arXiv subjects

Mihir Kulkarni

Publications and source records attributed to Mihir Kulkarni.

At least 37 records · Page 2Linked to original sources

arXiv2025

From Dark Matter Minihalos to Large-Scale Radiative Feedback: A Self-Consistent 3D Simulation of the First Stars and Galaxies using Neural Networks

A key obstacle to accurate models of the first stars and galaxies is the vast range of distance scales that must be considered. While star formation occurs on sub-parsec scales within dark matter (DM) minihalos, it is influenced by large-scale baryon-dark matter streaming velocities ($v_{\rm bc}$) and Lyman-Werner (LW) radiative feedback which vary significantly on scales of $\sim$100 Mpc. We present a novel approach to this issue in which we utilize artificial neural networks (NNs) to emulate the Population III (PopIII) and Population II (PopII) star formation histories of many small-scale cells given by a more complex semi-analytic framework based on DM halo merger trees. Within each simulation cell, the NN takes a set of input parameters that depend on the surrounding large-scale environment, such as the cosmic overdensity, $δ(\vec{x})$, and $v_{\rm bc}$ of the cell, then outputs the resulting star formation far more efficiently than is possible with the semi-analytic model. This rapid emulation allows us to self-consistently determine the LW background intensity on $\sim$100 Mpc scales, while simultaneously including the detailed merger histories (and corresponding star formation histories) of the low-mass minihalos that host the first stars. Comparing with the full semi-analytic framework utilizing DM halo merger trees, our NN emulators yield star formation histories with redshift-averaged errors of $\sim$7.3\% and $\sim$5.2\% for PopII and PopIII, respectively. When compared to a simpler sub-grid star formation prescription reliant on halo mass function integration, we find that the diversity of halo merger histories in our simulation leads to enhanced spatial fluctuations, an earlier transition from PopIII to PopII dominated star formation, and more scatter in star formation histories overall.

astro-ph.GA↗

arXiv2024

Neural Control Barrier Functions for Safe Navigation

Autonomous robot navigation can be particularly demanding, especially when the surrounding environment is not known and safety of the robot is crucial. This work relates to the synthesis of Control Barrier Functions (CBFs) through data for safe navigation in unknown environments. A novel methodology to jointly learn CBFs and corresponding safe controllers, in simulation, inspired by the State Dependent Riccati Equation (SDRE) is proposed. The CBF is used to obtain admissible commands from any nominal, possibly unsafe controller. An approach to apply the CBF inside a safety filter without the need for a consistent map or position estimate is developed. Subsequently, the resulting reactive safety filter is deployed on a multirotor platform integrating a LiDAR sensor both in simulation and real-world experiments.

cs.RO↗

arXiv2024

Maritime Vessel Tank Inspection using Aerial Robots: Experience from the field and dataset release

This paper presents field results and lessons learned from the deployment of aerial robots inside ship ballast tanks. Vessel tanks including ballast tanks and cargo holds present dark, dusty environments having simultaneously very narrow openings and wide open spaces that create several challenges for autonomous navigation and inspection operations. We present a system for vessel tank inspection using an aerial robot along with its autonomy modules. We show the results of autonomous exploration and visual inspection in 3 ships spanning across 7 distinct types of sections of the ballast tanks. Additionally, we comment on the lessons learned from the field and possible directions for future work. Finally, we release a dataset consisting of the data from these missions along with data collected with a handheld sensor stick.

cs.RO↗

arXiv2024

A Framework to Calibrate a Semi-analytic Model of the First Stars and Galaxies to the Renaissance Simulations

