Master of Science in Computer Science


Department of Computer Science

Faculty / School

Faculty of Computer Sciences (FCS)

Date of Submission



Dr. Muhammad Sarim, Visiting Faculty, Department of Computer Science

Document type

MSCS Survey Report


In recent years, there has been a wide spread use of miniature drones, also called unmanned aerial vehicles (UAVS) in different civilian applications. Few of the wide range use of these aerial robots include, but are not limited to, search & rescue, covering sports events, helping wildlife conservation efforts (Price, et al. 2018), aerial surveillance, goods transportation, surveying & inspection tasks (Li, et al. 2017), three dimensional mapping, monitoring & irrigating crops (Trujillano, et al. 2018), traffic monitoring etc.

Deep learning algorithms such as deep convolution neural networks or neural networks in general, due to being universal function approximators, have gained the attention of the aerial robotics research community for the last few years. Successfully applying the deep learning algorithms can accomplish the communities’ dream of automating the aforementioned tasks, which until few years back could only be done through a manual control of these drones.

The study is based on the research survey about deep learning applications in aerial robotics, how & what different algorithms can be used in the scenario, what are the specific challenges & solutions facing the implementation of autonomy in these systems. The study focuses on small light weight drones such as Quadrotors, which due to its constrained and limited resources, as compared to large military drones, require efficient algorithms to achieve automation tasks in indoor, outdoor settings.

The full text of this document is only accessible to authorized users.