Guohao Peng
Papers
13
Total Citations
144
H-Index
7
About
Guohao Peng is a robotics researcher whose work sits at the intersection of autonomous mobile robot localization, multi-sensor fusion, and robust perception. His research primarily addresses some of the most persistent challenges in robot navigation: operating reliably in geometrically repetitive environments, adverse weather conditions, and GPS-denied settings. A defining contribution is the NTU4DRadLM dataset (2023, 33 citations), which established a pioneering multi-modal benchmark centered on 4D radar, thermal cameras, and IMU — enabling SLAM research that holds up against rain, fog, and smoke where conventional LiDAR and visual approaches falter. Peng has also made significant strides in magnetic-field-assisted localization, developing frameworks such as the magnetic-assisted initialization method (22 citations) and MGLT (16 citations) that guide robots through featureless, degenerate environments like industrial warehouses and office corridors without pre-installed infrastructure. His work extends to visual place recognition, where LSDNet (12 citations) introduced a lightweight self-attentional distillation network balancing efficiency and accuracy. With contributions spanning sensor calibration, semantic map fusion, and last-mile delivery robot navigation, Peng's research collectively advances the resilience and practicality of autonomous systems across real-world deployment scenarios.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3
- 4
- 5
- 6Vision Based Sidewalk Navigation for Last-mile Delivery Robot12 citations · 2022
- 7
- 8Probabilistic 3D Semantic Map Fusion Based on Bayesian Rule6 citations · 2019
- 9
- 10