Radar helps locate vehicles using satellite images without gps
RadGeoLoc: Radar Geolocalization via Particle Filtering
Robotics
Summary
Global navigation systems like GPS sometimes don't work well outside or in tricky places. The authors present a way to find where a vehicle is by matching radar measurements with satellite pictures. They use a computer model that guesses possible locations and checks which one fits the radar data best. This method works in cities and offroad, even when the weather changes over months.
What this means in practice
- •For autonomous vehicle teams: Use radar and satellite images to localize vehicles accurately when GPS signals are poor or unavailable in diverse environments.
- •For field robotics operators: Improve robot navigation reliability in changing weather and offroad terrains by aligning radar data with georeferenced satellite maps.
Authors
Shaunak Kolhe, Peng Jiang, Kasi Viswanath, Srikanth Saripalli
Abstract
In real-time autonomous field applications, GNSS is often unreliable and inaccurate. A possible solution is localizing against a prebuilt map like a satellite image, as it is georeferenced and already surveyed. In this work, we localize a spinning radar measurement against overhead RGB imagery with a learned observation model. We use a radar representation pretrained by masked reconstruction and align it to features from a satellite image. A particle filter propagated by radar odometry proposes candidate poses, and our observation model provides a likelihood over these poses. We evaluate our method over 31 distinct routes, over 8 datasets spanning both urban and offroad conditions using only ground radar and a coarse local prior. We also examine the robustness over changing weather using repeated traversals over the same route over 13 months, in both urban and offroad conditions.