arXiv subjects

Aditya Narendra

Publications and source records attributed to Aditya Narendra.

At least 19 recordsLinked to original sources

arXiv2026

CSymPlan: Certified Symbolic Planning and Control for High-DOF Manipulators

Robot manipulators are commonly engineered around a decoupled motion-generation stack: a planner computes a collision-free path and a lower-level controller tracks the resulting reference. This separation is computationally convenient, but it can produce references that are difficult to execute under actuator limits, tracking error, model mismatch, and small obstacle clearances. We present CSymPlan, a certified symbolic planning and control framework for high-DOF manipulators with two complementary implementations: an offline implementation that precomputes certified reach-avoid feedback policies for known workspaces; and an online implementation that synthesizes or updates symbolic policies at runtime from changing task and perception information using parallelization. The offline implementation reduces the manipulator dynamics to a sampled perturbed double-integrator model in operational space through feedback linearization, treats torque-realization errors, modeling inaccuracies, and measurement uncertainty as bounded disturbances, and refines the synthesized symbolic policy to the Franka FR3 through a quantization--lookup--torque realization pipeline. The online implementation uses the same abstraction and refinement interface, but replaces the precomputed policy table with a runtime pFaces request--synthesis--execution loop. In randomized simulated benchmarks and perception-driven Franka FR3 experiments, both implementations complete reach-avoid tasks with zero safety violations; whenever no certified action exists, the robot holds, replans, or stops safely instead of executing an uncertified command.

cs.RO↗

arXiv2026

Knowledge-Guided Manipulation Using Multi-Task Reinforcement Learning

This paper introduces Knowledge Graph based Massively Multi-task Model-based Policy Optimization (KG-M3PO), a framework for multi-task robotic manipulation in partially observable settings that unifies Perception, Knowledge, and Policy. The method augments egocentric vision with an online 3D scene graph that grounds open-vocabulary detections into a metric, relational representation. A dynamic-relation mechanism updates spatial, containment, and affordance edges at every step, and a graph neural encoder is trained end-to-end through the RL objective so that relational features are shaped directly by control performance. Multiple observation modalities (visual, proprioceptive, linguistic, and graph-based) are encoded into a shared latent space, upon which the RL agent operates to drive the control loop. The policy conditions on lightweight graph queries alongside visual and proprioceptive inputs, yielding a compact, semantically informed state for decision making. Experiments on a suite of manipulation tasks with occlusions, distractors, and layout shifts demonstrate consistent gains over strong baselines: the knowledge-conditioned agent achieves higher success rates, improved sample efficiency, and stronger generalization to novel objects and unseen scene configurations. These results support the premise that structured, continuously maintained world knowledge is a powerful inductive bias for scalable, generalizable manipulation: when the knowledge module participates in the RL computation graph, relational representations align with control, enabling robust long-horizon behavior under partial observability.

cs.RO↗

arXiv2025

Probing evolution of Long GRB properties through their cosmic formation history aided by Machine Learning predicted redshifts

Gamma-ray Bursts (GRBs) are valuable probes of cosmic star formation reaching back into the epoch of reionization, and a large dataset with known redshifts ($z$) is an important ingredient for these studies. Usually, $z$ is measured using spectroscopy or photometry, but $\sim80\%$ of GRBs lack such data. Prompt and afterglow correlations can provide estimates in these cases, though they suffer from systematic uncertainties due to assumed cosmologies and due to detector threshold limits. We use a sample with $z$ estimated via machine learning models, based on prompt and afterglow parameters, without relying on cosmological assumptions. We then use an augmented sample of GRBs with measured and predicted redshifts, forming a larger dataset. We find that the predicted redshifts are a crucial step forward in understanding the evolution of GRB properties. We test three cases: no evolution, an evolution of the beaming factor, and an evolution of all terms captured by an evolution factor $(1+z)^\delta$. We find that these cases can explain the density rate in the redshift range between 1-2, but neither of the cases can explain the derived rate densities at smaller and higher redshifts, which may point towards an evolution term different than a simple power law. Another possibility is that this mismatch is due to the non-homogeneity of the sample, e.g., a non-collapsar origin of some long GRB within the sample.

