Looking for DDPG Quadcopter Helix Trajectory Example (MATLAB/Simulink or Python)

Hello everyone,

I am a Master’s student working on training a quadcopter to follow a helix trajectory using Deep Deterministic Policy Gradient (DDPG). The goal of my thesis is to develop a full reinforcement-learning-based controller for 3D trajectory tracking without traditional PID or cascaded controllers.

I have already built a working drone model and integrated DDPG for training, but I am struggling with:

• Reward shaping
• Normalization/scaling
• Training stability (divergence)
• Making the agent track a continuous 3D helix path
• Proper tuning of actor/critic networks

Before spending months debugging everything from scratch, I wanted to ask:❓ Has anyone successfully trained a DDPG quadcopter to follow a helix trajectory?

If so, it would help me a lot if you could share:

• Your quadcopter dynamics model
• DDPG training script or Simulink model
• Reward function
• Pretrained agent (if possible)
• Any notes on how you stabilized training

I will fully credit and cite any shared work in my thesis.

If you know any GitHub repo, paper, or example that includes DDPG + quadcopter + trajectory tracking, please comment. Even partial implementations are extremely valuable.

Thank you!