RISCBOT

1
RISCBOT

• Rajeev Sanyal, Sarosh Patel and Tarek Sobh
• RISC lab
• School of Engineering
• University of Bridgeport

2
RISCBOT

• Modular 802.11b enabled mobile autonomous
  robot.
• Controlled via the internet to navigate to
  desired rooms.
• Navigation based on a visual recognition
  algorithm.
• Riscbot coined from RISC lab(University of
  Bridgeport) and bot
• Online users receive real time video feedback

3
Telerobotics

• Controlling robotic devices from a distance
• Rich history
• a) Xavier
• b) Minerva
• c) Rhino

4
Riscbot hardware

• Onboard PC
• WLAN card
• NM6403 DSP board
• ATM 103 board
• 12V Panasonic SLA battery
• Cameras 2 USB, 1 NTSC
• Ultrasonic sensors
• 12V Pittman DC servomotor 2
• Inverter

5
Mechanical Design

• Initial design based on ProE software
• Frame built with aluminum extrusion rods
• Electronic circuitry mounted on acrylic sheet
• PC cabinet houses WLAN card and NM6403 DSP board
• Differential drive mechanism

6
Navigation module

• Wall clinging robot
• Waits for command from server
• Image processing program checks for doors
• DSP board checks for room ID
• PC acts as central decision maker

7
Image processing

• Edge detection employed to differentiate between
  wall and door
• Algorithm independent of varying light intensity
  and color
• Computationally fast
• Edge detection based on a Gaussian Laplacian
  filter with an order of 1.7

8
Steps - Locating Door Plate

• Thresholding
• Horizontal and vertical edge detection
• Hough transform
• Line segment extraction
• Candidate region extraction

9
Steps - Symbol recognition

• Extract characters with region growing method
• Compute profiles of extracted numeral
• Compare with stored library (trained)
• Assign Character

10
Web interface

• http//www.bridgeport.edu/sed/risc/html/proj/riscb
  ot/index.htm
• mobile robot connected to the Internet through an
  onboard WLAN 802.11b card.

11
Future Work

• 1.      Designing and building pan tilt units
  for the cameras.
• 2.      Implementing a dc dc (ATX power supply)
  converter circuit that will increase the power
  conversion efficiency and thereby the operational
  time for the robot.
• 3.      Permitting the robot to recharge itself
  by plugging into wall outlets.
• 4.      Mounting a manipulator on the mobile
  platform for implementing mobile manipulation
  tasks.