Pablo Lanillos

About

Pablo Lanillos is a pioneering researcher at the intersection of computational neuroscience, cognitive robotics, and artificial intelligence, whose work has fundamentally advanced how robots perceive themselves and interact with uncertain environments. His research centers on applying neuroscientifically-grounded principles — particularly the **free energy principle** and **active inference** — to enable robots to achieve human-like body awareness, adaptive perception, and robust action. Lanillos's most influential contributions include developing predictive coding frameworks for robotic body learning (75 citations), conducting landmark empirical studies of active inference on humanoid robots (65 citations), and pioneering deep learning extensions that allow robots to process raw visual input for body perception (57 citations). His comprehensive surveys on active inference in robotics (55 citations) and world models for developmental robotics (63 citations) have become key reference works shaping the field's direction. Beyond theoretical contributions, Lanillos has demonstrated practical impact through tactile-based object learning frameworks (63 citations) and early work on robotic self-perception through sensorimotor contingencies. His research bridges fundamental neuroscience with real-world robotic deployment, offering transformative tools for building machines that can safely and adaptively interact in complex, unpredictable environments — a challenge at the very heart of modern robotics.

Research Focus

Key Achievements

Top Papers

1. 1
   Adaptive Robot Body Learning and Estimation Through Predictive Coding

   75 citations · 2018
2. 2
   An Empirical Study of Active Inference on a Humanoid Robot

   65 citations · 2021
3. 3
   World models and predictive coding for cognitive and developmental robotics: frontiers and challenges

   63 citations · 2023
4. 4
   A Tactile-Based Framework for Active Object Learning and Discrimination using Multimodal Robotic Skin

   63 citations · 2017
5. 5
   End-to-End Pixel-Based Deep Active Inference for Body Perception and Action

   57 citations · 2020
6. 6
   Active Inference in Robotics and Artificial Agents: Survey and Challenges

   55 citations · 2021
7. 7
   How Active Inference Could Help Revolutionise Robotics

   40 citations · 2022
8. 8
   Yielding Self-Perception in Robots Through Sensorimotor Contingencies

   40 citations · 2016
9. 9
   Robotic technologies for fast deployment of industrial robot systems

   33 citations · 2016
10. 10
    Active inference body perception and action for humanoid robots

    29 citations · 2019

Key Collaborators

Contact & Links