Context-Specific Validation of Data-Driven Models

by   Somil Bansal, et al.

With an increasing use of data-driven models to control robotic systems, it has become important to develop a methodology for validating such models before they can be deployed to design a controller for the actual system. Specifically, it must be ensured that the controller designed for an abstract or learned model would perform as expected on the actual physical system. We propose a context-specific validation framework to quantify the quality of a learned model based on a distance metric between the closed-loop actual system and the learned model. We then propose an active sampling scheme to compute a probabilistic upper bound on this distance in a sample-efficient manner. The proposed framework validates the learned model against only those behaviors of the system that are relevant for the purpose for which we intend to use this model, and does not require any a priori knowledge of the system dynamics. Several simulations illustrate the practicality of the proposed framework for validating the models of real-world systems.


page 1

page 2

page 3

page 4


Active Model Learning using Informative Trajectories for Improved Closed-Loop Control on Real Robots

Model-based controllers on real robots require accurate knowledge of the...

Data-driven HVAC Control Using Symbolic Regression: Design and Implementation

The large amount of data collected in buildings makes energy management ...

On the Robustness of Data-Driven Controllers for Linear Systems

This paper proposes a new framework and several results to quantify the ...

Robust Data-Driven Predictive Control using Reachability Analysis

We present a robust data-driven control scheme for unknown linear system...

Data-Driven Optimization for Deposition with Degradable Tools

We present a data-driven optimization approach for robotic controlled de...

A Generalizable Knowledge Framework for Semantic Indoor Mapping Based on Markov Logic Networks and Data Driven MCMC

In this paper, we propose a generalizable knowledge framework for data a...

Online Modeling and Control of Soft Multi-fingered Grippers via Koopman Operator Theory

Soft grippers are gaining momentum across applications due to their flex...

Please sign up or login with your details

Forgot password? Click here to reset