Hello, I'm Ayoub HSAINE

State engineer in robotics and cobotics.

Preparatory Classes + Robotics,Cobotics and AI Engineering program.

CERTIFICATE:2022 Zhejiang University SDG Global Summer School, Smart City

This program was organized in two weeks, and it divides into two main sections: Intelligent Transportation & Smart Mobility Intelligent Internet of Things & Intelligent Service Robots

Baccaleureat year in physics sciences

EL JADIDA, Morocco

ON site Internship Trainee : Generation Robots Company

The internship focuses on developing and testing algorithms for the GR100 robot to verify the integrity of outdoor fences. During this internship, I undertook the following tasks:
• Conducted a state-of-the-art review.
• Tested and selected perception modalities for extracting the fence from the environment.
• Developed a deep learning model to assess the regularity of the fence pattern.
• Generated synthetic data using Blender, OpenCV, and computer vision techniques.
• Compressed and optimized the model.
• Converted the model to ensure compatibility with NVIDIA embedded systems.
• Deployed all mechanisms in Docker.
• Transferred the developed model to the GR100 robot.
• Conducted real-time inference with the robot to detect holes in the wire mesh.
• Collected data by the robot and retrieved it using ROS topics to enhance the model's accuracy, retrained the model, and relaunched the inference.
• Learned and applied ROS.

ON Site Research Internship : UVSQ Université de Versailles Saint-Quentin-en-Yvelines, VERSAILLES SYSTEMS ENGINEERING LABORATORY (LISV)

• Week 1: Comprehending articles related to EMG signals, feature extraction, and stages of EMG signal filtering.
• Week 2: Interpreting existing data containing recorded EMG signals from triceps and biceps muscles, gaining insights into signal characteristics.
• Week 3: Gathering data from 10 volunteers performing flexion-extension motions for triceps and biceps muscles, including with and without a 2 kg load, and MVC for both muscle groups.
• Week 4: Selecting complete and noise-free data recordings, discarding incomplete or unreliable entries, and applying filtering techniques and feature extraction methods to analyze the data.
• Week 5: Analyzing the data to calculate the elbow angle, comparing it with angles obtained through other measurement methods to assess correlation, and acquiring knowledge about interfacing with EMG signals, the Vicon Camera system, marker placement, and 3D Vicon software usage.

Remote Internship: Hassan II University Hospital, CHU Fez

• Purpose: Develop a mobile application to solve the problem of prescription validation, where certain drugs require pharmacist validation before nurses administer them due to a large number of patients.
• Goal: Develop a prototype application and create a functional database linking doctors, nurses, and pharmacists.
• Tools: Utilize Wamp Server, MySQL, PHPMyAdmin, PHP, and Postman for development purposes. Diagrams will also be employed to aid in the development process.
• Skills Required: Critical thinking, database administration, database design, proficiency in Figma software, and effective teamwork.

ON Site Initiation Internship : FLOQUET MONOPOLE

• Learning Experience: Acquired new skills and knowledge through hands-on experience.
• Professional Integration: Immersed myself in the professional world and learned effective team integration.
• Hands-on Experience: Worked on the production line of machining discs across various operands.
• Application of Theoretical Knowledge: Applied theoretical knowledge in practical settings, observing technologies like the Metrology Service and painting processes.
• Rewarding Experience: Developed analytical and problem-solving skills through analyzing different aspects of the production line.
• Insights into the Industry: Gained valuable insights into the industry by understanding various production processes and technologies.

Comparative Study: Path Planning Using Search Algorithms: DFS, BFS, and A*.

A comparative study was conducted using Python and Pmaze to evaluate the performance of DFS, BFS, and A* search algorithms in path planning for a maze. The main objective of the project was to compare the efficiency of these algorithms in terms of path cost and algorithmic complexity. No physical robot was used in this study as it focused solely on the analysis of search algorithms. This study contributes to the advancement of knowledge in the field of path planning by exploring different approaches for mazes using Python and Pmaze.

PepperChat: Integrating ChatGPT for Intelligent Conversations with Pepper Robot.

This project aims to integrate a GPT-3-based Chatbot into a Pepper robot, harnessing the capabilities of large language models and the diverse functionalities of the robot. The Chatbot will be equipped with the Watson Speech API by IBM, enabling seamless speech-to-text conversion. The primary goal of this project is to facilitate realistic and engaging conversations between users and the robot, while also enhancing the naturalness of various tasks. By leveraging advanced language processing and robotics, this project aims to deliver a memorable and immersive experience with the robot, bridging the gap between humans and intelligent machines.

SPIDER ROBOT

Spider Robot: is a robot belonging to the category of biorobotics, it is a robot inspired by nature, it contains four legs that allow it to move in the different sides(left,right,forward and back), these movements are controlled by an application through a bluetooth.

OpenCV and OCR for Parking Management.

