ReLU activated neural networks have a lot in common with non-smooth dynamical systems! Building off our prior work on frictional robotic systems, we analyze the stability of learned neural network control policies using convex optimization, specifically Linear Matrix Inequalities (LMIs). This efficient approach is made possible by drawing a clear connection between these neural networks and Linear Complementarity Systems. Feedback is welcome! The paper is below, with code to come shortly.
We’re excited to share a new preprint where we learn the dynamics of multi-contact interaction. Contact dynamics are notoriously difficult to model and identify, owing largely to the discontinuous nature of impacts and friction.
Common methods for learning implicitly assume motion is continuous, causing unrealistic predictions (e.g. penetration or floating). We resolve this conflict by learning a smooth, implicit encoding of contact-induced discontinuities, leading to data-efficient identification. Our method can predict realistic impact, non-penetration, and stiction when trained on 60 seconds of real-world data
How should force (tactile) sensors be used within reactive feedback loops? We have a new preprint available (an extended version of a 2020 ICRA publication), where we use bilinear matrix inequalities to synthesize provably stabilizing controllers for multi-contact systems, without combinatorial mode enumeration.
To help answer questions related to the upcoming fall semester, specifically for MEAM 517, there is now an FAQ available here.
Next week, it’s #icra2020! In “Optimal Reduced-order Modeling of Bipedal Locomotion” by Yu-Ming Chen and Michael Posa, we try to find the best low-dimensional model that captures high-performance walking. The solution combines trajectory optimization and stochastic gradient descient, as a bilevel optimization directly over potential models. Check out the paper or virtual talk.
We’re excited to present our work at #icra2020! In “Contact-Aware Controller Design for Complementarity Systems” by Alp Aydinoglu, Victor Preciado, and Michael Posa, we use bilinear matrix inequalities to synthesize controllers that use tactile feedback to stabilize systems through nearby contact events. Check out the paper or the virtual talk.
We are happy to have two papers accepted to ICRA this year! Check out the preprints “Optimal Reduced-order Modeling of Bipedal Locomotion” and “Contact-Aware Controller Design for Complementarity Systems.”
Prof. Aaron Johnson of Carnegie Mellon and I are excited to announce the 2020 Dynamic Walking Conference! Dynamic Walking will be held May 11-14 at Woodloch Resorts in the Poconos of Pennsylvania. Abstract submissions are due February 21. See you there!
Details at http://dynamicwalking.org
The GRASP Lab is excited to be hosting the Northeast Robotics Colloquium this year on October 12, 2019! See https://nerc.seas.upenn.edu/ for more details on abstract submission, registration, and other logistics.