Barry Kline, Captain

1

Team 008-4

Barry Kline, Captain

Jessica Falcon
Eric Lamberti

Dr. Paul Oh

DrexelUniversity

Mechanical Engineering and Mechanics

3141 Chestnut Street

PhiladelphiaPA19104

Re: A More Practical Solution to Disaster Management

Dear Dr. Oh:

Enclosed is a copy of our findings regarding the practical application of robots in disaster recovery. Through extensive research and real world modeling, we consider ourselves sufficiently informed to publish our conclusions on the matter.

Sincerely,

Team 008-4.

Barry Kline, Captain
Jessica Falcon
Eric Lamberti

ENGR 103

Freshman Engineering Design

Final Project

Design Team No. / Team 008-4
Submitted to: / Dr. Paul Oh

AND THE

ENGR 103 PROJECT DESIGN FACULTY OF DREXELUNIVERSITY

ENTITLED: / A More Practical Solution to Disaster Management

TEAM MEMBERS (include email addresses)

Barry Kline,
Jessica Falcon,
Eric Lamberti

Submitted in partial fulfillment of the requirements for

Freshman Engineering Design, ENGR 103, Design Project

Submitted on ______

ABSTRACT

In disaster management, advances in technology have allowed for a more agile and accurate way of rescuing those in need. The goal of our project is to design and create a robot that safely locates and rescues casualties in the most efficient manner. Keeping the danger of disaster sites in mind, our robot would rescue the casualty in a timely and secure manner while subsequently keeping other humans in safety.

Everyday, robots are becoming more accepted in society. The need for robots during disaster management will become more demanding as disaster sites become more hazardous. Casualties are unfortunately left for long periods of time due to the obstacles that disasters leave behind. Humans are not always the best way to search and rescue other humans under such conditions because of the harm that they may cause to themselves. Robots, however, can provide a quicker and more precise way of rescuing those in need. Several robots may be designed for search and rescue, but our robot focuses on rescuing the casualty by attaching a rope from the robot to the body, and pulling it out of the danger zone.

The final design of our robot exceeded expectations, both in function and cost. Although structurally simple, our robot was capable of rescuing two casualties at one time, all under fifteen seconds. The simplicity of our robot left little room for error and kept costs at the lowest possible.

EXECUTIVE SUMMARY

The operation of concise disaster management has posed as a difficulty in the past years. With technology advancing every day, the endangerments left behind by war, natural disasters, and other circumstances have become more difficult to overcome. For our freshman design project, we chose to design a robot, on a smaller scale, that is capable of searching and rescuing casualties from a mock disaster site. This report describes and evaluates the range of possible design solutions that we, as a group, generated, as well as our final prototype, and why it was chosen. While researching the history of search and rescue robots, we found that there is a lack in range of types of robots that are available. We believe, however, that robots will soon be used on a wider and more frequent scale. Therefore, while designing our robot, we first designed three possible robots before we created our final and most successful prototype. The purpose in designing three robots was so that we would test each of them in the mock disaster site, and find which performed best with more accuracy, dependability, and speed. Evaluation of the alternative robots identified the best design for the robot to be a “Tow Robot.” This also proved to be a very good strategy, because when our robot was put to the test, it succeeded every threshold and object of the project. Our “Tow Robot,” consisted of few parts, mainly being the NXT brick, two motors, and two strings, which is how the casualties were brought up to safety. A final evaluation and challenge proved that our “Tow Robot” will not only be a successful solution to disaster management, but also a viable solution.

TABLE OF CONTENTS

Absract…………………………………………………………………………………. i.

Executive Summary……………………………………………………………………...ii.

Table of Contents……………………………………………………………………….. 5

Introduction……………………………………………………………………………... 6

Problem Background……………………………………………………………………. 6

Survey of Literature……………………………………………………………………... 7

Objective and Criteria…………………………………………………………………… 9

Constraints of the Solution……………………………………………………………… 10

The Solution…………………………………………………………………………… 11

Statement of Work…………………………………………………………………….. 11

The Results – The solution…………………………………………………………….. 12

Discussion and Conclusion…………………………………………………………….. 13

Recommendation for the future work………………………………………………….. 15
References………………………………………………………………………………. 16
Acknowledgments……………………………………………………………………… 17

A More Practical Solution to Disaster Management

INTRODUCTION

1.1 Problem Background

The challenge is to design a robot that will transport and extract casualties from a mock disaster site at the bottom of a ramp and bring them back to safety. In our case, the casualties will be six inch action figures. In a disaster site, casualties are immersed in debris and difficult to locate. Although finding and extracting casualties in the shortest time frame is pivotal, endangering other human life is counterproductive. Robots are perfectly suited for such environments and are capable of accomplishing tasks that humans should not be attempting. For example, robots could enter very hot, very dark, or other hazardous, impenetrable areas allowing humans to be kept out of harms way.

