Due to their high maneuverability, flexible deployment, and low cost, unmanned aerial vehicles (UAVs) are expected to play a pivotal role in not only communication, but also sensing. Especially by exploiting the ultra-wide bandwidth of terahertz (THz) bands, integrated sensing and communication (ISAC)-empowered UAV has been a promising technology of 6G space-air-ground integrated networks. In this article, we systematically investigate the key techniques and essential obstacles for THz-ISAC-empowered UAV from a transceiver design perspective, with the highlight of its major challenges and key technologies. Specifically, we discuss the THz-ISAC-UAV wireless propagation environment, based on which several channel characteristics for communication and sensing are revealed. We point out the transceiver payload design peculiarities for THz-ISAC-UAV from the perspective of antenna design, radio frequency front-end, and baseband signal processing. To deal with the specificities faced by the payload, we shed light on three key technologies, i.e., hybrid beamforming for ultra-massive MIMO-ISAC, power-efficient THz-ISAC waveform design, as well as communication and sensing channel state information acquisition, and extensively elaborate their concepts and key issues. More importantly, future research directions and associated open problems are presented, which may unleash the full potential of THz-ISAC-UAV for 6G wireless networks.

Autonomous vehicles are expected to operate safely in complex environments, yet failures can still arise under rare and safety-critical conditions. Existing validation pipelines often rely on large-scale simulation and manual scenario design but lack the ability to semantically interpret failure events and respond with adaptive improvements. This article presents a self-evolving framework enhanced by AI capabilities, which leverages large language models (LLMs) to analyze failure cases, recommend semantically relevant test scenarios, and adapt affected modules through targeted few-shot learning. Structured execution logs are converted into semantic representations, which are used to retrieve safety-critical scenarios from a scenario bank. These scenarios enable reliable reproduction of failure behaviors and support few-shot adaptation of the affected module. A closed-loop evaluation confirms behavioral improvement while ensuring system stability. The framework has been extensively evaluated across a diverse set of driving tasks, consistently demonstrating enhanced robustness with minimal human supervision. By integrating semantic failure understanding, scenario-based testing, and data-efficient adaptation, this approach offers a scalable and generalizable solution for failure-aware improvement in safety-critical autonomous driving systems.

Airborne networks build upon the use of unmanned aerial vehicles (UAVs) and high-altitude platform stations (HAPSs) for wireless access and backhauling and are expected to be instrumental in providing global coverage and ubiquitous connectivity. They can offer a range of benefits, including lower latency and higher data rate capacity per unit area, making them an attractive alternative or complementary solution to low-earth orbit satellites in future non-terrestrial networks. The practical deployment of airborne nodes is restricted by onboard energy limitations, motivating the use of energy harvesting techniques. In this paper, we present an in-depth examination of the power consumption of HAPSs and rotary-wing UAVs. We delve into consumption patterns across various flight phases, shedding light on the multifaceted impact of diverse system and operational parameters on overall energy utilization. We then present a thorough analysis of energy harvesting methods. First, we examine solar energy harvesting and demonstrate its dependence on factors such as operational altitude, geographical location, climate conditions, and daylight duration. Subsequently, we introduce laser power beaming as a more predictable and controllable energy source. Thereafter, we discuss the feasibility of self-sustainable airborne networks based on these energy harvesting techniques and typical energy consumption patterns.

In this article, we propose the integration of the holographic multiple-input multiple-output (HMIMO) as a transformative solution for next-generation nonterrestrial networks (NTNs), addressing key challenges, such as high hardware costs, launch expenses, and energy inefficiency. Traditional NTNs are constrained by the financial and operational limitations posed by bulky and costly antenna systems, alongside the complexities of maintaining effective communications in space. HMIMO offers a novel approach utilizing compact and lightweight arrays of densely packed radiating elements with real-time reconfiguration capabilities, thus capable of optimizing system performance under dynamic conditions such as varying orbital dynamics and Doppler shifts. By replacing conventional antenna systems with HMIMO, the number of required radio-frequency (RF) chains can be cut to roughly one-tenth while achieving about threefold lower power consumption and an order-of-magnitude decrease in hardware cost. Thus, manufacturing and launch expenses can be substantially reduced, paving the way for more streamlined and cost-effective NTN deployments. This advancement holds significant potential to democratize space communications, making them accessible to a broader range of stakeholders, including smaller nations and commercial enterprises. Moreover, the inherent capabilities of HMIMO in enhancing energy efficiency, scalability, and adaptability position this technology as a key enabler of new use cases and sustainable satellite operations.

This article addresses electric vehicles (EVs) equipped with multiagent systems that integrate multiple motors and energy sources. The objective is to develop a novel energy management strategy (EMS) based on a unified adaptive network-based fuzzy inference system (ANFIS) framework. The unified framework simultaneously optimizes torque distribution regression between two motors and the driving mode classification in the current control between energy sources. By combining the learning of neural networks (NNs) with the interpretability of fuzzy logic, the ANFIS-based EMS effectively balances energy optimization with real-time capability. Experimental hardware-in-the-loop (HIL) validation results demonstrate significant improvements in energy efficiency and a reduction in battery stress. The article contributes to the broader adoption of learning-driven EMS designs in EV applications, particularly for multimotor multisource configurations.