Registration

Proposals
Interims
Final Reports

Dates

Kickoff
Evaluations
Expo
STI
Wiki

School Map
Sponsors

Mail

Challenge
Technical Guide


Past Participant Survey

GUTS

Challenge Team Interim Report

Team Number: 012 School Name: Bosque School Area of Science: Computer Science Project Title: Robotic Intervention in Fire Victim Retrieval

 
Problem Definition:
Although fire-fighting technology has consistently reduced the number of structure fires, since 1977, people are still focused on developing better ways to save human lives in fires. As reported by the National Fire Protection Association, “3,570 civilian deaths occurred in 1999, a moderate decrease of 11.5%.” More than 80% of these deaths occur in the home. The smoke leads to many of the deaths, because it suffocates its victims.  The smoke also inhibits the firefighters efforts to find victims because it decreases visibility.  Even though firefighters can enter a burning building with oxygen tanks, they must wait until the smoke clears or they must have expensive thermal imaging cameras to successfully locate all victims. Firefighters and other personnel are especially vulnerable when searching for unconscious victims in hazardous waste disasters. This project focuses on programming and enhancing an autonomous robot that can quickly locate victims in a fire and alert firefighters with an audible signal.
 
Problem Solution:
Our project is based on the assumption that the robot is inexpensive to build and can withstand fairly high temperatures.  By equipping the robot with infrared sensors it is not inhibited by poor visibility due to low light conditions, but we must assume that the infrared radiation coming from smoke would not interfere with the wavelengths coming from the victim or the “fire”. The movement of the robot is controlled by proximity sensors, which must be programmed so that the robot will not miss any areas within the home. This will require white lines to be placed at each doorway and that pieces of "furniture" will always be adjacent to one another and to the wall. Initially a candle will simulate our "fire" so that sensors can distinguish the difference in electromagnetic radio waves being emitted from the “fire” and the victim. As we continue to develop our project we will research methods to overcome this problem.
 
Progress to date:
First Mrs. Ashmore obtained an example of the robot, built from an inexpensive kit, originally designed by the Robotics Club at NM Tech from the summer class taught at the Albuquerque Academy.  We met to learn how the robot was built and how to program it.  We started to test the robot to see how the sensors worked and we found that one of the connections to one of the proximity sensors was loose.  In addition we have been collecting background information and reports, mainly from the Internet, on uses of robots, robotic design, competitions, the dynamics of fighting fires and statistics on victims.  Recently we visited New Mexico Tech where Dr. Bruder and Dr. Weiderward discussed many options for our project with us.  Particularly useful is the website they showed us that includes the specific information on the infrared sensors describing the wavelengths emitted by humans and fires. We got a copy of the cross compiler for the robot to use at school.  So, right now we have a working robot, aside from a few modifications. The program is going to be based on a standard search map algorithm, along with sub blocks for the infrared sensors and the human recognition.
 
Expected Results:
When completed, we expect the robot will be able to search through a small model of a house and find a simulated “unconscious human.”  We expect that when we finish the program, it will have a complex search algorithm that allows the robot to search all portions of the room. The robot also will use the same algorithm to avoid fire but still get around to check for victims on the other side of the fire. After we add a sound generator similar to that on a smoke detector, the robot will stop and emit a sound when it identifies a victim. To identify a victim, the robot’s infrared sensors will have to be sensitive enough to differentiate between the electromagnetic energy produced by human skin and a candle flame.

Team Members

Team Mail

• Kenny E. Sutherland
• Will J. McMain
• Kai P. Mammoser
• Nathan L. Perkins

Sponsoring Teachers

• Dorothy I. Ashmore
• Debra M. Loftin

Project Advisor(s)