Data Acquisition and Maps

Data Acquisition Maps

This work leverages the open source code published at this link [citation] to generate the training data. We use 8 vector maps of various buildings on campus at the University of Texas at Austin. The names and sizes (in meters) of the maps are as follows:

• SSB1: 67x76

• AHG1: 84x50

• AHG2: 84x50

• ASE1: 23x50

• ASE4: 23x50

• ETC2: 36x65

• GDC2: 85x56

• GDC3: 85x56

Images and links to the raw vectormap files of the maps can be found below:

** Note: each line in [mapname].vectormap.txt represents x1, y1, x2, y2 (i.e. the endpoints of a line in the map)

Experimental Maps

We evaluate the search method and the accuracy of the scan classifier in two unseen environments, which are shown below. Their names and sizes in meters are:

• AHG_Apartment: 30x20

• EER6: 38x15