We present a method that calibrates a semi-analytic model to the Renaissance Simulations, a suite of cosmological hydrodynamical simulations with high-redshift galaxy formation. This approach combines the strengths of semi-analytic techniques and hydrodynamical simulations, enabling the extension to larger volumes and lower redshifts inaccessible to simulations due to computational expense. Using a sample of Renaissance star formation histories (SFHs) from an average density region of the Universe, we construct a four parameter prescription for metal-enriched star formation characterized by an initial bursty stage followed by a steady stage where stars are formed at constant efficiencies. Our model also includes a treatment of Population III star formation where a minimum halo mass and log-normal distribution of stellar mass are adopted to match the numerical simulations. Star formation is generally well reproduced for halos with masses $\lesssim$$10^{9} M_{\mathrm{\odot}}$. Between $11<z<25$ our model produces metal-enriched star formation rate densities (SFRDs) that typically agree with Renaissance within a factor of $\sim$2 for the average density region. Additionally, the total metal-enriched stellar mass only differs from Renaissance by about $10\%$ at $z \sim 11$. For regions that are either more overdense or rarefied not included in the calibration, we produce metal-enriched SFRDs that agree with Renaissance within a factor of $\sim$2 at high-$z$, but eventually differ by higher factors for later times. This is likely due to environmental dependencies not included in the model. Our star formation prescriptions can easily be adopted in other analytic or semi-analytic works to match our calibration to Renaissance.

astro-ph.GA↗

arXiv2024

Reinforcement Learning for Collision-free Flight Exploiting Deep Collision Encoding

This work contributes a novel deep navigation policy that enables collision-free flight of aerial robots based on a modular approach exploiting deep collision encoding and reinforcement learning. The proposed solution builds upon a deep collision encoder that is trained on both simulated and real depth images using supervised learning such that it compresses the high-dimensional depth data to a low-dimensional latent space encoding collision information while accounting for the robot size. This compressed encoding is combined with an estimate of the robot's odometry and the desired target location to train a deep reinforcement learning navigation policy that offers low-latency computation and robust sim2real performance. A set of simulation and experimental studies in diverse environments are conducted and demonstrate the efficiency of the emerged behavior and its resilience in real-life deployments.

cs.RO↗

arXiv2024

Aerial Field Robotics

Aerial field robotics research represents the domain of study that aims to equip unmanned aerial vehicles - and as it pertains to this chapter, specifically Micro Aerial Vehicles (MAVs)- with the ability to operate in real-life environments that present challenges to safe navigation. We present the key elements of autonomy for MAVs that are resilient to collisions and sensing degradation, while operating under constrained computational resources. We overview aspects of the state of the art, outline bottlenecks to resilient navigation autonomy, and overview the field-readiness of MAVs. We conclude with notable contributions and discuss considerations for future research that are essential for resilience in aerial robotics.

cs.RO↗

arXiv2023

A Global Semi-Analytic Model of the First Stars and Galaxies Including Dark Matter Halo Merger Histories

We present a new self-consistent semi-analytic model of the first stars and galaxies to explore the high-redshift ($z{>}15$) Population III (PopIII) and metal-enriched star formation histories. Our model includes the detailed merger history of dark matter halos generated with Monte Carlo merger trees. We calibrate the minimum halo mass for PopIII star formation from recent hydrodynamical cosmological simulations that simultaneously include the baryon-dark matter streaming velocity, Lyman-Werner (LW) feedback, and molecular hydrogen self-shielding. We find an overall increase in the resulting star formation rate density (SFRD) compared to calibrations based on previous simulations (e.g., the PopIII SFRD is over an order of magnitude higher at $z=35-15$). We evaluate the effect of the halo-to-halo scatter in this critical mass and find that it increases the PopIII stellar mass density by a factor of ${\sim}1.5$ at $z{>}15$. Additionally, we assess the impact of various semi-analytic/analytic prescriptions for halo assembly and star formation previously adopted in the literature. For example, we find that models assuming smooth halo growth computed via abundance matching predict SFRDs similar to the merger tree model for our fiducial model parameters, but that they may underestimate the PopIII SFRD in cases of strong LW feedback. Finally, we simulate sub-volumes of the Universe with our model both to quantify the reduction in total star formation in numerical simulations due to a lack of density fluctuations on spatial scales larger than the simulation box, and to determine spatial fluctuations in SFRD due to the diversity in halo abundances and merger histories.

astro-ph.GA↗

arXiv2023

Autonomous Exploration and General Visual Inspection of Ship Ballast Water Tanks using Aerial Robots

