Autonomous Navigation Rover on ROS 2
A complete ROS 2 Humble autonomy stack integrating LiDAR SLAM, state estimation, sensor fusion, planning and motion control.
Problem
Autonomous systems need reliable localisation, mapping and obstacle-aware navigation before higher-level behaviour matters.
Engineered system
A complete ROS 2 Humble autonomy stack integrating LiDAR SLAM, state estimation, sensor fusion, planning and motion control.
Decisions
- Kept perception, estimation, planning and control modular so each layer could be tuned and validated independently.
- Used simulation-first validation to exercise repeatable failure cases before physical trials.
Verification
Repeated Gazebo simulations and RViz inspection checked maps, transforms, localisation stability, planned paths and obstacle avoidance across reruns.
Outcome
A repeatable simulation-validated autonomy stack with stable localisation and obstacle-aware navigation.
Evidence boundary
The evidence is deterministic simulated behaviour, not a fleet deployed in the field.