Intelligent Embedded Systems (Robotics)

50 %
50 %
Information about Intelligent Embedded Systems (Robotics)
Technology

Published on December 1, 2008

Author: foweja

Source: slideshare.net

Final Project By Adeyemi Fowe Intelligent Embedded Systems (Spring 2008) Dr Cang Ye Applied Science Department (Engineering Science and Systems) University of Arkansas at Little rock.

Project Title Dictionary said it all; Exploration means: A careful systematic search …. Autonomous Exploration Robot

The Concept; A Sketch

The Concept; Object Search This project idea is to achieve path navigation and object search within an environment. We utilize embedded systems programming that would enable a robot anonymously search for objects in an environment. There are a lot of desired environment that the human body and sensory organs are Incapacitated to reach. We believe intelligent embedded systems is the answer to harnessing this Type of Environments. Technologies involved include; Machine Learning/Computer vision, Path navigation, Obstacle avoidance, Embedded systems programming, Wireless TCP/IP Communication.

Robot Design

Systems Design

Systems Design The objective of the project can be achieved in so many ways. The design presented is one of the simplest ways to achieve object identification and search. It would make a good quick test system for robotic exploration work. The robot is equipped with two sensors (infrared and vision). The robots autonomous intelligence include; Locomotion & control, obstacle avoidance, search algorithms, self organization and re-initializations. The remote intelligence is for advanced computation, machine learning, Advanced computation, and larger storage space. The wireless cloud can be a Campus Area Wifi network, or even extended over a WAN (Wide Area Networks) like internet or intranet.

The robot is equipped with two sensors (infrared and vision).

The robots autonomous intelligence include; Locomotion & control,

obstacle avoidance, search algorithms, self organization and re-initializations.

The remote intelligence is for advanced computation, machine learning,

Advanced computation, and larger storage space.

The wireless cloud can be a Campus Area Wifi network, or even extended over a WAN (Wide Area Networks) like internet or intranet.

Application Areas This system is applicable to diverse exploration projects. E,g; Geology: A robotic Geologist for meteorite search etc. Space Exploration: Robot exploration to Moon, Mars and beyond. If smaller hardware, then its useful for search into small holes and corners that Cannot size human body. Also useful monitoring and Capturing activities in Severely cold regions. It can also be used for surveillance and security. Journalism in chaotic regions or war zones.

Systems & Devices iRobot Create http://www.irobot.com/ For Locomotion, Control & Infrared Sensors.

Gumstix Devices Gumstix Verdex XM4-BT Robostix NetwifimicroSD Wifi/Bluetooth Antenna http://www.gumstix.com/

Connectors S13-SERIAL-INT-CONN Gender Changer USB to Serial Converter

IP Wireless Camera •  802.11g Wireless Connectivity •  Compression Codec - MJPEG and MPEG-4 •  FTP Upload support •  Smooth Video Recording of Up to 30 FPS TrendNet TV-IP201W

Demo Implementation The author has implemented a simplest demonstration of this concept, using Available hardware devices and simple software programming. Vision Sensor; Trendnet Camera Capture periodic images in JPEG format every 1second. Environment; A small rectangular lab room. Machine Learning; C++ programming using TorchVision for RGB Pixel Color Analysis to detect RED Objects. Gumstix Verdex; Uboot version 1.2.0, Journalized File system(verdex.jffs2) Perl NET::FTP programming for automations. C++ programming for Player’s( Robots driver) client interface.

The author has implemented a simplest demonstration of this concept, using

Available hardware devices and simple software programming.

Vision Sensor; Trendnet Camera

Capture periodic images in JPEG format every 1second.

Environment; A small rectangular lab room.

Machine Learning; C++ programming using TorchVision for RGB Pixel Color Analysis

to detect RED Objects.

Gumstix Verdex; Uboot version 1.2.0, Journalized File system(verdex.jffs2)

Perl NET::FTP programming for automations.

C++ programming for Player’s( Robots driver) client interface.

