused Deposition Modelling) and the nature of plastics involved, such sensors are non-linear and inherently suffer from creep, drift and hysteresis.
Integrating the sensors in any given structure provides many new possibilities in the field of robotics, and especially in soft structures. But due to the aforementioned properties, the control of such a system poses a challenge.

The main goal of the thesis is to investigate the controllability of a system integrated with such sensors. A control task will be performed according to a chosen application. Some possible approaches will involve the usage of multiple sensors of the same type or a combination of different types of sensors as means to obtain usable information about the state of the system. A conclusion will be drawn on how it is possible for 3-D printed sensors to provide a feasible solution to real-world problems.