This paper presents a solution for the autonomous exploration and inspection of Ballast Water Tanks (BWTs) of marine vessels using aerial robots. Ballast tank compartments are critical for a vessel's safety and correspond to confined environments often connected through particularly narrow manholes. The method enables their volumetric exploration combined with visual inspection subject to constraints regarding the viewing distance from a surface. We present evaluation studies in simulation, in a mission consisting of 18 BWT compartments, and in 3 field experiments inside real vessels. The data from one of the experiments is also post-processed to generate semantically-segmented meshes of inspection-important geometries. Geometric models can be associated with onboard camera images for detailed and intuitive analysis.

cs.RO↗

arXiv2023

Task-driven Compression for Collision Encoding based on Depth Images

This paper contributes a novel learning-based method for aggressive task-driven compression of depth images and their encoding as images tailored to collision prediction for robotic systems. A novel 3D image processing methodology is proposed that accounts for the robot's size in order to appropriately "inflate" the obstacles represented in the depth image and thus obtain the distance that can be traversed by the robot in a collision-free manner along any given ray within the camera frustum. Such depth-and-collision image pairs are used to train a neural network that follows the architecture of Variational Autoencoders to compress-and-transform the information in the original depth image to derive a latent representation that encodes the collision information for the given depth image. We compare our proposed task-driven encoding method with classical task-agnostic methods and demonstrate superior performance for the task of collision image prediction from extremely low-dimensional latent spaces. A set of comparative studies show that the proposed approach is capable of encoding depth image-and-collision image tuples from complex scenes with thin obstacles at long distances better than the classical methods at compression ratios as high as 4050:1.

cs.CV↗

arXiv2023

Semantically-enhanced Deep Collision Prediction for Autonomous Navigation using Aerial Robots

This paper contributes a novel and modularized learning-based method for aerial robots navigating cluttered environments containing hard-to-perceive thin obstacles without assuming access to a map or the full pose estimation of the robot. The proposed solution builds upon a semantically-enhanced Variational Autoencoder that is trained with both real-world and simulated depth images to compress the input data, while preserving semantically-labeled thin obstacles and handling invalid pixels in the depth sensor's output. This compressed representation, in addition to the robot's partial state involving its linear/angular velocities and its attitude are then utilized to train an uncertainty-aware 3D Collision Prediction Network in simulation to predict collision scores for candidate action sequences in a predefined motion primitives library. A set of simulation and experimental studies in cluttered environments with various sizes and types of obstacles, including multiple hard-to-perceive thin objects, were conducted to evaluate the performance of the proposed method and compare against an end-to-end trained baseline. The results demonstrate the benefits of the proposed semantically-enhanced deep collision prediction for learning-based autonomous navigation.

cs.RO↗

arXiv2023

Aerial Gym -- Isaac Gym Simulator for Aerial Robots

Developing learning-based methods for navigation of aerial robots is an intensive data-driven process that requires highly parallelized simulation. The full utilization of such simulators is hindered by the lack of parallelized high-level control methods that imitate the real-world robot interface. Responding to this need, we develop the Aerial Gym simulator that can simulate millions of multirotor vehicles parallelly with nonlinear geometric controllers for the Special Euclidean Group SE(3) for attitude, velocity and position tracking. We also develop functionalities for managing a large number of obstacles in the environment, enabling rapid randomization for learning of navigation tasks. In addition, we also provide sample environments having robots with simulated cameras capable of capturing RGB, depth, segmentation and optical flow data in obstacle-rich environments. This simulator is a step towards developing a - currently missing - highly parallelized aerial robot simulation with geometric controllers at a large scale, while also providing a customizable obstacle randomization functionality for navigation tasks. We provide training scripts with compatible reinforcement learning frameworks to navigate the robot to a goal setpoint based on attitude and velocity command interfaces. Finally, we open source the simulator and aim to develop it further to speed up rendering using alternate kernel-based frameworks in order to parallelize ray-casting for depth images thus supporting a larger number of robots.

