Search

Reliable deep learning models require not only accurate predictions but also well-calibrated confidence estimates to ensure dependable uncertainty estimation. This is crucial in safety-critical applications like autonomous driving, which depend on rapid and precise semantic segmentation of LiDAR point clouds for real-time 3D scene understanding. In this work, we introduce a sampling-free approach for estimating well-calibrated confidence values for classification tasks, achieving alignment with true classification accuracy and significantly reducing inference time compared to sampling-based methods. Our evaluation using the Adaptive Calibration Error (ACE) metric for LiDAR semantic segmentation shows that our approach maintains well-calibrated confidence values while achieving increased processing speed compared to a sampling baseline. Additionally, reliability diagrams reveal that our method produces underconfidence rather than overconfident predictions, an advantage for safety-critical applications. Our sampling-free approach offers well-calibrated and time-efficient predictions for LiDAR scene semantic segmentation.

The objective assessment of gait kinematics is crucial in evaluating human movement, informing clinical decisions, and advancing rehabilitation and assistive technologies. Assessing gait symmetry, in particular, holds significant importance in clinical rehabilitation, as it reflects the intricate coordination between nerves and muscles during human walking. In this research, a dataset has been compiled to improve the understanding of gait kinematics. The dataset encompasses motion capture data of the walking patterns of eleven healthy participants who were tasked with completing various activities on a circular path. These activities included normal walking, walking with a weighted dominant hand, walking with a braced dominant leg, and walking with both weight and brace. The walking tasks involving weight and brace were designed to emulate the asymmetry associated with common health conditions, shedding light on irregularities in individuals' walking patterns and reflecting the coordination between nerves and muscles. All tasks were performed at regular and fast speeds, offering valuable insights into upper and lower body kinematics. The dataset comprises raw sensor data, providing information on joint dynamics, angular velocities, and orientation changes during walking, as well as analyzed data, including processed data, Euler angles, and joint kinematics spanning various body segments. This dataset will serve as a valuable resource for researchers, clinicians, and engineers, facilitating the analysis of gait patterns and extracting relevant indices on mobility and balance., Comment: All datasets generated during the current study are publicly available on figshare (doi:10.6084/m9.figshare.26090200.v1.)

By introducing ansatz and using a PDF at the $N^3LO$ approximation in this paper, we have studied the structure of the nucleons in generalized parton distributions (GPDs). For this purpose, the form factors and the electric radius of nucleons, as well as the quark's form factors, are calculated in this paper. These calculations give important new information about the nucleons and the distribution of partons within them. The results are analyzed, and to validate them, the abstract emphasizes the importance of comparing the results with existing research and experimental data. By varying the free appropriate parameters of the new ansatz, the study seeks to optimize the model, ensure its suitability for accurately describing the nucleon's internal structure, and provide valuable insights about the distribution of partons.

Current autonomous building research primarily focuses on energy efficiency and automation. While traditional artificial intelligence has advanced autonomous building research, it often relies on predefined rules and struggles to adapt to complex, evolving building operations. Moreover, the centralized organizational structures of facilities management hinder transparency in decision-making, limiting true building autonomy. Research on decentralized governance and adaptive building infrastructure, which could overcome these challenges, remains relatively unexplored. This paper addresses these limitations by introducing a novel Decentralized Autonomous Building Cyber-Physical System framework that integrates Decentralized Autonomous Organizations, Large Language Models, and digital twins to create a smart, self-managed, operational, and financially autonomous building infrastructure. This study develops a full-stack decentralized application to facilitate decentralized governance of building infrastructure. An LLM-based artificial intelligence assistant is developed to provide intuitive human-building interaction for blockchain and building operation management-related tasks and enable autonomous building operation. Six real-world scenarios were tested to evaluate the autonomous building system's workability, including building revenue and expense management, AI-assisted facility control, and autonomous adjustment of building systems. Results indicate that the prototype successfully executes these operations, confirming the framework's suitability for developing building infrastructure with decentralized governance and autonomous operation., Comment: 40 pages, 22 figures

