Mini Grand Challenge

''For more on the AUVSI IGVC competition, see the IGVC article. This project, for the year 2009-2010, is owned by Jeremy Bridon.''

Modeled after the DARPA Grand Challenge, the Mini Grand Challenge is an outdoor autonomous vehicle competition hosted by the Abington branch campus of the Pennsylvania State University. The Mini Grand Challenge consists of two parts. In part 1, the robot has to stay on an asphalt path while navigating through a series of waypoints. If an obstacle is detected, the robot must stop within 2 feet and wait for the obstacle to be removed. Branches in the path are blocked with orange cones. In part 2, the robot must move off the path and navigate towards a final GPS waypoint and stop within 30 feet of that location. In this stage of the competition, the robot must navigate around any obstacles it encounters. Scoring for the competition is based on a number of different criteria, including total waypoints reached by the robot and also through crowd interaction. There are also several mandatory sub-tasks: The robot must stay within 1 to 5 MPH as well as carry one gallon of water. The vehicle may support crowd interaction (voice, conversation, music, etc..) for bonus points.

The Penn State Robotics Club has a modified Power-Wheels chassis that is used as the vehicle base, interfaced with an Arduino microcontroller. Other hardware components included are a laser-range finder, cameras, GPS, compass, wheel encoders, and on-board computers.

= 2009/2010 Solution =

Vehicle Course
Below is an overview of the campus with way points and path marked. The blue line directly connects these points in order, with the green line representing the path to be traveled. The starting waypoint is marked as 1, with the final destination being waypoint 6. Note how the grassy field is between waypoint 5 and 6.

Source(s): Mini Grand Challenge solutions

Solution Overview
The 2009/2010 solution is based on a laser/vision merger of data with forward vehicle projection. The vehicle will have a camera at waist level, facing forward. Below this, a laser-range finder device is mounted, but slightly pitched down. The collision data points will be within the camera's field of view. This is done such that we can apply a laser/vision fusion to gather the color of the possible drivable surface.

Laser collision data will be used to find what the robot may use as a surface to drive on, as well as general obstacles. This is merged with camera image data to produce a color range that the robot considers a "drivable surface". The image is then segmented using this color range, producing a collision map (a black and white image of drivable and non-drivable surfaces). A projection transform is applied to help choose the best forward path for the vehicle. This last component will find the optimal path for all near-future vehicle positions, acting as the "main controller" of the vehicle.

A more in-depth explanation is provided in the MGC Project Abstract article.

Solution Dependencies
This solution will require certain software and hardware tools to accomplish the task. The current hardware platform from previous years (a modified Power-Wheels vehicle) contains all the needed hardware and electronics; meeting those needs. Software tools, on the other hand, have significantly changed for this year's solution. Player/Stage is officially being removed as a tool due to its complex nature and inefficient implementation. To replace the parallelism that was provided through that tool, we are now using the CMU IPC library. Hardware interfacing will be done by extracting the relevant code from previous solutions from the club. Our target operating system is Linux, with the target language C/C++, compiled using GCC. The only other major software dependency would be OpenGL for graphics and OpenCV for our computer vision library.

Team Resources
The team has access to several tools to help the development of the project. This includes a mailing list, a source management system (SVN), and access to source documentation. Much of this is discussed in our MGC Development Tools article.

Team Roster

 * Jeremy Bridon - Team leader
 * Adam Brockett
 * Tony Jones

= 2009-2010 Online Project Documentation =


 * Mini Grand Challenge
 * MGC Development Tools
 * MGC Project Abstract
 * Mini Grand Challenge Architecture
 * Mini Grand Challenge Action Items
 * Mini Grand Hardware

= Previous Solution Documentation =
 * The 2008/2009 Mini Grand Challenge project outline [[media:Minigrand09Outline.pdf|PDF]]
 * A report on the 2007/2008 vehicle steering modification [[media:F150steering.pdf|PDF]]
 * A report on the 2007/2008 designs for both the Mini Grand and IGVC [[media:design_3-19-08.pdf|PDF]]

= External Links =


 * Official Contest Page
 * Old Robotics Club Mini Grand Challenge Main Page - (No longer updated)
 * Rich's 2007 Success
 * The current online source browser and direct SVN address