To create this robot, we used the Lego NXT set. In order to be effective, this robot is capable of identifying and extracting casualties from a disaster site. The Lego NXT set is equipped with all of the necessary parts that are needed to build the robot. These parts include electrical motors, sensors and structural pieces. Our purpose was, however, to build a robot with the least amount of parts, in order to reduce simulated cost and complexity of construction.

1.2 Survey of Literature

In order to build the most sufficient robot, we had to keep in mind the obstacles that disaster situations create. In most circumstances, humans are incapable of or not prepared for the rubble, immense fires, unstable structural conditions, and not to mention the scattered, hidden human causalities. Thus, robots must be built that are capable of withstanding these harsh conditions.

Rubble causes a great difficulty for evacuators because it obscures casualties from visual recognition and inhibits rapid removal from the disaster site. Rubble ranges in size, and therefore presents a variety of challenges. Smaller rubble, which blankets the disaster site, creates dust in the air limiting the visual range and accuracy of emergency technicians. Larger pieces of debris inhibit the speed at which rescuers can remove obstacles and obtain access to casualties beneath. Hence, rubble makes the process of searching and rescuing casualties a life-threatening endeavor. In addition to debris, fires can also inhibit the rescue operation.

The collapse of systems, such as buildings or communities, can cause fires due to electrical malfunctions and the destruction of utilities, such as gas and electrical lines. Fires are uniquely detrimental to the rescue effort due to their inaccessibility and level of danger. Fires are a progressive problem, which require immediate attention. If the attention is not met, it rapidly descends into chaos and does not allow for the extraction of casualties.

In most circumstances, debris, fires, and unstable structural conditions cause the identification and hence extraction of casualties to be nearly impossible. Robots however, can replace human rescuers because they are uniquely qualified to handle hazardous conditions.

Robots can be smaller and more agile then humans. They can easily navigate through labyrinths of debris that humans cannot attempt to brave. Their non-organic structure makes them mostly, or nearly immune to fire and extreme temperature. In addition, robots are equipped with precise sensors that allow them the capability of locating casualties and relaying position information. If robots such as these had been fully developed during the 9/11 catastrophe, the rescue operation would have proceeded more smoothly and would have had less of an endangerment on the lives of emergency technicians. When 9/11 occurred in 2001, three robots were tested. These robots were equipped with different sensors that were able to detect a variety of characteristics of casualties. “All of the robots have microphones to detect voices or other sounds of possible human presence within the ruins. Some of the robots carry thermal cameras that can detect body heat; others have cameras that search for colors distinctive from the gray dust that has blanketed the debris. Everything is gray and computers are really good at looking for color. A tiny dot of red, whether it is fabric or blood, can be easily identified and used to alert a rescue team” (National Geographic 1). Although these robots seem to be ideal for search and rescue, at the time, the robots were not sophisticated enough to embark on the terrain that 9/11 left. Only three of the eight robots that were brought to the disaster sight were capable to actually roaming Ground Zero (National Geographic 2). Although the robots did not succeed in finding any new casualties, however, they did open the doors to the acceptance of robots as mechanisms for search and rescue. As technology progresses, so will the sophistication of these types of robots. As previously mentioned, debris, fires, and other unstable structural conditions inhibit humans and animals to fully search sites. “In this disaster, officials have found that even dogs trained for search and rescue have not been able to climb across much of the debris, and the dust-laden air has diminished the dogs' keen sense of smell” (National Geographic 3). One day, robots will be able to take the place of humans and animals and allow for more accurate searches that will then lead to more precise information conveyed to rescue teams.

1.3 The Objective and Criteria

There are two options that our robot can meet in order to be considered as completing the task of searching and rescuing casualties from a disaster site: the threshold and the objective. In order to create a more realistic robot, our group built a robot with the goal of completing the objective. The difference between the threshold and the objective is the level of difficulty and the risk taken of putting our robot in a more difficult obstacle. For example, the ideal robot that meets the objective criteria would consist of less then 100 parts that are only NXT, less then ten programming steps, be able to go down a ramp in thirty seconds, retrieve six inch figures, and return them back to safety up a 20 degree ramp. On the other hand, the criteria for the threshold, is slightly simpler to accomplish. In order to meet the threshold criteria, the robot must be able to retrieve a ball at the bottom of a five degree ramp, in under 90 seconds, and consist of 200 parts in which ten percent can be non-NXT, and lastly, in 20 programming steps.

