By studying the computational model of the human hippocampus, researchers at CNR-IBF have developed a robotic system capable of understanding how to move in an environment, similar to how a human would. The research is published in the Neural Networks paper. Applying research conducted on the simulation of the human hippocampus, researchers at the Institute of Biophysics of the National Research Council (CNR-IBF) in Palermo have developed a robotic system capable of learning how to move in an environment similarly to how a human being would. "The implemented neural network outperforms the 'deep learning' model. A robot equipped with this network is able to self-organize its synaptic connections as it moves, and is able to learn how to reach a specific destination with a single learning session, remembering the path in subsequent sessions," explains Michele Migliore of CNR-IBF.
|