ETD system

Electronic theses and dissertations repository


Tesi etd-04162020-172157

Thesis type
Tesi di laurea magistrale
Design and development of a real-time automatic mapping and localization system for a Surface Contamination Monitor for large-area surveys
Corso di studi
relatore Prof. Fanucci, Luca
tutor Ing. Rollins, Micah
Parole chiave
  • odometry
  • lidar
  • SLAM
  • mapping
  • localization
  • ROS
  • radiation safety
  • radiation detection
Data inizio appello
Secretata d'ufficio
Data di rilascio
Riassunto analitico
The Surface Contamination Monitor (SCM) is a Position Sensitive Proportional Counter (PSPC) based alpha/beta detection system, designed to measure and record 100% of areas surveyed ideally with a minimum of human factors that contribute to survey inaccuracies. The current system, originally developed by Radiation Safety and Control Services (RSCS, Inc.) in the 1990s, uses a combination of a position-sensitive detector (x-axis) and wheel encoders (y-axis) to collect and localize data. The data then are combined in linear acquisitions and processed in a complex post-processing phase. While effective, the system is outdated and inefficient in both data localization and analysis, requiring significant human interaction throughout the process.
Over the past several years, there have been substantial advances in positioning and mapping technologies, often related to challenges such as autonomous driving, obstacle avoidance, and localization. This thesis project aims at integrating such technologies into the SCM to increase its accuracy and production efficiency, reducing the need for survey site preparation by technicians, and minimizing human-driven errors in post-survey data processing.
At first an introduction about radiation safety and detection is given, focusing on the SCM working principle and the reasons for the development of this thesis project. The requirements of the system have been defined and an exhaustive study on localization and mapping technologies has been carried out, to find the most suitable solution for our purpose. The proposed approach is based on the use of a LiDAR sensor and odometry data to enhance Simultaneous Localization and Mapping (SLAM). The implementation of a prototype as a Proof of Concept is then described, focusing on the generation of different type of maps and on system localization. The system along with the map generation algorithms was developed on the Robot Operating System (ROS). The prototype has been then integrated into the SCM and several tests have been run localizing event and generating accurate and analyzable maps of the environment.