Modeling and implementation of a networked system of bio-inspired autonomous mobile sensors with applications to real time angiogenesis, drug delivery, and environmental monitoring.

点击翻译按钮获取中文摘要

Systems of interacting living organisms display a variety of cooperative strategies and collective behaviours, as well as fine adaptations to their environments, as individuals sense each other and the environment through different communication channels. The wide range of spatial and temporal scales at which these systems operate motivates multidisciplinary studies to understand and model their peculiar characteristics. At the evolutionary time scales, all organisms from very simple to complex, adapt to their environment through behaviours that allow them to navigate through their environments. The mechanics of biological systems locomotion is studied to understand the structure of interactions with diverse environments, which in turn can inform robotic design in the attempt of adopting the same solutions that are often characterized by robustness, adaptability, and extensibility to a variety of scenarios. In the context of this research, the physics and mechanics of millipedes and centipedes has been used to elaborate a mathematical model for a robotic device, targeting the following main characteristics: (1) Autonomous operation; (2) Capability of navigating in different environments; (3) Capability of sensing properties of the environment, as for example stiffness and density of the substrate. Millipedes and centipedes posses elongated flexible bodies, sustained by distributions of legs in contact with the environment. The redundancy of the legs ensures persistent contact and therefore the capability of moving in uneven terrains. Moreover, the bodies' flexibility allows for shape morphing with respect to the substrate, facilitating the navigation.***This set of features are desirable in a robotic device autonomously negotiating unknown, unstructured environments. In recent work from my research group, a mathematical model based on these characteristics has been derived and demonstrated in simulation, and a first generation prototype is currently being manufactured. Building on these premises, I propose to extend theoretical results to the study of a networked system of these devices, in which several of them are coupled (for example through mechanical actions transmitted through a common substrate), and take advantage of multitasking and collective capabilities of the group, that emerge from cooperation. I envision tailoring the system to different important applications, such as real time angiogenesis, in which miniaturized versions of the devices continually monitor the health status of cardiovascular tissues in critical areas, and induce tissue repair when critical conditions are detected; drug delivery targeting specific functions and using physiological data as feedback to assess the effectiveness and adjust to dynamic conditions; and environmental monitoring with real time intervention and disaster prevention.**