Skip to main content

​
Topics

​
/hazards/front

Type: sensor_msgs/msg/PointCloud2
Publisher: hazard_detection_node
Source: src/lunabot_perception/lunabot_perception/hazard_detection.py:22 Point cloud data representing detected obstacles (boulders, craters, and other hazards) from the front depth camera. The raw depth camera data is processed using Open3D to:
  1. Downsample the point cloud (5cm voxel size)
  2. Remove statistical outliers
  3. Segment ground plane using RANSAC
  4. Publish obstacle points for navigation costmap integration
Used by the navigation stack to identify and avoid hazards in the lunar terrain.

​
/tag_pose

Type: geometry_msgs/msg/PoseWithCovarianceStamped
Publisher: tag_pose_publisher
Source: src/lunabot_localisation/lunabot_localisation/tag_pose_publisher.py:30 Robot pose estimate in the map frame derived from AprilTag detection. This topic provides an independent localization source by combining:
  • Static map->tag transform (known tag location)
  • Dynamic camera->tag transform (AprilTag detection)
  • Static camera->base_footprint transform (URDF)
The covariance is distance-dependent:
  • Position covariance: (distance × 0.1)² with minimum 0.01
  • Yaw covariance: (distance × 0.05)² + 0.01
Published at 10 Hz when the AprilTag is visible.

​
/camera_front/points

Type: sensor_msgs/msg/PointCloud2
Subscriber: hazard_detection_node
Source: src/lunabot_perception/lunabot_perception/hazard_detection.py:25 Raw point cloud data from the front-facing depth camera. This is the input for hazard detection processing.

​
Actions

​
/mission/excavate

Type: lunabot_interfaces/action/Excavate
Server: material_action_server
Source: src/lunabot_control/lunabot_control/material_action_server.py:31 Controls the excavation mechanism to collect lunar regolith during the digging phase. Goal Fields:
  • timeout_s (float64): Maximum duration before aborting (0 = no timeout)
Feedback Fields:
  • phase (uint8): Current excavation phase
    • PHASE_PRECHECK = 0: Pre-operation safety checks
    • PHASE_SPINUP = 1: Starting excavation motors
    • PHASE_DIGGING = 2: Active material collection
    • PHASE_RETRACT = 3: Retracting bucket
  • elapsed_s (float64): Time since goal acceptance
  • fill_fraction_estimate (float64): Estimated bucket fill level (0.0-1.0)
  • excavation_motor_current_a (float64): Motor current draw in amperes
  • jam_detected (bool): Mechanism jam detection flag
  • estop_active (bool): Emergency stop status
Result Fields:
  • success (bool): Whether excavation completed successfully
  • reason_code (uint8): Result category
    • REASON_SUCCESS = 0: Nominal completion
    • REASON_TIMEOUT = 1: Goal timeout exceeded
    • REASON_CANCELED = 2: Client canceled the goal
    • REASON_FORCED_FAILURE = 3: Testing failure mode
    • REASON_SHUTDOWN = 4: Node shutdown during execution
  • failure_reason (string): Human-readable error description
  • collected_mass_kg_estimate (float64): Estimated mass collected
  • duration_s (float64): Total execution time

​
/mission/deposit

Type: lunabot_interfaces/action/Deposit
Server: material_action_server
Source: src/lunabot_control/lunabot_control/material_action_server.py:40 Controls the deposition mechanism to unload collected material into the collection bin. Goal Fields:
  • dump_duration_s (float64): How long to hold dump position
  • timeout_s (float64): Maximum duration before aborting (0 = no timeout)
Feedback Fields:
  • phase (uint8): Current deposition phase
    • PHASE_PRECHECK = 0: Pre-operation safety checks
    • PHASE_OPENING = 1: Opening bin door
    • PHASE_RAISING = 2: Raising dump bed
    • PHASE_DUMPING = 3: Material flowing out
    • PHASE_CLOSING = 4: Returning to transport configuration
  • elapsed_s (float64): Time since goal acceptance
  • actuator_current_a (float64): Actuator current draw in amperes
  • door_open (bool): Bin door position
  • bed_raised (bool): Dump bed position
  • estop_active (bool): Emergency stop status
Result Fields:
  • success (bool): Whether deposition completed successfully
  • reason_code (uint8): Result category
    • REASON_SUCCESS = 0: Nominal completion
    • REASON_TIMEOUT = 1: Goal timeout exceeded
    • REASON_CANCELED = 2: Client canceled the goal
    • REASON_FORCED_FAILURE = 3: Testing failure mode
    • REASON_SHUTDOWN = 4: Node shutdown during execution
  • failure_reason (string): Human-readable error description
  • residual_fill_fraction_estimate (float64): Estimated material remaining (0.0-1.0)
  • duration_s (float64): Total execution time

​
TF Frames

The robot’s coordinate frame tree follows REP-105 conventions for mobile robot localization.

​
Frame Hierarchy

map
 └─ odom
     └─ base_footprint
         └─ base_link
             └─ [sensor frames...]

​
map → odom

Type: EKF Transform
Publisher: ekf_filter_node_map
Source: src/lunabot_localisation/config/ekf.yaml The map-to-odom transform corrects for odometry drift using absolute positioning sources (AprilTag detections, GPS). This transform is:
  • Discontinuous: Jumps when new absolute measurements are fused
  • Globally consistent: Maintains accurate long-term positioning
  • Published by the map-frame EKF filter node

​
odom → base_footprint

Type: EKF Transform
Publisher: ekf_filter_node_odom
Source: src/lunabot_localisation/config/ekf.yaml The odom-to-base_footprint transform provides smooth, continuous motion estimates by fusing:
  • Wheel odometry
  • IMU data
  • (Optionally) Visual odometry
This transform is:
  • Continuous: No sudden jumps, smooth motion
  • Locally accurate: Drifts over long distances
  • Published by the odom-frame EKF filter node
Type: Static Transform (URDF Joint)
Source: src/external/leo_common-ros2/leo_description/urdf/macros.xacro Fixed offset between the ground projection point (base_footprint) and the robot’s physical center (base_link). This accounts for the robot’s chassis height and ensures 2D navigation operates at ground level while 3D sensors reference the physical robot body.
  • Static: Never changes
  • Purpose: Separates 2D navigation (footprint) from 3D perception (link)

​
Frame Usage Guidelines

FrameUse For
mapLong-term goals, waypoint navigation, mission planning
odomShort-term control, local planning, velocity commands
base_footprintGround-level 2D navigation, costmap obstacles
base_linkSensor mounting, 3D perception, robot geometry

​
Deferred Topics

The following topics are referenced in the interface contracts but not yet implemented:
  • /mission/state - Current mission state machine phase
  • /mission/autonomy_mode - Manual vs autonomous control mode
  • /mission/time_remaining_s - Competition time remaining
  • /mission/cycle_count - Number of excavation cycles completed
  • /mission/last_failure_reason - Most recent failure diagnostic
These topics are planned for future implementation as the mission autonomy system develops.

Build docs developers (and LLMs) love

Get started for freeTalk to us