Browse publications on Google Scholar (top-right) ↗
Resume (English only)
Academic Achievements
VENTURA: Adapts pre-trained image diffusion models for unified task-conditioned navigation, generating precise long-range visual plans with only 10 hours of real-world data (Under Review, Summer 2025)
CREStE: First approach to solve the full mapless navigation problem using internet-scale priors and counterfactual guidance, achieving SOTA performance over 2km in unseen urban environments (Accepted at RSS 2025, Spring 2025)
UT Campus Object Dataset (UT CODa): Largest multiclass, multimodal urban robotics dataset to date, with 1.3M 3D bounding boxes and 204M annotated terrain points (Accepted at T-RO 2024, Fall 2023)
Looking Inside Out: Transformer and LSTM-based models fusing in-cabin eye tracking and external features for improved driver intent prediction (Accepted at ICRA 2024, Fall 2023)
Convolutional Bayesian Kernel Inference for 3D Semantic Mapping: Learnable convolutional kernel enabling real-time Bayesian inference on 3D point clouds (Accepted at ICRA 2023, Fall 2022)
MotionSC: Novel outdoor dataset and spatiotemporal network for semantic scene completion in dynamic environments (Accepted at RAL 2022, Spring 2022)