Ofertes treballs fi d'estudis

Títol: Deep Learning Methods for the Implementation of Remote UAV Applications

Director/a: Soria Perez, Jose antonio

Email del professor/a: JASORIA@EEL.UPC.EDU

Departament del professor/a: 710-EEL

• 2009 - GRAU EN ENGINYERIA ELECTRÒNICA INDUSTRIAL I AUTOMÀTICA
• 2009 - GRAU EN ENGINYERIA ELECTRÒNICA INDUSTRIAL I AUTOMÀTICA/GRAU EN ENGINYERIA ELÈCTRICA
• 2010 - GRAU EN ENGINYERIA EN TECNOLOGIES AEROESPACIALS
• 2023 - GRAU EN ENGINYERIA EN TECNOLOGIES INDUSTRIALS
• 2010 - GRAU EN ENGINYERIA EN VEHICLES AEROESPACIALS
• 2012 - MÀSTER UNIVERSITARI EN ENGINYERIA DE SISTEMES AUTOMÀTICS I ELECTRÒNICA INDUSTRIAL
• 2013 - MÀSTER UNIVERSITARI EN ENGINYERIA INDUSTRIAL
• 2014 - MÀSTER UNIVERSITARI EN ENGINYERIA AERONÀUTICA
• 2016 - MÀSTER UNIVERSITARI EN ENGINYERIA ESPACIAL I AERONÀUTICA

Identificador de l'oferta: 205-04149

Modalitat: Universitat

Possibilitat de beca/finançament: No

Paraules clau:

Descripció:

The integration of deep learning algorithms into UAV systems has become a rapidly growing field of research in recent years. This is due to the success of deep learning algorithms in various applications, including video and image processing and speech recognition systems, as well as the increasing demand for UAVs in various industries, just to name few examples. However, there are still several challenges and limitations that need to be addressed in order to fully realize the potential of deep learning and its deploiment in UAVs. These include issues related to the energy consumption of real-time processing, and the size and weight of processing of deep learning algorithms, the hardware required to implement them, and the limited amount of training data available for certain applications. Additionally, there is a need for more effective and efficient deep learning algorithms that can handle the complexity and variability of real-world data and environments.

Objectius:

This project covers the implementation or customization of UAVs and the integration of deep learning techniques that can be used to enhance the capabilities of UAVs in areas such as autonomous flight, object detection and tracking, energy efficiency, agricultural monitoring, and image recognition. It also covers the development of web-based communication applications that serve to run remote operations asynchronously.

Tasques a realitzar / Característiques:

The following is a list is of tasks to be carried out by the successful engineer student applying to this offer: 1. Conduct a comprehensive review of the state-of-the-art research in the field of UAVs and deep learning. 2. Implementation of a customized UAV (quadcopter, hexacopter or an aircraft with wings) 3. Development of deep learning algorithms for remote UAV applications for, at least one of the following application fields: autonomous flight, object detection and tracking, energy efficiency, agricultural monitoring, and image recognition. 4. Training for the developed deep learning algorithms using real-world data and UAV systems. 5. Inference tests and validation of the deep learning algorithms in various environments and conditions to assess their robustness and reliability. 6. Write a comprehensive report and thesis documenting the results of the research, including an analysis of the performance of the developed deep learning algorithms and a discussion of the potential future applications and developments in the field. 7. Present the results of the research to academic and industrial communities through technical report. 8. Continuously stay informed of the latest developments and trends in the field of UAVs and deep learning, and actively engage in ongoing research and development activities.

Tema: ENGINYERIA AERONÀUTICA

Tipus: Estudi

Càrrega de treball:

The workload will be adatped according to the student needs in trems of the ECTS trhe student is enrolled in (12, 14, 15, 24, 30 or 48 ECTS)

Requisits:

The successfull applicant for this offer should meet most of the following requirements: 1- High School or Bachelor degree (if you're coursing a Master) in electrical engineering, electronics, mechatronics or a related field in computer science. 2. Experience working with embedded computer systems such as Raspberry, Arduino or Jetson Nano 3. Proficiency in programming languages such as Python and C++, Linux driver development and experience with deep learning libraries such as TensorFlow, Keras, scikitLearn and PyTorch. 4. Experience with IoT applications and related web programming languages and communication protocols: HTML-CSS-JavaScript and Websockets, Socket-IO, MQTT LoraWan or NB-IoT among other wireless communication protocols 5. Technical skills with hardware systems and mechatronic, prefereably with flight controllers, ESCs, rotors, communication systems, and vision cameras. 6. Strong mathematical skills, including linear algebra, calculus, and probability theory. The following skills are not strictly necessary but are a plus for candidate consideration: 1. Knowledge of computer vision and image processing techniques. Particularly, artificial neural networks (ANNs), recurrent neural networks (R-NNs) and Convolutional Neural Networks (CNNs) 2. Experience with previous deep learning models implemented in real-world applications. 3. Strong passion for the application of deep learning and UAVs. 4. Ability to work effectively in a team environment, as well as independently. 5. Ability to effectively manage time and prioritize tasks to meet project deadlines. 6. Ability to work effectively under pressure and adapt to changing circumstances. 7. Strong attention to detail and commitment to producing high-quality work. 8. Ability to continuously learn and stay informed of the latest developments and trends in the field of UAVs and deep learning.

Temàtica ambiental: No

Confidencial (informatiu): No

Ambit de cooperació: No

Publicació: 07/02/2023

Caducitat: 03/02/2033