Academia.eduAcademia.edu

Outline

for Wireless Mobile Robots

2008

Abstract

A client-server application for the remote control of a wireless mobile robot (a small four-wheel vehicle) is described. The application requires an onboard computer (e.g. a PC-104 or a laptop) running a Java Virtual Machine (JVM) on the vehicle. The mobile robot is the client, which is controlled by the server. The server can be any desktop machine or another mobile robot. We show how mobile robots can be controlled using any computer on the internet through a wireless Ethernet network. This project uses commercial off-the-shelf (COTS) hardware and software. The ultimate goal of this effort is to build autonomous or semi-autonomous airborne vehicles which can run with little or no human intervention, but for the time being we are working with ground-base vehicles to develop the hardware and software. A small traditional radio-controlled aircraft (e.g. 2 meter wing span) could easily carry the computer required by the client. The approach we are using essentially gives the power of large desktop or Beowulf clusters to very small mobile robots, since the client is a complete internet-based PC. The client is basically an intelligent agent. This approach can be extended to allow multiple vehicles to communicate with each other and with other computers on the internet. This is not an approach that can be used for real-time remote control of airborne vehicles, but it is expected to be useful for autonomous and semi-autonomous vehicles, where it would be useful to occasionally send commands to the vehicle. This approach will also allow very powerful computers such as Beowulf clusters or parallel supercomputers to be used to control numerous mobile robots, since they too can act as servers.

References (6)

  1. Horstmann, C. and Cornell, G., "Core Java," Sun, 1999.
  2. Modi, L. Long, N. Sezer-Uzol, and P. Plassmann, "Scalable Computational Steering System for Vizualization of Large-Scale CFD Simulations," AIAA Fluids Conference, AIAA paper 2002-2750, St. Louis, 2002
  3. Modi, L. Long, and P. Plassmann, "Real-Time Visualization of Wake-Vortex Simulations using Computational Steering and Beowulf Clusters," Parallel Computing Conference, Portugal, 2002
  4. Modi, A. "Real-Time Visualization of Aerospace Simulations using Computational Steering and Beowulf Clusters," Ph.D. Dissertation, Penn State Univ., Aug., 2002.
  5. Genovesi, D. and Long, Lyle N., "An Object Oriented Approach to the Direct Simulation Monte Carlo Method," Java Grande, Nov., 2002.
  6. Genovesi, D., "An Object Oriented Approach to the Direct Simulation Monte Carlo Method," M.S. Thesis, Penn State Univ., May, 2002.