In the new design, each of the four units has a single flapping wing pointing away from the robot's center, stabilizing the wings and boosting their lift forces. The researchers also improved the way the wings are connected to the actuators, or artificial muscles, that flap them. In previous designs, when the actuators' movements reached the extremely high frequencies needed for flight, the devices often started buckling. That reduced the power and efficiency of the robot. Thanks in part to a new, longer wing hinge, the actuators now experience less mechanical strain and can apply more force, so the bots can fly faster, longer, and in more precise paths.

Now that Apple's recent slew of hardware releases are behind us, we got some news on the software side last week. First, the company publicly announced that it was delaying the smarter, more personal version of Siri that'll be powered by Apple Intelligence. Then, rumors sprang up again that Apple was giving an extensive visual update to its software platforms, including iOS 19 and macOS 16 which are expected to be revealed at WWDC in June. The sources for this redesign rumor are solid. Jon Prosser dropped a video on his YouTube channel Front Page Tech back in January where he said that he had seen a redesigned Camera app for the next version of iOS that had a number of interface changes that made it feel more like a visionOS app. His thinking is that Apple wouldn't redesign a core app like Camera without bringing changes to some of the rest of the OS, as well.

Offline model-based optimization (MBO) aims to maximize a black-box objective function using only an offline dataset of designs and scores. A prevalent approach involves training a conditional generative model on existing designs and their associated scores, followed by the generation of new designs conditioned on higher target scores. However, these newly generated designs often underperform due to the lack of high-scoring training data. To address this challenge, we introduce a novel method, Design Editing for Offline Model-based Optimization (DEMO), which consists of two phases. In the first phase, termed pseudo-target distribution generation, we apply gradient ascent on the offline dataset using a trained surrogate model, producing a synthetic dataset where the predicted scores serve as new labels. A conditional diffusion model is subsequently trained on this synthetic dataset to capture a pseudo-target distribution, which enhances the accuracy of the conditional diffusion model in generating higher-scoring designs. Nevertheless, the pseudo-target distribution is susceptible to noise stemming from inaccuracies in the surrogate model, consequently predisposing the conditional diffusion model to generate suboptimal designs. We hence propose the second phase, existing design editing, to directly incorporate the high-scoring features from the offline dataset into design generation. In this phase, top designs from the offline dataset are edited by introducing noise, which are subsequently refined using the conditional diffusion model to produce high-scoring designs. Overall, high-scoring designs begin with inheriting high-scoring features from the second phase and are further refined with a more accurate conditional diffusion model in the first phase. Empirical evaluations on 7 offline MBO tasks show that DEMO outperforms various baseline methods.

The majority of existing post-hoc explanation approaches for machine learning models produce independent, per-variable feature attribution scores, ignoring a critical inherent characteristics of homogeneously structured data, such as visual or text data: there exist latent inter-variable relationships between features. In response, we develop a novel model-agnostic and permutation-based feature attribution approach based on the relational analysis between input variables. As a result, we are able to gain a broader insight into the predictions and decisions of machine learning models. Experimental evaluations of our framework in comparison with state-of-the-art attribution techniques on various setups involving both image and text data modalities demonstrate the effectiveness and validity of our method.

This new technology has evident social and economic impacts on our lifestyle and represents one of the most vital future trends [16]. In this context, Suab et al. have generally explained how different UAVs could be employed in forests and farming [7]. On the other hand, the mobility and flexibility of UAVs as well as their cheap costs represent their most powerful features for developing a new generation of cellular communication flying terminals as in [8][9][10], acquiring images rich of details for executing several geological applications as suggested in [11], or monitoring and tracking wildlife features [12]. Many engineers and researchers look at commercial multicopters as practical devices that could be adapted to perform a wide range of applications, even though they are unable to carry considerable payloads, and their batteries may sustain up to two successive flying hours if their maximal weight, including their payload, is around ten kilos. In such a case, building advanced types of flying robots that satisfy any special requirements regarding their payload or their maximal flying range requires new creative ideas. This paper presents a few novel designs with many technical details including computer vision systems and some modern AI-based algorithms that are required to perform smart missions. The organization of this paper is as follows: first, we present several recently designed and presented platforms for serving in different areas.

The Industrial Internet of Things, or IIoT, has recently gained popularity due to its ability to create communication networks between different components of an industry and bring about the new revolution--Industry 4.0. Powered by wireless 5G connectivity and artificial intelligence (AI), IIoT holds the ability to analyze critical problems and provide solutions that can improve the operational performance of industries ranging from manufacturing to health care. IoT is highly user-centric--it connects TVs, voice assistants, refrigerators, etc.--whereas IIoT deals with enhancing the health, safety, or efficiency of larger systems, bridging hardware with software, and carrying out data analysis to provide real-time insights. However, while IIoT does have many advantages, it also comes with its share of vulnerabilities such as security threats in the form of attacks trying to disturb the network or siphoning resources. As IIoT is getting more popular in industries, it is becoming crucial to develop an efficient system to handle such security concerns.

A team of researchers at the University of Zurich, has developed a highly agile quadrotor drone that is able to avoid obstacles and carry out trajectory tracking. In their paper published in the journal Science Robotics, the group describes how they designed their drone, what they put into it and how well it worked when tested. Quadrotor drones can be very agile fliers, most particularly when they have a human pilot guiding their movements. Autonomous quadrotors, on the other hand, have suffered from agility issues, particularly when traveling at high speeds. In this new effort, the team in Switzerland has improved the agility of a quadrotor drone with their new design built using a variety of technologies.

Now MIT engineers have taken a step toward that modular vision with a LEGO-like design for a stackable, reconfigurable artificial intelligence chip. The design comprises alternating layers of sensing and processing elements, along with light-emitting diodes (LED) that allow for the chip's layers to communicate optically. Other modular chip designs employ conventional wiring to relay signals between layers. Such intricate connections are difficult if not impossible to sever and rewire, making such stackable designs not reconfigurable. The MIT design uses light, rather than physical wires, to transmit information through the chip.

Estimated to be worth $3T by the end of the decade, per CB Insights' Industry Analyst Consensus, the fashion industry is growing at a fast pace, led by cutting-edge technologies. From robots that sew and cut fabric to AI algorithms that predict style trends, VR mirrors in dressing rooms, shopping off the runway and a number of other innovations show how technology is automating and evolving the industry. In 2016, Google collaborated with online fashion platform Zalando and production company Stinkdigital to launch predictive design engine, Project Muze. The algorithm consisted of a set of aesthetic parameter and trained a neural network to comprehend colours, textures and styles derived from Google Fashion Trends Report and data sourced by Zalando -- to create designs in sync with with style preferences identified by the network. Amazon is taking an algorithmic approach to fashion as well.

Competition is tight in the manufacturing sector, and the ability to bring new products to market fast is often key to success. This is where VR, MR, and AR can enhance product design, essentially by helping to speed up the creative process. Indeed, many manufacturers are already using XR technologies to improve their product design and development. Ford is one such manufacturer. At Ford, designers use the Microsoft HoloLens headset to design cars in mixed reality.