Subject

A flight scheduler is a program that collects and processes data from multiple sources to plan flight paths for Unmanned Aerial Vehicles (UAVs). Current versions of flight schedules are not able to deal with the dynamism required for UAV flight planning. The leveraging of a deep reinforcement learning (RL) based flight plan and scheduler can help overcome limitations that exist with more traditional schedulers. These schedulers should be able to provide and update flight plans in a quicker and more efficient way, all in an automated way.
A key component of the scheduler is the routing algorithm, which defines the optimal route that the UAV should follow to go from one point to another. These routing algorithms typically require data from multiple sources, including, but not limited to, power consumption, weather data and crowd data. The scheduler in this case is a combination of A* and RL agents, which work together to define the best path.
A key challenge is to ensure that the data being received is reliable (sources) and represents reality with a reasonable degree of accuracy and completeness. A geo-hashed data storage model has been developed to efficiently store (in a scalable way) and expose a multitude of 4D datasets towards the routing algorithms/AI. Furthermore, if the incoming data is reliable, the processing and aggregation of these sets of data should be checked, to ensure that the algorithms are operating in the intended way.

Kind of work

The auditing of several components of the scheduler:
• Evaluate and improve quality of data sources (e.g. obstacle, crowd, traffic & weather data)
• Evaluate and identify (additional) data sources that play a role in path definition
• Review the efficacy/efficiency of the geo-hashed data storage model for path definition
• Propose improvements
• The outputs of the data aggregation and processing agents

Framework of the Thesis

Support will be provided from a functional and IT technical side from Helicus and its core partners. A first conceptual study on this topic has been created as part of a Helicus coordinated ICON project HAI-SCS.

Number of Students

1 or 2 (preferably in the form of a duo thesis)

Expected Student Profile

Applied Computer Science or Electrical Engineering students with an interested in Data Science and in routing optimization