Session E1: Emerging Technologies for Alternative, Resilient, and Intelligent PNT Systems

LiDAR, UWB and IMU-Based Cross-Agent Relative Pose Estimation
Andrea Masiero, Alberto Guarnieri, and Antonio Vettore, University of Padua
Location: Peale (First Floor)
Date/Time: Wednesday, Sep. 10, 11:48 a.m.

GNSS (Global Navigation Satellite System) typically enables outdoor positioning and navigation almost everywhere, satellite-based positioning cannot be used for precise navigation in certain challenging environments, such as, for instance, indoors, in tunnels and urban canyons. Despite GNSS can still be considered as the main localization technology for autonomous platforms, it has to be integrated with other approaches for the usability of such platforms in the previously mentioned challenging environments. This aspect is critical in particular when different platforms have to properly interact and cooperate: precisely determining the relative pose between them in such conditions is fundamental in order to make them properly complete their tasks in a cooperative manner. This work considers different alternatives for relative pose estimation between ground robots in a 2D environment, analyzing in particular the use of IMU (Inertial Measurement Unit) measurements and UWB (ultra-wide band) ranging, and their use supporting a spatial feature-based approach. Tests have been conducted on the second floor of a building of the University of Padua. Preliminary results show that the performance of the IMU-UWB approach quickly degenerates increasing the distance between the robots, whereas, UWB used to support feature-based method allow to reduce the chances of large errors.