Leveraging mobility in 3D space and received wireless signals to emulate a "virtual antenna array"
We develop the analytical framework for a novel Wireless signal-based Sensing capability for Robotics (WSR) by leveraging robots' mobility. It allows robots to measure relative direction, or Angle-of-Arrival (AOA), to other robots, while operating in non-line-of-sight, unmapped environments and without requiring external infrastructure. We do so by capturing all of the paths that a wireless signal traverses as it travels from a transmitting to a receiving robot in the team, which we term an AOA profile. The key intuition behind our approach is to enable a robot to emulate antenna arrays as it moves freely in 2D and 3D space, processing small differences in signal phase together with the robot's local displacement — a method akin to Synthetic Aperture Radar (SAR).
A wireless signal-based sensing framework for robotics
Our framework accommodates arbitrary 3D trajectories and continuous mobility of all robots while computing AOA profiles, and provides an accompanying analysis with a lower bound on the variance of AOA estimation as a function of the robot trajectory geometry, based on the Cramér–Rao bound. This formally characterizes the informativeness of a trajectory — a computable quantity with a closed form. All theoretical developments are substantiated by extensive simulation and hardware experiments, and the framework is open-sourced as the WSR Toolbox.
Active rendezvous for multi-robot pose graph optimization using sensing over Wi-Fi
We present a framework for collaboration amongst a team of robots performing pose graph optimization (PGO) that addresses two important challenges in multi-robot SLAM: enabling information exchange "on-demand" via active rendezvous without using a map or the robots' locations, and rejecting outlying measurements. Our key insight is to exploit relative position data present in the communication channel between robots to improve the ground truth accuracy of PGO. The approach is distributed and applicable in low-lighting or featureless environments where traditional sensors often fail. In experiments on actual robots, active rendezvous results in a 64% reduction in ground truth pose error, and using channel state information for outlier rejection reduces it by a further 32%.