Design team members: Jackie Leng & Ben Olsen
Supervisor: Professor David Wang
Background
For the past nine years, the Association for Unmanned Vehicle Systems (AUVS) has organized a series of collegiate level robotics competitions called the International Aerial Robotics Competition. This competition focuses on the development of autonomous robots (mainly aerial) to navigate through a hazardous disaster area to accomplish the mission of human search and rescue. Tasks may include finding the location of objects (i.e. fire, water, humans), correctly identifying the objects, and determining whether the animatronic humans are dead or alive (where survivors emit sounds and have simulated hand waving motions). Points are awarded according to the importance and accuracy of each discovery within a one hour period.
The year 2000 competition will allow the use of autonomous ground robots that work in collaboration with the aerial robots. The ground robots can provide assistance to air vehicles by helping in locations where it may not be easy or convenient for the air vehicle to travel to, or by searching and navigating the site to observe and collect information about conditions at ground level. Suggested tasks that may be carried out by ground robots include putting out small fires, tracing out a line to lead rescuers to the survivors by way of a safe path through and then back out of the wreckage, or rescuing a survivor itself.
The Waterloo Aerial Robotics Group (WARG) is a team of Engineering students from the University of Waterloo who have designed and constructed a helicopter that has participated in this competition with many flourishing results. WARG has successfully made it through the first two qualifiers, allowing them to compete in the year 2000 competition. This event is scheduled to take place on June 27 through June 30, 2000.
Project description
The goal of this workshop is to create a design for an autonomous ground support robot that will supplement the University of Waterloo's entry to the Aerial Robotics Competition for the year 2000. The design will primarily focus on searching for simulated human survivors by recognizing waving arms and then signaling the location to another member of the autonomous robot force.
A diagram of our existing design can be found at: Robot Diagram. It includes a sonar, contact sensors, heat detectors, a microphone, tank tracks and a 16 bit microprocessor made by Motorola.
Design methodology
The design of the robot will follow a problem decomposition methodology similar to those used in computer algorithm design. The advantage of using this type of methodology is that it produces modularized subsystems where interdependencies between modules can be strategically controlled. Not only does modularization allow the reuse of robot subsystems on other WARG robots, it also reduces the complexity of the design and thus reduces errors and helps reduce the cascading of failures.
This methodology also provides a solid frame work for setting goals and measuring goal completion. Unfortunately, since the methodology takes a top down hierarchical approach, any modification of the main goals of the project will invalidate many of the sub goals that are derived from it. So far, this has not been necessary in this workshop.
The process began by choosing a workshop objective. Next, a general strategy was derived from it. This general strategy was made more specific by analyzing various problems and issues that would arise in trying to meet the original general strategy. Solutions to these problems were proposed in terms of more specific strategies. This process was continued until a hardware or software design to solve the problem was suggested. At this point, criteria and constraints were created for each of the sub problems and various solutions generated. Additional criteria were added where many options still remained after the initial analysis.
This type of methodology would not be beneficial for others who may lack the experience necessary to evaluate the strategies in terms of whether they are possible or not with current technology.