astro-ph.HE↗

arXiv2025

Probing Evolution of Long Gamma-Ray Burst Properties through Their Cosmic Formation History

The astrophysics of Long GRB (LGRB) progenitors as well as possible cosmological evolution in their properties still poses many open questions. Previous studies suggest that the LGRB rate density (LGRB-RD) follows the cosmic star formation rate density (SFRD) only at high-z and attribute this to the metallicity evolution of progenitor stars. For low z, opinions differ on whether the uptick in the LGRB RD is due to a distinct class of low-luminosity GRBs or perhaps even a different progenitor subclass. To investigate these questions, we utilize data from the Neil Gehrels Swift Observatory and ground-based observatories (redshift). To test the hypothesis that the observations can be mapped (with/without evolution) to the well-established cosmic SFRD, we consider three cases: no evolution, beaming angle evolution, and a simple power-law evolution. The comparison shows that the 'no evolution' case can be ruled out. Our study highlights that the beaming angle evolution or the simple power law evolution are also not sufficient to obtain a good match between the LGRB-RD and SFRD. Rather, the inclusion of multiple evolving properties of LGRBs in combination appears to be required to match the two rate densities in their entirety.

astro-ph.HE↗

arXiv2025

Mind and Motion Aligned: A Joint Evaluation IsaacSim Benchmark for Task Planning and Low-Level Policies in Mobile Manipulation

Benchmarks are crucial for evaluating progress in robotics and embodied AI. However, a significant gap exists between benchmarks designed for high-level language instruction following, which often assume perfect low-level execution, and those for low-level robot control, which rely on simple, one-step commands. This disconnect prevents a comprehensive evaluation of integrated systems where both task planning and physical execution are critical. To address this, we propose Kitchen-R, a novel benchmark that unifies the evaluation of task planning and low-level control within a simulated kitchen environment. Built as a digital twin using the Isaac Sim simulator and featuring more than 500 complex language instructions, Kitchen-R supports a mobile manipulator robot. We provide baseline methods for our benchmark, including a task-planning strategy based on a vision-language model and a low-level control policy based on diffusion policy. We also provide a trajectory collection system. Our benchmark offers a flexible framework for three evaluation modes: independent assessment of the planning module, independent assessment of the control policy, and, crucially, an integrated evaluation of the whole system. Kitchen-R bridges a key gap in embodied AI research, enabling more holistic and realistic benchmarking of language-guided robotic agents.

cs.RO↗

arXiv2025

M3PO: Massively Multi-Task Model-Based Policy Optimization

We introduce Massively Multi-Task Model-Based Policy Optimization (M3PO), a scalable model-based reinforcement learning (MBRL) framework designed to address sample inefficiency in single-task settings and poor generalization in multi-task domains. Existing model-based approaches like DreamerV3 rely on pixel-level generative models that neglect control-centric representations, while model-free methods such as PPO suffer from high sample complexity and weak exploration. M3PO integrates an implicit world model, trained to predict task outcomes without observation reconstruction, with a hybrid exploration strategy that combines model-based planning and model-free uncertainty-driven bonuses. This eliminates the bias-variance trade-off in prior methods by using discrepancies between model-based and model-free value estimates to guide exploration, while maintaining stable policy updates through a trust-region optimizer. M3PO provides an efficient and robust alternative to existing model-based policy optimization approaches and achieves state-of-the-art performance across multiple benchmarks.

cs.LG↗

arXiv2025

The Machine Learning to reconstruct GRB lightcurves

The current knowledge in cosmology deals with open problems whose solutions are still under investigation. The main issue is the so-called Hubble constant ($H_0$) tension, namely, the $4-6 \sigma$ discrepancy between the local value of $H_0$ obtained with Cepheids+Supernovae Ia (SNe Ia) and the cosmological one estimated from the observations of the Cosmic Microwave Background (CMB). For the investigation of this problem, probes that span all over the redshift $z$ ranges are needed. Cepheids are local objects, SNe Ia reached up to $z=2.9$, and CMB is observed at $z=1100$. In this context, the use of probes at intermediate redshift $z>3$ is auspicious for casting more light on modern cosmology. The Gamma-ray Bursts (GRBs) are particularly suitable for this task, given their observability up to $z=9.4$. The use of GRBs as standardizable candles requires the use of tight and reliable astrophysical correlations and the presence of gaps in the GRB time-domain data represents an obstacle in this sense. In this work, we propose to improve the precision of the lightcurve (LC) parameters through a reconstruction process performed with the functional forms of GRB LCs and the Gaussian Processes (GP). The filling of gaps in the GRB LCs through these processes shows an improvement up to $41.5\%$ on the precision of the LC parameters fitting, which lead to a reduced scatter in the astrophysical correlations and, thus, in the estimation of cosmological parameters.

