DEVELOPMENT OF AN INTERACTIVE SIMULATION

WITH REAL-TIME ROBOTS FOR SEARCH AND RESCUE

 

D. A. Campos and J. T. Wunderlich

Elizabethtown College Computer Engineering Program

 

 

The goal of this research is to implement a simulation that models search and rescue mobile robots while running concurrently and interactively with several real-time robots.  By having arranged the two systems as symbiotic, we are able to improve the search and rescue effort. 

The simulation functions by first modeling a mobile robot scouting an enclosed disaster area for life, hazards, or fire.  Because the robot doesn’t have any predisposed knowledge of its environment, it must have some way to recognize random objects.  The simulation can also adapt to information fed to it.  Elements in a real-time robot environment such as friction and deformation of terrain may be encountered and thus anticipated.  By having a learning simulation anticipate these faults, a planned path may be fed to the robot in real-time.  When the simulation is not receiving information, it may work on creating maps of the robot’s environment.

A large part of this effort focuses on the operations available to the mobile robots.  Because they are free moving within the walls of the environment, we allow them to communicate with one another.  The feedback input/output algorithm works by implementing the communication modules of the separate systems (CPU, and robots). The robots are able to add into their routines new data that is gained (such as holes or debris).  As part of our research, we have three different mobile robots, each dedicated to a separate task; one for searching, one for fire suppression, and one for medical assistance.  Both the simulation and real-time rescue robots find their targets through communicated coordinates passed by the scout.  Then the rescue robots perform their tasks as the scout continues searching the environment. The present effort includes three fully functional mobile robots and a concurrently running simulation.

  In real world situations where small robots may be used to search small crevices or fragile structures, having robots search for signs of life is extremely beneficial.  Furthermore, a simulation that can provide better paths for robots to maneuver can reduce tardiness and protect the real-time robots from hazards.  The use of robots will eliminate life-threatening situations where humans cannot risk going into unstable structures or fire-engulfed environments.