Accelerating the Development of Mobile Robotics Technology

 The Competition

Pushing the Envelope

Because the primary function of this competition is to advance the state of the art, we encourage the participation of experienced engineers as well as university-level student teams.   Unlike most other robot competitions, participation will not be limited to young students who lack the needed engineering skills and experience. That these events will inspire young people will be a fringe benefit.

Competition problems will favor robust, versatile solutions, rather than problem-specific throw-away code that can only work in highly-constrained environments. 

For instance, although contest tasks will be documented in advance, the environment in which they will be run, such as lighting or surrounding objects, might not be disclosed beforehand.  Event organizers will be entirely at liberty to include "smoke and mirrors" in order to create challenging problems.  There will be no trivial tasks such as line-following or searching for a bright red ball against a black background. 

Nomad Antarctic Meteorite Search Robot - CMU and NASA

3D Occupancy Grid  Vision - SeeGrid

Focus on Software

Although mechanical design is an important aspect of any robot, these competitions will instead focus exclusively upon mobile robotics software.

The competitions will be limited to problems that can be solved with standard, inexpensive mobile robots and will not require any specialized mechanical components.

Entries will not be rewarded for purely mechanical characteristics such as physical size, maximum velocity, or traction systems that are more effective over irregular terrain.

These competitions will include events that can be conducted on flat surfaces such as gymnasium floors or parking lots that can be easily traversed by any mobile robot.

 

Encouraging Participation

These competitions will be announced in publications such as the IEEE Robotics and Automation Magazine.

Compared with the DARPA Grand Challenge, the costs for vehicles and sensors will be minimal.  Teams will not be required to devote time and energy to seeking financial sponsors.

Event organizers will not required financial backing.  A team entry fee of $100 should suffice to pay for facilities rental and a modest financial prize.

To minimize transportation costs, local competitions can be organized by IEEE RAS chapters and other groups across the USA and around the world.

Multiple Autonomous Robots Project at U of Penn

 Georgia Tech Robotics Lab - ActivMedia Pioneer ATs

Continuous Improvement and Cooperation

Similar challenge problems will be available in successive competitions to encourage continuous improvement.

Participants at each competition will also make a technical presentation of their results to provide an opportunity for constructive critique from other teams.  

All contestants will be encouraged to provide complete, documented source code that will be published at the end of each competition, for the benefit of all subsequent competitors.  Teams that are unwilling to share source code will be ineligible for the full financial prizes.

From this shared source code a development library of robotics algorithms, as well as basic sensor and control drivers, will be created.

A Mobile Robot Competition Competition

Defining the most effective tasks for The Mobile Robot Software Challenge is in itself a non-trivial problem.

There will be an ongoing competition to define the best challenge tasks that will drive improved mobile robot software.

Winning ideas will become future competition events.  These events will be named after the individuals who proposed them.  (Akin to the Turing Test.)

 

K-Team Koala Robot with Pan-Tilt Cameras