Upper limb exoskeletons (ULEs) are robotic devices that can mobilize a severely disabled individual's arm and help the user perform some activities of daily living independently. Despite advancement in the mechanical design of ULEs, a versatile control interface that enables full and continuous control of a ULE with multiple degrees of freedom for a user with disability in both arms and legs (tetraplegia) still requires further research and development. In this study, ten individuals with tetraplegia used a tongue-based interface to fully control a five degrees of freedom ULE for a drinking and a snacking task. This required moving the ULE hand from a wheelchair armrest position to grasp an object (bottle or strawberry) placed on a table in front of the participant, moving the object towards them until it touched a face shield, and placing the object on the table. All participants successfully controlled the exoskeleton and completed the tasks. The drinking task lasted 149.6 s on the first day and 122.9 s (median) on the second day of using the exoskeleton. The participants performed the snacking task only on the first day of ULE use and achieved a median task time of 167.0 s. The study showed that the tongue interface could provide effective, efficient, and safe control of the exoskeleton.

We propose a novel multi-pose loss function to train a neural network for 6D pose estimation, using synthetic data and evaluating it on real images. Our loss is inspired by the VSD (Visible Surface Discrepancy) metric and relies on a differentiable renderer and CAD models. This novel multi-pose approach produces multiple weighted pose estimates to avoid getting stuck in local minima. Our method resolves pose ambiguities without using predefined symmetries. It is trained only on synthetic data. We test on real-world RGB images from the T-LESS dataset, containing highly symmetric objects common in industrial settings. We show that our solution can be used to replace the codebook in a state-of-the-art approach. So far, the codebook approach has had the shortest inference time in the field. Our approach reduces inference time further while a) avoiding discretization, b) requiring a much smaller memory footprint and c) improving pose recall.3

In this work we present a novel publicly available stereo based 3D RGB dataset for multi-object zebrafish tracking, called 3D-ZeF. Zebrafish is an increasingly popular model organism used for studying neurological disorders, drug addiction, and more. Behavioral analysis is often a critical part of such research. However, visual similarity, occlusion, and erratic movement of the zebrafish makes robust 3D tracking a challenging and unsolved problem. The proposed dataset consists of eight sequences with a duration between 15-120 seconds and 1-10 free moving zebrafish. The videos have been annotated with a total of 86,400 points and bounding boxes. Furthermore, we present a complexity score and a novel open-source modular baseline system for 3D tracking of zebrafish. The performance of the system is measured with respect to two detectors: a naive approach and a Faster R-CNN based fish head detector. The system reaches a MOTA of up to 77.6%. Links to the code and dataset is available at the project page http://vap.aau.dk/3d-zef.