Personal tools

Study and Development of Intelligent Capability for Small-Size UAVs

From iis-projects

Jump to: navigation, search
Red-Based object-tracking application with the nano-size Crazyflie 2.0.


State-Of-The-Art (SOA) algorithms using sensors (i.e images, pressure, etc) for autonomous navigation of Unmanned Aerial Vehicles (UAVs) have reached very high capability and complexity. Unfortunately, such capabilities are nowadays present only on big-size, power-angry machines. Due to the limited power-envelope a major challenge is represented by bringing such algorithms on tiny size aerial vehicles such as the Bitcraze CarzyFlie 2.0 [1]. This project focuses on the feasibility study and the development of high level cognitive skills (e.g. “follow-me” capability, object detection, etc.) for such constrained small-size quad-rotors. The project will investigate both in algorithms that uses camera and in other sensors or techniques. In particular the student will investigate on the possibility to use beacons sent by an external source (anchor or user's smart phone) to follow a specific direction. Starting from the existing open-source/open-hardware project, CrazyFlie 2.0, the student has to evaluate first the current architecture and organize the software/hardware modifications required to enable high level cognitive capabilities. The CrazyFlie 2.0 is characterized by an on-board MCU (i.e. STM32F405) that will be used as initial processing device for the on-board tasks computation. After the initial evaluations, the student will apply the identified modifications to the existing system and develop the additional software tasks. Lastly, an evaluation of the proposed solution, for other architectures (e.g. other MCUs, PULP platform, etc) will be performed in order to study the portability and to overcome eventual limitations in the initial MCU.


The key milestones of this project should be:

  • Establish a first autonomous navigation capability where the UAV is able to collected data from sensor deployed in a field.
  • Design, implementation of techniques to achieve the follow me application using, BTLE 3D direction detection, camera, other sensors or a combination of them.
  • Implement and profiling feature-tracking technique on the on-board MCU (STM32F405) and on the PULP platform for the Follow me Applications.
  • Testing and evaluation of the final prototype and write a project report.

Status: Completed

Master Thesis by Jaskirat Singh
Supervision: Daniele Palossi, Michele Magno


  • Familiarity with embedded system programming in C.
  • Basic knowledge of FreeRTOS [2] and STM32F4 MCU family [3] is favorable.


33% Theory and SoA study
33% C embedded programming
33% Verification and experimental evaluation


Luca Benini

Detailed Task Description

Meetings & Presentations

The student(s) and advisor(s) agree on weekly meetings to discuss all relevant decisions and decide on how to proceed. Of course, additional meetings can be organized to address urgent issues.


[1] Bitcraze Crazyflie2.0

[2] Free RTOS

[3] STM32F405/7

Practical Details

↑ top