no code implementations • 8 Apr 2024 • Yusuf Umut Ciftci, Zeyuan Feng, Somil Bansal
The algorithm abstracts the imitation error as an adversarial disturbance in the system dynamics, injects it during data collection to expose the expert to safety critical states, and collects corrective actions.
no code implementations • 14 Dec 2023 • Albert Lin, Somil Bansal
In this work, we propose two verification methods, based on robust scenario optimization and conformal prediction, to provide probabilistic safety guarantees for neural reachable tubes.
1 code implementation • 23 Sep 2023 • Aryaman Gupta, Kaustav Chakraborty, Somil Bansal
Our results show the efficacy of the proposed approach in identifying and handling system-level anomalies, outperforming methods such as prediction error-based detection, and ensembling, thereby enhancing the overall safety and robustness of autonomous systems.
no code implementations • 4 Nov 2022 • Kaustav Chakraborty, Somil Bansal
Our approach blends simulation-based analysis with HJ reachability methods to compute an approximation of the backward reachable tube (BRT) of the system, i. e., the set of unsafe states for the system under vision-based controllers.
no code implementations • 25 Sep 2022 • Albert Lin, Somil Bansal
A recently proposed method called DeepReach overcomes this challenge by leveraging a sinusoidal neural PDE solver for high-dimensional reachability problems, whose computational requirements scale with the complexity of the underlying reachable tube rather than the state space dimension.
1 code implementation • 4 Nov 2020 • Somil Bansal, Claire Tomlin
Its advantages include compatibility with general nonlinear system dynamics, formal treatment of bounded disturbances, and the ability to deal with state and input constraints.
1 code implementation • 20 Mar 2020 • Varun Tolani, Somil Bansal, Aleksandra Faust, Claire Tomlin
Videos describing our approach and experiments, as well as a demo of HumANav are available on the project website.
no code implementations • L4DC 2020 • Anjian Li, Somil Bansal, Georgios Giovanis, Varun Tolani, Claire Tomlin, Mo Chen
In Bansal et al. (2019), a novel visual navigation framework that combines learning-based and model-based approaches has been proposed.
no code implementations • 29 Oct 2019 • Somil Bansal, Andrea Bajcsy, Ellis Ratner, Anca D. Dragan, Claire J. Tomlin
We construct a new continuous-time dynamical system, where the inputs are the observations of human behavior, and the dynamics include how the belief over the model parameters change.
no code implementations • 12 Sep 2019 • Thomas Beckers, Somil Bansal, Claire J. Tomlin, Sandra Hirche
In this work, we present a framework to optimize the kernel and hyperparameters of a kernel-based model directly with respect to the closed-loop performance of the model.
no code implementations • 1 May 2019 • Andrea Bajcsy, Somil Bansal, Eli Bronstein, Varun Tolani, Claire J. Tomlin
Our safety method is planner-agnostic and provides guarantees for a variety of mapping sensors.
Robotics
no code implementations • 6 Mar 2019 • Somil Bansal, Varun Tolani, Saurabh Gupta, Jitendra Malik, Claire Tomlin
Model-based control is a popular paradigm for robot navigation because it can leverage a known dynamics model to efficiently plan robust robot trajectories.
no code implementations • 14 Feb 2018 • Somil Bansal, Shromona Ghosh, Alberto Sangiovanni-Vincentelli, Sanjit A. Seshia, Claire J. Tomlin
We propose a context-specific validation framework to quantify the quality of a learned model based on a distance measure between the closed-loop actual system and the learned model.
1 code implementation • 21 Sep 2017 • Somil Bansal, Mo Chen, Sylvia Herbert, Claire J. Tomlin
Hamilton-Jacobi (HJ) reachability analysis is an important formal verification method for guaranteeing performance and safety properties of dynamical systems; it has been applied to many small-scale systems in the past decade.
Systems and Control Dynamical Systems Optimization and Control
no code implementations • 10 Sep 2017 • Somil Bansal, Roberto Calandra, Kurtland Chua, Sergey Levine, Claire Tomlin
Reinforcement Learning is divided in two main paradigms: model-free and model-based.
no code implementations • 27 Mar 2017 • Somil Bansal, Roberto Calandra, Ted Xiao, Sergey Levine, Claire J. Tomlin
Real-world robots are becoming increasingly complex and commonly act in poorly understood environments where it is extremely challenging to model or learn their true dynamics.
no code implementations • 21 Mar 2017 • Sylvia L. Herbert, Mo Chen, SooJean Han, Somil Bansal, Jaime F. Fisac, Claire J. Tomlin
We propose a new algorithm FaSTrack: Fast and Safe Tracking for High Dimensional systems.
Robotics