I study how humans and machines exchange information through the body - how movement intent can be read out of muscle activity, and how touch can be delivered back.
The through-line is infrastructure rather than one-off demonstrations: instruments that other labs can build, trust, and cite. Postdoc at Imperial College London, PhD in bioengineering, working across haptics, tactile sensing, and biosignal acquisition.
Active projects and collaborations with research groups.
A backpack that gives you two extra robotic arms - first as a demo you could wear for hours, now as a research platform for studying how people work with limbs they can't see.
A bioinspired electronic skin with layered sensing for robotics, prosthetics, and haptic applications.
Peer-reviewed work in haptics, tactile sensing, and human-robot interaction.
Hover or tap a paper for its abstract.
Xuanyun Qiu, Dorian Verdel, Hector Cervantes-Culebro, Alexis W.M. Devillard, Etienne Burdet·arXivPDF ↗︎abstract +
A hierarchical framework for maintaining balance when humans are augmented with supernumerary robotic limbs, featuring prediction, planning, and control layers tested with ten participants performing bending tasks.
Anirvan Dutta, Alexis W.M. Devillard, Zhihua Zhang, Xiaoxiao Cheng, Etienne Burdet·npj Robotics (Nature)abstract +
Investigating how sensing embodiment affects tactile perception using a modular e-Skin with interchangeable mechanical compliance and an unsupervised, action-conditioned deep state-space model to interpret soft object properties.
Alexis W.M. Devillard, Aruna Ramasamy, Xiaoxiao Cheng, Damien Faux, Etienne Burdet·Scientific Data (Nature)PDF ↗︎abstract +
A comprehensive multi-modal dataset capturing concurrent haptic, audio, and visual signals from participants interacting with textured surfaces, addressing gaps in human tactile exploration research.
Henrik Jörntell, Alexis W.M. Devillard, Etienne Burdet·Multisensory Research (Brill)abstract +
Examining mechanical spatiotemporal invariants emerging from skin-body-environment interactions and corresponding neural mechanisms, with applications to robotics.
Alexis W.M. Devillard, Anirvan Dutta, Zhihuan Zhang, Xiaoxiao Cheng, Etienne Burdet·IEEE World Haptics Conference (WHC)abstract +
A bioinspired electronic skin with layered sensing architecture for robotics, prosthetics, and haptic applications.
Jenireth Torrealba Molina, Toqa AbuBaker, Alexis W.M. Devillard, Yanpei Huang, Xiaoxiao Cheng, Etienne Burdet·IEEE International Conference on Robotics and Automation (ICRA)abstract +
A modular robotic platform designed and evaluated for precision microsurgical applications.
Alexis W.M. Devillard, Luca Mattioli, Etienne Burdet·ICRA Workshop on Supernumerary Limbsabstract +
Exploring multi-sensory integration approaches for sensory augmentation in virtual and robotic interaction scenarios.
Alexis W.M. Devillard, Aruna Ramasamy, Damien Faux, Vincent Hayward, Etienne Burdet·IEEE World Haptics Conference (WHC)abstract +
The conference presentation of the multi-modal tactile exploration dataset, later extended in Nature Scientific Data.
MSc and MEng projects at Imperial College London, mostly built on the open platforms I maintain - so the work outlasts the academic year.
Currently no open projects.
New topics are usually listed ahead of the academic year. If you have a strong idea in haptics, biosignals, or wearable robotics, get in touch anyway.
A piezo actuator on the fingernail, a custom high-voltage driver, and a rendering method tuned per material - evaluated against six real textures.
Vibration on the nail of the finger to use, leaving the fingertip free to feel the key - faster and less error-prone than matched visual cues.
How far can a modular electronic skin grow before its electronics give out? Measured on a four-module prototype and extrapolated.
Muscle and motion sensors built into backpack straps as a command channel for a worn robotic arm: what the shoulders can and cannot tell a controller.
A third arm that watches the tool and repositions the workpiece by itself - tested against a plain stand in a soldering-like task.
Steer a robotic finger with your own finger and feel what it touches through a vibration on the nail - compared against letting the robot explore by itself.
Muscle and motion sensors on the shoulders as a hands-free command channel for a robot arm, tested for discrete positions, continuous tracking and use while the hands are busy.
A robot pressing, rotating and sliding a multi-layer e-skin over objects, and an unsupervised model that learns to tell them apart from touch alone.
Open to research collaborations, joint student projects, and labs wanting to build on the CleverHand platform. Get in touch.