We’ve previously talked about the importance of robot localization. Kiran Mohan, a master student from Worcester Polytechnic Institute, developed a prototype system using Wi-Fi data to improve robot localization in several ways.
Wi-Fi data has proven most useful in allowing a robot to determine an initial estimate of its position within the map. FetchCore already includes the ability to build a Wi-Fi strength map for your facility using the wireless card embedded in the robot. Once such a map exists, the robot can then use a “fingerprint” of the visible Wi-Fi access points to roughly determine the location of the robot within this map. This rough estimate will someday replace the need for users to manually localize robots when they are first powered on.
While this project was only a first step towards a production system, it is typical of the type of projects interns can work on at Fetch Robotics. If you are still in school, or a recent grad, and interested in learning more about software engineering for robots, check out internship openings for this fall semester.
*Wi-Fi is a registered trademark of the Wi-Fi Alliance.