https://vicltz.github.io/Recent content in Victor Lutz, Robotics & ML Engineer on Victor LutzHugoen-ushttps://vicltz.github.io/contact/Mon, 01 Jan 0001 00:00:00 +0000https://vicltz.github.io/contact/<p>Feel free to reach out for research discussions, collaborations, or any inquiries. <strong>I am currently looking for new opportunities.</strong></p> <h2 id="email">Email</h2> <p>victor [dot] lutz [at] outlook [dot] fr</p> <h2 id="links">Links</h2> <ul> <li><a href="https://github.com/vicltz">GitHub</a></li> <li><a href="https://www.linkedin.com/in/victor-lutz-5a4931186">LinkedIn</a></li> </ul>https://vicltz.github.io/cv/Mon, 01 Jan 0001 00:00:00 +0000https://vicltz.github.io/cv/<h2 id="experience">Experience</h2> <p><strong>Research Engineer in Humanoid Robotics</strong> — LAAS-CNRS Gepetto, Toulouse, France <em>(Nov. 2024 – Nov. 2025)</em></p> <ul> <li>Deep Reinforcement Learning and Optimal Control for Bipedal Locomotion.</li> <li>Modeling and Control of a Humanoid with Parallel Mechanisms.</li> <li>Software Development for Control, State Estimation, System Identification and Safety on a Bipedal Robot.</li> </ul> <p><strong>Research Internship in Robotics</strong> — LAAS-CNRS Gepetto, Toulouse, France <em>(2024, 6 months)</em></p> <ul> <li>Parkour Trajectory Optimization on the Solo12 Quadruped Robot.</li> <li>Modeling and Control of a Humanoid with Parallel Mechanisms.</li> </ul> <p><strong>Engineering Internship</strong> — PAL Robotics, Barcelona, Spain <em>(2023, 3 months)</em></p>https://vicltz.github.io/projects/Mon, 01 Jan 0001 00:00:00 +0000https://vicltz.github.io/projects/<h2 id="deep-reinforcement-learning-nanodegree-projects">Deep Reinforcement Learning Nanodegree Projects</h2> <p><em>Udacity / NVIDIA Institute, March 2026</em></p> <p>Three deep RL projects covering navigation, continuous control, and multi-agent collaboration, completed as part of the Deep Reinforcement Learning Nanodegree (Udacity / NVIDIA Institute). <strong>Navigation:</strong> trained a DQN agent to collect yellow bananas while avoiding blue ones in a large 3D environment. <strong>Continuous Control:</strong> trained a double-jointed arm (Reacher) to follow a target location using PPO. <strong>Collaboration:</strong> trained two agents to play tennis cooperatively using Multi-Agent PPO, maximizing the number of consecutive volleys.</p>https://vicltz.github.io/publications/Mon, 01 Jan 0001 00:00:00 +0000https://vicltz.github.io/publications/<h2 id="conference-papers">Conference Papers</h2> <p><strong>Control of Humanoid Robots with Parallel Mechanisms using Differential Actuation Models</strong> Victor Lutz, Ludovic De Matteïs, Virgile Batto, Nicolas Mansard <em>IEEE ICRA 2026 — International Conference on Robotics and Automation</em>, Vienna, Austria, 2026. <a href="https://arxiv.org/abs/2503.22459">arXiv:2503.22459</a> · <a href="https://scholar.google.com/citations?user=UTf6kc8AAAAJ&amp;hl=fr">Google Scholar</a></p> <blockquote> <p>Several recently released humanoid robots employ actuator configurations in which the motors are displaced from the joints to reduce leg inertia. This paper introduces a compact analytical formulation for the two standard knee and ankle mechanisms that captures the exact non-linear transmission while remaining computationally efficient. The model is fully differentiable up to second order, enabling low-cost evaluation of dynamic derivatives for trajectory optimization and of the apparent transmission impedance for reinforcement learning.</p>