Rohitashva Singh Saurabh

About

Rohitashva Singh Saurabh is a robotics researcher whose work lies at the intersection of computer vision, simultaneous localization and mapping (SLAM), and reinforcement learning. His most-cited contribution, "Learning to Prevent Monocular SLAM Failure using Reinforcement Learning" (2018, 8 citations), addresses a critical challenge in autonomous navigation: the fragility of monocular SLAM systems when integrated with trajectory planning. Rather than treating SLAM as a passive module, Saurabh pioneered a reinforcement learning framework that actively learns to predict and avoid conditions that would cause SLAM failure—such as low-texture environments or rapid motions—enabling more robust, long-term autonomous operation. This work bridges the gap between classical geometric estimation and modern learning-based control, offering a practical solution for robots operating in unstructured environments with only a single camera. While his citation count reflects a focused, early-stage career, the novelty of his approach—using RL to safeguard a core perception system—marks him as a researcher tackling foundational problems in field robotics. His work is particularly relevant for students and engineers working on vision-based autonomous systems where reliability is paramount.

Research Focus

Key Achievements

Top Papers

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

   8 citations · 2018

Key Collaborators

Contact & Links