Safe navigation in new environments requires autonomous vehicles and robots to accurately interpret their surroundings, relying on LiDAR scene segmentation, out-of-distribution (OOD) obstacle detection, and uncertainty computation. We propose a method to distinguish in-distribution (ID) from OOD samples and quantify both epistemic and aleatoric uncertainties using the feature space of a single deterministic model. After training a semantic segmentation network, a Gaussian Mixture Model (GMM) is fitted to its feature space. OOD samples are detected by checking if their squared Mahalanobis distances to each Gaussian component conform to a chi-squared distribution, eliminating the need for an additional OOD training set. Given that the estimated mean and covariance matrix of a multivariate Gaussian distribution follow Gaussian and Inverse-Wishart distributions, multiple GMMs are generated by sampling from these distributions to assess epistemic uncertainty through classification variability. Aleatoric uncertainty is derived from the entropy of responsibility values within Gaussian components. Comparing our method with deep ensembles and logit-sampling for uncertainty computation demonstrates its superior performance in real-world applications for quantifying epistemic and aleatoric uncertainty, as well as detecting OOD samples. While deep ensembles miss some highly uncertain samples, our method successfully detects them and assigns high epistemic uncertainty., Comment: Accepted for publication in the Proceedings of the European Conference on Computer Vision (ECCV) 2024

The design of solid state batteries with lithium anodes is attracting attention for the prospect of high capacity and improved safety over liquid electrolyte systems. The nature of transport with lithium as the current carrier has as a consequence the accretion or stripping away of the anode with every charge-discharge cycle. While this poses challenges from the growth of protrusions (dendrites) to loss of contact, there lurks an opportunity: Morphogenesis at the anode-electrolyte interface layer can be studied, and may ultimately be controlled as a factor in solid state battery design. The accessible interface morphologies, the dynamic paths to them, and mechanisms to control them expand considerably if lithium alloys are introduced in the anode. The thermodynamics and kinetics of lithium intermetallics present principled approaches for morphogenic interface design. In this communication we adopt a computational approach to such an exploration. With phase field models that are parameterized by a combination of first principles atomistic calculations and experiments, we present phenomenological studies of two lithium intermetallics: Li-Mg and Li-Zn. An array of parametric investigations follows on the influence of kinetics, charge-discharge rate, cycling, transport mechanisms and grain structure. The emphasis across these computations is on the dynamic morphogenesis of the intermetallic interface. Specifically, the plating, segregation and smooth distribution of Li, Mg and Zn, the growth and disappearance of voids, evolution of solid electrolyte-anode contact area, and grain boundary structure are investigated. The computational platform is a framework for future studies of morphogenic electrolyte-anode interfaces with more extensive inputs from first principles atomistics and experiments.

Automating activities through robots in unstructured environments, such as construction sites, has been a long-standing desire. However, the high degree of unpredictable events in these settings has resulted in far less adoption compared to more structured settings, such as manufacturing, where robots can be hard-coded or trained on narrowly defined datasets. Recently, pretrained foundation models, such as Large Language Models (LLMs), have demonstrated superior generalization capabilities by providing zero-shot solutions for problems do not present in the training data, proposing them as a potential solution for introducing robots to unstructured environments. To this end, this study investigates potential opportunities and challenges of pretrained foundation models from a multi-dimensional perspective. The study systematically reviews application of foundation models in two field of robotic and unstructured environment and then synthesized them with deliberative acting theory. Findings showed that linguistic capabilities of LLMs have been utilized more than other features for improving perception in human-robot interactions. On the other hand, findings showed that the use of LLMs demonstrated more applications in project management and safety in construction, and natural hazard detection in disaster management. Synthesizing these findings, we located the current state-of-the-art in this field on a five-level scale of automation, placing them at conditional automation. This assessment was then used to envision future scenarios, challenges, and solutions toward autonomous safe unstructured environments. Our study can be seen as a benchmark to track our progress toward that future., Comment: arXiv admin note: text overlap with arXiv:2312.07843, arXiv:2402.05741 by other authors

Cognitive load assessment is crucial for understanding human performance in various domains. This study investigates the impact of different task conditions and time constraints on cognitive load using multiple measures, including subjective evaluations, performance metrics, and physiological eye-tracking data. Fifteen participants completed a series of primary and secondary tasks with different time limits. The NASA-TLX questionnaire, reaction time, inverse efficiency score, and eye-related features (blink, saccade, and fixation frequency) were utilized to assess cognitive load. The study results show significant differences in the level of cognitive load required for different tasks and when under time constraints. The study also found that there was a positive correlation (r = 0.331, p = 0.014) between how often participants blinked their eyes and the level of cognitive load required but a negative correlation (r = -0.290, p = 0.032) between how often participants made quick eye movements (saccades) and the level of cognitive load required. Additionally, the analysis revealed a significant negative correlation (r = -0.347, p = 0.009) and (r = -0.370, p = 0.005) between fixation and saccade frequencies under time constraints., Comment: Accepted for publication at the 30th National and 8th International Iranian Conference on Biomedical Engineering (ICBME 2023), Tehran, Iran, Nov. 2023

