feat(perception): Dynamic obstacle tracking — moving object prediction for path planning #176

Closed
opened 2026-03-02 10:37:08 -05:00 by sl-jetson · 0 comments
Collaborator

Summary

Track and predict trajectories of moving objects (people, pets, vehicles) for proactive path planning.

Requirements

  • Moving object detection: Differentiate static vs moving objects from consecutive LIDAR/depth scans
  • Trajectory prediction: Linear + Kalman prediction of object future position (2-3s horizon)
  • Velocity estimation: Compute speed and heading of each tracked moving object
  • Nav2 integration: Publish predicted obstacle positions as time-parameterized costmap layer
  • Multi-object: Track up to 20 simultaneous moving objects
  • ROS2 topics: /saltybot/moving_objects (MovingObjectArray with id, pose, velocity, predicted_path)
  • Performance: 10Hz update rate, <50ms latency

Agent: sl-perception

Labels: phase-2, navigation

## Summary Track and predict trajectories of moving objects (people, pets, vehicles) for proactive path planning. ## Requirements - **Moving object detection**: Differentiate static vs moving objects from consecutive LIDAR/depth scans - **Trajectory prediction**: Linear + Kalman prediction of object future position (2-3s horizon) - **Velocity estimation**: Compute speed and heading of each tracked moving object - **Nav2 integration**: Publish predicted obstacle positions as time-parameterized costmap layer - **Multi-object**: Track up to 20 simultaneous moving objects - **ROS2 topics**: /saltybot/moving_objects (MovingObjectArray with id, pose, velocity, predicted_path) - **Performance**: 10Hz update rate, <50ms latency ## Agent: sl-perception ## Labels: phase-2, navigation
Sign in to join this conversation.
No Label
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: seb/saltylab-firmware#176
No description provided.