MISEL stands for Multispectral Intelligent vision System with Embedded Low-power neural computing
The word “misel” also means “a thought” in Slovenian – what a lucky coincidence for the project that aims at an intelligent standalone system, a system that “thinks”!
Animals, even small insects have superior capabilities of perceiving/understanding the environment and effortlessly perform tasks in unknown and changing terrain. This has led to a widespread interest in trying to mimic the way biology computes – neuromorphic computing.
The MISEL project aims at bringing artificial intelligence to the edge computing (decisions made on-device) through a low-power bio-inspired standalone vision system with adaptable multi-band sensing and in-sensor spatio-temporal neuromorphic processing based on complex events. The expected outcome is a device featuring smaller size, weight and power consumption than what is possible with existing components.
The science-to-technology breakthrough is the heterogeneous integration of a neuromorphic computing scheme featuring three different abstraction levels (cellular, cerebellar, and cortical processors) with high-density memory arrays and adaptive photodetector technology for fast and energy efficient operation. MISEL employs a holistic approach where everything from devices to use case is adopted to serve as a whole. This allows context-aware, low power and distributed computation on an edge device, challenging the current mainstream approach relying on massive-data transfers and centralized computational resources. The objective is a compact, brain-mimicking, low-power system exploiting a wide range of biological principles: human eye like adaptivity with cellular processor, as well as data fusion, learning and reasoning performed by cerebellar processor (reflexes) and cortical processor (conscious decisions). Finally, by demonstrating and evaluating the system operation in challenging tasks (end-to-end use scenarios), the project targets outperforming conventional state of the art systems.
