Three-Dimensional Extended Object Tracking and Shape Learning Using Gaussian Processes

In this article, we investigate the problem of tracking objects with unknown shapes using 3-D point cloud data. We propose a Gaussian process-based model to jointly estimate object kinematics, including position, orientation, and velocities, together shape for online offline applications. describe by radial function in 3-D, induce correlation structure via process. Furthermore, an efficient algorithm reduce computational complexity working This is accomplished casting into projection planes, which are attached object's local frame. The resulting algorithms can process data accomplish dynamic object. they provide analytical expressions representation confidence intervals. intervals, quantify uncertainty estimate, later be used solving gating association problems inherent tracking. performance methods demonstrated both on simulated real results compared existing random matrix model, commonly extended literature.