Columbus Maximus


Project Domains Mentors
Embedded Systems, CAD, ROS2, SLAM, UWB Technology, Robotics, Sensor Fusion, Autonomous Navigation Siddharth Mishra, Vedant Malkar

Project Description

GPS fails indoors yet robots in warehouses and factories navigate with remarkable precision. Columbus Maximus solves this by fusing three sensing technologies: UWB anchors for absolute position, LiDAR-based SLAM for real-time mapping, and wheel odometry for motion tracking. Since no single sensor is fully reliable, an Extended Kalman Filter (EKF) continuously blends all three, correcting drift and noise to produce a stable, accurate pose estimate. Built on ROS2, the system visualizes the robot’s live position and generates a map in RViz, the same workflow used in professional autonomous navigation systems.

Beyond building a working robot, the project compares SLAM-only, UWB-only, and full EKF fusion approaches giving you real experimental insight into sensor trade-offs and robustness. Columbus Maximus gives you hands-on experience with sensor fusion, probabilistic estimation, embedded hardware, and autonomous navigation, the core skills behind self-driving robots and industrial AGVs.


Resources

Research Paper
What is UWB?
What is SLAM?
UWB Dev Kit
Playlist For Example Project Pipeline