Next-Best-Sense: A Multi-Criteria Robotic Exploration Strategy for RFID Tags Discovery

Abstract

Automated exploration is one of the most relevant applications of autonomous robots. In this paper, we suggest a novel online coverage algorithm called Next-Best-Sense (NBS), an extension of the Next-Best-View class of exploration algorithms that optimizes the exploration task balancing multiple criteria. This novel algorithm is applied to the problem of localizing all Radio Frequency Identification (RFID) tags with a mobile robotic platform that is equipped with a RFID reader. We cast this problem as a coverage planning problem by defining a basic sensing operation -- a scan with the RFID reader -- as the field of “view” of the sensor. NBS evaluates candidate locations with a global utility function which combines utility values for travel distance, information gain, sensing time, battery status and RFID information gain, generalizing the use of Multi-Criteria Decision Making. We developed an RFID reader and tag model in the Gazebo simulator for validation. Experiments performed both in simulation and with a real robot suggest that our NBS approach can successfully localize all the RFID tags while minimizing navigation metrics such sensing operations, total traveling distance and battery consumption. The code developed is publicly available on the authors' repository.