In this paper, we propose and experimentally investigate an optical sensor based on a novel combination of a long-period fiber grating (LPFG) with a permanent magnet to measure electrical current in unmanned aerial vehicles (UAVs). The proposed device uses a neodymium magnet attached to the grating structure, which suffers from an electromagnetic force produced when the current flows in the wire of the UAV engine. Therefore, it causes deformation on the sensor and thus, different shifts occur in the resonant bands of the transmission spectrum of the LPFG. Finally, the results show that it is possible to monitor electrical current throughout the entire operating range of the UAV engine from 0 A to 10 A in an effective and practical way with good linearity, reliability and response time, which are desirable characteristics in electrical current sensing.

The autonomous robotic system accurate localization is a challenging step in robot navigation field once the mobile device should avoid dangerous situations, such as unsafe conditions and collisions. In this context, the present paper proposes a localization method using the Extended Kalman Filter (EKF) to fuse the information coming from two different sensors (i.e. odometry and computer vision). The localization results present with known and unknown starting points and are tested in a simulated environment.

Bin picking is a challenging task involving many research domains within the perception and grasping fields, for which there are no perfect and reliable solutions available that are applicable to a wide range of unstructured and cluttered environments present in industrial factories and logistics centers. This paper contributes with research on the topic of object segmentation in cluttered scenarios, independent of previous object shape knowledge, for textured and textureless objects. In addition, it addresses the demand for extended datasets in deep learning tasks with realistic data. We propose a solution using a Mask R-CNN for 2D object segmentation, trained with real data acquired from a RGB-D sensor and synthetic data generated in Blender, combined with 3D point-cloud segmentation to extract a segmented point cloud belonging to a single object from the bin. Next, it is employed a re-configurable pipeline for 6-DoF object pose estimation, followed by a grasp planner to select a feasible grasp pose. The experimental results show that the object segmentation approach is efficient and accurate in cluttered scenarios with several occlusions. The neural network model was trained with both real and simulated data, enhancing the success rate from the previous classical segmentation, displaying an overall grasping success rate of 87.5%.