astro-ph.CO↗

arXiv2024

GRB Redshift Estimation using Machine Learning and the Associated Web-App

Context. Gamma-ray bursts (GRBs), observed at redshifts as high as 9.4, could serve as valuable probes for investigating the distant Universe. However, this necessitates an increase in the number of GRBs with determined redshifts, as currently, only 12% of GRBs have known redshifts due to observational biases. Aims. We aim to address the shortage of GRBs with measured redshifts, enabling us to fully realize their potential as valuable cosmological probes Methods. Following Dainotti et al. (2024c), we have taken a second step to overcome this issue by adding 30 more GRBs to our ensemble supervised machine learning training sample, an increase of 20%, which will help us obtain better redshift estimates. In addition, we have built a freely accessible and user-friendly web app that infers the redshift of long GRBs (LGRBs) with plateau emission using our machine learning model. The web app is the first of its kind for such a study and will allow the community to obtain redshift estimates by entering the GRB parameters in the app. Results. Through our machine learning model, we have successfully estimated redshifts for 276 LGRBs using X-ray afterglow parameters detected by the Neil Gehrels Swift Observatory and increased the sample of LGRBs with known redshifts by 110%. We also perform Monte Carlo simulations to demonstrate the future applicability of this research. Conclusions. The results presented in this research will enable the community to increase the sample of GRBs with known redshift estimates. This can help address many outstanding issues, such as GRB formation rate, luminosity function, and the true nature of low-luminosity GRBs, and enable the application of GRBs as standard candles

astro-ph.HE↗

arXiv2024

Dynamic Neural Potential Field: Online Trajectory Optimization in the Presence of Moving Obstacles

Generalist robot policies must operate safely and reliably in everyday human environments such as homes, offices, and warehouses, where people and objects move unpredictably. We present Dynamic Neural Potential Field (NPField-GPT), a learning-enhanced model predictive control (MPC) framework that couples classical optimization with a Transformer-based predictor of footprint-aware repulsive potentials. Given an occupancy sub-map, robot footprint, and optional dynamic-obstacle cues, our NPField-GPT model forecasts a horizon of differentiable potentials that are injected into a sequential quadratic MPC program via L4CasADi, yielding real-time, constraint-aware trajectory optimization. We additionally study two baselines: NPField-StaticMLP, where a dynamic scene is treated as a sequence of static maps; and NPField-DynamicMLP, which predicts the future potential sequence in parallel with an MLP. In dynamic indoor scenarios from BenchMR and on a Husky UGV in office corridors, NPField-GPT produces more efficient and safer trajectories under motion changes, while StaticMLP/DynamicMLP offer lower latency. We also compare with the CIAO* and MPPI baselines. Across methods, the Transformer+MPC synergy preserves the transparency and stability of model-based planning while learning only the part that benefits from data: spatiotemporal collision risk. Code and trained models are available at https://github.com/CognitiveAISystems/Dynamic-Neural-Potential-Field

cs.RO↗

arXiv2024

Gamma-ray Bursts as Distance Indicators by a Statistical Learning Approach

Gamma-ray bursts (GRBs) can be probes of the early universe, but currently, only 26% of GRBs observed by the Neil Gehrels Swift Observatory GRBs have known redshifts ($z$) due to observational limitations. To address this, we estimated the GRB redshift (distance) via a supervised statistical learning model that uses optical afterglow observed by Swift and ground-based telescopes. The inferred redshifts are strongly correlated (a Pearson coefficient of 0.93) with the observed redshifts, thus proving the reliability of this method. The inferred and observed redshifts allow us to estimate the number of GRBs occurring at a given redshift (GRB rate) to be 8.47-9 $yr^{-1} Gpc^{-1}$ for $1.9<z<2.3$. Since GRBs come from the collapse of massive stars, we compared this rate with the star formation rate highlighting a discrepancy of a factor of 3 at $z<1$.

