Photovoltaic (PV) cells, which can generate energy from the sun, could be very useful in tackling the current environmental crisis. Perovskite PV cells, cells made of metal halide perovskite semiconductors, have recently proved to be particularly promising, as...
TECHXPLORE
VRKitchen: An interactive virtual environment to train and test AI agents
A team of researchers at the University of California Los Angeles (UCLA), under the supervision of Prof. Song-Chun Zhu, has developed an interactive 3-D virtual environment for training and evaluating AI systems. Their platform, called VRKitchen, was recently...
Three convolutional neural network models for facial expression recognition in the wild
Two researchers at Shanghai University of Electric Power have recently developed and evaluated new neural network models for facial expression recognition (FER) in the wild. Their study, published in Elsevier's Neurocomputing journal, presents three models of...
Using a simulation framework to study spine behaviors of quadruped robots
Researchers at the Robert Bosch center for cyber physical systems in Bangalore, India, have recently proposed a simulation framework to systematically study the effects of spinal joint actuation on the locomotion performance of quadruped robots. In their study,...
Using deep learning to predict parameters of batteries on electric vehicles
The batteries used to power electric vehicles have several key characterizing parameters, including voltage, temperature, and state of change (SOC). As battery faults are associated with abnormal fluctuations in these parameters, effectively predicting them is of...
How the Avengers assemble: Ecology-based metrics model effective cast sizes for Marvel movies
In a recent study, researchers at the ARC Centre of Excellence for Mathematical and Statistical Frontiers (ACEMS) in Adelaide have tried to use ecology-related concepts to model effective cast sizes for movies, focusing on characters from the Marvel Cinematic Universe...
Using game theory to model poisoning attack scenarios
Poisoning attacks are among the greatest security threats for machine learning (ML) models. In this type of attack, an adversary tries to control a fraction of the data used to train neural networks and injects malicious data points to hinder a model's performance.
Using imitation and reinforcement learning to tackle long-horizon robotic tasks
Reinforcement learning (RL) is a widely used machine-learning technique that entails training AI agents or robots using a system of reward and punishment. So far, researchers in the field of robotics have primarily applied RL techniques in tasks that are completed...
Using laser beams for communication and coordination of spacecraft swarms
Swarms of small spacecraft could have a variety of interesting applications, particularly in terms of Earth observation, global positioning and communications. Compared to large spacecraft, small spacecraft provide greater apertures, allowing for better observation of...
Using sensors to improve the interaction between humans and robots walking together
Researchers at the BioRobotics Institute of Scuola Superiore Sant"Anna, Co-Robotics srl and Sheffield Hallam University have recently proposed a new approach to improve interactions between humans and robots as they are walking together. Their paper, published in...