Papers
1
Total Citations
2
H-Index
1
About
Swayam Agrawal is a rising researcher at the intersection of computer vision, robotics, and natural language processing, with a primary focus on enabling intelligent agents to understand and navigate 3D environments through language. His key research areas include vision-language navigation (VLN), semantic mapping, and open-world scene understanding. Agrawal’s most notable contribution is the development of O3D-SIM (Open-set 3D Semantic Instance Maps), a pioneering framework that allows robots to construct open-set, instance-level semantic maps of their surroundings. This work, presented in 2024, builds upon his earlier SI-Maps project, which introduced instance-level semantic mapping for VLN. By enabling agents to recognize and reason about novel objects not seen during training, O3D-SIM represents a significant step toward more flexible, human-like navigation systems. The paper has already garnered early citations, reflecting its timely impact on the field. Agrawal’s research addresses a critical gap in embodied AI: how to bridge the gap between static object databases and the dynamic, open-world environments that robots must operate in. His work is particularly relevant for applications in service robotics, autonomous exploration, and human-robot interaction, where understanding object relationships and following natural language commands is essential.
Research Focus
Key Achievements
Top Papers
- 1