The adoption of Autonomous Mobile Robots (AMRs) for internal logistics is accelerating, with most solutions emphasizing decentralized, onboard intelligence. While AMRs in indoor environments like factories can be supported by infrastructure, involving external sensors and computational resources, such systems remain underexplored in the literature. This paper presents a comprehensive overview of infrastructure-based AMR systems, outlining key opportunities and challenges. To support this, we introduce a reference architecture combining infrastructure-based sensing, on-premise cloud computing, and onboard autonomy. Based on the architecture, we review core technologies for localization, perception, and planning. We demonstrate the approach in a real-world deployment in a heavy-vehicle manufacturing environment and summarize findings from a user experience (UX) evaluation. Our aim is to provide a holistic foundation for future development of scalable, robust, and human-compatible AMR systems in complex industrial environments.
The steady, mirror-symmetric, two-dimensional secondary flow in a loosely coiled curved pipe - the classical Dean problem solved numerically by Collins and Dennis (1975) for Dean numbers D up to 5000 - is reanalysed through the objective transport-barrier framework of Haller et al. (2020). The paper has two aims, kept separate throughout. The methodological aim is to establish this flow as the simplest non-trivial setting in which the active-barrier machinery becomes exactly solvable: in 2D steady incompressible flow the active momentum and active vorticity barriers are exact level curves of the streamwise vorticity Omega and of Delta Omega, so the numerically computed active finite-time Lyapunov exponent (aFTLE) and active polar rotation angle (aPRA) can be calibrated against an analytically known answer across a parameter sweep. The physical aim is to ask what objective barrier structure the curved-pipe base flow possesses and how it differs from the straight pipe. The straight-pipe reference state is degenerate: both barrier Hamiltonians vanish identically, and the full threedimensional momentum barrier field of Hagen-Poiseuille flow is spatially uniform, so every radius is equally a barrier, no radius can be ranked above another, and the family carries no Reynolds-number dependence. Curvature converts this into a stratified ladder. Sweeping D is an element of [96, 5000] we find four characteristic radial loci that satisfy a strict nested ordering at every D >= 500, with inter-family gaps that widen monotonically with D; the passive recirculation envelope shrinks from 74% to 45% of the cross-section while its cells migrate outward and across the upper pole, falsifying the initial hypothesis that shear at the recirculation-cell boundaries supplies the dominant barrier. The active barriers instead sharpen and specialise radially, and their scaling is predicted rather than merely fitted. The prediction follows from the way the integration window of the diagnostic is set: the window is chosen so that a representative fluid element travels a prescribed distance along the barrier field, a choice we call travel-target normalisation, and it makes the aFTLE proportional to the magnitude of the barrier field. The whole-domain momentum exponent 3 approximate to 1 then follows exactly from the classical Dean core and boundary-layer scales, O(D1/3) and O(D-1/3) - the same pair of scales that produces the Collins-Dennis asymptotic flux law Qc/Qs similar or equal to 8.12 D-1/3-16.7 D-2/3, so the objective diagnostic and the classical flux reduction are two readings of one asymptotic structure. The active vorticity field shows a four-lobe topology whose inner-versus-outer area crossover at D approximate to 2000 is the intersection of two smooth power laws of opposite sign, marking the Dean number at which the viscous vorticity source passes from wall-layer-dominated to core-dominated; it has no signature in the streamwise-velocity statistics. Whether this curvature-generated barrier structure contributes causally to the delayed transition remains untested; the present steady analysis identifies the candidate structure and, in the straight-pipe contrast, establishes that it is present wherever, and only where, the curvature that produces the delay is present.
Sustainability transitions involve genuine uncertainty, competing values, and persistent stakeholder disagreement about problems, solutions, goals and policy interventions. Understanding these contested dynamics requires methodological approaches that can systematically capture subjectivity while maintaining analytical rigor. In this perspective, we argue that Q methodology, a mixed-methods approach that combines qualitative depth with quantitative rigor, is an underutilized yet highly valuable tool for transitions research. Drawing on our experience from Q studies on industrial decarbonization and electrification in Sweden and Norway, as well as a broader reading of transitions-related Q research, we reflect on the method's strengths, limitations, and untapped potential. We highlight how Q methodology enables the identification of distinct stakeholder narratives, reveals areas of alignment and disjointment, and brings into view discursive and normative dimensions of transitions that research drawing on frameworks such as Technological Innovation Systems and the Multi-Level Perspective has often left underexamined. We also discuss important limitations, including challenges related to statement design, participant selection, and the interpretation of factors, and share practical lessons learned from applying the method across different empirical contexts. Beyond research, we suggest that Q could complement established participatory approaches in transition governance, for instance by helping to select participants who span the range of perspectives and by feeding identified narratives back into stakeholder dialogue. We conclude by outlining a research agenda to advance Q methodology in transition studies, including longitudinal designs, cross-country comparisons, mixed-methods integration and methodological refinements.
Electrification offers the clearest pathway for emission reduction in the transport sector, causing rapidly growing demand for lithium-ion batteries. Developing battery packs that improve performance while reducing environmental impacts presents a key challenge. This study investigates how battery pack design and cell chemistry selection affect pack specific energy and, in consequence, cradle-to-gate environmental impacts. Primary teardown-based life cycle inventory data are provided for eight electric vehicle battery packs utilising NMC622, NMC811, NCA and LFP cell chemistries, covering around 50 materials and material groups. Cradle-to-gate environmental impacts are analysed with focus on climate change, highlighting the links between pack design and cell chemistry. Cell-to-pack ratios ranged from 56% to 82%, with baseline cradle-to-gate climate change impacts for the rest-of-pack ranging from 21 to 44 kg CO2-eq/kWh, contributing 19–33% of total pack impact. These findings underscore the importance of pack-level considerations in establishing energy dense and environmentally sustainable batteries for electric vehicles.
Semi-natural grasslands have great cultural and biodiversity value, but are threatened by land use change and abandonment. Grazing is essential to successful management of semi-natural grasslands, but its implementation for targeted nature conservation requires temporally and spatially varied management intensity. Virtual fence technology, where grazing animals wear GPS-tracked collars and are controlled by invisible digital boundaries, represents a potentially valuable tool for nature conservation management of these environments. This case study explores how virtual fences can be used for ecologically-tailored management of semi-natural grasslands. By combining GPS-based movement data from cattle with drone-based NDVI analyses, and field inventories of flowering plants and pollinators, the relationships between grazing patterns, vegetation, and conservation goals were analysed. The results show that virtual fences can regulate grazing pressure and create variation in vegetation structure and flower abundance in the short term. We found no clear link between pollinators and grazed and temporarily ungrazed areas with lower and higher flower abundance respectively, possibly due to high variation in pollinator numbers between consecutive survey days, and the mobility of flying insects across the virtual fence line. The combination of GPS data on livestock residence patterns and drone technology on changes in vegetation over this time period provides important information about interactions between the animals and the vegetation. The study demonstrates that virtual fencing has potential value for precision-oriented conservation and adaptive grazing planning.