MIT Lincoln Labs - Controls Engineering Intern
Overview:
MIT Lincoln Labs has been working on a autonomous driving a sensor that uses very high frequency (VHF) radar reflections of underground features to generate a baseline map of a road's subsurface. The localizing ground penetrating radar (LGPR) is used to generate a map during a vehicle's drive along a roadway. When revisiting the location, the car compares the measurements for the LGPR and with GPS and the map stored in memory. This system allows provides extra robustness because it does not rely on optical systems for localization.
My role was implement and tune an Extended Kalman Filter in ROS (link) for location estimation. This first involved creating a simulation platform in ROS to simulate driving from previously collected data. This tool was used to validate that the EKF worked and tune the EKF the characteristic parameters.
Please see the link below for even more information about the project.
MIT Lincoln Labs has been working on a autonomous driving a sensor that uses very high frequency (VHF) radar reflections of underground features to generate a baseline map of a road's subsurface. The localizing ground penetrating radar (LGPR) is used to generate a map during a vehicle's drive along a roadway. When revisiting the location, the car compares the measurements for the LGPR and with GPS and the map stored in memory. This system allows provides extra robustness because it does not rely on optical systems for localization.
My role was implement and tune an Extended Kalman Filter in ROS (link) for location estimation. This first involved creating a simulation platform in ROS to simulate driving from previously collected data. This tool was used to validate that the EKF worked and tune the EKF the characteristic parameters.
Please see the link below for even more information about the project.