Table 1

1.4 Constraints on the Solution

We originally set out to create a robot that consisted of the least amount of parts possible. However, the requirement that only 10% of our parts could be non-NXT created a problem for us. Our plan was to extract three casualties simultaneously, using three separate strings. These three strings required at least thirty NXT parts to abide by the necessary part ratio. In order to comply with restrictions, unnecessary, nonfunctional NXT parts had to be added to our robot. This is a restraint because in addition to maintain our NXT to non- NXT ratio, it is also a goal to minimize overall part count. However, we managed to keep our part list to the smallest number possible, of thirty. We could have, for instance, used only one string or two, but this would only allow the robot to rescue one casualty at a time, rather then three. We wanted to keep simplicity in mind, but ultimately, our main goal was to rescue as many casualties as possible in the shortest amount of time. The follow table is a list of our final parts list.

Quantity / Part / Number
1 / Long Axle #12 / 3708
1 / NXT Brain
4 / Technic Axle Pins / 3749
2 / Technic Brick 1x16 / 3703
12 / Technic Pin Long / 6558
2 / Technic Angle Connector / 32034
2 / NXT motors
2 / NXT Wires
5 / Technic Bush / 3713

Another constraint to the success of our robot was the strings. During our practice trial, we noticed that the strings were too thick and were in fact getting tangled together when they were winding up. Although the robot still worked properly, the speed at which the robot rescued was not up to our par. Fortunately, we were able to recognize this problem before the actual critical design review, and were able to fix it. Instead of our original string, we chose to use fishing wire that is equally as strong, yet thinner, which allows the robot to wind up quicker without tangling, thus bringing the casualties back to safety more rapidly.

THE SOLUTION

1.1 Statement of Work

In order to design the most optimal robot for our project, we went through a series of evaluations of other robots that were created before finalizing our idea. We first had to evaluate the situation; we needed to design a robot that was capable of extracting first a ball, then an action figure, from the bottom of two ramps, on being at 5 degrees, and the other being at 20. We also had time limitations and part count limitations. Keeping all of the criteria in mind, we contemplated between two types of robots, a stationary robot, and a mobile robot and created a table to asses the risks of each.

parts / ease of program / reliability / feasibility / speed / only NXT parts / Total
stationary robot / 3 / 3 / 1 / 3 / 2 / 1 / 13
mobile robot / 1 / 1 / 2 / 3 / 2 / 2 / 11
1-very bad
2-exceptable
3-excellent

As shown, we found that the best design for our robot would be a stationary one. This then allowed us to narrow down our options as to how we were going to extract the casualties from the disaster site. After deciding that our robot would be stationary, we came up with two ways that we could possibly extract the casualties. First, we could have a long, rake type end that would drag up the casualties up. However, we found this idea to be primitive and not accurate because of the possibility of leaving a causality behind, having one fall out of the rake, etc. Our second solution, which also became our final solution, was to tie two strings from the robot, to the casualties, thus extracting them by pulling this up. Due to our situation being only a mock disaster site, this solution was plausible because we had 60 seconds to prep our disaster site, which allowed us to tie the strings onto the casualty. Ina real world disaster however, a second and smaller robot would be needed to locate, tie, and connect the casualty to the main robot.

1.2 Results – The solution

Our solution to a more practical way of disaster management is our tow robot. Our finalized robot consists of only thirty-three NXT parts, four steps, and has the capability of extracting two casualties at a time, rather than one. By keeping our robot as simplistic as possible, we have created the most ideal robot for this simulated challenge. After analyzing the risk assessments and contemplating alternative robots, we feel that our robot is best qualified to undergo and extract casualties from the disaster site. In true life, however, our robot would need to be paired with another robot that would locate the casualties and attach the harness. In our case, we are allowed to affix the harness onto the figure before beginning. This reduces the challenge, but increases the chance of completion. On the other hand, our robot does have practical implication for searching and rescuing.

DISCUSSION AND CONCLUSION

Although error could have easily occurred during the challenge, our robot was designed with keeping the risk of error very low, hence, leading to our robot successfully accomplishing the challenge. Our risk assessment of our original robots also allowed us to evaluate the importance of certain aspects of the robot, such as stability, accuracy and speed. Without having designed these prototypes, and testing them, our final product would not have been as successful as it was. Our robot validates itself with its performance in the challenge. The following picture shows our robots extracting two casualties.

The cost of our project was simply the cost of an NXT Lego set, which is $250 dollars. However, our robot only consisted of about 7.4% of the kit, therefore, the price is in fact less then $250 dollars. The benefit our doing our project on this smaller scale was that we were able to recreate our robot several times before deciding on our final prototype. This allowed us to test out different options and design the final robot that would cost the least, however work the best, and benefit the most. The simplicity of our robot also allows the robot to be easily controlled and maintained during disaster management. The less parts and steps allow for better use of the robot. Another key aspect of our tow robot is its availability. The few assembly steps allow the cost to be minimal, which then extends the robots use to not only commercial nations, but to poorer nations as well. Our robot not only follows the criteria of the job, to search and rescue, but also benefits its user with its lost cost and low error rate.