Laser Drawn Map of 2029

In order for localization to work, two maps must be drawn. The first is a more traditional, measured map of the room. The second is drawn with the robot itself. The robot must be walked around the room, with the laser taking measurements on the static items. We must not include the dynamic items, such as the chairs and tables on wheels. These two maps are laid on top of each other, along with where the robot thinks it is, to correctly determine the exact position of the robot.

Analyze Andrew Lamonica’s code

Andrew developed an example program that uses the Markov Localization libraries and routines to perform some basic tasks. We will study this code and follow up by interviewing Andrew. Although the tasks we will be performing are not something he is completely familiar with, he may have some valuable advice.

Preliminary map

Using the “sickLogger” program provided by ActivMedia, we will use the robot’s Laser range finder to generate a map of EB2029. There are several steps involved with editing the map and preparing it for usage. In addition, a considerable amount of testing will have to be done to ensure we have a quality map of the room.

Interview Andrew

This time will be spent clearing up issues with example source code and our generated map of EB2029. Andrew’s experience with the mapping tools will help us reveal any inadequacies in our map of EB2029.

Final Map

Considering our test results and Andrew’s advice, we will construct our stable map that will be used during localization programming.