AIConfPaper
← Search

Thomas Lew

9 accepted papers

2026

First, Learn What You Don't Know: Active Information Gathering for Driving at the Limits of Handling

Alexander Davydov, Franck Djeumou, Marcus Greiff, Makoto Suminaka, Michael Thompson, John Subosits +1
ICRA 2026poster

Combining data-driven models that adapt online and model predictive control (MPC) has enabled effective control of nonlinear systems. However, when deployed on unstable systems, online adaptation may not be fast enough to ensure reliable simultaneous learning and control. For example, a controller o…

Cited by 0Source ↗PDF ↗Scholar ↗
2026

Generalized-CVO: Fast and Correspondence-Free Local Point Cloud Registration with Second Order Riemannian Optimization

Ray Zhang, Marcus Greiff, Thomas Lew, John Subosits
CVPR 2026

We propose a fast and correspondence-free local point cloud registration method that leverages geometric surface structure and reproducing kernel Hilbert space (RKHS) embeddings. The method represents point clouds as continuous functions with point-wise anisotropic kernels that encode local geometry

Cited by 0Source ↗Scholar ↗
2025

First, Learn What You Don't Know: Active Information Gathering for Driving at the Limits of Handling

Alexander Davydov, Franck Djeumou, Marcus Greiff, Makoto Suminaka, Michael Thompson, John K. Subosits +1
RA-L 2025

Combining data-driven models that adapt online and model predictive control (MPC) has enabled effective control of nonlinear systems. However, when deployed on unstable systems, online adaptation may not be fast enough to ensure reliable simultaneous learning and control. For example, a controller o

Cited by 8Source ↗Scholar ↗
2025

Risk-Averse Model Predictive Control for Racing in Adverse Conditions

Thomas Lew, Marcus Greiff, Franck Djeumou, Makoto Suminaka, Michael Thompson, John K. Subosits
ICRA 2025

Model predictive control (MPC) algorithms can be sensitive to model mismatch when used in challenging nonlinear control tasks. In particular, the performance of MPC for vehicle control at the limits of handling suffers when the underlying model overestimates the vehicle's performance capabilities. I

Cited by 7Source ↗Scholar ↗
2023

Robotic Table Wiping via Reinforcement Learning and Whole-body Trajectory Optimization

Thomas Lew, Sumeet Singh, Mario Prats, Jeffrey Bingham, Jonathan Weisz, Benjie Holson +8
ICRA 2023poster

We propose a framework to enable multipurpose assistive mobile robots to autonomously wipe tables to clean spills and crumbs. This problem is challenging, as it requires planning wiping actions while reasoning over uncertain latent dynamics of crumbs and spills captured via high-dimensional visual o…

Cited by 24Source ↗Scholar ↗
2021

Control Barrier Functions for Cyber-Physical Systems and Applications to NMPC

Jan Schilliger, Thomas Lew, Spencer M. Richards, Severin Hänggi, Marco Pavone, Christopher Harald Onder
RA-L 2021

Tractable safety-ensuring algorithms for cyber-physical systems are important in critical applications. Approaches based on Control Barrier Functions assume continuous enforcement, which is not possible in an online fashion. This letter presents two tractable algorithms to ensure forward invariance

Cited by 15Source ↗Scholar ↗
2020

Sampling-based Reachability Analysis: A Random Set Theory Approach with Adversarial Sampling

Thomas Lew, Marco Pavone
CoRL 2020

Reachability analysis is at the core of many applications, from neural network verification, to safe trajectory planning of uncertain systems. However, this problem is notoriously challenging, and current approaches tend to be either too restrictive, too slow, too conservative, or approximate and th

Cited by 0Source ↗PDF ↗code ↗Scholar ↗
2019

Trajectory Optimization on Manifolds: A Theoretically-Guaranteed Embedded Sequential Convex Programming Approach

Riccardo Bonalli, Abhishek Cauligi, Andrew Bylard, Thomas Lew, Marco Pavone
RSS 2019poster

Sequential Convex Programming (SCP) has recently gained popularity as a tool for trajectory optimization due to its sound theoretical properties and practical performance. Yet, most SCP-based methods for trajectory optimization are restricted to Euclidean settings, which precludes their application…

Cited by 58Source ↗PDF ↗code ↗Scholar ↗