Code for "Propeller Damage Detection: Adapting Models to Diverse UAV Sizes" Submitted to IEEE Robotics and Automation Letters.

This manuscript introduces a transfer learning method for adapting propeller fault detection neural networks to different unmanned aerial vehicles (UAVs). After training a simple model for detecting if any propeller in a specific vehicle has a failure (in this case, a chipped tip), a domain adaptation based in the vehicles' physics is performed in order to use the same model to detect failures in vehicles with different structures, weights, or motor-propeller sets. A key feature is that the detection model uses only inertial sensors that are standard in commercial UAVs, making it broadly applicable without the need for additional hardware.

https://www.youtube.com/watch?v=MXNc1VfQ0Y8

1. Clone this repository.

2. Install Python3 dependencies, matplotlib, numpy, pytorch, scikit-learn, scipy, seaborn, torch, lightning, tqdm

3. Run model_train.py with the following configuration for each experiment:

Use the quadcarbono_datamodule inside model_train.py

data_module = quadcarbono_datamodule

Use the affine_datamodule inside model_train.py

affine_datamodule = affine_DataModule(
    source_path="data/garage_manual/",
    target_path="data/quadcarbono/",
    source_fs=222,
    target_fs=222,
)
data_module = affine_datamodule

For the naive use no scaling, set self.scale_factor = 1.0 inside affine.py

For the scaled experiment, set self.scale_factor = 0.734 inside affine.py

For this experiment use the data in quadcarbono_weight_shift as test instead of Hasymm164_inertial

Use the affine_datamodule inside model_train.py

affine_datamodule = affine_DataModule(
    source_path="data/quadcarbono/",
    target_path="data/quadcarbono_1000/",
    source_fs=222,
    target_fs=222,
)
data_module = affine_datamodule

For the naive use no scaling, set self.scale_factor = 1.0 inside affine.py

For the scaled experiment, set self.scale_factor = 1.12 inside affine.py

For this experiment use quadcarbono_water as test

affine_datamodule = affine_DataModule(
    source_path="data/quadcarbono/",
    target_path="data/hexaF550/",
    source_fs=222,
    target_fs=222,
)
data_module = affine_datamodule