LIRA

We work on methods for autonomous robot navigation and perception in unstructured and uncertain environments, like outdoor areas. We also work on image-based anomaly detection for complex, multi-modal datasets.

Relevant publications

WayFAST: Traversability Predictive Navigation for Field Robots

RA-L'22

A robot navigating through outdoor terrain.

D-AnoGAN: Anomaly Detection in Disconnected Data Manifolds with Generative Adversarial Networks

IJCNN'22

A visualization of latent clustering for anomaly detection of MNIST.

This work was funded in part by ARL W911NF2020184, Army W9132T-19-C-0006, the MITRE Corporation research agreement 126392, and ONR N00014-20-1-2249.

Robot Navigation and Perception

Learning for autonomous robot navigation and perception in the field

Last updated: February 24, 2026

Relevant publications

WayFAST: Traversability Predictive Navigation for Field Robots

D-AnoGAN: Anomaly Detection in Disconnected Data Manifolds with Generative Adversarial Networks