Autonomous Robot Navigation
Applied research on robust perception and navigation for autonomous mobile robots in unstructured real-world environments.
Project Overview
We develop and validate navigation systems for autonomous mobile robots operating in real, unstructured environments. The goal is to move beyond controlled lab conditions to reliable deployment in offices, warehouses, and outdoor spaces.
Research Topics
Simultaneous Localization and Mapping (SLAM)
Real-time 2D and 3D SLAM using LiDAR and camera data. We focus on robustness to sensor noise and dynamic environments.
Path Planning
Combining classical planning (A*, RRT) with learned heuristics for faster, more adaptive navigation.
Obstacle Avoidance
Dynamic obstacle prediction and avoidance at operational speeds, including handling moving people and objects.
Current Results
- Reliable navigation in 200m² indoor environments
- Map building with <5cm error at 50m range
- Dynamic obstacle avoidance at 1.2 m/s operational speed
Platform
Experiments conducted on the ARL Autonomous Robot Vehicle platform.