Minahil Raza
Papers
1
Total Citations
6
H-Index
1
About
Minahil Raza is a robotics researcher whose work lies at the intersection of reinforcement learning and real-world autonomous navigation. Her primary contributions focus on developing collision avoidance systems for mobile robots, specifically through her work on SACPlanner—a Soft Actor Critic local planner that leverages polar state representations. This approach, detailed in her most-cited paper, demonstrates how recent enhancements to the SAC algorithm, including RAD and DrQ, can achieve near-perfect training performance after just 10,000 episodes, bridging the gap between simulated training and real-world deployment. With 6 citations for this key work, Raza’s research is gaining traction among roboticists seeking practical, data-efficient solutions for autonomous navigation in dynamic environments. Her findings highlight the potential of reinforcement learning to produce robust, collision-free trajectories on physical robots, addressing a critical challenge in field robotics. By validating these methods on real-world platforms, Raza is helping to move RL-based planners from theory into practice, making her a rising contributor to the growing body of work on safe, learning-driven robot control.
Research Focus
Key Achievements
Top Papers
- 1