Vieri Giuliano Santucci

About

Vieri Giuliano Santucci is a pioneering researcher at the intersection of autonomous robotics, intrinsic motivation, and developmental learning. His work centers on a fundamental challenge in artificial intelligence: how robots can learn autonomously, without explicit external rewards, in the same open-ended way that biological agents do. His most cited contribution, the GRAIL architecture (2016, 91 citations), introduced a landmark four-level robotic system capable of autonomously discovering environmental changes, forming goal representations, and directing its own learning — a significant step toward truly self-directed machines. Complementing this, his comparative study of intrinsic motivation signals (2013, 65 citations) provided the field with crucial empirical guidance on which internal drives best support multi-skill acquisition. Santucci has also explored sensorimotor contingencies, body self-modeling, and the role of dopamine-like prediction errors in robotic learning, bridging computational neuroscience with robotics. More recently, he has expanded into ethical dimensions of artificial autonomy (2022), arguing that interdependence should underpin autonomous systems design. With over 300 cumulative citations and an editorial role shaping the open-ended learning community, Santucci stands as a leading voice in building robots that genuinely grow through curiosity.

Research Focus

Key Achievements

Top Papers

1. 1
   GRAIL: A Goal-Discovering Robotic Architecture for Intrinsically-Motivated Learning

   91 citations · 2016
2. 2
   Which is the best intrinsic motivation signal for learning multiple skills?

   65 citations · 2013
3. 3
   Phasic dopamine as a prediction error of intrinsic and extrinsic reinforcements driving both action acquisition and reward maximization: A simulated robotic study

   36 citations · 2013
4. 4
   Sensorimotor Contingencies as a Key Drive of Development: From Babies to Robots

   32 citations · 2019
5. 5
   Know Your Body Through Intrinsic Goals

   27 citations · 2018
6. 6
   Intrinsically motivated discovered outcomes boost user's goals achievement in a humanoid robot

   25 citations · 2017
7. 7
   Editorial: Intrinsically Motivated Open-Ended Learning in Autonomous Robots

   23 citations · 2020
8. 8
   Cumulative Learning Through Intrinsic Reinforcements

   13 citations · 2013
9. 9
   Autonomous selection of the “what” and the “how” of learning: An intrinsically motivated system tested with a two armed robot

   12 citations · 2014
10. 10
    Interdependence as the key for an ethical artificial autonomy

    11 citations · 2022

Key Collaborators

Contact & Links