Bayesian Probabilistic Stopping Test and Asymptotic Shortest Time Trajectories for Object Reconstruction with a Mobile Manipulator Robot

Heikel Yervilla-Herrera, Israel Becerra, Rafael Murrieta-Cid, Luis Enrique Sucar, Eduardo F. Morales

Abstract: This paper proposes a Bayesian probabilistic stopping test for 3D object reconstruction with a mobile manipulator robot. Additionally, this work also presents an objective function to be optimized, which requires to find the time-optimal trajectories in the presence of obstacles for a non-holonomic robotic base satisfying visibility constraints at the sensing locations. The termination test together with a whole method for object reconstruction, is validated through comparisons with other termination criteria. Experiments in both simulation and with a physical mobile manipulator robot equipped with a sensor mounted on the arm’s end effector, are presented. On the basis of an experimental evaluation, it was found that the proposed objective function reduces the total reconstruction time, and that the proposed termination test achieves a good balance between the total number of sensing operations and the reconstruction percentage.