Robotic Through-Wall Imaging

The overall goal of this paper is to present the new possibilities created at the intersection of robotic path planning and inverse scattering for through-wall imaging with everyday RF signals. More specifically, we are interested in the throughwall imaging of a completely unknown space using unmanned vehicles and with everyday RF transceivers. We first focus on robotic through-wall imaging based on only WiFi received signal power measurements (RSSI). In our second case, we then focus on robotic imaging based on only UWB signals, motivated by the new commercially-available chipsets that can provide a lightweight solution for small robotic platforms. The paper then shows that through proper path planning, sparse signal processing, and wave modeling, high-resolution throughwall imaging of a completely unknown space is possible with ubiquitous RF transceivers. We present several experimental results for robotic through-wall imaging of several different unknown spaces, comparing WiFi and UWB-based approaches. We then extensively discuss the impact of robotic path design on imaging, which provides general guidelines for robotic path planning in the context of through-wall imaging.