Peter Werner

Graduate Student in Robotics

Massachusetts Institue of Technology

Hi there!

My name is Peter Werner. I am a 4th-year Ph.D. student in Prof. Daniela Rus’ lab at the Computer Science and Artificial Intelligence Laboratory (CSAIL) at MIT.

My research interests are broadly centered around Mathematics, Control Theory, Optimization, and Robotics. While my interests tend more towards theory and software, I have a deep appreciation for hardware. I find nothing more convincing than an actual hardware system executing an approach - there is no hiding or glossing over details.

Anyways, if I am not in the lab you can probably find me running along the Charles River, hiking in the Whites, skiing back home in Switzerland, or playing badminton. I also enjoy drawing inspiration from great engineers and inventors such as Wile E. Coyote (SuPeR gEnIouS), André Franquin’s Gaston, and Calvin from Calvin and Hobbes by Bill Watterson.

If you have any questions regarding my work or funny memes to share, feel free to contact me at {wernerpe} at {mit.edu}!

If you can guess what artist the favicon is referencing I will buy you a coffee =)

Publications

Faster Algorithms for Growing Collision-Free Convex Polytopes in Robot Configuration Space

We propose two new algorithms for constructing collision-free convex polytopes in robot configuration space. Our key insight is that finding near-by configuration-space obstacles using sampling is inexpensive and greatly accelerates region generation. Our two algorithms use such samples to either employ nonlinear programming more efficiently (IRIS-NP2 ) or circumvent it altogether using a massively-parallel zero-order optimization strategy (IRIS-ZO). We show that IRIS-ZO achieves an order-of-magnitude speed advantage over its predecessor IRIS-NP. IRISNP2, also significantly faster than IRIS-NP, builds larger polytopes using fewer hyperplanes, enabling faster downstream computation.
[arXiv, 2024]

Peter Werner, Thomas Cohn*, Rebecca H. Jiang*, Tim Seyde, Max Simchowitz, Russ Tedrake, Daniela Rus

Approximating Robot Configuration Spaces with few Convex Sets using Clique Covers of Visibility Graphs

We propose an algorithmic appraoch to approximating robot configuration space with a small collection of convex sets. First, the algorithm constructs a visibility graph using sampling, then a small clique cover of the visibility graph is computed, and finally the cliques are inflated to full-dimensional polytopes that are collision free.
[ICRA, 2024]

Peter Werner, Alexandre Amice, Tobia Marcucci, Daniela Rus, Russ Tedrake

Approximating Robot Configuration Spaces with few Convex Sets using Clique Covers of Visibility Graphs

Certifying Bimanual RRT Motion Plans in a Second

We present an efficient method for certifying noncollision for piecewise-polynomial motion plans in algebraic reparametrizations of configuration space. Such motion plans include those generated by popular randomized methods including RRTs and PRMs, as well as those generated by many methods in trajectory optimization. Based on Sums-of-Squares optimization, our method provides exact, rigorous certificates of non-collision; it can never falsely claim that a motion plan containing collisions is collision-free.
[ICRA, 2024]

Alexandre Amice, Peter Werner, Russ Tedrake

Certifying Bimanual RRT Motion Plans in a Second

Dynamic Multi-Team Racing: Competitive Driving on 1/10-th Scale Vehicles via Learning in Simulation

Autonomous racing is a challenging task that requires vehicle handling at the dynamic limits of friction. While single-agent scenarios like Time Trials are solved competitively with classical model-based or model-free feedback control, multi-agent wheel-to-wheel racing poses several challenges including planning over unknown opponent intentions as well as negotiating interactions under dynamic constraints. In this work, we address these challenges via self-play reinforcement learning to enable zero-shot sim-to-real transfer of highly dynamic policies.
[CORL, 2023 + Best Paper Award ICRA Multi-Robot Learning Workshop, 2023]

Peter Werner, Tim Seyde, Paul Drews, Thomas Balch, Igor Gilitschenski, Wilko Schwarting, Guy Rosman, Sertac Karaman, Daniela Rus

Dynamic Multi-Team Racing: Competitive Driving on 1/10-th Scale Vehicles via Learning in Simulation

Solving Continuous Control via Q-learning

We show that action discretization and value decomposition, enables us to scale and apply Q-learning to continuous control tasks. With bang-bang actions alone, the performance of this critic-only approach matches state-of-the-art continuous actor-critic methods when learning from features or pixels.
[ICLR, 2023]

Tim Seyde, Peter Werner, Igor Gilitschenski, Wilko Schwarting, Martin Riedmiller, Daniela Rus, Markus Wulfmeier

See all publications

Gallery

Alma learns badminton with PPO

Emergent teaming behavior from multi-agent RL (red team)

Swing up with tactile feedback (Gain scheduled LQR + Catching MPC)