This project aims to develop an automated system for detecting and recognizing car license plates. By employing image processing and character recognition techniques, the system will be able to extract and identify license plate numbers from images or videos. This project has the potential to enhance road safety and traffic management through advanced automation.

Safety deposit box using LabVIEW MyRIO.

The objective of this project is to develop a safety deposit box using LabVIEW MyRIO, which enables users to store valuable items and secure them with a four-digit PIN code. The major challenge of this project involves designing a finite state machine to manage the different states of the safety deposit box, including "open," "locking in progress," "locked," and "opening in progress." To facilitate user interaction, the system features a keyboard and LCD display for entering the PIN code and displaying instructions and alerts in case of suspicious activities. Additionally, the safety deposit box controller incorporates a servo for locking and unlocking the box and a mode for resetting the code for a new client or in case the user forgets the code. The system is programmed using LabVIEW to monitor and control all components and ensure that the finite state machine detects only unique key presses. Furthermore, an LED indicator can be used to signify the locking status of the safety deposit box.

Detection of COVID-19 using classification algorithms.

This project focuses on the use of machine learning algorithms for binary classification of COVID-19 data. Several algorithms were applied, including k-Nearest Neighbors (KNN), Linear Support Vector Machine (SVM), Gaussian SVM, and Perceptron. The performance of these algorithms was evaluated on a dataset of COVID-19 cases and compared to determine the most effective method for classifying positive and negative cases. The results of this study demonstrate the potential of these machine learning techniques for accurately identifying COVID-19 cases and could be useful in the development of diagnostic tools and tracking the spread of the disease.

ROVERBOT

Roverbot is a mechanum wheels robotic vehicle that is designed to move in all directions, thanks to its advanced algorithms that plan the most efficient routes to reach its destination. Its unique feature is its ability to move in tight spaces with precision, thanks to its mechanum wheels. Additionally, it is equipped with a navigation system to locate its position and follow a predetermined route, as well as an obstacle detection system to avoid collisions. Our goal is to create a robot that can move effectively through a maze using advanced autonomous navigation techniques, while being able to detect and avoid obstacles. We hope that Roverbot will be able to perform useful tasks such as surveillance, mapping, search and rescue, and delivery. Our algorithm for this mobile robot project consists of using a matrix to represent the maze, where each cell is numbered based on its distance to the destination. The destination itself is marked with a "0", the adjacent cells are marked with a "1", the cells next to the cells marked with "1" are marked with "2" and so on, until all the cells are filled. Then, the robot is placed in a random cell and uses this matrix to plan the shortest path to reach the destination while avoiding obstacles, which are defined as cells with high values in the matrix.

Smart Home

Introducing our latest Smart Home project using Arduino technology! This project allows you to control and monitor various aspects of your home remotely using a simple and user-friendly interface. The system includes a central hub that connects to various sensors and actuators placed around the home. The sensors can detect changes in temperature, humidity, motion, and more. The actuators, such as LED lights and motors, can be used to control appliances and devices. Some examples of how this system can be used include: Automatically turning on the lights when you enter a room. Monitoring the temperature in your home and adjusting the thermostat accordingly. Getting a notification when motion is detected in a specific room. Our smart home project demonstrates the power of Arduino and how it can be used to create innovative and useful solutions for everyday life. It's a great way to learn about programming and electronics while also having fun and make something practical. We are excited to see how this technology will continue to shape the future of home automation.

Real Time Object Detection using MATLAB

Our project is an application developed using MATLAB, in this project we tried to integrate image processing techniques and technologies such as: Preprocessing, Segmentation, extraction and real-time object detection, The project consists in taking real-time photos of each object passes in front of the camera, the first photo of each object will be qualified as reference and the calculation of their intensity , then with each movement of this object the application calculates the intensity of the image it will find that it is larger than the reference image which implies the presence of a movement, these movements are presented in a two-dimensional graphed , then each SIGN in the graph represents a movement of this object.

Controlling a Differential-Drive Mobile Robot Using Fuzzy Logic.

The objective of this project is to create a fuzzy controller for a differential drive robot. The aim is to familiarize ourselves with fuzzy logic and to explore the capabilities of Simulink. The project involves modeling the kinematics of the robot in a block and linking it to a block of fuzzy logic. The robot has two speeds, which are both the outputs of our fuzzy controller and the inputs for the robot model. Our system is looped : we initially provide the positions that the robot should reach, these positions enter a subtract block along with the estimated model measurements to provide errors that are used as inputs for the fuzzy controller. According to the rules and inference used, the controller provides the necessary speeds for the robot to navigate towards the destination. The robot will reach its destination when the errors cancel each other out.

Dictionary AttackDictionary Attack

In the part of the projects of parallel computing module and application reparties, the professor proposed several projects for parallelised, we chose as project Attack by dictionary, the main objective is to familiarize with OPEN MP and to master parallelism tools.

