Ilyass Taouil

I am a first year PhD student at the Technical University of Munich (TUM), co-supervised by Prof. Majid Khadiv and Prof. Angela Dai . My research is at the interesection of robotics and computer vision with a special focus on trajectory generation for complex humanoid loco-manipulation tasks leveraging the 3D understanding of the environment. I graduated with an MSc in Computer Science from the University of Bonn, and a BSc in Computer Science from the University of Leeds.

Previously, I was a research engineer with the Dynamic Legged Systems Lab, headed by Claudio Semini at the Istituto Italiano di Tecnologia, working on mapping, motion planning, and contact planning for legged locomotion. I was also part of the Autonomous Intelligent Systems Lab , the Huamanoid Robotics Lab , the Humanoid Sensing and Perception Group , and Fotokite .

Feel free to drop me an email to get in touch: ilyass [dot] taouil [at] tum [dot] de

news

Jun 21, 2023	My work on quadrupedal motion planning was accepted at IROS 2023.
Jul 1, 2022	Joined the Dynamic Legged Systems Lab as a Research Engineer.
Apr 23, 2022	Graduated from the University of Bonn with a 1.5/4.0 (i.e. 3.7/4.0 GPA).
Jun 29, 2021	Participated in the ETH Robotics Summer School
Oct 1, 2020	Joined the Autonomous Intelligent Systems Lab as a Research Assistant.
Feb 1, 2020	Joined the Huamanoid Robotics Lab as a Research Assistant.
Mar 1, 2019	Joined Fotokite as an Engineering Intern.
Sep 1, 2018	Joined the Humanoid Sensing and Perception Group as a Deep Learning Engineer.
Jul 1, 2018	Graduated from the University of Leeds as the 3rd best student in the cohort.

selected publications

MotionDisco: Motion Discovery for Extreme Humanoid Loco-Manipulation

Ilyass Taouil*, Michal Ciebielski*, Shafeef Omar*, and 4 more authors

arXiv preprint arXiv:2606.06139, 2026

Abs arXiv Bib Website Video

We present MotionDisco, a framework that discovers contact-rich, long-horizon humanoid loco-manipulation motions from scratch, without relying on teleoperation or motion retargeting from human demonstrations. This is challenging because the space of possible contact interactions grows combinatorially with the task horizon and the number of objects in the scene. MotionDisco enables rapid discovery of novel motions by coupling a large language model (LLM) guided evolutionary search over sequences of interactions with an efficient sequential kinodynamic trajectory optimizer and pruning strategy. Through extensive ablation studies, we show that our LLM-guided search discovers successful whole-body trajectories across several challenging long-horizon tasks. Finally, by training reinforcement learning tracking policies on the discovered trajectories, we transfer the motions to a real humanoid robot. This is the first work to discover and deploy long-horizon humanoid loco-manipulation skills entirely through automated evolutionary search.
@article{taouil26motiondisco, title = {MotionDisco: Motion Discovery for Extreme Humanoid Loco-Manipulation}, author = {Taouil*, Ilyass and Ciebielski*, Michal and Omar*, Shafeef and Zhao, Haizhou and Dai, Angela and Johnson, Aaron M. and Khadiv, Majid}, journal = {arXiv preprint arXiv:2606.06139}, year = {2026}, }

DynaRetarget: Dynamically-Feasible Retargeting using Sampling-Based Trajectory Optimization

Victor Dhedin*, Ilyass Taouil*, Shafeef Omar*, and 4 more authors

arXiv preprint arXiv:2602.06827, 2026

Abs arXiv Bib Website Video

@article{dhedin26dynaretarget,
  title = {DynaRetarget: Dynamically-Feasible Retargeting using Sampling-Based Trajectory Optimization},
  author = {Dhedin*, Victor and Taouil*, Ilyass and Omar*, Shafeef and Yu, Dian and Tao, Kun and Dai, Angela and Khadiv, Majid},
  journal = {arXiv preprint arXiv:2602.06827},
  year = {2026},
}

Quadrupedal Footstep Planning using Learned Motion Models of a Black-Box Controller

Ilyass Taouil, Giulio Turrisi, Daniel Schleich, and 3 more authors

IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2023

Abs arXiv Bib Website

Legged robots are increasingly entering new domains and applications, including search and rescue, inspection, and logistics. However, for such a systems to be valuable in real-world scenarios, they must be able to autonomously and robustly navigate irregular terrains. In many cases, robots that are sold on the market do not provide such abilities, being able to perform only blind locomotion. Furthermore, their controller cannot be easily modified by the end-user, requiring a new and time-consuming control synthesis. In this work, we present a fast local motion planning pipeline that extends the capabilities of a black-box walking controller that is only able to track high-level reference velocities. More precisely, we learn a set of motion models for such a controller that maps high-level velocity commands to Center of Mass (CoM) and footstep motions. We then integrate these models with a variant of the A* algorithm to plan the CoM trajectory, footstep sequences, and corresponding high-level velocity commands based on visual information, allowing the quadruped to safely traverse irregular terrains at demand.
@article{taouil2023quadrupedal, title = {Quadrupedal Footstep Planning using Learned Motion Models of a Black-Box Controller}, author = {Taouil, Ilyass and Turrisi, Giulio and Schleich, Daniel and Barasuol, Victor and Semini, Claudio and Behnke, Sven}, journal = {IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)}, year = {2023}, }