UAV Path Prediction (Guideline)

1. Introduction

We utilize collected data from unmanned aerial vehicle (UAV) movements in the x, y, and z axes as input for model training. The Long Short-Term Memory (LSTM) architecture is employed to train the model, and the OSC AIMLFW framework is used as the training model framework. Through this process, a model capable of predicting the path of UAVs is trained, providing a solution for various use cases.

Ensure that the Influx database is operational for this use case, which is based on the OSC AI/ML Framework (Release-J).

2. Suggested Requirements

Getting Started (Start from data insertion)

• Step 1. Query influx token

  # Search the influx "token" value in below output cat bitnami/influxdb/influxd.bolt | tr -cd "[:print:]"

  
  
  

• Step 2. Create “UAVData” bucket (Inside Influx DB container)

  influx bucket create -n UAVData -o primary -t <token>

  
  
  

• Step 3. Fill the file config (UAV_insert.py)

  DATASET_PATH = '/path/to/dataset.csv' # Replace to the UAV dataset path INFLUX_IP = 'localhost' # Influx IP INFLUX_TOKEN = 'VJpoNpqeVnjzvhpPm8jZ' # Influx token

  
  
  

• Step 4. Excute the insert processing to insert data into Influx DB

  
  
  

• Step 5.  Upload UAV_pipeline.ipynb to aiml-notebook and generate a UAV_pipeline.yaml
  
  

  Open

  

  
  
  
  

• Step 6. Start a UAV model training job
  
  

  Open

  

  
  
  
  

• Step 7. Load UAV model
  

  
  
  

• Step 8. Start model prediction
  1. obtain Ingress port for Kserve (kubectl get svc istio-ingressgateway -n istio-system)
  
  

  
  2. source uav.sh

File List

• UAV_dataset.csv (Download) , (NEW)_UAV_dataset(source)
  The file contains collected UAV movement path data.
  
  
  

• UAV_insert.py (Download)
  The file processes the UAV_dataset and inserts the data into InfluxDB.
  (Changed required: DATASET_PATH , INFLUX_IP , INFLUX_TOKEN)
  
  
  

• UAV_pipeline.ipynb (Download)
  The file defines the model structure and training process.
  
  
  

• UAV_deploy.yaml (Download)
  The yaml file is used for deploying model inference service.
  
  
  

• UAV_input.json (Download)
  The json file is the sample data for the prediction.
  
  
  

• UAV_predict.sh (Download)
  The script used for excuting the model prediction.
  
  
  

Example

• Input:
  This input data represents a collection of points in a three-dimensional space, with each point defined by a set of three coordinates corresponding to the x, y, and z axes (After normalization).

  • Example input

    UAV_input

    
    
    

• Output:
  The output should be next xyz-axis path prediction (After normalization).
  
  
  

  • Example output

    UAV_output

    
    
    

Contributors

• Joseph Thaliath - Samsung

• Antony Wang - NTUST MITLab

• Jasmine Lee - NTUST MITLab

• Tuck-Wai Choong - NTUST MITLab