Programmable Interfacial Mechanics in Soft Robotics: From Mechanical Adhesion to Friction-Based Magnetic Actuation

In biological systems, adhesion and friction are not fixed material properties. They emerge from the interplay of local geometry, hierarchy, and compliance. This reflects a basic functional need: the same interface may need to resist separation under some conditions and allow motion under others. Starting from this perspective, this work treats contact mechanics as a design variable for soft robotic systems. The underlying idea is that, by shaping surface morphology and controlling deformation, it becomes possible to regulate normal and tangential interfacial forces in a predictable and useful way. This work lies at the intersection of tribology and soft actuation, two areas that have often developed along separate lines. On the one hand, contact mechanics offers increasingly refined descriptions of adhesion, friction, and rough contact. On the other hand, magnetically active soft systems have mostly been studied in terms of deformation and locomotion. The connection between actuation, structural response, and the way forces are ultimately transmitted across an interface is therefore still only partially understood. This thesis addresses that gap by examining from a theoretical and experimental perspective two classes of active surfaces, each chosen to explore a different mode of interface control: mechanically interlocking patterned substrates for the regulation of adhesion, and magnetically actuated cilia-like carpets for the modulation of friction and transport. The first part of the work focuses on how interlocking microstructures can be designed to achieve tunable, switchable, and selective mechanical adhesion, including in wet and dusty environments where conventional surface forces are reduced or absent. Their behavior is described through a mean-field continuum contact formulation that links the mechanics of individual interlocking pairs to the effective work of adhesion of patterned interfaces. The theoretical predictions are then examined experimentally. The second part introduces hard-magnetic elastomeric cilia-like elements that deform under external magnetic fields and generate directional friction asymmetry, depending on their induced morphological asymmetry. Their response is described within a hard-magnetic elastica framework that also accounts for non-constant cross-section elements and non-uniform actuation fields, while ratcheting contact models and directional friction measurements explain how oscillatory magnetic excitation can be transformed into net locomotion. These principles are then validated and extended to multiple transport systems based on interfacial asymmetric dissipation phenomena. The thesis follows this progression closely. Chapter 1 introduces the biological inspiration and the theoretical foundations of adhesion, friction, magnetic actuation, and transport at small scales. Chapter 2 presents the fabrication routes, characterization methods, actuation platforms, biological assays, and numerical tools developed for the study. Chapter 3 investigates the fundamental mechanisms governing adhesion control and friction-based locomotion, with particular emphasis on the design rules that connect geometry, material response, and actuation to interfacial performance. Chapter 4 extends these principles to system-level demonstrations: pneumatic modulation for adhesion control on soft interlocking substrates; untethered motion of robots and passive payloads based on friction asymmetry; fluid transport based on drag asymmetry; and remote mechanobiological stimulation in isolated environments. Overall, the thesis proposes a unifie d mechanics-based approach to programmable interfacial behavior in soft robotics. Its main contribution is to show that adhesion and friction need not be treated only as passive constraints. They can instead be engineered as functional mechanical responses, enabling force transmission, motion generation, and multifunctional interaction with complex environments.