Numerous studies on adversarial attacks targeting self-driving policies fail to incorporate realistic-looking adversarial objects, limiting real-world applicability. Building upon prior research that facilitated the transition of adversarial objects from simulations to practical applications, this paper discusses a modified gradient-based texture optimization method to discover realistic-looking adversarial objects. While retaining the core architecture and techniques of the prior research, the proposed addition involves an entity termed the 'Judge'. This agent assesses the texture of a rendered object, assigning a probability score reflecting its realism. This score is integrated into the loss function to encourage the NeRF object renderer to concurrently learn realistic and adversarial textures. The paper analyzes four strategies for developing a robust 'Judge': 1) Leveraging cutting-edge vision-language models. 2) Fine-tuning open-sourced vision-language models. 3) Pretraining neurosymbolic systems. 4) Utilizing traditional image processing techniques. Our findings indicate that strategies 1) and 4) yield less reliable outcomes, pointing towards strategies 2) or 3) as more promising directions for future research.

Embodied agents is a term used to denote intelligent agents, which are a component of devices belonging to the Internet of Things (IoT) domain. Each agent is provided with sensors and actuators to interact with the environment, and with a 'controller' that usually contains an artificial neural network (ANN). In previous publications, we introduced three software approaches to design, implement and test IoT embodied agents. In this paper, we propose a reference model based on statecharts that offers abstractions tailored to the development of IoT applications. The model represents embodied agents that are controlled by neural networks. Our model includes the ANN training process, represented as a reconfiguration step such as changing agent features or neural net connections. Our contributions include the identification of the main characteristics of IoT embodied agents, a reference model specification based on statecharts, and an illustrative application of the model to support autonomous street lights. The proposal aims to support the design and implementation of IoT applications by providing high-level design abstractions and models, thus enabling the designer to have a uniform approach to conceiving, designing and explaining such applications.

Everything working and connected – in perfect symbiosis." I help our customers and partners with their digital journey, which involves innovation, business growth, and digital transformation. In this blog series of 6 posts, I look at the universal framework and the "building blocks" of smart cities in several contexts, trying to answer and interpret some of the questions that arise when thinking of digital transformation and smart city construction. In this third post, I touch upon the intelligent city ecosystem and industry convergence. We can think of "smart" at more of a technological level – sensor, actuators, data collection, and a reactive response; for example, a smart street light that switches on and off when it senses a pedestrian.

Agent-based Internet of Things (IoT) applications have recently emerged as applications that can involve sensors, wireless devices, machines and software that can exchange data and be accessed remotely. Such applications have been proposed in several domains including health care, smart cities and agriculture. However, despite their increased adoption, deploying these applications in specific settings has been very challenging because of the complex static and dynamic variability of the physical devices such as sensors and actuators, the software application behavior and the environment in which the application is embedded. In this paper, we propose a modeling approach for IoT analytics based on learning embodied agents (i.e. situated agents). The approach involves: (i) a variability model of IoT embodied agents; (ii) feedback evaluative machine learning; and (iii) reconfiguration of a group of agents in accordance with environmental context. The proposed approach advances the state of the art in that it facilitates the development of Agent-based IoT applications by explicitly capturing their complex and dynamic variabilities and supporting their self-configuration based on an context-aware and machine learning-based approach.

You can be forgiven if your first reaction to hearing the term "smart cities" is an eye roll. Sure, we have smart diapers, smart toothbrushes and smart faucets, but cities? How is that even possible? While the word is a bit amorphous, there's no question that smart cities are a thing and an important one at that. Cities are the locus of much of the world's population and economic activity. By 2050, a full 66 percent of the world is expected to reside in one, according to the United Nations.

Amazon's Prime Air drone delivery service, if it ever gets off the ground, could one day use the top of street lights, cell towers, and even church steeples as docking stations for its flying machine. The stations would serve as charging points for the drones, enabling them to stop off at multiple points for a battery boost thereby giving them a much greater flying range. Such a system could, in theory, open up pretty much the whole of the country to the possibility of drone delivery, as a single drone could hop from point to point on its way to an address. The docking stations could also shelter the drones from harsh weather conditions that may develop after they leave the distribution center to begin their delivery run. The new ideas are outlined in a patent awarded this month by the United States Patent and Trademark Office (USPTO) under the title, "Multi-use UAV (unmanned aerial vehicle) docking station systems and methods."

Amazon's Prime Air delivery drones already have a glaring problem: how do you keep them charged and sheltered when dedicated facilities are likely to be few and far between? The company has an idea. It recently received a patent for a "UAV docking station" concept that would offer a temporary perch for drones in need. If a drone runs low on battery or needs to take shelter from an impending storm, it would only have to travel to a station on top of a street light, cell tower, church steeple or another high-up location. The drone could even drop off a package for another drone, turning a delivery into an aerial relay race.

Airlines around the world are eager to take advantage of rapidly emerging technologies to improve their passengers' experience and become more efficient. But while executives recognize the opportunities, they know they can't do it alone. The two industry leaders in aircraft engines and technology are collaborating to offer carriers their expertise and ideas in a business where cutting 1 percent of fuel usage amounts to 250,000 in annual savings per plane. A recent PricewaterhouseCoopers report estimates digital tools in aircraft maintenance could save more than 100 million a year for a large carrier with a fleet of about 500 planes. "Our TotalCare maintenance program was revolutionary in the '90s, so we're pioneers ourselves, and by collaborating with a fellow pioneer like Microsoft, we can absolutely bring innovative digital solutions to airlines now," says Alex Dulewicz, head of marketing for services at Rolls-Royce's civil aerospace division.