Gimbal Tracking

Problem Statement

Gimbal pointing control keeps a camera line-of-sight centered on a moving target while the UAV body rotates and translates.

Model and Formulation

Pan-tilt angle errors:

$$ e_{\psi} = \psi_{des}-\psi,\quad e_{\theta} = \theta_{des}-\theta $$

Basic control law:

$$ u = K_p e $$

Desired angles are generated from look-at geometry between camera and target coordinates.

Algorithm Procedure

1. Compute target direction in camera/body frames.
2. Derive desired pan and tilt angles.
3. Apply proportional gimbal control with rate limits.
4. Feed stabilized image stream to vision modules.

Tuning and Failure Modes

• Low gains under-track fast targets.
• High gains induce jitter near setpoint.
• Inaccurate extrinsics lead to persistent pointing offsets.

Implementation and Execution

python -m uav_sim.simulations.sensors.gimbal_tracking

Evidence

References

• Shin and Park, PID-based Gimbal Stabilization for UAV Camera Systems
• Siciliano et al., Robotics: Modelling, Planning and Control

Related Algorithms

• Visual Servoing
• Gimbal BBox Tracking