Digital Theses Archive


Tesi etd-09262019-121033

Type of thesis
Biomechanically inspired artificial mechanosensing for soft robotics
INGEGNERIA - Biorobotics
relatore Prof.ssa LASCHI, CECILIA
  • Nessuna parola chiave trovata
Exam session start date
During mechanical sensing in Nature, stimuli induce deformations and/or displacements of tissues and specialized sensory organs, which are soft either through materials themselves, or through deformable architectures. Stimuli are only later detected and transduced in electrochemical signals. In the artificial world, complex issues related to mechanical sensing remain to be solved, in particular in soft robotics. To contribute to their solution, the mechanical deformation of the sensor can be further exploited. To this end, bioinspired approaches can provide innovative solutions. Therefore, this thesis first investigates the relevance of biomechanical aspects in mechanical sensing, both in case studies from the natural and the artificial world. As a result, mechanical principles are identified, as well as transversal mechanical aspects to different morphology types and environments. Deliberate tailoring of the according structural features in artificial sensors could gradually lead to new insights, and lead to the elaboration of a toolbox to be used for the future methodical design of sensors. As a second step, a natural case study is selected to unveil its working mechanism. The tactile blep, which is found on tendrils of plants, is hypothesized to be a specialized tactile sensory organ and sensitive to shear force. The hypothesis is tested and confirmed under certain assumptions by the results of finite element analyses. Inspired from this natural structure, a simplified soft dome-shaped bilayer structure based on a capacitive transduction principle is designed. Modelling results show that the variation of its geometrical and material parameters permit to tune the sensitivity to normal and shear force, and, with particular configurations, to discriminate between both types of force with a simple electrical layout and no signal processing. Going beyond a fixed configuration, a design of a shape shifting skin based on Hall-effect transduction is implemented and tested. By exploiting the switch between different shape states with a gradual application of inner pressure, it is enabled with active touch. This is investigated to detect the range of stiffness of a fixed object in close proximity. Finally, magnetorheological elastomeric membranes are investigated as smart materials capable of sensing deformation in a soft actuator. A device acting both as a soft pneumatic actuator and a sensor is presented, based on an inductive transduction principle and developed through a collaborative work. The magnetorheological elastomeric membrane on its top functions both as the skin of the actuator and as the target of the sensing system. By using pressure information, it is possible to discriminate between the deformation of the skin caused due to internal driving and due to external loads. Consequently, this sensing system gives the possibility of discriminating between proprioception and tactile sensing. It is implemented in a soft pneumatic bending actuator, demonstrating the feasibility of the concept for soft robots.