Building an Offline-First Robot Control App with Kotlin

In this tutorial, you will build a practical Kotlin/Android component for a robotics or Physical AI system. The design emphasizes asynchronous processing, lifecycle-aware state, real-time data handling, observability, and safe separation between the Android interface and physical robot control.

Android Kotlin + Jetpack Compose
             ↓
       ViewModel / Flow
             ↓
      Repository / API
             ↓
   ROS 2 / Jetson / AI Backend
             ↓
        Robot System
data class LocalRobotState(
    val connected: Boolean = false,
    val pendingCommands: List<String> = emptyList()
)

Use a local persistence layer appropriate to your application so the UI can open without a network connection.

data class PendingCommand(
    val id: String,
    val command: String,
    val createdAt: Long
)
if (connected) {
    pendingCommands.forEach { send(it) }
}

Use unique command IDs so a reconnect does not accidentally execute the same command twice.

Do not allow dangerous physical actions to queue indefinitely while disconnected. Some commands should be disabled entirely when the robot cannot be verified as connected.

Make it obvious whether the user is operating the live robot, viewing cached information, or waiting for synchronization.

StateFlow for observable UI state. The resulting Kotlin layer can be extended with real ROS 2 bridges, NVIDIA Jetson services, computer vision models, smart-glasses SDKs, or multimodal AI backends. Keep hardware-specific code behind interfaces so the Android application remains maintainable as the robotics stack evolves.

Website: www.v-modal.com

SDK Flutter: https://github.com/v-modal/vmodal_sdk_flutter

SDK Android: https://github.com/v-modal/vmodal_sdk_android

Discord: https://discord.gg/K72z28KUx