UESim_feedback

Setting up the sim

Start the Estimator(Keypoint network)

• Start term
• Navigate to CNN folder: cd Documents/projects/NN_Attitude
• Start the conda environment: conda activate nn_attitude
• Run the estimator: python3 predict_server.py\newline This script waits for an image to appear in folder Documents/projects/ue_images and saves the result in a csv file in cnn_estimate.txt

Start Rendring Engine

• Start Unreal Engine
  • Start terminal from desktop and run the startup shell script : ./unreal.sh
  • Alternatively from term, run : /opt/Linux_Unreal_Engine_5.1.1/Engine/Binaries/Linux/./UnrealEditor
  • From Recent Projects tab, select Browse and navigate to home/mist/Documents/UnrealProjects/RainSIm folder. Select and open the RainSIm.uproject file to run the project in UE.\newline This project contains all the assets required to run single agent or multiagent target satellite tracing problem.
• Navigate to Content Drawer folder from a tab on the bottom left. Run the level: Earth_EditorMode3 by double clicking. Set window pane to Output Log to view status.

Start the dynamics Engine (Python/Matlab)

• From term run matlab: matlab
• Navigate to /home/mist/Documents/projects/Matlab-trial-codes and open and run the script traj_HCW_rotated_dt_adi.m.

Run the python UE executor

• Navigate back to UE tab
• Goto the Tools menu, navigate to the bottom and run Execute Python Scripts. Navigate to folder : home/mist/Documents/UnrealProjects/RainSIm. Run the script matlab_mover_socket_debug.py

Wait