What is happening worldwide in digital twins, mathematical modelling and the management of large-scale systems: research, press releases, product news. From the US, China, ASEAN and Africa — translated and summarised.
300 items · 38 more in the archive
Growing automation makes collaborative robots work in more variable environments, increasing the need for adaptation. We propose a human-in-the-loop online training framework combining a digital twin (DT), reinforcement learning (RL), and human demonstrations.
Realistic household simulation must capture not only diverse environments but also the lived-in object arrangements and spatial constraints that shape robot motion and interaction. Existing resources often trade off scale, real-world correspondence, and interaction readiness, leaving a gap in…
Industrial environments are increasingly characterized by the tight interaction among physical processes, communication infrastructures, and intelligent applications. In this context, Digital Twins (DTs) have emerged as a key technology for system analysis and optimization.
arXiv:2610.09888v2 Announce Type: new Abstract: We develop a hybrid pore-network model (PNM) for discrete microbubble transport coupled to connected two-phase flow in porous media. We apply the model to simulate flow in the porous transport layer of a rotating proton exchange membrane (PEM)…
arXiv:2308.00505v4 Announce Type: replace-cross Abstract: Agent Based Models (ABMs) often deal with systems where there is a lack of quantitative data or where quantitative data alone may be insufficient to fully capture the complexities of real-world systems.
arXiv:2610.09789v1 Announce Type: new Abstract: Cooperative behavior can improve environmental quality, whereas selfish defection can degrade shared resources. In a recent agent-based study of the Aksum civilization (Vasellini et al., 2025), a population of agents moving over a reconstructed…
A ruleless digital twin approach is proposed where decisions are generated automatically from purely declarative user specifications instead of hand-written rules. This reduces the complexity of building and maintaining twins under dynamically changing conditions such as weather or dynamic energy pricing.
The PCDT framework closes the loop from network observation to predictive cognition and proactive resource control. Unlike reactive twins that act after performance risk appears, it aims to pre-empt degradation.
A calibration and control method is developed for a multiscale bioprocess digital twin using sparse observations, quasi-likelihood estimation and adjoint sensitivity analysis. Parameter uncertainty then feeds into policy optimization and adaptive experiment design.
arXiv:2608.04265v2 Announce Type: replace Abstract: LLM-agent evaluations commonly measure task success or agreement with a declared plan. In strategic cyber-physical systems, an architecture must also remain appropriate after autonomous participants respond and physics constrains outcomes.
arXiv:2610.06976v1 Announce Type: new Abstract: Safety-critical Model Predictive Control (MPC) formulations based on Control Barrier Functions (CBFs) often require multiple tuning parameters and may become conservative or difficult to keep feasible, particularly for high-relative-degree…
arXiv:2610.07232v1 Announce Type: cross Abstract: Accurate short-term load forecasting (STLF) is essential for the reliable and efficient operation of modern power systems. While time series foundation models (TSFMs) have recently demonstrated remarkable performance across a wide range of…
AVEVA Day Nigeria 2026 Explores Digital Twins, Industrial Intelligence and Connected Ecosystems Tech Build Africa
A gateway admits an update to a twin's persistent state only after provenance, contradiction and namespace checks pass. Evaluation covers three attack families over 20 seeded splits of a semi-synthetic stream derived from the open eICU database, with roughly 29,000 candidate updates per seed.
The work treats population-scale adverse-event surveillance as a distinct belief layer of a surgical digital twin, evaluated with a federated Bayesian protocol under formal privacy guarantees: sites exchange only differentially private natural-parameter updates. Benchmarking uses the full FDA MAUDE cohort for thrombus-retrieval catheters — 8,617 reports, 6,491 classified into five complication classes across eight manufacturer sites.
The study examines how far a DT-R2S2R pipeline can substitute for real target-region data: recorded driving clips are reconstructed in a georeferenced digital twin and a diffusion model is conditioned on geometrically aligned simulator renderings, yielding a digital-twin-grounded Sim2Real model.
ONDT is a modular, virtualized, cloud-native experimentation framework for O-RAN introduced within the SUNRISE-6G project. It automates rApp/xApp onboarding, RAN scenario configuration, KPM collection and closed-loop control validation, letting AI/ML control algorithms be tested without the risks and costs of live networks.
arXiv:2610.04222v1 Announce Type: new Abstract: Life support equipment coupled to the human lungs forms a complex, nonlinear dynamic system. Current respiratory mask test standards rely on sinusoidal breathing machines and surrogate trachea geometries that substantially reduce physiologically…
A Model Predictive Control approach approximates infinite-horizon feedback Nash equilibrium trajectories by repeatedly solving finite-horizon open-loop games. It targets multi-agent control settings, including physical human-machine interaction, where agent responses and state/input constraints must be respected.
Using Willems' fundamental lemma, a single offline input-output trajectory serves as an implicit model of a linear time-invariant system, with online operation needing only past measurements. Bounded measurement noise is handled in both stages, and recursive feasibility plus practical exponential stability are established.
A network testbed demonstration shows a Digital Twin-aided multi-agent LLM system for autonomous class-aware optical network management. The system detects soft failures and autonomously adapts modulation format per service class requirements, ensuring continuous, marginless channel quality with minimal human intervention.
APOD is a language-model agent that iteratively reasons over biological knowledge and fit diagnostics in an open-ended search space to discover a population digital twin, a shared ODE system with between-subject variability. On synthetic pharmacokinetic and tumor-dynamics benchmarks it recovered ground-truth structures in 94-100% of runs, 12-fold faster in median than a library-based search.
A demo on a Unitree G1 robot and NVIDIA Sionna uses a vision-language model to classify materials and set conductivity priors, while gradient descent refines them from received signal strength measurements. Online waypoint planning picks the next measurement location, cutting initialization sensitivity and travel costs.
The paper introduces a trading strategy for battery storage on the continuous intraday market in which forecast-driven round-trip trading takes priority and the battery provides a second route to close positions that would otherwise be closed at a distressed price. The agent opens positions using price quantile forecasts and closes them closer to delivery, replacing the myopic rolling intrinsic approach.