Implementation Design

Implementation Steps Device setup: Connect all devices and see that all networks and interfaces are working. 3) Build Software Environment : Loading boot/Flash image, creating C++ programming env. 5) C++ programming: Image Detection using pixel RGB Color summation. 6) Programming: Perl programming for real-time automations. 7) Software compilation, building and deployment. 8) Systems Testing: Learn environment ->Learn Object type -> Search for object 9) Testing and Demo 10) Systems Presentation &Lunch!

Device setup: Connect all devices and see that all networks and interfaces are working.

3) Build Software Environment : Loading boot/Flash image, creating C++ programming env.

5) C++ programming: Image Detection using pixel RGB Color summation.

6) Programming: Perl programming for real-time automations.

7) Software compilation, building and deployment.

8) Systems Testing: Learn environment ->Learn Object type -> Search for object

9) Testing and Demo

10) Systems Presentation &Lunch!

System Setup

System Setup

Demo Concept

Systems Function

Pixel Analysis Source Code For Red Color Detection

Perl Program For Rea-time Automation

Conclusion This presentation has shown the design of a prototype autonomous exploration robot. The project is utilizing the power of small embedded devices like Gumstix, Which allow for remote and autonomous intelligence for the robot. A demo of how a robot can search and detect a red colored object is presented. Extensions: This project can be extended into Swamp Robotics. The power of wireless and small hardware can be extended into mission critical Projects that utilizes teams of swamp robots. Either flying or crawling. All the swamp bots can be controlled from the same image server for synergy and advanced computations. This project can also be extended into Ubicomp (Ubiquitous Computing).

Reference http://images.tomshardware.com/2007/05/15/irobot_create.jpg http://www.acroname.com/robotics/parts/S13-SERIAL-INT-CONN.jpg http://roboticsprimer.sourceforge.net/workbook/Main_Page http://www.si87.com/images/gender-bender.jpg http://www.eham.net/data/reviews/images/3704.jpeg

Add a comment

Related presentations

Related pages

Model-based programming of intelligent embedded systems ...

Model-Based Programming of Intelligent Embedded Systems and Robotic Space Explorers BRIAN C. WILLIAMS, MEMBER, IEEE, MICHEL D. INGHAM, SEUNG H. CHUNG, AND
Read more

Multi-Robot Systems - USC Robotics Research Lab

Center for Robotics and Embedded Systems ... Naval Research Laboratory Multi-Robot Systems ... for Robotics and Intelligent Systems ...
Read more

MSc Intelligent Systems and Robotics :: University of Essex

Our MSc Intelligent Systems and Robotics is delivered by our ... intelligent environments, robotics, ... software for embedded systems and robotics.
Read more

Robotics Education using Embedded Systems and Simulations

Robotics Education using Embedded Systems and Simulations Thomas Bräunl The University of Western Australia and Technical University München
Read more

Robotic Embedded Systems Laboratory - USC RESL

The Robotic Embedded Systems Laboratory (RESL) is part of the Robotics and Autonomous Systems Center (RASC) at USC. The research mission of RESL is to ...
Read more

Intelligent Systems, Robotics and Cybernetics

Our unmatched facilities enable leading edge robotic and intelligent systems research ... communications systems, embedded ... & Robotics? Contact us ...
Read more

Robotics and Embedded Systems | School of Electrical and ...

Overview. Robots are intelligent electrical-mechanical devices that can sense, think, actuate and interact with humans. Their motion behavior can be ...
Read more

Master's programme in Intelligent Embedded Systems, 120 ...

Master's programme in Intelligent Embedded Systems, 120 credits. HT17, Västerås, 100%. Application code. MDH-44951
Read more

CS 599 Intelligent Embedded Systems - HOMEPAGE

CS 599: Intelligent Embedded Systems (soon to become CS 546) Spring 2002: W 3:30 - 6:20 p.m. in THH 114 (note new room) Note: This webpage is outdated.
Read more