astro-ph.HE↗

arXiv2024

Inferring the redshift of more than 150 GRBs with a Machine Learning Ensemble model

Gamma-Ray Bursts (GRBs), due to their high luminosities are detected up to redshift 10, and thus have the potential to be vital cosmological probes of early processes in the universe. Fulfilling this potential requires a large sample of GRBs with known redshifts, but due to observational limitations, only 11\% have known redshifts ($z$). There have been numerous attempts to estimate redshifts via correlation studies, most of which have led to inaccurate predictions. To overcome this, we estimated GRB redshift via an ensemble supervised machine learning model that uses X-ray afterglows of long-duration GRBs observed by the Neil Gehrels Swift Observatory. The estimated redshifts are strongly correlated (a Pearson coefficient of 0.93) and have a root mean square error, namely the square root of the average squared error $\langle\Delta z^2\rangle$, of 0.46 with the observed redshifts showing the reliability of this method. The addition of GRB afterglow parameters improves the predictions considerably by 63\% compared to previous results in peer-reviewed literature. Finally, we use our machine learning model to infer the redshifts of 154 GRBs, which increase the known redshifts of long GRBs with plateaus by 94\%, a significant milestone for enhancing GRB population studies that require large samples with redshift.

astro-ph.CO↗

arXiv2023

Chaurah: A Smart Raspberry Pi based Parking System

The widespread usage of cars and other large, heavy vehicles necessitates the development of an effective parking infrastructure. Additionally, algorithms for detection and recognition of number plates are often used to identify automobiles all around the world where standardized plate sizes and fonts are enforced, making recognition an effortless task. As a result, both kinds of data can be combined to develop an intelligent parking system focuses on the technology of Automatic Number Plate Recognition (ANPR). Retrieving characters from an inputted number plate image is the sole purpose of ANPR which is a costly procedure. In this article, we propose Chaurah, a minimal cost ANPR system that relies on a Raspberry Pi 3 that was specifically created for parking facilities. The system employs a dual-stage methodology, with the first stage being an ANPR system which makes use of two convolutional neural networks (CNNs). The primary locates and recognises license plates from a vehicle image, while the secondary performs Optical Character Recognition (OCR) to identify individualized numbers from the number plate. An application built with Flutter and Firebase for database administration and license plate record comparison makes up the second component of the overall solution. The application also acts as an user-interface for the billing mechanism based on parking time duration resulting in an all-encompassing software deployment of the study.

cs.CV↗

arXiv2023

GRB Optical and X-ray Plateau Properties Classifier Using Unsupervised Machine Learning

The division of Gamma-ray bursts (GRBs) into different classes, other than the "short" and "long", has been an active field of research. We investigate whether GRBs can be classified based on a broader set of parameters, including prompt and plateau emission ones. Observational evidence suggests the existence of more GRB sub-classes, but results so far are either conflicting or not statistically significant. The novelty here is producing a machine-learning-based classification of GRBs using their observed X-rays and optical properties. We used two data samples: the first, composed of 203 GRBs, is from the Neil Gehrels Swift Observatory (Swift/XRT), and the latter, composed of 134 GRBs, is from the ground-based Telescopes and Swift/UVOT. Both samples possess the plateau emission (a flat part of the light curve happening after the prompt emission, the main GRB event). We have applied the Gaussian Mixture Model (GMM) to explore multiple parameter spaces and sub-class combinations to reveal if there is a match between the current observational sub-classes and the statistical classification. With these samples and the algorithm, we spot a few micro-trends in certain cases, but we cannot conclude that any clear trend exists in classifying GRBs. These microtrends could point towards a deeper understanding of the physical meaning of these classes (e.g., a different environment of the same progenitor or different progenitors). However, a larger sample and different algorithms could achieve such goals. Thus, this methodology can lead to deeper insights in the future.

