Aashi Manglik
Papers
1
Total Citations
11
H-Index
1
About
Aashi Manglik is a researcher advancing the frontiers of autonomous navigation and human-robot interaction, with a particular focus on safety-critical perception systems. Her work centers on developing vision-based methods that enable robots to anticipate and avoid collisions in crowded, dynamic environments. In her highly cited 2019 paper, "Forecasting Time-to-Collision from Monocular Video: Feasibility, Dataset, and Challenges," Manglik introduced a novel deep learning approach that directly estimates time-to-collision from a single monocular camera—eliminating the need for expensive depth sensors. This work, which has garnered 11 citations, demonstrated the feasibility of using purely image-based models to predict imminent collisions between a suitcase-shaped robot and nearby pedestrians. By creating a new dataset and establishing benchmarks, she provided the research community with both the tools and the challenge of real-time collision forecasting. Manglik's contributions are particularly impactful for assistive robotics and autonomous navigation in human environments, where reliable, low-cost perception is essential. Her research continues to push the boundaries of what is possible with monocular vision, making robots safer and more intuitive companions in our daily lives.
Research Focus
Key Achievements
Top Papers
- 1