6 January 2023 in IEEE Robotics and Automation Letters
For the safe and successful navigation of autonomous vehicles in unstructured environments, the traversability of terrain should vary based on the driving capabilities of the vehicles. Actual driving experience can be utilized in a self-supervised fashion to learn vehicle-specific traversability. However, existing methods for learning self-supervised traversability are not highly scalable for learning the traversability of various vehicles. In this work, we introduce a scalable framework for learning self-supervised traversability, which can learn the traversability directly from vehicle-terrain interaction without any human supervision. We train a neural network that predicts the proprioceptive experience that a vehicle would undergo from 3D point clouds. Using a novel PU learning method, the network simultaneously identifies non-traversable regions where estimations can be overconfident. With driving data of various vehicles gathered from simulation and the real world, we show that our framework is capable of learning the self-supervised traversability of various vehicles. By integrating our framework with a model predictive controller, we demonstrate that estimated traversability results in effective navigation that enables distinct maneuvers based on the driving characteristics of the vehicles. In addition, experimental results validate the ability of our method to identify and avoid non-traversable regions.
1 October 2014 in Journal of Luminescence
1 September 2022 in Journal of Korea Robotics Society
14 January 2026 in IEEE Robotics and Automation Letters
Semantic mapping aims to construct a 3D semantic representation of the environment, providing essential knowledge for robots operating in complex outdoor settings. While Bayesian Kernel Inference (BKI) addresses discontinuities of map inference from sparse sensor data, existing semantic mapping methods suffer from various sources of uncertainties in challenging outdoor environments. To address these issues, we propose an uncertainty-aware semantic mapping framework that handles multiple sources of uncertainties, which significantly degrade mapping performance. Our method estimates uncertainties in semantic predictions using Evidential Deep Learning and incorporates them into BKI for robust semantic inference. It further aggregates noisy observations into coherent Gaussian representations to mitigate the impact of unreliable points, while employing geometry-aligned kernels that adapt to complex scene structures. These Gaussian primitives effectively fuse local geometric and semantic information, enabling robust, uncertainty-aware mapping in complex outdoor scenarios. Comprehensive evaluation across diverse off-road and urban outdoor environments demonstrates consistent improvements in mapping quality, uncertainty calibration, representational flexibility, and robustness, while maintaining real-time efficiency.
29 August 2026 in SSRN Electronic Journal
Creating high-quality fused images from visible and infrared modalities remains challenging due to substantial modality differences and residual cross-modal misalignment. In addition, the reliance on accurately aligned visible--infrared image pairs limits the practical applicability of existing fusion methods. We propose UniFusion, a visible--infrared fusion framework that uses a shared representation for content-aware fusion and residual cross-modal registration. The shared representation is constructed through an SSIM-based decomposition and refinement process that emphasizes structurally meaningful content while reducing modality-specific appearance variations. It provides a common basis for content-aware fusion and can also support residual registration when misalignment is present, without requiring separate representations for the two tasks. The proposed framework can be trained with unpaired visible and infrared images from unrelated scenes, eliminating the need for accurately aligned training pairs. Extensive experiments on public benchmarks demonstrate improved fusion quality and robustness to residual misalignment compared with state-of-the-art methods.