Deadline to register is Jun 30, 2019.
Teams can still edit your proposals during judging period.

Category: Sort By:
📁Machine Learning
👤Vadim Havryluk (entrepreneur (BestSoft))
📅Jun 30, 2019
Birds helped hunters to find prey. I want to create a neural network similar to the brain of a bird and train it to manage the drone to find free parking spaces. 15 minutes before the parking place, the driver launches a drone to search for parking, the drone finds and places a barrier piece on the parking space, and sends the location coordinates to the driver, the driver navigator marks it on the map.
details »
👀 1404   💬 0
📁High Performance Computing
👤Amarzish Qadeer (Riphah International University, Faisalabad)
📅Jul 01, 2019
In our project, we will be provided with a black box and we have to determine its response by using pseudo random binary sequence method which is one of the many methods used for system identification. The input which was being generated and output which is obtained from plant are then transferred to computer for computation in MATLAB.
System identification is the science of building mathematical models of dynamic systems from observed input-output data. It can be seen as the interface between the real world of applications and the mathematical world of control theory and model abstractions. As such, it is an ubiquitous necessity for successful applications.
In this case we are considering Black-box identification – determine model structure and
estimate parameters from data.
PRBS is generated in the FPGA and applied as input to the system under test through a
comparator. Comparator gives +1V if PRBS signal is high and -1V if PRBS signal is low.
The systems responds to this signal and gives an analog output. ADC samples the output of
the system at a frequency of 249kHz. ADC used was the embedded ADC on the Spartan 3E kit. The sampled output and the corresponding input is sent in a single packet to PC via the Ethernet port.
The C# application on PC gathers all the packets and save the input and output data in different files.
details »
👀 1458   💬 2
📁Internet of Things
👤Ebasa Girma (AAiT)
📅Jul 05, 2019
Imagine how life would be easier if we could build J.A.R.V.I.S from Iron man’s movie. Well, we can’t build the hologram yet but we are working with a smart home assistant that almost resemble Jarvis. Our aim here is to design home assistant that mimics Jarvis in different ways. Its assistance ranges from simply home automation's like turning on and off light, opening TVs and music using speech command, to maintaining room’s temperature and other home appliance control. Our home assistance can provide a home security system that detect and alarm users through their phone or PC. And with camera placed on front door our system can learn the Identified of who is standing on the door, showing their picture on the phone or PC and then later identify them and inform the user to late them in or not or [Even call the cop’s if the user maybe in danger]. We plan our system to be able to fully automate most home appliances and to have some level of intelligence that can assist the user in different status. i.e. It can assist the user in making appointments, remind them to take medication, suggesting an outfit to the user consulting weather for-casts. All in all, the aim of this project is to relieve the user from repetitive and time-consuming tasks at home as much as possible.
details »
👀 2368   💬 29
📁Digital Design
👤John Poole (Wynberg Girls High Schhoney (and University of Cape Town))
📅Jun 30, 2019
High value products such as honey have been adulterated on a large scale to defraud consumers. This project will use nuclear magnetic relaxometry to distinguish natural product from diluted product with added sugars from other sources. A SoC with integrated FPGA provides the ideal platform to build a low field NMR relaxometer forthis purpose (placeholder: proposal to be expanded)
details »
👀 1356   💬 0
📁Internet of Things
👤John Poole (Wynberg Girls High School (and University of Cape Town))
📅Jun 30, 2019
Barcodes represent an inexpensive method to reliably capture small amounts of data. Two dimensional symbologies such as QR Codes and PDF417 have become universal for puposes as varied as address and contents labels to providing quick access to information on the Internet via URLs and mediating financial transactions in retail settings. This project will use a SoC with integrated FPGA to accelerate the detection, image processing and decoding of a QR Code and PDF417 symbol. The parallel processing potential of the FPGA will provide significant acceleration over serial algorithms. OpenCL will be primary development language. (placeholder: proposal will be expanded)
details »
👀 1528   💬 0
📁Machine Learning
👤Liudmila Zenina (Bauman Moscow State Univercity)
📅Jun 30, 2019
The goal of the project is the creation of a control device that can provide dynamic control of manipulator with a complex structure.
details »
👀 52645   💬 0
📁Machine Learning
👤Asmaa Ibrahim (Zewail City of Science and Technology)
📅Jul 01, 2019
In this Project we aim to use the powerful capabilities of fpgas to build a brand new type of glasses that can provide users with a visual representation of what they think. Those who can benefit from an application like this can be graphic designers, crime witness or maybe a gamer. It can also provide a means of communication for those who cannot verbally communicate. This can have endless uses in multiple fields and can even be developed for further applications.
details »
👀 5714   💬 22
📁Machine Learning
👤Elisha Odemakinde (Obafemi Awolwo University, ile-ife)
📅Jul 08, 2019
In our world today there are tasks humans can't achieve or accomplish but can be achieved when ability is given to machines to do so. This project is all about having a collaborative interaction between humans and robots via hand gesture recognition (Machine learning) to accomplish a desirable task. This project is implemented In such a way that the machine learns co-currently from the human and gets used to such task for future purposes. We therefore believe this can be achieved using a computational device (The intel FPGA DE 10-Nano Kit).
details »
👀 1324   💬 0
📁Machine Learning
👤mohamed osama (None)
📅Jul 01, 2019
Our project is a firefighter robot that use AI and image processing to detect fire and deal with it, without human interaction
details »
👀 1350   💬 0
📁Machine Learning
👤Mohamed Osman (None)
📅Jul 01, 2019
The Aim of the project is to create a modular-based low-energy digital design to implement Convolutional Neural Networks CNN using Stochastic-Computing SC approach. SC has a big benefit over binary computing in its low-energy requirements. Hence, enabling using sustainable energy-supplies like Solar photo-voltaic cells with batteries.
details »
👀 1441   💬 0

1 ... 4 5