Advancements in neurosurgical robotics have improved medical procedures, particularly deep brain stimulation, where robots combine human and machine intelligence to precisely implant electrodes in the brain. While effective, this procedure carries risks and side effects. Noninvasive deep brain stimulation (NIDBS) offers promise by making brain stimulation safer, more affordable, and accessible. However, NIDBS lacks guidelines for electrode placement. This study explores adapting robotic principles to enhance the accuracy of NIDBS targeting and provides preliminary guidelines for transcranial electrode placement. Safety is also emphasized, ensuring a balance between therapeutic effectiveness and patient safety by maintaining electric fields within safe limits., Comment: Preprint version of the work accepted for publication at the 11th RSI International Conference on Robotics and Mechatronics (ICRoM), Tehran, Iran, Dec. 2023

This article explores natural interaction modalities for human-cyber-physical systems (CPS) interaction in construction. CPS has been applied in construction for many purposes with the promise of improving the safety and productivity of construction operations. However, there is little research on human-CPS interaction in construction. This study proposes two methodologies for human-CPS interactions for construction progress monitoring - a) hand gesture interaction using transfer learning, and b) voice command interaction using natural language processing. User studies with thirty-two users validated the generalizability of the proposed methodologies. The proposed hand gesture recognition method achieved higher accuracy (99.69% vs 87.72%) and speed (36.05ms vs 578.91ms) than the proposed voice command recognition method, though users performed the progress monitoring task more correctly with voice commands than hand gestures (88% vs 66.1%). The main contribution of the study is the development of an ML pipeline and computational framework to recognize hand gestures and voice commands without the need for a large training dataset for human-CPS interaction., Comment: 15 pages

In this study, we show that the background of primordial gravitational waves (PGWs) predicted by the inflationary scenario induce a decrease of the spatial coherence length of an electromagnetic field propagating over cosmological distances, leading the van Citter-Zernike correlations to eventually be unobservable, an affect called as blurring. Since this spatial correlation is actually observed in VLBI measurements of distant quasars, it imposes a constraint on the level of the primordial gravitational waves background. In this paper, we precisely evaluate this blurring effect by considering the primordial gravitational waves background to be in the so-called two-mode squeezed state, which is the standard quantum state predicted by the simplest scenario of inflation. We then exploit a sample of compact radio quasars located at redshift range $0.46\leq z \leq 2.73$ observed by very long baseline interferometry (VLBI) means. These quasars constitute a standard ruler and the angular size-distance $\theta-z$ relation is determined thanks to the measurement of their spatial coherence. These spatial coherence observations set an upper limit on the squeezing parameters of the PGWs background, which turns into an upper limit for the tensor-to-scalar ratio $\text{r}_{k_0}$. One finds $\text{r}_{k_0} < 10^{-7}$, five orders of magnitude better than present constraints on this parameter with the CMB. This result is nevertheless compatible with the possible detection of a cosmic background by the pulsar timing array measurement. Further issues and caveats that potentially affect the results are reviewed. In particular, the possible effect of quantum-to-classical transition of PGWs is discussed. Ultimately, we promote the idea of using high-precision VLBI measurement of angular size-redshift of distant sources as a new possible way of constraining the background of primordial tensor perturbations., Comment: 61 pages, 7 figures, submitted to JCAP

Most recently the formation of boron monoxide (BO) in the two-dimensional (2D) form has been confirmed experimentally (J. Am. Chem. Soc. 2023, 145, 14660). Motivated by the aforementioned finding, herein we theoretically explore the key physical properties of the single-layer and suspended BO. Density functional theory (DFT) results reveal that BO monolayer yields a large indirect band gap of 3.78 (2.18) eV on the basis of HSE06(PBE) functional. Ab-initio molecular dynamics results reveal the remarkable thermal stability of the BO monolayer at 1000 K. The thermal and mechanical properties at room temperature are furthermore investigated using a machine learning interatomic potential (MLIP). The developed MLIP-based model close to the ground state could very precisely reproduce the DFT predictions for the mechanical properties of the BO monolayer. The elastic modulus, tensile strength and lattice thermal conductivity of the BO monolayer at room temperature are predicted to be 107 GPa, 25 GPa and 5.6 W/mK, respectively. At the room temperature the BO monolayer is noticeably predicted to yield an ultrahigh negative thermal expansion coefficient, by almost 17 folds larger than that of the single-layer graphene. The presented results reveal the large indirect electronic band gap, decent thermal and dynamical stability, anomalously low elastic modulus to tensile strength ratio, ultrahigh negative thermal expansion coefficients and low lattice thermal conductivity of the BO monolayer.