Songpengcheng Xia

Papers

Total Citations

H-Index

About

Songpengcheng Xia is a rising researcher at the forefront of embodied AI, autonomous navigation, and 3D scene understanding. His work bridges the gap between perception and decision-making in complex, real-world environments. Xia’s most influential contribution is **Thermal-NeRF**, which extends Neural Radiance Fields (NeRFs) to the infrared spectrum, enabling high-fidelity 3D reconstruction from thermal cameras—a breakthrough for low-light and adverse-weather autonomy, already garnering 16 citations since 2024. In robotics, he developed **UA-LIO**, an uncertainty-aware LiDAR-inertial odometry system that enhances localization reliability for autonomous driving in cluttered urban settings (6 citations). For off-road robotics, Xia introduced a **learning-based traversability costmap** that predicts terrain passability from raw sensor data, enabling safe navigation beyond structured roads (4 citations). His work uniquely integrates probabilistic reasoning with deep learning, advancing both the theoretical rigor and practical deployment of autonomous systems. As a young scholar, Xia’s research is already shaping next-generation perception and navigation stacks for self-driving cars, field robots, and AR/VR platforms.