⚙️ Configuration Reference

A comprehensive glossary of all configuration parameters available in config_resolved.yaml.

🧭 General Settings

Parameter	Type	Default	Description
`run_name`	`string`	`RUN_001`	Unique identifier for the benchmark instance.
`output_dir`	`path`	`./results`	Directory where logs, bags, and metrics are saved.
`headless`	`bool`	`false`	If `true`, suppresses all GUI windows (for CI/CD).

Parameters defining the robot's physical constraints and sensors.

Example robot.yaml

robot:
  radius: 0.25
  max_speed: 0.5
  lidar:
    range: 12.0
    fov: 360

radius : (float) Radius of the robot's footprint in meters. Used for collision checking.

max_speed : (float) Maximum linear velocity in m/s.

lidar.range : (float) Maximum effective range of the laser scanner.

Common parameters tuned by the Auto-Tuner.

linearUpdate : (float) Process a scan each time the robot translates this far (meters). * Recommended: 0.2 - 0.5

angularUpdate : (float) Process a scan each time the robot rotates this far (radians). * Recommended: 0.1 - 0.5

particles : (int) Number of particles in the filter. Higher = better accuracy but more CPU. * Typical: 30 (fast) to 100 (accurate).

TRAJECTORY_BUILDER_2D.use_imu_data : (bool) Whether to use IMU data for scan matching. Critical for 3D SLAM, optional for 2D.

submaps.num_range_data : (int) Number of scans to insert before creating a new submap.

Settings for the metrics engine.

Parameter	Description
`align_trajectories`	Use Umeyama alignment to match estimated path to ground truth.
`eval_rate_hz`	Frequency of pose recording (default: 10Hz).
`save_images`	Generate PNGs of coverage and error maps.