https://omerkurkutlu.github.io/tags/mobile-robotics/Mobile RoboticsHugoBlox Kit (https://hugoblox.com)en-usThu, 02 Jul 2026 00:00:00 +0000https://omerkurkutlu.github.io/media/icon_hu_eee4a95885829ab2.pnghttps://omerkurkutlu.github.io/tags/mobile-robotics/https://omerkurkutlu.github.io/projects/tunnel-robot/Thu, 02 Jul 2026 00:00:00 +0000https://omerkurkutlu.github.io/projects/tunnel-robot/<h2 id="overview">Overview</h2> <p>The Autonomous Tunnel Burrowing Robot is an intelligent robotic excavation platform designed to autonomously construct underground tunnels based on user-defined dimensions. The system combines a mobile robot, a gantry-mounted robotic arm, environmental perception, and autonomous planning to perform excavation tasks in challenging environments.</p> <p>This project was developed during my research at the University of Notre Dame under the supervision of <strong>Dr. Yasemin Ozkan Aydin</strong>.</p> <h2 id="motivation">Motivation</h2> <p>Underground construction is often dangerous, labor-intensive, and difficult to automate. This project explores how autonomous robotics can improve safety, precision, and efficiency in tunnel excavation by integrating perception, manipulation, and autonomous decision-making.</p> <h2 id="key-features">Key Features</h2> <ul> <li>Autonomous tunnel excavation</li> <li>Mobile robotic platform</li> <li>Gantry-mounted robotic manipulator</li> <li>User-defined tunnel generation</li> <li>ROS and Gazebo simulation</li> <li>Sim-to-real deployment</li> <li>Soil interaction analysis</li> <li>Autonomous task execution</li> </ul> <h2 id="hardware">Hardware</h2> <ul> <li>Custom mobile robot</li> <li>Gantry robotic arm</li> <li>Intel RealSense Depth Camera</li> <li>Torque sensor</li> <li>Custom excavation end-effectors</li> </ul> <h2 id="software">Software</h2> <ul> <li>ROS</li> <li>Gazebo</li> <li>Python</li> <li>C++</li> <li>OpenCV</li> <li>Point Cloud Processing</li> <li>Robot Manipulation</li> </ul> <h2 id="research-contributions">Research Contributions</h2> <p>This project investigates:</p> <ul> <li>Autonomous excavation planning</li> <li>Robot manipulation for construction tasks</li> <li>Perception using depth cameras</li> <li>Soil interaction analysis</li> <li>Sim-to-real robotic deployment</li> <li>Autonomous navigation during excavation</li> </ul> <h2 id="experimental-evaluation">Experimental Evaluation</h2> <p>The system was evaluated using multiple excavation scenarios with different soil conditions and interchangeable digging tools. Both simulation and real-world experiments demonstrated reliable autonomous excavation and tunnel generation.</p> <h2 id="future-work">Future Work</h2> <p>Future research directions include:</p> <ul> <li>Learning-based excavation strategies</li> <li>Vision-language robot control</li> <li>Multi-robot cooperative excavation</li> <li>Foundation models for autonomous construction robotics</li> </ul> <h2 id="project-status">Project Status</h2> <p><strong>Status:</strong> ✅ Completed Research Project</p>https://omerkurkutlu.github.io/projects/vision-language-action/Thu, 02 Jul 2026 00:00:00 +0000https://omerkurkutlu.github.io/projects/vision-language-action/<h2 id="overview">Overview</h2> <p>Vision-Language-Action (VLA) Navigation is an ongoing research project that aims to develop a universal autonomous navigation framework capable of controlling both aerial and ground robots directly from visual observations and natural language instructions.</p> <p>The goal is to eliminate traditional modular navigation pipelines by learning an end-to-end policy that maps perception and language directly to robot actions.</p> <h2 id="research-motivation">Research Motivation</h2> <p>Current robotic navigation systems typically consist of separate perception, localization, mapping, planning, and control modules. While effective, these pipelines are often difficult to generalize across different robot platforms and environments.</p> <p>Inspired by recent advances in foundation models and embodied AI, this project investigates whether a single Vision-Language-Action model can generalize navigation behaviors across multiple robotic platforms.</p> <h2 id="research-objectives">Research Objectives</h2> <ul> <li>Universal navigation across multiple robot platforms</li> <li>End-to-end visual navigation</li> <li>Natural language robot control</li> <li>Cross-domain generalization</li> <li>Sim-to-real deployment</li> <li>Resource-efficient inference</li> </ul> <h2 id="target-platforms">Target Platforms</h2> <ul> <li>Autonomous Tiny Drones</li> <li>Mobile Robots</li> <li>Autonomous Vehicles</li> <li>Legged Robots</li> </ul> <h2 id="input-modalities">Input Modalities</h2> <ul> <li>RGB Images</li> <li>Video Streams</li> <li>Natural Language Instructions</li> <li>Robot State Information</li> </ul> <p>Example instructions include:</p> <ul> <li>&ldquo;Navigate to the red chair.&rdquo;</li> <li>&ldquo;Avoid pedestrians.&rdquo;</li> <li>&ldquo;Move to the charging station.&rdquo;</li> <li>&ldquo;Fly through the open doorway.&rdquo;</li> </ul> <h2 id="output">Output</h2> <p>The model directly predicts navigation actions such as:</p> <ul> <li>Linear velocity</li> <li>Angular velocity</li> <li>Steering commands</li> <li>Drone velocity commands</li> <li>Waypoints</li> </ul> <h2 id="planned-architecture">Planned Architecture</h2> <p>The proposed framework combines:</p> <ul> <li>Vision Encoder</li> <li>Language Encoder</li> <li>Multimodal Fusion</li> <li>Action Decoder</li> </ul> <p>to produce end-to-end navigation commands.</p> <h2 id="simulation-platforms">Simulation Platforms</h2> <ul> <li>NVIDIA Isaac Sim</li> <li>NVIDIA Omniverse</li> <li>CARLA</li> <li>Webots</li> <li>ROS 2</li> </ul> <h2 id="potential-applications">Potential Applications</h2> <ul> <li>Autonomous driving</li> <li>Drone navigation</li> <li>Warehouse automation</li> <li>Search and rescue</li> <li>Industrial inspection</li> <li>Service robotics</li> </ul> <h2 id="current-status">Current Status</h2> <p>This project is currently under active development as part of my PhD research at the University of Illinois Chicago.</p> <h2 id="future-directions">Future Directions</h2> <p>Future work includes:</p> <ul> <li>Multi-robot collaboration</li> <li>Long-horizon navigation</li> <li>Outdoor deployment</li> <li>Continual learning</li> <li>Foundation models for autonomous robotics</li> <li>Deployment on embedded robotic platforms</li> </ul>