cs.RO↗

arXiv2023

A universal equation to predict $Ω_{\rm m}$ from halo and galaxy catalogues

We discover analytic equations that can infer the value of $Ω_{\rm m}$ from the positions and velocity moduli of halo and galaxy catalogues. The equations are derived by combining a tailored graph neural network (GNN) architecture with symbolic regression. We first train the GNN on dark matter halos from Gadget N-body simulations to perform field-level likelihood-free inference, and show that our model can infer $Ω_{\rm m}$ with $\sim6\%$ accuracy from halo catalogues of thousands of N-body simulations run with six different codes: Abacus, CUBEP$^3$M, Gadget, Enzo, PKDGrav3, and Ramses. By applying symbolic regression to the different parts comprising the GNN, we derive equations that can predict $Ω_{\rm m}$ from halo catalogues of simulations run with all of the above codes with accuracies similar to those of the GNN. We show that by tuning a single free parameter, our equations can also infer the value of $Ω_{\rm m}$ from galaxy catalogues of thousands of state-of-the-art hydrodynamic simulations of the CAMELS project, each with a different astrophysics model, run with five distinct codes that employ different subgrid physics: IllustrisTNG, SIMBA, Astrid, Magneticum, SWIFT-EAGLE. Furthermore, the equations also perform well when tested on galaxy catalogues from simulations covering a vast region in parameter space that samples variations in 5 cosmological and 23 astrophysical parameters. We speculate that the equations may reflect the existence of a fundamental physics relation between the phase-space distribution of generic tracers and $Ω_{\rm m}$, one that is not affected by galaxy formation physics down to scales as small as $10~h^{-1}{\rm kpc}$.

astro-ph.CO↗

arXiv2022

If dark matter is fuzzy, the first stars form in massive pancakes

Fuzzy dark matter (FDM) is a proposed modification for the standard cold dark matter (CDM) model motivated by small-scale discrepancies in low-mass galaxies. Composed of ultra-light (mass $\sim 10^{-22}$ eV) axions with kpc-scale de Broglie wavelengths, this is one of a class of candidates that predicts that the first collapsed objects form in relatively massive dark matter halos. This implies that the formation history of the first stars and galaxies would be very different, potentially placing strong constraints on such models. Here we numerically simulate the formation of the first stars in an FDM cosmology, following the collapse in a representative volume all the way down to primordial protostar formation including a primordial non-equilibrium chemical network and cooling for the first time. We find two novel results: first, the large-scale collapse results in a very thin and flat gas "pancake"; second, despite the very different cosmology, this pancake fragments until it forms protostellar objects indistinguishable from those in CDM. Combined, these results indicate that the first generation of stars in this model are also likely to be massive and, because of the sheet morphology, do not self-regulate, resulting in a massive Pop III starburst. We estimate the total number of first stars forming in this extended structure to be $10^4$ over 20 Myr using a simple model to account for the ionizing feedback from the stars, and should be observable with JWST. These predictions provide a potential smoking gun signature of FDM and similar dark matter candidates.

astro-ph.GA↗

arXiv2022

Predicting Treatment Adherence of Tuberculosis Patients at Scale

Tuberculosis (TB), an infectious bacterial disease, is a significant cause of death, especially in low-income countries, with an estimated ten million new cases reported globally in $2020$. While TB is treatable, non-adherence to the medication regimen is a significant cause of morbidity and mortality. Thus, proactively identifying patients at risk of dropping off their medication regimen enables corrective measures to mitigate adverse outcomes. Using a proxy measure of extreme non-adherence and a dataset of nearly $700,000$ patients from four states in India, we formulate and solve the machine learning (ML) problem of early prediction of non-adherence based on a custom rank-based metric. We train ML models and evaluate against baselines, achieving a $\sim 100\%$ lift over rule-based baselines and $\sim 214\%$ over a random classifier, taking into account country-wide large-scale future deployment. We deal with various issues in the process, including data quality, high-cardinality categorical data, low target prevalence, distribution shift, variation across cohorts, algorithmic fairness, and the need for robustness and explainability. Our findings indicate that risk stratification of non-adherent patients is a viable, deployable-at-scale ML solution. As the official AI partner of India's Central TB Division, we are working on multiple city and state-level pilots with the goal of pan-India deployment.

