This paper introduces a model-agnostic approach designed to enhance uncertainty estimation in the predictive modeling of soil properties, a crucial factor for advancing pedometrics and the practice of digital soil mapping. For addressing the typical challenge of data scarcity in soil studies, we present an improved technique for uncertainty estimation. This method is based on the transformation of regression tasks into classification problems, which not only allows for the production of reliable uncertainty estimates but also enables the application of established machine learning algorithms with competitive performance that have not yet been utilized in pedometrics. Empirical results from datasets collected from two German agricultural fields showcase the practical application of the proposed methodology. Our results and findings suggest that the proposed approach has the potential to provide better uncertainty estimation than the models commonly used in pedometrics.

The proposed solution enables safe and feasible trajectories while accommodating heterogeneous time-bound constraints and vehicle physical limits. An optimization problem is formulated to meet mission objectives and temporal requirements encoded as Signal Temporal Logic (STL) specifications. Additionally, an event-triggered replanner is introduced to address unforeseen events and compensate for lost time. Furthermore, a generalized robustness scoring method is employed to reflect user preferences and mitigate task conflicts. The effectiveness of the proposed approach is demonstrated through MATLAB and Gazebo simulations, as well as field multi-robot experiments in a mock-up scenario.

Stochastic optimization problems in large-scale multi-stakeholder networked systems (e.g., power grids and supply chains) rely on data-driven scenarios to encapsulate complex spatiotemporal interdependencies. However, centralized aggregation of stakeholder data is challenging due to the existence of data silos resulting from computational and logistical bottlenecks. In this paper, we present SplitVAEs, a decentralized scenario generation framework that leverages variational autoencoders to generate high-quality scenarios without moving stakeholder data. With the help of experiments on distributed memory systems, we demonstrate the broad applicability of SplitVAEs in a variety of domain areas that are dominated by a large number of stakeholders. Our experiments indicate that SplitVAEs can learn spatial and temporal interdependencies in large-scale networks to generate scenarios that match the joint historical distribution of stakeholder data in a decentralized manner. Our experiments show that SplitVAEs deliver robust performance compared to centralized, state-of-the-art benchmark methods while significantly reducing data transmission costs, leading to a scalable, privacy-enhancing alternative to scenario generation.

This paper investigates payload grasping from a moving platform using a hook-equipped aerial manipulator. First, a computationally efficient trajectory optimization based on complementarity constraints is proposed to determine the optimal grasping time. To enable application in complex, dynamically changing environments, the future motion of the payload is predicted using physics simulator-based models. The success of payload grasping under model uncertainties and external disturbances is formally verified through a robustness analysis method based on integral quadratic constraints. The proposed algorithms are evaluated in a high-fidelity physical simulator, and in real flight experiments using a custom-designed aerial manipulator platform.

As Chile's electric power sector advances toward a future powered by renewable energy, accurate forecasting of renewable generation is essential for managing grid operations. The integration of renewable energy sources is particularly challenging due to the operational difficulties of managing their power generation, which is highly variable compared to fossil fuel sources, delaying the availability of clean energy. To mitigate this, we quantify the impact of increasing intermittent generation from wind and solar on thermal power plants in Chile and introduce a hybrid wind speed forecasting methodology which combines two custom ML models for Chile. The first model is based on TiDE, an MLP-based ML model for short-term forecasts, and the second is based on a graph neural network, GraphCast, for medium-term forecasts up to 10 days. Our hybrid approach outperforms the most accurate operational deterministic systems by 4-21% for short-term forecasts and 5-23% for medium-term forecasts and can directly lower the impact of wind generation on thermal ramping, curtailment, and system-level emissions in Chile.

In July, Elon Musk made a bold prediction: that his artificial intelligence startup xAI would release "the most powerful AI in the world," a model called Grok 3, by this December. The bulk of that AI's training, Musk said, would happen at a "massive new training center" in Memphis, which he bragged had been built in 19 days. But many residents of Memphis were taken by surprise, including city council members who said they were given no input about the project or its potential impacts on the city. And in the months since, an outcry has grown among community members and environmental groups, who warn of the plant's potential negative impact on air quality, water access, and grid stability, especially for nearby neighborhoods that have suffered from industrial pollution for decades. These activists also contend that the company is illegally operating gas turbines.

