To support our research activities, we recently built and inaugurated a quadrocopter platform MART-i. This platform will soon be used to implement experimental algorithms, and to verify the applicability of several scientific methods. However, this requires us to modify the currently-stable platform, which bears the risk of introducing defects. In general, any defect can lead to failure, which in an aerial vehicle may cause danger, injury or loss/crash. This is especially true for light-weight vehicles, where both safety-critical and mission-critical systems tend to share some hardware, to save weight and cost.

Your task is to develop and Emergency Recovery System (ERS) for the quadrocopter, which can detect major malfunctions and trigger an external rescue mechanism, e.g., a parachute. Towards that, first you should analyze how onboard components can be monitored efficiently and intrusion-free (heartbeats? independent measurements of acceleration?), i.e., without risking that the monitoring itself influences the other components.

Then, you design and implement the ERS system using commercial components, such as the MARS mini parachute, and a small processing platform, e.g., the Raspberry Pi computer (http://www.raspberrypi.org/product/model-b/).

Eventually, an identification of failure scenarios and their influence on the ERS-equipped system becomes necessary, to prove that major faults can be caught. This requires using techniques from reliability engineering, such as Fault Tree Analysis (FTA) or Failure Modes and Effects Analysis (FMEA).

The goal is to obtain a mostly self-contained ERS, that can be installed at our quadcopter for the case of highly-experimental flight tests, enabling us to modify core components without risking that an unexpected failure leads to catastrophic events.