GPS-Denied Drone Navigation Development
Budget: $1,500 – $3,000 USD
Project Title:
Development of a GPS-Denied Drone Navigation System Using SLAM and MAVLink Integration
Project Overview:
I am developing a drone navigation system capable of operating in GPS-denied environments using onboard sensors and visual-inertial SLAM. The project is being built on a Jetson Nano (Ubuntu 18.04) with ROS Melodic, and leverages a stereo camera, IMU, and barometer for localization and obstacle avoidance.
The system streams video from a Raspberry Pi CSI camera to the Jetson Nano, which processes the data using VINS-Fusion. Commands are sent to a Cube Orange flight controller via MAVLink for real-time autonomous control (roll, pitch, yaw). The final objective is fully autonomous navigation from point A to point B, with dynamic obstacle avoidance and waypoint updates.
Scope of Work:
I am seeking an experienced ROS and embedded systems developer to assist with the following:
Integration and optimization of VINS-Fusion SLAM on Jetson Nano
Writing or debugging MAVROS/MAVLink ROS nodes to communicate with Cube Orange
Developing or improving real-time obstacle detection and avoidance logic
Ensuring reliable sensor data integration (stereo camera, IMU, barometer)
Optional: System documentation and ROS architecture structuring
Technical Environment:
Jetson Nano (Ubuntu 18.04)
ROS Melodic
Stereo camera (IMX219)
Raspberry Pi (for camera stream)
Cube Orange flight controller
MAVLink/MAVROS
VINS-Fusion
Python and C++ (for ROS node development)
Ideal Candidate:
Strong experience with ROS (preferably ROS 1 – Melodic)
Prior work with VINS-Fusion, RTAB-Map, or similar SLAM systems
Proficient in MAVLink protocol and drone communication
Familiarity with Jetson Nano or similar embedded AI platforms
Demonstrated ability in integrating sensor data and working with real-time autonomous navigation systems
Please reach out if you require any additional context such as diagrams, current implementation status, or log outputs. I am looking for efficient and reliable collaboration to advance this project toward a robust field-tested prototype.
Development of a GPS-Denied Drone Navigation System Using SLAM and MAVLink Integration
Project Overview:
I am developing a drone navigation system capable of operating in GPS-denied environments using onboard sensors and visual-inertial SLAM. The project is being built on a Jetson Nano (Ubuntu 18.04) with ROS Melodic, and leverages a stereo camera, IMU, and barometer for localization and obstacle avoidance.
The system streams video from a Raspberry Pi CSI camera to the Jetson Nano, which processes the data using VINS-Fusion. Commands are sent to a Cube Orange flight controller via MAVLink for real-time autonomous control (roll, pitch, yaw). The final objective is fully autonomous navigation from point A to point B, with dynamic obstacle avoidance and waypoint updates.
Scope of Work:
I am seeking an experienced ROS and embedded systems developer to assist with the following:
Integration and optimization of VINS-Fusion SLAM on Jetson Nano
Writing or debugging MAVROS/MAVLink ROS nodes to communicate with Cube Orange
Developing or improving real-time obstacle detection and avoidance logic
Ensuring reliable sensor data integration (stereo camera, IMU, barometer)
Optional: System documentation and ROS architecture structuring
Technical Environment:
Jetson Nano (Ubuntu 18.04)
ROS Melodic
Stereo camera (IMX219)
Raspberry Pi (for camera stream)
Cube Orange flight controller
MAVLink/MAVROS
VINS-Fusion
Python and C++ (for ROS node development)
Ideal Candidate:
Strong experience with ROS (preferably ROS 1 – Melodic)
Prior work with VINS-Fusion, RTAB-Map, or similar SLAM systems
Proficient in MAVLink protocol and drone communication
Familiarity with Jetson Nano or similar embedded AI platforms
Demonstrated ability in integrating sensor data and working with real-time autonomous navigation systems
Please reach out if you require any additional context such as diagrams, current implementation status, or log outputs. I am looking for efficient and reliable collaboration to advance this project toward a robust field-tested prototype.