CARTOS: Charging-Aware Real-Time Operating System for Intermittent Batteryless Devices

This repository contains the source code and implementation of CARTOS, a charging-aware real-time operating system designed for intermittently-powered batteryless devices (IPDs) in IoT applications. CARTOS ensures reliable real-time scheduling and energy-efficient operation of devices that harvest energy from ambient sources, overcoming the challenges posed by intermittent power supply.

Features

• Mixed-preemption Scheduling Model: Tasks are classified into preemptible (computation) and non-preemptible (peripheral) tasks, allowing the system to handle intermittent power efficiently.
• Just-In-Time (JIT) Checkpointing: Computational tasks are dynamically checkpointed when the system detects low energy levels, ensuring forward progress without unnecessary overhead.
• Peripheral Task Management: Non-preemptible peripheral tasks are executed atomically, ensuring that critical operations like sensor reading or data logging are not interrupted by power loss.
• Portability: CARTOS is built on FreeRTOS and can be easily ported to other embedded real-time operating systems.
• Energy Prediction and Adaptation: CARTOS adapts task scheduling to environmental energy availability using a neural network-based energy predictor, ensuring continuous operation in fluctuating conditions.

System Architecture

The system consists of the following key components:

• Energy-Harvesting Unit: Collects energy from ambient sources such as solar, RF, or thermoelectric power.
• Power Management Unit: Regulates and manages energy storage, ensuring efficient use of harvested energy.
• Microcontroller Unit (MCU): Executes tasks based on the mixed-preemption model and handles checkpointing, scheduling, and peripheral task management.
• Memory: Includes volatile (RAM) and non-volatile (FRAM/MRAM) memory for state saving and checkpointing during power loss.

Installation

1. Clone this repository:

   git clone https://github.com/rtenlab/CARTOS.git

2. Ensure you have the necessary development environment set up, including:

   • A compatible MCU development board (e.g., ARM Cortex-M4)
   • FreeRTOS or any other supported RTOS
   • Required peripherals (sensors, capacitors for energy storage, etc.)

3. Build and flash the code to your device using your preferred IDE (e.g., Keil, GCC, IAR).

Usage

1. Configuring Tasks: CARTOS provides APIs for defining tasks as preemptible or non-preemptible. Tasks can be scheduled periodically and grouped into processing chains. Refer to the task_config.h file for task configuration options.

   Example:

   void Task1(void const * arguments) {
       setTaskPeriod(10000, 2000);  // Period = 10s, Offset = 2s
       while (1) {
           // Task code here
           waitForNextPeriod();
       }
   }

2. Running the System: Once the tasks are defined, compile and run the system on your device. CARTOS will handle energy monitoring, checkpointing, and scheduling based on the available energy.

3. Energy Prediction: CARTOS uses an internal energy prediction model to adapt the scheduling strategy based on real-time energy availability.