ROB498 EngSci Capstone Project

Contributors

• Savanna Blade
• Nicole Amenta
• Kohava Mendelson

Project Description

Welcome to the ROB498 EngSci Capstone Project! In this project, our team was tasked with constructing a drone and endowing it with a suite of intelligent capabilities, including stationkeeping and waypoint navigation. This README file provides an overview of the project and its components.

Hardware Setup

To achieve the above capabilities, the drone requires an accurate 3D pose estimate relative to its starting point. We utilized Visual-Inertial Odometry (VIO) to obtain this estimate, which determines the pose of a vehicle by combining information from camera images and inertial measurements. As part of our hardware setup, we integrated the Jetson Nano and RealSense T265 tracking camera into our existing hardware suite.

Jetson Nano: The Jetson Nano serves as a companion computer to the PX4 flight controller. We installed and configured MAVROS on the Jetson Nano to establish MAVLink communication with the PX4.

RealSense T265 Tracking Camera: The RealSense T265 tracking camera provides odometry information to the drone, which is crucial for VIO. We used a pre-existing ROS node, available at this GitHub repository, to bridge the camera with ROS running on the Jetson Nano. It's important to note that the camera was mounted on our chassis with forward-facing lenses, requiring us to configure the camera orientation within the VIO bridge. To achieve this, we modified the bridge_mavros.launch file to specify a transformation of 0.127 0 -0.0762 0 0 0 between the PX4 and the camera.

Software Frameworks

The software frameworks that we chose to use in our design are detailed below.

ROS: ROS offers a standardized infrastructure for communication and data exchange between various components of a robotic system. In our design, we set up a publish-subscribe messaging system that allowed our drone to make real-time control decisions based on information from its various sensors. We also made use of ROS services which facilitate more sophisticated interactions between nodes beyond basic messaging. Python was used to integrate ROS into our software solution for all of the challenge tasks.

MAVROS: MAVROS is a middleware that facilitates communication between the Jetson Nano and PX4, allowing data exchange between the two systems. One such example is the exchange of odometry information from the PX4, which is transmitted to the Jetson Nano through MAVROS on the /mavros/vision_pose/pose topic. Additionally, the PX4 is configured to receive MAVLink messages that control the drone's flight mode. This functionality allows for remote control of the drone and is highly relevant to all challenge tasks.

Project Files

This project includes two main Python files:

1. challenge_2.py: This script accomplishes the following task:

   For this task, teams must provide their drone with the ability to perform autonomous stationkeeping. In the context of inspection drones, stationkeeping is a task that requires the drone to maintain a stable hover at a fixed altitude. This challenge goes beyond ordinary stationkeeping and demands that the drone also maintain its pose (position and orientation) relative to a fixed external frame (e.g., MY580). The pose that the drone should maintain is determined by its starting position and a prescribed altitude of 150 cm. Since stationkeeping is often required for prolonged observation, the drone must demonstrate a stable hover at this pose for a test duration of 30 seconds.

2. challenge_3.py: This script accomplishes the following task:

   For this task, teams must enable their drones to navigate through several waypoints. Waypoint navigation allows the drone to visit several variable or pre-programmed sites accurately and quickly. As in Challenge #2, the task begins with a 150 cm hover; upon reaching the hover, a set of waypoint coordinates are sent to the drone (represented in the global Vicon frame). The waypoints are defined as spherical volumes around the given coordinates with a radius of 35 cm. There is no constraint regarding how long the drone must remain in the waypoint; it is sufficient to simply enter the bounding area. Additionally, this is a timed challenge, with a limit of 60 seconds for a full traversal of all given waypoints.

Usage

Contributions

We welcome contributions to this project! If you'd like to contribute, please follow these guidelines:

1. Fork the repository on GitHub.
2. Clone your forked repository to your local machine.
3. Create a new branch for your feature or bug fix.
4. Make your changes and test thoroughly.
5. Commit your changes and push them to your GitHub repository.
6. Create a pull request against the main branch of the original repository.

License

This project is licensed under the MIT License. You are free to use, modify, and distribute this software as long as you adhere to the terms of the license.

---

Thank you for your interest in the ROB498 EngSci Capstone Project! If you have any questions or need further assistance, please feel free to contact us. We look forward to your contributions and hope this project will be a valuable learning experience for all involved.