Hello, my name is
Bridging the gap between high-level AI reasoning and low-level robotic control for human-like grasping and dexterous multi-fingered manipulation.
Dexterous Robotic Manipulation
Bridging artificial intelligence with embodied physical robotics.
As a 25-year-old PhD student at CEA-List and Paris-Saclay University, I specialize in the intersection of Artificial Intelligence and Robotics through dexterous manipulation of any objects using anthropomorphic robotic hands. My research aims to bridge the gap between high-level AI reasoning and low-level robotic control to achieve human-like grasping and manipulation. I combine a strong theoretical foundation with a hands-on commitment to shaping the future of intelligent robotics. Beyond my academic research, I am a big sports fan (runner, cyclist, black-belt judoka) and a guitarist.
Disciplined mindset shaped by competitive athletics and musical creativity:
Peer-reviewed publications in top-tier computer vision and robotics venues.
Julien Mérand, Boris Meden, Liming Chen, and Mathieu Grossard
We propose CoToGrasp, a novel generative framework that synthesizes diverse, stable grasps strictly conditioned on specific contact topologies.
Julien Mérand, Boris Meden, Mathieu Grossard, and Liming Chen
We propose GOAG, a novel deep generative model that learns a compact latent representation of a specific gripper's contact surface distribution, enabling the efficient sampling of valid grasp configurations without relying on object-specific training data.
Julien Mérand, Boris Meden and Mathieu Grossard
This paper presents an efficient approach for determining the joint configuration of a multifingered gripper solely from the point cloud data of its poly-articulated chain, as generated by visual sensors, simulations or even generative neural networks.
Research and engineering roles in industry and research institutions.
Dual degree engineering, Master of Science, and doctoral studies.
Open-source robotics, hardware prototypes, and embedded systems.
Ongoing personal project inspired by BCN3D's Moveo robot. Developed various applications in Python and C++ on ROS2 Humble.
Proposed a dynamic obstacle avoidance solution for mobile robots, leveraging machine learning for velocity estimation and cost mapping.
Designed and built an analog guitar effects pedal, gaining hands-on experience in analog electronics and understanding how electronic components manipulate audio signals.
Wrote algorithms for an embedded product designed to make learning Braille fun. The project has now become an independent company.