Close this search box.
Close this search box.

Faculty of Aerospace Engineering

Field of Research
The main research idea is to develop an autonomous agent capable of per- forming the task (piloting and landing a RAM air parachute) and convert the computed solution (control variables) into learning aids, utilized in a training simulator. Many aspects that are currently learned by skydivers by trial and error, thus introducing many dangerous situations and accidents, can be trained in VR:
  • adjusting the landing pattern to the weather conditions: wind strength and direction, air humidity and density, etc.
  • adjusting the landing pattern to the traffic situation, taking a safe slot in the stack, preventive flying, dealing with obstacles
  • performing the flare – canopy stalling procedure during landing
  • adjusting the flare procedure to the weather conditions, and different canopy loading and canopy model
  • recovery after canopy stall and canopy collision
  • piloting and landing a canopy with partial malfunction/ double-canopy mal-function
  • high performance landings
At the second stage this research can be extended to training cooperative maneuvers: when multiple agents, both human and autonomous, are simultaneously flying their landing patterns towards the same landing area.

This position is for MSc and PhD students

This research includes the following major steps:

• Developing a dynamic model of RAM air parachute, driven by user inputs applied to the steering toggles.
The model should be configurable to fit different canopy models, wing loading, and weather conditions.
• Validating the model in experiments.
• Developing a VR world showing the inertial motion of the skydiver under canopy from his perspective given
his steering inputs, and the task parameters (e.g. obstacles and the landing area)
• Building the experimental setup in the Lab including the dynamic simulationand VR world running on
PC, VR goggles for displaying the output, and the mechanical imitation of steering lines for providing
the simulator input.
• Conducting experiments with human subjects for developing and validatinga training strategy for
canopy pilots
• Designing control and path planning algorithms for an autonomous canopy pilot in order to gain an
insight into piloting challenges and convert the computed flying pattern and control variables into
motor learning aids.

Requirements: A good background, or a strong desire to acquire knowledge in dynamic systems, control theory, optimal control, and optimization; experience in Matlab/Python/C++; open mind and enthusiasm.

Start Date: Immediate

Apply to:

Research Fellow Anna Clarke

Contact Us

For open questions, please contact us and we will get back to you.



Skip to content