Postgraduate Study Topics
Characterising Bird Flight in Turbulence and Gusts
The project aims to discover how birds perceive and cope with turbulence in order to greatly increase the steadiness of aircraft flight through turbulent air. Initial research showed that by adapting novel sensors, inspired by the sensory functions of birds, the flight performance and safety in turbulence is significantly improved. As part of this proposal, other avian turbulence mitigation strategies will be discovered through wind tunnel experimentation in repeatable gust(s) and turbulence. The discovered avian strategies will be adapted for man-made aircraft, to enable ultra-stable flight in turbulence. Measurements are proposed that will document bird’s motion and the flow fields impinging on the birds during flight in repeatable turbulence. An image tracking system will be used for measuring the wing, body and head motions of the bird using markers on the bird.
Tethered Fixed-Wing Drone Delivery
Civil package delivery via drones is a realistic concept currently under exploration by both the academia and industry alike. Most current research is focused on multi-rotor Unmanned Aerial Systems (UAS) which are designed to carry a payload from one destination to another through autonomous guidance and control systems. Due to the endurance limitation of current multi-rotor UAS (commonly less than 30 minutes) the capable range is limited and thus restricted to deliveries that are relatively near to the ground station. City and suburban environments are well suited for limited-range package delivery however long range package delivery can be explored using fixed-wing UAS technology. This becomes particularly important for deliveries in country areas where delivery costs can be significantly minimized. Fixed-wing aircraft do not have the vertical take-off and landing (VTOL) capability of rotary UAS systems and thus provides a fundamental problem in using fixed-wing UAS for package delivery missions.
Using tethered system technology, a UAV equipped with a tethered system can deploy a package down a cable while adopting a circular flight pattern above the target area. The cable will form into a trailing helix formation and allow a geometrically stationary package to be lowered to the ground target area and released (Refer to the illustration below). Past research has simulated and verified the helix formation under various payloads and aircraft scales. Simulations have also suggested that inflight parcel pick up can also be performed using this method. RMIT are currently developing the full system while exploring other advantages this system may have for applications that require rapid deployment and drop off of packages of various sizes and weight. The influence of the system on a generic postal delivery system is also under investigation however initial studies have revealed significant cost savings in fuel and staff costs due to the high levels of UAV automation and reduced fuel consumption of the UAV
Micro-Sensor Development for Turbulence Detection by UAVs
The operational capability of Small unmanned air vehicles (UAVs) is severely limited in windy conditions. The undesirable aircraft motion caused by turbulence in the wind blur image data, curtail the number of flying days per year and result in aborted flights (e.g., crashes). Reducing the size of UAVs increases the challenges of holding a steady flight path. A patented, biomimetic technique of ’feeling’ a way through turbulent air has demonstrated enormous promise, with far steadier flight being demonstrated than is possible with existing inertial-based stabilization: https://youtu.be/m_2-bblBmQY . The technique involves sensing upstream wind gusts and providing control inputs to counteract the impending undesirable motions much earlier than current technology permits. To-date we have “sensed” the upstream turbulent air using pressure probes forward of each wing. This PhD project proposes to build on this prior success, extending the technique to incorporate improved sensing using new, non-invasive sensors (e.g. micro LIDAR and / or RADAR). The systems will be evaluated via wind-tunnel flight tests of small UAVs in turbulent flows, followed by outdoor flight trials under a range of adverse turbulent winds.
Bio-inspired Autonomous Soaring Drones
Small Unmanned Air Vehicles (UAVs) suffer from limited range and endurance. However naturally occurring phenomenon in the atmosphere (such as updrafts) can be harvested to sustain flight. The goal of this project is to develop an energy biased path-planning algorithms for real-time implementation on a UAV to minimise energy expenditure.
Turbulence Mitigation for Micro Air Vehicles
Micro Air Vehicles can be particularly sensitive to atmospheric flow disturbances. The undesirable aircraft motion caused by turbulence in the wind blurs image data, curtail the number of flying days per year and result in aborted flights (e.g., crashes). Reducing the size of UAVs increases the challenges of holding a steady flight path. A patented, biomimetic technique of ’feeling’ a way through turbulent air has demonstrated enormous promise, with far steadier flight being demonstrated than is possible with existing inertial-based stabilization: https://youtu.be/m_2-bblBmQY . The technique involves sensing upstream wind gusts and providing control inputs to counteract the impending undesirable motions much earlier than current technology permits. To-date we have “sensed” the upstream turbulent air using pressure probes forward of each wing allowing a fixed wing UAV to achieve significant improvements in disturbance rejection performance. However there is potential for further improvements through considering non-conventional aircraft designs and actuation techniques enabling steadier MAVs.
Swarming Drone Flight in Turbulence
Multiple Small Unmanned Air Vehicles (UAVs) are well-suited for formation flight in the Atmospheric Boundary Layer (ABL) for cooperative missions. However their operational capability can be severely limited in windy conditions resulting in mid-air collisions. A research opportunity exists to characterize effects of gust(s) on a flock of UAVs in RMIT’s Wind Tunnel facility whereby repeatable turbulence can be generated. This understanding can be used to develop suitable flocking/swarming architectures, which best enable a group of UAVs to navigate through turbulent conditions. This project will further evolve to consider suitable to mitigation techniques to ensure safe operation of the drones
Risk-based Sensor Management for Unmanned Aerial Vehicles’ Safe Emergency Landing
UAM involves safe and efficient transportation of everything from small parcels to human passengers over urban environments. The concept of air taxi services whereby a UAV can take off, fly and land almost without passenger intervention represents an important achievement in aviation, and is yet to be realized. As UAV operation becomes routine in populated environments, significant effort is required to safe and
efficient operation in an uncontrolled real-world uncertainty-rich environment. With advancements in
autonomy, UAVs are capable of operating with minimal human intervention, thus warranting the
commercial viability of UAM. Further advancements in autonomy can also alleviate the current regulatory
restrictions associated with flight over populated areas. Current regulatory restrictions are mainly due to
current UAV technologies which are not able to demonstrate an Equivalent Level of Safety to manned
aircraft, particularly in the case of an engine failure which would require an emergency or forced landing.
Safety systems currently available to UAVs only allow the aircraft to fly towards a pre-defined safe landing
area from a database of known safe landing locations. However, these systems must be preprogramed with
up to-date information, thus requiring a continuous communications link between a human operator and
the UAV to ensure that the latter will not attempt to land at an unsuitable location leading to a hazardous
situation. An alternative would be to have a system onboard the UAV that can process information in a
similar way to a human pilot in emergency situations that require the aircraft to land immediately. This project involves developing such a system using vision based computing systems