cs.LG↗

arXiv2022

Robust field-level inference with dark matter halos

We train graph neural networks on halo catalogues from Gadget N-body simulations to perform field-level likelihood-free inference of cosmological parameters. The catalogues contain $\lesssim$5,000 halos with masses $\gtrsim 10^{10}~h^{-1}M_\odot$ in a periodic volume of $(25~h^{-1}{\rm Mpc})^3$; every halo in the catalogue is characterized by several properties such as position, mass, velocity, concentration, and maximum circular velocity. Our models, built to be permutationally, translationally, and rotationally invariant, do not impose a minimum scale on which to extract information and are able to infer the values of $Ω_{\rm m}$ and $σ_8$ with a mean relative error of $\sim6\%$, when using positions plus velocities and positions plus masses, respectively. More importantly, we find that our models are very robust: they can infer the value of $Ω_{\rm m}$ and $σ_8$ when tested using halo catalogues from thousands of N-body simulations run with five different N-body codes: Abacus, CUBEP$^3$M, Enzo, PKDGrav3, and Ramses. Surprisingly, the model trained to infer $Ω_{\rm m}$ also works when tested on thousands of state-of-the-art CAMELS hydrodynamic simulations run with four different codes and subgrid physics implementations. Using halo properties such as concentration and maximum circular velocity allow our models to extract more information, at the expense of breaking the robustness of the models. This may happen because the different N-body codes are not converged on the relevant scales corresponding to these parameters.

astro-ph.CO↗

arXiv2022

Team CERBERUS Wins the DARPA Subterranean Challenge: Technical Overview and Lessons Learned

This article presents the CERBERUS robotic system-of-systems, which won the DARPA Subterranean Challenge Final Event in 2021. The Subterranean Challenge was organized by DARPA with the vision to facilitate the novel technologies necessary to reliably explore diverse underground environments despite the grueling challenges they present for robotic autonomy. Due to their geometric complexity, degraded perceptual conditions combined with lack of GPS support, austere navigation conditions, and denied communications, subterranean settings render autonomous operations particularly demanding. In response to this challenge, we developed the CERBERUS system which exploits the synergy of legged and flying robots, coupled with robust control especially for overcoming perilous terrain, multi-modal and multi-robot perception for localization and mapping in conditions of sensor degradation, and resilient autonomy through unified exploration path planning and local motion planning that reflects robot-specific limitations. Based on its ability to explore diverse underground environments and its high-level command and control by a single human supervisor, CERBERUS demonstrated efficient exploration, reliable detection of objects of interest, and accurate mapping. In this article, we report results from both the preliminary runs and the final Prize Round of the DARPA Subterranean Challenge, and discuss highlights and challenges faced, alongside lessons learned for the benefit of the community.

cs.RO↗

arXiv2022

Autonomous Teamed Exploration of Subterranean Environments using Legged and Aerial Robots

This paper presents a novel strategy for autonomous teamed exploration of subterranean environments using legged and aerial robots. Tailored to the fact that subterranean settings, such as cave networks and underground mines, often involve complex, large-scale and multi-branched topologies, while wireless communication within them can be particularly challenging, this work is structured around the synergy of an onboard exploration path planner that allows for resilient long-term autonomy, and a multi-robot coordination framework. The onboard path planner is unified across legged and flying robots and enables navigation in environments with steep slopes, and diverse geometries. When a communication link is available, each robot of the team shares submaps to a centralized location where a multi-robot coordination framework identifies global frontiers of the exploration space to inform each system about where it should re-position to best continue its mission. The strategy is verified through a field deployment inside an underground mine in Switzerland using a legged and a flying robot collectively exploring for 45 min, as well as a longer simulation study with three systems.

cs.RO↗

arXiv2022

RMF-Owl: A Collision-Tolerant Flying Robot for Autonomous Subterranean Exploration

This work presents the design, hardware realization, autonomous exploration and object detection capabilities of RMF-Owl, a new collision-tolerant aerial robot tailored for resilient autonomous subterranean exploration. The system is custom built for underground exploration with focus on collision tolerance, resilient autonomy with robust localization and mapping, alongside high-performance exploration path planning in confined, obstacle-filled and topologically complex underground environments. Moreover, RMF-Owl offers the ability to search, detect and locate objects of interest which can be particularly useful in search and rescue missions. A series of results from field experiments are presented in order to demonstrate the system's ability to autonomously explore challenging unknown underground environments.

cs.RO↗