Cosmic: A Modular Autonomous Drone Delivery System

Abstract

Introducing \"Cosmic,\" an innovative modular drone delivery platform designed to autonomously transport payloads in environments where traditional logistics falter, such as disaster- stricken areas and rural locations. At its core lies the Cosmo v1 embedded flight controller based on the STM32H7 microcontroller, which handles low-level sensor integration, motor control, and stabilization. Accompanying it is a Raspberry Pi Compute Module 5 (CM5), enabling high-level autonomy, mission logic, and edge processing. With a dual-processor architecture, Cosmic bridges the gap between reliable embedded control and adaptable decision-making, allowing autonomous aerial logistics even under infrastructure constraints. This paper details the system\'s architecture, hardware design, software implementation, comparative evaluation, and future development roadmap.

Introduction

The rising need for reliable logistics in disaster zones, rural areas, and medical emergencies has spurred advancements in autonomous drone technologies. Traditional UAVs depend on remote operators or centralized servers, limiting functionality in low-infrastructure areas. The Cosmic system overcomes these barriers with a modular, autonomous UAV that supports real-time decision-making, edge-level intelligence, and fault tolerance. It's built around a custom flight controller (Cosmo v1) and a Raspberry Pi CM5, combining robust hardware and intelligent software.

2. Key Features of Cosmic

• Autonomy: Operates without external infrastructure, ideal for harsh environments.

• Modularity: Customizable payloads and flexible architecture for various missions.

• Hardware:

  • Cosmo v1 Flight Controller: Powered by an STM32 microcontroller, includes IMU, barometer, GPS, and DShot ESCs.

  • CM5 Companion Computer: Manages mission planning, computer vision, and communication.

• Electronics:

  • Dual-PCB design for separation of control and power systems.

  • 920KV BLDC motors, custom power distribution board (PDB), passive cooling systems.

3. Software and Firmware

• Cosmo v1 Firmware (C):

  • Kalman filter-based IMU fusion.

  • PID-based stabilization.

  • Safety routines for power/signal loss.

• CM5 Software (Python + ROS):

  • GPS navigation, mission planning, and image processing.

  • Optional LTE/LoRa telemetry.

  • Communication via UART and custom telemetry protocol.

4. Use Case Scenarios

• Disaster Relief: Deliver emergency supplies to isolated regions.

• Medical Logistics: Transport vaccines and blood to rural clinics.

• Search and Rescue: Conduct visual sweeps and drop supplies in remote areas.

Each mission is customizable with modular hardware and software settings.

5. Performance Benchmarks

Simulation and field tests validate strong performance in real-world and emulated conditions.

6. Limitations and Ethical Considerations

• Current Limitations:

  • Lacks obstacle avoidance (planned in v2).

  • Affected by high wind conditions (>20 km/h).

  • Thermal throttling risk under heavy CM5 loads.

• Ethical Concerns:

  • Potential misuse for surveillance.

  • Requires geo-fencing and regulatory compliance.

7. Future Enhancements

• Obstacle detection via stereo cameras or LiDAR.

• Swarm coordination with ROS2.

• Secure telemetry using encrypted LTE/LoRa.

• AI vision using onboard YOLOv5.

• Global NGO partnerships for real-world deployment.

References

[1] Achitya Kushwaha \"Cosmo v1: Modular UAV Flight Controller Design,\" IJRASET, 2025 [2] PX4 Autopilot: https://px4.io/ [3] Raspberry Pi CM5 Docs: https://www.raspberrypi.com/documentation/cm5/ [4] Bosch Sensortec BMP390 Datasheet [5] STMicroelectronics STM32H7 Reference Manual [6] ChatGPT OpenAI https://chatgpt.com/

Copyright

Copyright © 2025 Achitya Kushwaha. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.