Abstract
Current collaborative augmented reality (AR) systems establish a common localization coordinate frame among users by exchanging and comparing maps comprised of feature points. However, relative positioning through map sharing struggles in dynamic or feature-sparse environments. It also requires that users exchange identical regions of the map, which may not be possible if they are separated by walls or facing different directions. In this paper, we present Cappella11Like its musical inspiration, Cappella utilizes collaboration among agents to forgo the need for instrumentation, an infrastructure-free 6-degrees-of-freedom (6DOF) positioning system for multi-user AR applications that uses motion estimates and range measurements between users to establish an accurate relative coordinate system. Cappella uses visual-inertial odometry (VIO) in conjunction with ultra-wideband (UWB) ranging radios to estimate the relative position of each device in an ad hoc manner. The system leverages a collaborative particle filtering formulation that operates on sporadic messages exchanged between nearby users. Unlike visual landmark sharing approaches, this allows for collaborative AR sessions even if users do not share the same field of view, or if the environment is too dynamic for feature matching to be reliable. We show that not only is it possible to perform collaborative positioning without infrastructure or global coordinates, but that our approach provides nearly the same level of accuracy as fixed infrastructure approaches for AR teaming applications. Cappella consists of an open source UWB firmware and reference mobile phone application that can display the location of team members in real time using mobile AR. We evaluate Cappella across mul-tiple buildings under a wide variety of conditions, including a contiguous 30,000 square foot region spanning multiple floors, and find that it achieves median geometric error in 3D of less than 1 meter.

Abstract
Label noise is omnipresent in the annotations process and has an impact on supervised learning algorithms. This work focuses on the impact of label noise on the performance of learning models by examining the effect of random and class-dependent label noise on a binary classification task: quality assessment for photoplethysmography (PPG). PPG signal is used to detect physiological changes and its quality can have a significant impact on the subsequent tasks, which makes PPG quality assessment a particularly good target for examining the impact of label noise in the field of biomedicine. Random and class-dependent label noise was introduced separately into the training set to emulate the errors associated with fatigue and bias in labeling data samples. We also tested different representations of the PPG, including features defined by domain experts, 1D raw signal and 2D image. Three different classifiers are tested on the noisy training data, including support vector machine (SVM), XGBoost, 1D Resnet and 2D Resnet, which handle three representations, respectively. The results showed that the two deep learning models were more robust than the two traditional machine learning models for both the random and class-dependent label noise. From the representation perspective, the 2D image shows better robustness compared to the 1D raw signal. The logits from three classifiers are also analyzed, the predicted probabilities intend to be more dispersed when more label noise is introduced. From this work, we investigated various factors related to label noise, including representations, label noise type, and data imbalance, which can be a good guidebook for designing more robust methods for label noise in future work.

Abstract
Legged-wheeled locomotion systems are a particular case of robot types that can be characterized by an increase in the degrees of freedom. To increase safety and robustness in the performance of industrial robots, while reducing the risk of damage to the robot joints and injure to human operators, the use of non-rigid joints is growing in the literature and in the industry. Realistic simulators are tools capable of detecting rigid bodies interactions through physics engines. This paper presents the simulation model of a hybrid legged-wheeled robot, built in the SimTwo simulator. The proposed algorithms for path following control are detailed, along with the tests performed to them. These showed that the errors in linear paths are at most 1 cm. For circular paths, the maximum error is 3 cm.

Abstract
This article focuses on a sensorial glove prototype capable of acquiring hand motion and estimating its pose. The presented solution features twelve inertial measurement units (IMUs) to track hand orientation. The sensors are attached to a glove to decrease the project cost. The system also focuses on sensor fusion algorithms for the IMUs and further implementations, presenting the algebraic quaternion algorithm (AQUA), used because of its modularity and intuitive implementation. An adaptation of a human hand model is proposed, explaining its advantages and its limitations. Considering that the calibration is a very important process in gyroscope performance, the online and offline calibration data was analyzed, pointing out its challenges and improvements. To better visualize the model and sensors a simulation was conducted in Unity.

Abstract
Throughout this article the execution of the motion planning for a robotic manipulator by means of Reinforcement Learning methods is studied. Towards this, an implementation based on a Wire and loop game is used as an example case to be solved. The loop is controlled in a single plane as the endeffector of the manipulator. The modeling of the problem and the process of training the agent is detailed. This allowed for the verification of the capacity of a learning based method, having produced, under the considered abstractions, satisfying results by gaining the capability of completing the path imposed by the wire in 23 seconds.

Abstract