Publications

Google Scholar

Pre-prints (Under Review)

LaserAttention pipeline: VLM-based human-intent extraction and 3D segmentation feeding an LLM spatial reasoning and control module, demonstrated on a Tiago++ humanoid pointing at a target on a table

LaserAttention: Vision-Language Models for Joint Attention in Human-Robot Interaction

Dionis Totsila, Clemente Donoso, Quentin Rouxel, Jean-Baptiste Mouret, Serena Ivaldi

Submitted to IEEE Transactions on Robotics (T-RO)

Peer-reviewed Journal Papers

From Vocal Instructions to Household Tasks: The Inria Tiago++ in the euROBIN Service Robots Coopetition

Fabio Amadio*, Clemente Donoso*, Dionis Totsila*, Raphael Lorenzo, Quentin Rouxel, Olivier Rochel, Enrico Mingo Hoffman, Jean-Baptiste Mouret, Serena Ivaldi
(* indicates equal contribution)

IEEE Robotics and Automation Practice (RA-P), 2026

Autonomous Neural Dynamic Policy on a quadruped robot

Sensorimotor Learning with Stability Guarantees via Autonomous Neural Dynamic Policies

Dionis Totsila, Konstantinos Chatzilygeroudis, Valerio Modugno, Denis Hadjivelichkov, and Dimitrios Kanoulas

IEEE Robotics and Automation Letters (RA-L), 2025

RobotDART: a versatile robot simulator for robotics and machine learning researchers

Konstantinos Chatzilygeroudis, Dionis Totsila, and Jean-Baptiste Mouret

Journal of Open Source Software, 2024

Peer-reviewed Conference Papers

Safe bimanual teleoperation concept diagram

Safe Bimanual Teleoperation with Language-Guided Collision Avoidance

Dionis Totsila*, Clemente Donoso*, Enrico Mingo Hoffman, Jean-Baptiste Mouret, Serena Ivaldi
(* indicates equal contribution)

IEEE Conference on Telepresence, 2025

Words2Contact: Identifying Support Contacts from Verbal Instructions Using Foundation Models

Dionis Totsila, Quentin Rouxel, Jean-Baptiste Mouret, and Serena Ivaldi

IEEE-RAS International Conference on Humanoid Robots, 2024

Peer-reviewed Workshop Papers

Autonomous Neural Dynamic Policy simulation diagram

End-to-End Stable Imitation Learning via Autonomous Neural Dynamic Policies

Dionis Totsila*, Konstantinos Chatzilygeroudis*, Denis Hadjivelichkov, Valerio Modugno, Ioannis Hatzilygeroudis, and Dimitrios Kanoulas
(* indicates equal contribution)

Life-Long Learning with Human Help (L3H2) Workshop, ICRA, 2023

Patents

Patent

Système d'interaction robotique (Robotic Interaction System)

D. Totsila, S. Ivaldi, J.-B. Mouret, and C. Donoso Krauss

Patent FR2513437 · Filed Nov 17, 2025

Talks & Presentations

Sensorimotor learning with stability guarantees via autonomous neural dynamic policies — Short talk and discussion, Embodied Intelligence Conference, 2023
Sensorimotor learning with stability guarantees via autonomous neural dynamic policies — Abstract presentation, 17th Learning and Intelligent Optimization Conference (LION17), 2023