Projects

We propose LBGP, a follow ahead method that uses both reinforcement learning and point based navigation. We address the limitations of classical methods and end-to-end approaches by combining Deep RL and classical motion planner

• Source Code on GitHub
• Paper

Designed a relational graph learning approach for robotic crowd navigation using model-based deep reinforcement learning that plans actions by looking into the future. Tools: Pytorch, ROS, Stage, Python, OpenCV

• Source Code on GitHub
• Paper

Built an occupancy grid map during online and at the same time use a neural network (Based on ResNet34 and YOLOv2) to detect, locate and say the target classes around the robot. Tools: Pytorch, ROS, Stage, Python, OpenCV

• Source Code on GitHub
• Paper

Designed a fast human pose and orientation estimator based on the OpenPose structure

Crated a dataset from youtube videos to train an audio-visal lipreading system

Presented a human model for an autonomous mobile robot that follows a walking user while staying ahead of them. Used multi-modal person detection and a human-motion model that considers obstacles to predict the future path of the user.

• Source Code on GitHub
• Paper

Use a combination of Deep RL and transfer learning to train an agent to do maximal map coverage in an unknown map.

• Paper

Desined a Attention neural model to describe navigation environemnt featured for visually impaired people.

Re-identifying people using 3D point cloud data.

• Source Code on GitHub

presented a classical method to solve collecting food from unknown sources.

Controll the mouse curser by tracking the user's hand using a RGB camera

Controlled the mouse pointer by detecting user's gaze direction

Created a pear to pear software for sharing files

Designed a multiplayer first person shooter game in Unity3d with the capibility of 100 concurrent players.

Created a multiplayer game using DirectX

Used OpenCL to speed up image processing algorithms

Implimented a book shpping website using Django and MVC pattern.

Made a simple game consule with two games using Atmega 32, touch pad, AVR and GLCD

Used Minmax algorithm to solve the PacMan and Hex game.

Designed a course management system using J2EE, Hibernate, HTML and JS.

Created a clinet server system

Designed a 2D snake game with AI. This project ranked first for it's AI (artificial inteligence) in Advanced programming class.

Built a snake robot using servo motor and ATMEGA32