Robotic Project
Course duration: 16h
During this course we will construct robot that will be able to discover terrain, build its map, and go from one specified point to another – that involves solving quite a set of problems for computer vision, decision making, control etc. Final test of the robot will be performed in the KPI park.
We will discuss following topics:
- Construction – small practice on handling aluminium, basic soldering, attaching servos
- Electronics – how to make it work with minimal set of components
- Arduino programming – interaction with sensors, servos, smartphones, PC
- Application of control theory under restrictions of practically available hardware
- Computer vision – various problems, including
- stereovision
- camera and objects motion detection
- object recognition
- terrain reconstruction
- Decision making under uncertainty:
- pathfinding with possible object recognition errors
- advanced obstacles avoiding, including humans movement prediction
- world model construction and application for decision making
As the result of the class, fully functional robot will be constructed. Of course it will use some software that will be written before the class, and part of the hardware also will be pre-assembled, but all important problems will be discussed during the class, all source code and blueprints will be available for class members, and all algorithms and engineering solutions will be developed strictly during summer 2013, so it's not just a presentation of some long-term sophisticated project.
The goal of this class is to make listeners capable of creating mechanically and intellectually advanced robots from scratch without enormous efforts.
Tutor
Mr. Dmitry Dziuba
Country: Ukraine
Place of employment: Institute of Mathematical Machines and Systems, Kyiv, Ukraine
Academic degrees: M.Sc. in Applied Physics, NTUU "KPI", 2007
Current position: Robotics engineer, Researcher at Institute of Mathematical Machines and Systems
Email: This email address is being protected from spambots. You need JavaScript enabled to view it.
Phone: +38 063 5749723