Neuromorphological Computation

NeuCompAct – Neuro-Morphological Computation in Active Vision

In BioRobotics, a robotics area that gets inspiration from biological organisms, the control of movements cannot be computed centrally for all degrees of freedom, but part of the required computation has to be incorporated into the morphology of the agent, a concept called morphological computation. Indeed, this principle is intimately related to the contribution of movement in perception. Specifically, in active vision, tasks like reading, visual search and scene perception are tightly linked with eye movements and therefore both motor and sensory processes must be considered at the same time.

Do the morphology and dynamics of the eye plant and related neural areas facilitate the control and execution of eye movements and thus active perception?

If so, two consequences arise, which this project is going to take advantage of: firstly, we cannot design actuators without taking into consideration the overall system morphology. One example is how the eye muscles are connected to the ocular bulb. We then speak of actuation system; the second consequence is relative to the type of control signal that the actuation system will use, which needs to be spike-like. Current models of brain function, simulated on von Neumann architectures, poorly match the massively parallel and active sensory-motor processing systems of the biological counterpart and rely on a huge processing power. The neuromorphic circuits emulate the organization and function of thousands of neurons in electronic devices. The advantages of such computational chips must be exploited with actuation solutions that can be directly driven by the spike-like signals.

This project aims at applying computational neural models of attentive eye movements to neuro-inspired information processing and actuation hardware, for producing a physical implementation of a sensory-motor Active Vision system. It will lead to improve both robot performances and understanding of how movement and body structure augment sensing.





Zambrano, D., Cianchetti, M., & Laschi, C. (2014). The Morphological Computation Principles as a New Paradigm for Robotic Design, in E-book on Opinions and Outlooks on Morphological Computation, H. Hauser, R. Fuchslin, and R. Pfeifer, pp. 214–225, no. 19, isbn 978-3-033-04515-6,

Corradi, F., Zambrano, D., Raglianti, M., Passetti, G., Laschi, C., & Indiveri, G. (2014). Towards a Neuromorphic Vestibular System. IEEE Transactions on Biomedical Circuits and Systems, 1–1. DOI:

Zambrano, D., Rombouts, J., Laschi, C., & Bohte, S. M. (2014). Spiking AGREL. Presented at the European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning (ESANN) 2014. Available from

Passetti G, Corradi F, Raglianti M, Zambrano D, Laschi C, and Indiveri G. (2013) Implementation of a neuromorphic vestibular sensor with analog VLSI neurons. IEEE Biomedical Circuits and System Conference, BIOCAS Oct. 31-Nov. 2, 2013, Rotterdam, The Netherlands, DOI:


Our team

Dr.  Zambrano , Davide


Dr.  Cianchetti , Matteo

Assistant Professor

Dr.  Raglianti , Marco


 Licofonte , Alessia


 Passetti , Giovanni

Ph.D. Student

 Rogai , Francesco

Former Research Assistant