Integrated bio-inspired systems: An event-driven design framework

Abstract

Human robotics is an emerging research field aiming at the conjugation of mechanical and electrical engineering with biology; neuromorphic engineering has shown that it is possible to develop effective systems mimicking key aspects of biological systems, such as spiking (event driven) information processing and communication, adaptive and learning behavior. However, most of these solutions are tightly focused on one biological aspect, often aiming to precisely replicate it with a one-to-one mapping between the electronic and biological domain. We believe that, to successfully “port” biological strengths to the engineering world, such a tight relationship has to be loosened, and a new multi-domain co-design approach has to be devised. By exploiting the properties of the information processing, we are able to better design and integrate information transmission subsystems, with particular regard to the power consumption; by taking into account low-power requirements/constraints, we can develop energy-aware information processing modules. Designing the whole systems in a modular way, allows us to easily re-design a full-fledged system for new scenarios/applications. The research regards the implementation of discrete component systems as well, so that integrated circuits or proof-of-concept implementations can be easily validated.