Black-Bee-Drones/nectar-sdk

Black-Bee-Drones/nectar-sdk has 37 GitHub stars and 2 forks. GitHot measured +2 in 7 days and +2 in 15 days. Last checked 2026-11-11 02:18 UTC. Last push 2026-10-22.

How these counts are measured

ROS2 SDK designed to simplify drone control and computer vision tasks

37 stars 2 forks

What you can do

Control — firmware-agnostic flight interface, navigation, PID, obstacles, GPS
Vision — cameras, ArUco, color, line, distance, optical flow, MediaPipe
AI — detection, segmentation, and classification (YOLO, DETR, RF-DETR, ViT), training, evaluation
Sensors — companion rangefinders and MAVLink bridges

Measured star growth

in 24h: Update delayed
in 7 days: +2
in 15 days: +2
in 1 month: Collecting history

Net star changes over overlapping periods, not daily totals. Missing measurements are left blank. Stars measure attention, not software quality.

README.md

Original

ROS 2 software development kit for autonomous aerial systems. Unified interfaces for flight control, computer vision, and AI — one mission across vehicles, sensors, and simulators.

ROS 2 Distro	Build & Test	Docker
Humble
Jazzy
Kilted

Documentation

Please visit the Nectar SDK documentation (en | pt-BR) for installation, tutorials, module reference, Python API, simulation, and Docker.

Features

ArduPilot and PX4 over MAVROS, direct MAVLink, or uXRCE-DDS, plus Bebop and Crazyflie — modular install via uv. The same stack across indoor and outdoor competition missions (IMAV, CBR, SAE), with perception overlays and matched sim-to-real frames:

Depth for each area lives on the docs site:

Control — firmware-agnostic flight interface, navigation, PID, obstacles, GPS
Vision — cameras, ArUco, color, line, distance, optical flow, MediaPipe
AI — detection, segmentation, and classification (YOLO, DETR, RF-DETR, ViT), training, evaluation
Sensors — companion rangefinders and MAVLink bridges
Localization — indoor VSLAM and vision-pose integration
Interface — Qt6 operator GUI
Simulation — ArduPilot/PX4 SITL with Gazebo

End-to-end architecture and design patterns: Architecture. With/without API comparisons: with-without.md. Tagged fixtures and LoC / swap counts: scripts/effort/.

Quick start

Your starting point	Run
Fresh machine, no ROS 2	`bash <(curl -fsSL https://raw.githubusercontent.com/Black-Bee-Drones/nectar-sdk/main/scripts/bootstrap.sh)`
Have ROS 2, no SDK yet	`cd ~/ros2_ws/src && git clone [email protected]:Black-Bee-Drones/nectar-sdk.git && cd nectar-sdk && make setup`
Have the SDK cloned	`make setup`
Want zero host setup	`make docker-build && make docker-run`

Full install paths, drivers, and compatibility: Installation guide.

Example

import nectar
from nectar.control import DroneFactory, MavrosConfig, PoseSource

nectar.init()
drone = DroneFactory.create("mavros", MavrosConfig(pose_source=PoseSource.GPS))

drone.takeoff(altitude=2.0)
drone.move_to(x=5.0, y=0.0, z=0.0, precision=0.3)
drone.move_to_gps(lat=-22.413, lon=-45.449, alt=15.0)
drone.land()
nectar.shutdown()

More examples: Get started

Contributing

We welcome contributions. See the Contributing guide for setup, code style, and the PR process.

Check GitHub Issues for existing discussions
Follow Conventional Commits for commit messages
Read our Code of Conduct

Black Bee Drones

Black Bee Drones is Latin America's first academic autonomous drone team, founded at the Federal University of Itajubá (UNIFEI) in 2014. The team competes nationally and internationally — including IMAV (3rd place indoor 2023 and 2025), CBR RoboCup Flying Robots, and SAE Eletroquad.

Nectar SDK started in 2023 as shared infrastructure for competition missions and is now open source under Apache 2.0.

Acknowledgments

Built on open source across robotics, perception, and AI:

Robotics & flight: ROS 2, ArduPilot, PX4, MAVLink, MAVROS
Perception & AI: OpenCV, PyTorch, Ultralytics, HuggingFace, Roboflow, MediaPipe
Localization & sensors: NVIDIA Isaac ROS, Intel RealSense, Luxonis

License

This project is licensed under the Apache-2.0 License — see the LICENSE file for details.