Swapnil Daga

About

Swapnil Daga is a roboticist whose research lies at the intersection of visual perception, autonomous navigation, and reinforcement learning. His most impactful work tackles a critical bottleneck in mobile robotics: the fragility of monocular SLAM (Simultaneous Localization and Mapping). In his highly cited 2018 paper, Daga introduced a novel framework that uses reinforcement learning to actively prevent monocular SLAM failures during trajectory planning. Rather than passively relying on a camera feed, his system learns to anticipate and avoid conditions that cause tracking loss—such as low-texture environments or aggressive motion—thereby enabling more robust, long-term autonomous navigation with a single camera. This work, which has garnered 8 citations, represents a significant step toward closing the loop between perception and planning in resource-constrained robots. Daga’s contributions are particularly valuable for applications in drones, micro-rovers, and other platforms where weight and cost preclude stereo or LiDAR sensors. By merging learning-based decision-making with classical geometric estimation, he has opened new pathways for resilient, self-aware visual odometry systems.

Research Focus

Key Achievements

Top Papers

1. 1
   Learning to Prevent Monocular SLAM Failure using Reinforcement Learning

   8 citations · 2018

Key Collaborators

Contact & Links