1 Pathfinding Developed By Joe Churchwell

I don’t yet have a great example of a project that I am truly interested in these days since it is dependent on the micro-controller that I use. Software with machine interaction is really what I care about and what I would like to do. Robots that can move autonomously through an unknown environment most certainly fits the bill.

1.1 Modified A* Rover Simulation

The video below was generated from the GUI application I built to show the path that an autonomous rover will traverse. The video is a little choppy as compared to running the application. You can download all the source code, application files, and directions for this project here. Please note the GUI code is a bit raw due to time constraints and I really placed more emphasis on the C++ code when doing this for my class assignment as the GUI was just extra credit. It should be noted that the rover is constantly discovering objects in the environment as it moves and the algorithm doesn’t try to optimize the shortest path in terms of Euclidian distance but attempts to minimize the cost of traversal. For this reason, the simulation that reproduces the Wiki example does not have the same optimized path.

Figure 1.1 - Larger Map

Note: The best way to run the application (Windows Only) is to download the run folder and load the GUI using a pre-configured file. You can load these files using the ‘L’ keyboard key which is what I did in the video.

GUI Command from Keyboard:

L – Load a file to run

R – Setup a randomized board layout. An input box will be provided to enter in the blocked percentage

S – Stop the simulation

Escape – Clears the colored cells

Enter – Run the algorithm and animate the result

Figure 1.2 - 50 x 50 Map