CUDA Center Using Robots to Build Real Time Maps

by Chandra Cheij

Need a map? Don’t bother asking for directions. Just send in the robots.

Imagine sending a robot into a new environment – say, a lush tropical jungle brimming with rare creatures or an unexplored building – and being able to get a map or digital blueprint anywhere in the world in just a few moments.

A research team at Poland’s Institute of Mathematical Machines, led by Janusz Bedkowski, wants to do just that. their plan is to develop new robotic applications that will use photos to build three-dimensional maps and electronic blueprints that can be served up on phones and tablets.

It’s a data-intensive task, and Bedkowski and his colleagues want to build these maps in real time as new environments are explored. It’s a perfect application for NVIDIA’s CUDA technology, which puts the parallel processing power of the GPUs used by millions of gamers to work solving problems for scientists and businesspeople.

Bedkowski’s work is an example of the challenges being tackled by the 26 institutions added this past quarter to our roster of CUDA Research Centers and CUDA Teaching Centers. CUDA Teaching Centers help tens of thousands of students graduating learn how to use the parallel processing power of GPUs (see “What Is CUDA?”).

They get free teaching kits, textbooks, software licenses, NVIDIA CUDA GPUs for teaching lab computers and academic discounts for additional hardware.  CUDA Research Centers have access to events with key researchers and academics, a designated NVIDIA technical liaison and specialized training sessions.

As part of a project dubbed ICARUS — short for Integrated Components for Assisted Rescue and Unmanned Search operations – IMM’s team is collecting data from unmanned vehicles, pouring it into a common reference coordinate system, and using CUDA-based algorithms and NVIDIA’s GRID technology to distribute the data over networks. From there, it could be accessed on mobile devices.

Another project, funded by the Polish National Centre of Research and Development, is using mobile robots to build models of the environment around them, serve up that data via GRID, and build a 3D CAD model of a building that can be accessed by architects anywhere in the world within an hour.

There are now more than 344 CUDA Research and Teaching centers in 44 countries.

Additional New CUDA Research Centers:

  • Imperial College London (UK)
  • Laboratório Nacional de Computação Científica (Brazil)
  • Nirma University (India)
  • Perm State University (Russia)
  • The George Washington University (U.S.)
  • Tsinghua University (China)
  • University of Michigan (U.S.)

New CUDA Teaching Centers:

  • Al-Balqa Applied University (Jordan)
  • Columbia University (U.S.)
  • COMSAT Islamabad (Pakistan)
  • Harbin Institute of Technology (China)
  • Indian Institute of Technology Guwahati (India)
  • Institute of Management and Career Courses (India)
  • Manipal Universeity Jaipur (India)
  • Nankai University (China)
  • Riga High Tech University (Latvia)
  • Southern Federal University (Russia)
  • Technical University of Braunschweig (Germany)
  • Thapar University (India)
  • Trinity College Dublin (Ireland)
  • Universidad Católica San Antonio (Spain)
  • Universidad Tecnológica de Pereira (Colombia)
  • University of Winnipeg (Canada)
  • Veermata Jijabai Technological Institute (India)
  • University of Seville (Spain)