Robotics Reinforcement Learning Character Animation Motion Tracking Computer Graphics Computer Vision

I work as a Research Scientist at Meta Reality Labs in San Francisco. I study the physical principles of human and animal movement to replicate them with robots or virtual agents. This work combines Reinforcement learning, Computer Vision, Character animation, Motion Tracking and Robotics. I did my PhD at the Agile and Dexterous Robotics Lab at ETH Zurich, where I worked on robot motion-planning, trajectory optimization, and model-based control for quadruped robots. I received my Masters in Mechanical Engineering at the Karlsruhe Institute of Technology.

These Youtube videos give an overview of my work.

Google Scholar / Twitter / Youtube / Github / LinkedIn

Publications

2024
Real-Time Simulated Avatar from Head-Mounted Sensors, CVPR
Zhengyi Luo, Jinkun Cao, Rawal Khirodkar, Alexander Winkler, Jing Huang, Kris Kitani, Weipeng Xu
Project page
RoHM: Robust Human Motion Reconstruction via Diffusion, CVPR
Siwei Zhang, Bharat Bhatnagar, Yuanlu Xu, Alexander Winkler, Petr Kadlecek, Siyu Tang, Federica Bogo
Project page
Humanoid Motion Representation for Physics-Based Control, ICLR
Zhengyi Luo, Jinkun Cao, Josh Merel, Alexander Winkler, Jing Huang, Kris Kitani, Weipeng Xu
Project page / Paper
2023
Perpetual Humanoid Control for Real-time Simulated Avatars, ICCV
Zhengyi Luo, Jinkun Cao, Alexander Winkler, Kris Kitani, Weipeng Xu
Project page / Paper
Physics-based Motion Retargeting from Sparse Inputs, SCA
Daniele Reda, Jungdam Won, Yuting Ye, Michiel van de Panne, Alexander Winkler
Project page / Paper
Environment-Aware Simulated Motion Tracking, SIGGRAPH
Sunmin Lee, Sebastian Starke, Yuting Ye, Jungdam Won, Alexander Winkler
Project page / Paper
2022
QuestSim: Human Motion Tracking from Sparse Sensors, SIGGRAPH (A)
Alexander Winkler, Jungdam Won, Yuting Ye
Project page / Paper
Human Motion Reconstruction from Sparse IMUs, SIGGRAPH (A)
Yifeng Jiang, Yuting Ye, Deepak Gopinath, Jungdam Won, Alexander Winkler, Karen Liu
Project page / Paper
2018
Gait & Trajectory Optimization for Legged Systems, RAL
Alexander Winkler, Dario Bellicoso, Marco Hutter, Jonas Buchli
Project page / Paper
Optimization-based Motion-Planning for Legged Robots, PhD Thesis ETH Zurich
Alexander Winkler
Thesis
2017
Trajectory Optimization for with Vertex-based ZMP Constraints, RAL
Alexander Winkler, Farbod Farshidian, Diego Pardo, Michael Neunert, Jonas Buchli
Paper / Video / Link
Robust Whole-Body Motion Control of Legged Robots, IROS
Farbod Farshidian, Edo Jelavic, Alexander Winkler, J Buchli
Paper / Video / Link
Direct Collocation & Control for Dynamic Legged Locomotion, RSS
Diego Pardo, Michael Neunert, Alexander Winkler, Ruben Grandia, Jonas Buchli
Paper / Video / Link
Motion & Foothold Optimization Quadruped Locomotion, ICRA
Alexander Winkler, Farbod Farshidian, Michael Neunert, Diego Pardo, Jonas Buchli
Paper / Video / Link
Optimal Planning and Control For Quadrupedal Locomotion, ICRA
Farbod Farshidian, Michael Neunert, Alexander Winkler, Gonzalo Rey, Jonas Buchli
Paper / Video / Link
Trajectory Optimization Through Contacts, RAL
Michael Neunert, Farbod Farshidian, Alexander Winkler, Jonas Buchli
Paper / Video / Link
Optimal and Learning Control for Autonomous Robots
Jonas Buchli, Farbod Farshidian, Alexander Winkler, Timothy Sandy, Markus Giftthaler
Paper
2016
Direct Transcription & Nonlinear Opt. for Motion Planning, RAL
Diego Pardo, Lukas Moller, Michael Neunert, Alexander Winkler, Jonas Buchli
Paper / Video / Link
2015
Whole-Body Locomotion for Robots on Challenging Terrain, ICRA
Alexander Winkler, Carlos Mastalli, Ioannis Havoutis, Michele Focchi, Darwin Caldwell, Claudio Semini
Paper / Video / Link
Planning for Quadruped Locomotion using On-board Perception, TePRA
Carlos Mastalli, Ioannis Havoutis, Alexander Winkler, Darwin Caldwell, Claudio Semini
Paper
2014
Path Planning w/ Force-based Foothold Adaptation, ICRA
Alexander Winkler, Ioannis Havoutis, Stephane Bazeille, Jesus Ortiz, Michele Focchi, Ruediger Dillmann, Darwin Caldwell, Claudio Semini
Paper / Video / Link

Software

All code is published on Github.

Light-weight and extensible C++ library for trajectory optimization for legged robots.

Star Download

$ apt install ros-melodic-towr

Eigen-based and light-weight C++ Interface for NLP Solvers (Ipopt, Snopt).

Star Download

$ apt install ros-melodic-ifopt

Visualization of legged robots, forces, support areas, ZMP, CoM etc. in ROS rviz.

Star Download

$ apt install ros-melodic-xpp

Blog

Six months backpacking

and mastering the Bum Gun

2019-3-30  ·  7 min read

I am sitting on the toilet throwing my shit-stained toilet paper in the trash can under the sink whe... read more

Collect memories or things?

Why they're not so different

2018-12-24  ·  4 min read

It's Christmas Eve and I'm waiting for my plane at LAX scrolling through my Instagram feed. An image ... read more

Living like a Buddhist monk

10 days of silent meditation

2018-10-27  ·  10 min read

My ass hurt, my concentration was gone and time seemed to stand still. I squinted open my eyes just e... read more

Why I'm travelling

Exploration vs exploitation

2018-10-09  ·  4 min read

Nowadays, the reasons for travel are assumed to be so obvious, the benefits so universally agreed-u... read more

How to postpone your dreams

Why money can help

2018-09-12  ·  3 min read

Of a 168-hour week, 56 hours are typically spent sleeping and 40 hours earning money. That leaves 72 ... read more