Tracking and Estimation of a Swaying Payload Using a LiDAR and an Extended Kalman Filter

A Ground Penetrating Radar (GPR) has become an important tool for remote sensing studies in the Arctic for numerous applications, such as imaging ice sheets, making avalanche predictions and measuring the snow to ground boundary which can be used to forecast freshwater supply. Flying the GPR over a remote terrain such as the Arctic allows access to otherwise inaccessible Arctic regions. This can be achieved by suspending the GPR from a drone. However, the flight stability may be impacted by the nonlinear motion of the GPR. Minimizing the motion of the suspended payload is key to obtaining a stable flight and requires an accurate estimate for the position of the payload. This study uses a Velodyne VLP-16TM Light Detection and Ranging (LiDAR) sensor to measure the position of a suspended payload and an Extended Kalman filter to obtain an accurate estimate of the position of the payload. An experiment was conducted on a stationary drone with a swinging cablesuspended payload to test the feasibility of the proposed tracking and estimation system. The experimental results are presented to show the efficacy of the proposed solution. Vicon motion capture system was used to provide truth measurements and verify the experimental results.