Automated Asset Locating System (AALS)
The goal of this project is to develop a “proof of concept” system for automated asset tracking. The system must be capable of self-localization and navigation while mapping the location of discovered asset tags.
NATHAN (Nascent AALS Testing, Honing, and Authentication of NATHAN), our development platform, will accomplish this by fusing multiple types of data, including odeometry, LIDAR, and radio frequency identification (RFID).
NATHAN is built upon the iRobot Create. It utilizes a combination of odometry, RFID, and LIDAR for localization. The LIDAR also enables obstacle detection and avoidance.
LIDAR: Hokyuo URG 04LX
RFID: Skyetek M9
The AALS uses Monte Carlo Localization to autonomously navigate a given environment. By merging data from a floor plan and its onboard sensors, the AALS is able to accurately locate itself based on physical landmarks, as well as avoid unplanned obstacles. There is a video of a simulated AALS doing just this, here //vader.
- Red particles: Estimated robot positions
- Green particle: Simulated robot
- Yellow circles: Landmark RFID tags, used to aid localization
- Blue circle: Most likely robot position, based on LIDAR and RFID readings
AALS employs an ultra high frequency (UHF) RFID reader to locate objects marked with a passive RFID tag.
Center for Engineering Logistics and Distribution (CELDi)
- Thomas Miller
- John Spletzer
- Mooi-Choo Chuah