This is a mobile application for paper Autopilot drone in construction: A proof of concept for handling lightweight instruments and materials.
This paper presents a proof of concept for automating the lifting procedure of lightweight instruments and materials in construction sites.
The below chart shows the system overview that consist of a drone, a moblie application and machine vision algorithms.
This is the application overview. This application is designed for defining a mission and automatically controlling the UAV through the transportation operation.
This application is built based on DJI Mobile SDK for Android.
We integrate the Monodepth2 algorithm into the system in order to avoid unpredicted collisions during flight.
- Download the Monodepth2 from here.
- Extract the
monodepth2-master
intoapp/src/main/java/pythonPackage/monodepth2
directory. - Download the
mono_640x192
precomputed disparity predictions from the given link in their repository. - Extract the file into the
monodepth2/models
-
Build from Android Studio:
- Click the "Build" menu and select "Make Project"
-
Build from Command Line:
- Open a terminal window and navigate to the project's root directory.
Run
./gradlew build
(orgradlew build
on Windows) to execute the build process from the command line.
- Open a terminal window and navigate to the project's root directory.
Run
-
Run from Android Studio:
- Click the green "Run" button (play icon) in the toolbar. Select a connected device or emulator from the dropdown. Android Studio deploys the APK to the chosen device and launches the app.
-
Run from Command Line:
- After a successful build, run the app directly on a connected device using:
adb shell am start -n com.dji.sdk.autopilot/MainActivity
- After a successful build, run the app directly on a connected device using: