Muhammad Mudassir Ejaz

About

Muhammad Mudassir Ejaz is a researcher at the forefront of autonomous robotics, specializing in vision-based navigation and deep reinforcement learning (DRL). His work addresses a critical challenge: enabling tracked robots to navigate safely and autonomously in complex, changing environments without human intervention. Ejaz’s major contribution is the development of novel end-to-end deep reinforcement learning architectures that allow robots to learn collision-free navigation directly from raw visual input. His most cited paper (2020, 28 citations) introduces a pioneering DRL-based approach for tracked robots, significantly improving autonomous steering capabilities. He has also advanced the field by proposing methods to accelerate the DRL training process, reducing the substantial computational power typically required—a key step toward practical, real-time deployment. With a total of over 43 citations across his core works, Ejaz’s research bridges the gap between theoretical reinforcement learning and real-world robotic applications. His overview paper on autonomous visual navigation using DRL (2019, 13 citations) serves as a valuable resource for students and researchers entering this domain. Through his innovative network designs and focus on training efficiency, Ejaz is helping to shape the next generation of intelligent, self-navigating robots.

Research Focus

Key Achievements

Top Papers

1. 1
   Vision-Based Autonomous Navigation Approach for a Tracked Robot Using Deep Reinforcement Learning

   28 citations · 2020
2. 2
   Autonomous Visual Navigation using Deep Reinforcement Learning: An Overview

   13 citations · 2019
3. 3
   A fast learning approach for autonomous navigation using a deep reinforcement learning method

   2 citations · 2021

Key Collaborators

Contact & Links