Enterprise Network

An enterprise network is the backbone for facilitating an organization’s communications and connecting computers and devices throughout departments. An enterprise network environment is usually configured to facilitate access to data. Enterprise networking refers to the physical, virtual and logical design of a network, and how the various software, hardware and protocols work together to transmit data. When it comes to enterprise networking, every organization has different needs. For more details visit my page.

System Hotel Management (QT C++)

The project consists in creating a complete Qt application, a graphical interface that manages customer booking. Our application is contain four Buttons each one plays precise role by Example our customer will check the available rooms then he will choose, once he has chosen he goes on to the next step, that is the recording of the necessary information in room booking, once he records this information, our database will receive this data and when he leaves he must check out.

VHDL design and simulation of an electronic combination lock

As part of the final grade of embedded system module and connected object, our professor proposed a project in VHDL, the main objective of this project is to model in VHDL language a lock circuit with electronic combination, and then validate the description developed by performing a simulation on the Modelsim software. This project is an opportunity to review the approaches of descriptions provided by the VHDL language, as well as to develop a good mastery of the different algorithmic structures that are the basis of these approaches.

Skills

• Programming Languages: C/C++, Java, Python, and MATLAB
• C++ GUI with QT
• Operations Research
• Fuzzy Logic
• Ubuntu OS( Networking, Security(basics of ethical hacking)
• Arduino, Raspberry Pi, MyRio
• VHDL (modelsim software)
• LabVIEW
• Control Systems: Ladder, Grafcet, GEMMA
• Sensor Technology
• ML, DL and OpenCV
• Signal and Image Processing
• Localization and Navigation Methodology (Kalman Filter, SLAM)
• Robotics System Design and Modeling
• Computer Vision
• Android Studio
• CAO:3D Experience
• Blender 3D
• Drone Design and Modeling
• Knowledge in ROS

Management

• Project Management
• Strategic Management
• Innovation Management

Programming language Software

• Matlab(Image & Signal Processing, modelling robotics system)
• Ellipse for java
• Spyder python(ML, Opencv)
• Anaconda(ML,DL)
• Visaul studio(Html,css,js) for web development
• QT Creator(C++, IHM)
• Maxima for Computer Algebra System
• Altova,MongoDB(Data Base, XML NoSQL)

3D printing Software

• Simplify3D

Design Software CAO

• 3Dexperience

Animation & simulation software

• Blender
• Gazebo
• Proteus(cirucit simulator, sensor simulation)

Languages

• Arabic
• French
• English
• Spanish

Exploring Maze Navigation: A Comparative Study of DFS, BFS, and A* Search Algorithms.

A comparative study was conducted using Python and Pmaze to evaluate the performance of DFS, BFS, and A* search algorithms in path planning for a maze. The main objective of the project was to compare the efficiency of these algorithms in terms of path cost and algorithmic complexity. No physical robot was used in this study as it focused solely on the analysis of search algorithms. This study contributes to the advancement of knowledge in the field of path planning by exploring different approaches for mazes using Python and Pmaze.

OpenAI Stack Hackathon

In the recent OpenAI Stack Hackathon organized by lablab.ai, I had the privilege to collaborate with a talented team from February 24th to March 4th. Together, we crafted 'HealthyAI,' a cross-platform application powered by OpenAI's GPT-3 model. This innovative tool provides precise medical diagnoses based on patient symptoms and history, offering practical home remedies as well. Grateful for the expertise of my colleagues, Mohammed Machrouh and Tanishq Kr. Kaushal, our project utilized React for frontend, Django for backend, GPT-3 for AI, and Redis as our database. Our journey continues as we refine the app, adding features like a health dashboard. Explore our project at and the source code at . Thanks to lablab.ai, this experience fuels my enthusiasm for future hackathons and collaborative tech ventures.

The 5th edition of the national robotics competition in Bengrir, organized by Emines Technology Club at UM6P - Mohammed VI Polytechnic University.

This year, the challenge was to build a six-legged SPIDER robot, with each leg being powered by 2 servo motors. The robot had to be capable of moving within a defined circuit.

AfricanRobotics Brains22

We participated in the first 'Road Assistant Robot' challenge. The objective of this challenge was to assemble the robot from scratch and program it to overcome obstacles and reach the end of the maze. Here is an overview of the activity: Initially, the robots were placed in their stations. Upon receiving an RFID signal (each RFID signal was assigned to a specific destination), the robot had to react and identify the type of fire present and decide which type of fire to respond to, in order to reach its destination. The robot had to navigate through static and dynamic obstacles, as well as a maze, to reach its destination. The challenge consisted of three elimination phases, and fortunately, my team qualified for the second phase, securing the fifth position. Alhamdulillah

Alten AMSA5

My participation in the semi-final of the 5th edition of the national competition ALTEN MOROCCO SOFTWARE AWARDS (AMSA) under the theme 'Artificial Intelligence Applied to Eco-Mobility,' organized by ALTEN Morocco in partnership with the Euro-Mediterranean University of Fès as a hackathon, took place from December 8, 2020, at 8:30 AM to December 9, 2020, at 8:30 PM