Showing 1–1 of 1 results for author: Cavero-Vidal, M

Advances on Affordable Hardware Platforms for Human Demonstration Acquisition in Agricultural Applications

Authors: Alberto San-Miguel-Tello, Gennaro Scarati, Alejandro Hernández, Mario Cavero-Vidal, Aakash Maroti, Néstor García

Abstract: This paper presents advances on the Universal Manipulation Interface (UMI), a low-cost hand-held gripper for robot Learning from Demonstration (LfD), for complex in-the-wild scenarios found in agricultural settings. The focus is on improving the acquisition of suitable samples with minimal additional setup. Firstly, idle times and user's cognitive load are reduced through the extraction of individ… ▽ More This paper presents advances on the Universal Manipulation Interface (UMI), a low-cost hand-held gripper for robot Learning from Demonstration (LfD), for complex in-the-wild scenarios found in agricultural settings. The focus is on improving the acquisition of suitable samples with minimal additional setup. Firstly, idle times and user's cognitive load are reduced through the extraction of individual samples from a continuous demonstration considering task events. Secondly, reliability on the generation of task sample's trajectories is increased through the combination on-board inertial measurements and external visual marker localization usage using Extended Kalman Filtering (EKF). Results are presented for a fruit harvesting task, outperforming the default pipeline. △ Less

Submitted 11 June, 2025; originally announced June 2025.

Comments: 7 pages, 2 figures

Journal ref: European Robotics Forum 2025. ERF 2025. Springer Proceedings in Advanced Robotics, vol 36. Springer