Augmented reality (AR) is used to enhance the perception of the real world by integrating virtual objects to an image sequence acquired from various camera technologies. Numerous AR applications in robotics have been developed in recent years. The aim of this paper is to provide an overview of AR research in robotics during the five year period from 2015 to 2019. We classified these works in terms of application areas into four categories: (1) Medical robotics: Robot-Assisted surgery (RAS), prosthetics, rehabilitation, and training systems; (2) Motion planning and control: trajectory generation, robot programming, simulation, and manipulation; (3) Human-robot interaction (HRI): teleoperation, collaborative interfaces, wearable robots, haptic interfaces, brain-computer interfaces (BCIs), and gaming; (4) Multi-agent systems: use of visual feedback to remotely control drones, robot swarms, and robots with shared workspace. Recent developments in AR technology are discussed followed by the challenges met in AR due to issues of camera localization, environment mapping, and registration. We explore AR applications in terms of how AR was integrated and which improvements it introduced to corresponding fields of robotics. In addition, we summarize the major limitations of the presented applications in each category. Finally, we conclude our review with future directions of AR research in robotics. The survey covers over 100 research works published over the last five years.

No abstract

"Horizon-T" is an innovative detector system located at Tien Shan high-altitude Science Station (TSHASS) at approximately 3340 meters above the sea level. It consists of eight detection points separated by the distance up to one kilometer that can measure time characteristics of the Extensive Air Showers (EAS) and record signal shapes with time resolution of ~10 ns. It was constructed to register EAS in the energy range above 10 16 eV coming from a wide range of zenith angles (0 o -85 o ). The system includes both the plastic scintillator particle detectors as well as the Vavilov -Cerenkov radiation detectors subsystem to observe the Cerenkov light from the EAS in the atmosphere directly. The time resolution and signal shape analysis capabilities of the detection points are used to study EAS development in the atmosphere.The development of the EAS is a process that can be studied both spatially and temporally. For the spatial part, a distributed network of detection points is required. For the time part, a signal shape must be recorded and analysed at each point with time resolution on the order of ~10 ns. In this paper, the current system description and performance level are described. Additionally, the latest data examples showing the unusual EAS examples above 10

Recognizing objects and estimating their poses have a wide range of application in robotics. For instance, to grasp objects, robots need the position and orientation of objects in 3D. The task becomes challenging in a cluttered environment with different types of objects. A popular approach to tackle this problem is to utilize a deep neural network for object recognition. However, deep learning-based object detection in cluttered environments requires a substantial amount of data. Collection of these data requires time and extensive human labor for manual labeling. In this study, our objective was the development and validation of a deep object recognition framework using a synthetic depth image dataset. We synthetically generated a depth image dataset of 22 objects randomly placed in a 0.5 m × 0.5 m × 0.1 m box, and automatically labeled all objects with an occlusion rate below 70%. Faster Region Convolutional Neural Network (R-CNN) architecture was adopted for training using a dataset of 800,000 synthetic depth images, and its performance was tested on a real-world depth image dataset consisting of 2000 samples. Deep object recognizer has 40.96% detection accuracy on the real depth images and 93.5% on the synthetic depth images. Training the deep learning model with noise-added synthetic images improves the recognition accuracy for real images to 46.3%. The object detection framework can be trained on synthetically generated depth data, and then employed for object recognition on the real depth data in a cluttered environment. Synthetic depth data-based deep object detection has the potential to substantially decrease the time and human effort required for the extensive data collection and labeling.

No abstract

No abstract

No abstract

No abstract