The Avionics section is responsible for the design and implementation of our aircraft onboard electrical, flight control and communications systems, as well as ground-based support hardware and plane propulsion.
Vision is an advanced aerial rescue system designed to detect individuals in need and determine their precise GPS coordinates during rescue operations. Utilising high-resolution onboard cameras and AI-driven person detection through a convolutional neural network, Vision accurately identifies people and pinpoints their location with sub-meter accuracy. Real-time video transmission allows ground station operators to monitor and coordinate rescue efforts effectively. Powered by a cutting-edge onboard computer, Vision ensures rapid and reliable performance, enhancing the efficiency of search and rescue missions and ultimately saving lives.
Project Lifeline was launched in June 2023 with a vision to develop a payload deployment system for the Albatross. It currently incorporates concepts of winching down a 2 kg payload from the drone, releasing it a few metres off the ground with our unique latch mechanism and winching the cables back into the Albatross. A special mechanism, called Nichrome Emergency Release Failsafe (NERF), is being used to cut the cables should any emergencies arise. This project utilises Python code to run a Raspberry Pi that will operate the payload deployment stages as quickly as possible and communicate with base station regarding its progress.
As part of the Albatross, we must be able to identify and provide the location of unique paper targets on the ground from the air.
This project combines a machine-learning algorithm for finding these targets and advanced geo-locating techniques to provide highly accurate GPS coordinates of the identified targets.
The avionics stack manages the power and communication channels of the plane. The stack team is responsible both designing and managing the wiring harness for the plane and well as performing pre-flight checks to ensure that it can take off and land safely.
Detect and Avoid (D&A), launched in May 2024, focuses on creating and implementing a system capable of detecting other users of a shared airspace and preventing mid-air collisions. It integrates with Mission Management to calculate flight paths that avoid any potential collisions, ensuring the safety of the aircraft and any surrounding aircraft.
The Power Distribution Board (PDB) project revolves around efficiently connecting battery power and regulating it to suit the voltage and current requirements of peripherals and motors. Primarily undertaken within Altium, the project encompasses circuit and PCB design to develop a streamlined power management solution. In addition to its core focus on power distribution, the project also involves creating other necessary custom hardware development such as LED lighting, custom CAN nodes and protection circuitry
DroneLink is UAS’s custom-built, in-house mission management software tailored to UAS’s needs. It automates the drone’s mission, reducing the risk of human error. It features an intuitive general user interface that displays flight-critical information such as live aircraft speed, location of the drone, on-board camera feeds, battery data and much more, allowing our users to easily monitor and analyse the mission. This UI can be simultaneously run on multiple devices, with the data updating dynamically on all of them.
The key feature of Dronelink is its ability to take gps coordinates as input to set a search area boundary, and generate a splined, optimal serach path within that area. Our custom waypoint-splining algorithm takes into consideration the plane’s turning radius, the camera’s field of view, and tries to minimise turning as much as possible in order ot provide a more stable feed for the onboard camera. On top of this, a settings menu for further fine-tuning of the computed search path is provided so that the optimal settings can be configured for every mission.
DroneLink also automates the actions of the plane upon detection of a patient on the ground. If patient detection is confirmed, DroneLink commands the plane to travel to the patient and orbit them, and switches to an in-built interface so that the pilot can select a safe location to deploy medical supplies. The plane will automatically travel towards the deployment location, deploy the supplies, and then return to orbit the patient for monitoring.
This year, DroneLink will be expanding to automate the drone maneuvers for the 2025 NFC competition. We plan to add a function that instructs the drone to execute complicated aerobatics such as a barrel roll, or a full 360 degree circle turn.