Amid an outbreak of recent wildfires in California, Google announced a commitment to spend 13 million to improve satellite imaging to help track and detect wildfires, starting as early as next year. FireSat, a constellation of more than 50 satellites, will be able to detect wildfires as small as the size of a classroom, about 16 by 16 feet, and the first satellite will launch in early 2025, the media giant announced Monday. Firefighting authorities currently rely on satellite imagery that detects wildfires but only when they reach about the size of a football field, or more than an acre. "We realized that if we can pair satellites with machine learning and artificial intelligence, it was the perfect platform to generate real-time operational intelligence on fires," Christopher Van Arsdale, who leads the Google Research Climate & Energy group and is chairman of the Earth Fire Alliance, said in a video announcement. The initiative is being led by the Earth Fire Alliance, a nonprofit that was launched in May to create FireSat and develop wildfire datasets, with funding from Google and the Gordon and Betty Moore Foundation.

A graph neural network (GNN) approach is introduced in this work which enables mesh-based three-dimensional super-resolution of fluid flows. In this framework, the GNN is designed to operate not on the full mesh-based field at once, but on localized meshes of elements (or cells) directly. To facilitate mesh-based GNN representations in a manner similar to spectral (or finite) element discretizations, a baseline GNN layer (termed a message passing layer, which updates local node properties) is modified to account for synchronization of coincident graph nodes, rendering compatibility with commonly used element-based mesh connectivities. The architecture is multiscale in nature, and is comprised of a combination of coarse-scale and fine-scale message passing layer sequences (termed processors) separated by a graph unpooling layer. The coarse-scale processor embeds a query element (alongside a set number of neighboring coarse elements) into a single latent graph representation using coarse-scale synchronized message passing over the element neighborhood, and the fine-scale processor leverages additional message passing operations on this latent graph to correct for interpolation errors. Demonstration studies are performed using hexahedral mesh-based data from Taylor-Green Vortex flow simulations at Reynolds numbers of 1600 and 3200. Through analysis of both global and local errors, the results ultimately show how the GNN is able to produce accurate super-resolved fields compared to targets in both coarse-scale and multiscale model configurations.

A core strength of Model Predictive Control (MPC) for quadrupedal locomotion has been its ability to enforce constraints and provide interpretability of the sequence of commands over the horizon. However, despite being able to plan, MPC struggles to scale with task complexity, often failing to achieve robust behavior on rapidly changing surfaces. On the other hand, model-free Reinforcement Learning (RL) methods have outperformed MPC on multiple terrains, showing emergent motions but inherently lack any ability to handle constraints or perform planning. To address these limitations, we propose a framework that integrates proprioceptive planning with RL, allowing for agile and safe locomotion behaviors through the horizon. Inspired by MPC, we incorporate an internal model that includes a velocity estimator and a Dreamer module. During training, the framework learns an expert policy and an internal model that are co-dependent, facilitating exploration for improved locomotion behaviors. During deployment, the Dreamer module solves an infinite-horizon MPC problem, adapting actions and velocity commands to respect the constraints. We validate the robustness of our training framework through ablation studies on internal model components and demonstrate improved robustness to training noise. Finally, we evaluate our approach across multi-terrain scenarios in both simulation and hardware.

This paper investigates the problem of multi-UAV uniform sweep coverage, where a homogeneous swarm of UAVs must collectively and evenly visit every portion of an unknown environment for a sampling task without having access to their own location and orientation. Random walk-based exploration strategies are practical for such a coverage scenario as they do not rely on localization and are easily implementable in robot swarms. We demonstrate that the Mergeable Nervous System (MNS) framework, which enables a robot swarm to self-organize into a hierarchical ad-hoc communication network using local communication, is a promising control approach for random exploration in unknown environments by UAV swarms. To this end, we propose an MNS-based random walk approach where UAVs self-organize into a line formation using the MNS framework and then follow a random walk strategy to cover the environment while maintaining the formation. Through simulations, we test the efficiency of our approach against several decentralized random walk-based strategies as benchmarks. Our results show that the MNS-based random walk outperforms the benchmarks in terms of the time required to achieve full coverage and the coverage uniformity at that time, assessed across both the entire environment and within local regions.