astro-ph.HE↗

arXiv2023

Fermi LAT AGN classification using supervised machine learning

Classifying Active Galactic Nuclei (AGN) is a challenge, especially for BL Lac Objects (BLLs), which are identified by their weak emission line spectra. To address the problem of classification, we use data from the 4th Fermi Catalog, Data Release 3. Missing data hinders the use of machine learning to classify AGN. A previous paper found that Multiple Imputation by Chain Equations (MICE) imputation is useful for estimating missing values. Since many AGN have missing redshift and the highest energy, we use data imputation with MICE and K-nearest neighbor (kNN) algorithm to fill in these missing variables. Then, we classify AGN into the BLLs or the Flat Spectrum Radio Quasars (FSRQs) using the SuperLearner, an ensemble method that includes several classification algorithms like logistic regression, support vector classifiers, Random Forests, Ranger Random Forests, multivariate adaptive regression spline (MARS), Bayesian regression, Extreme Gradient Boosting. We find that a SuperLearner model using MARS regression and Random Forests algorithms is 91.1% accurate for kNN imputed data and 91.2% for MICE imputed data. Furthermore, the kNN-imputed SuperLearner model predicts that 892 of the 1519 unclassified blazars are BLLs and 627 are Flat Spectrum Radio Quasars (FSRQs), while the MICE-imputed SuperLearner model predicts 890 BLLs and 629 FSRQs in the unclassified set. Thus, we can conclude that both imputation methods work efficiently and with high accuracy and that our methodology ushers the way for using SuperLearner as a novel classification method in the AGN community and, in general, in the astrophysics community.

astro-ph.HE↗

arXiv2023

A Stochastic Approach To Reconstruct Gamma Ray Burst Lightcurves

Gamma-Ray Bursts (GRBs), being observed at high redshift (z = 9.4), vital to cosmological studies and investigating Population III stars. To tackle these studies, we need correlations among relevant GRB variables with the requirement of small uncertainties on their variables. Thus, we must have good coverage of GRB light curves (LCs). However, gaps in the LC hinder the precise determination of GRB properties and are often unavoidable. Therefore, extensive categorization of GRB LCs remains a hurdle. We address LC gaps using a 'stochastic reconstruction,' wherein we fit two pre-existing models (Willingale 2007; W07 and Broken Power Law; BPL) to the observed LC, then use the distribution of flux residuals from the original data to generate data to fill in the temporal gaps. We also demonstrate a model-independent LC reconstruction via Gaussian Processes. At 10% noise, the uncertainty of the end time of the plateau, its correspondent flux, and the temporal decay index after the plateau decreases, on average, by 33.3% 35.03%, and 43.32%, respectively for the W07, and by 33.3%, 30.78%, 43.9% for the BPL. The slope of the plateau decreases by 14.76% in the BPL. After using the Gaussian Process technique, we see similar trends of a decrease in uncertainty for all model parameters for both the W07 and BPL models. These improvements are essential for the application of GRBs as standard candles in cosmology, for the investigation of theoretical models and for inferring the redshift of GRBs with future machine learning analysis.

astro-ph.HE↗

arXiv2023

From Robots to Books: An Introduction to Smart Applications of AI in Education (AIEd)

The world around us has undergone a radical transformation due to rapid technological advancement in recent decades. The industry of the future generation is evolving, and artificial intelligence is the following change in the making popularly known as Industry 4.0. Indeed, experts predict that artificial intelligence(AI) will be the main force behind the following significant virtual shift in the way we stay, converse, study, live, communicate and conduct business. All facets of our social connection are being transformed by this growing technology. One of the newest areas of educational technology is Artificial Intelligence in the field of Education(AIEd).This study emphasizes the different applications of artificial intelligence in education from both an industrial and academic standpoint. It highlights the most recent contextualized learning novel transformative evaluations and advancements in sophisticated tutoring systems. It analyses the AIEd's ethical component and the influence of the transition on people, particularly students and instructors as well. Finally, this article touches on AIEd's potential future research and practices. The goal of this study is to introduce the present-day applications to its intended audience.

cs.CY↗

arXiv2022

Using Multivariate Imputation by Chained Equations to Predict Redshifts of Active Galactic Nuclei

Redshift measurement of active galactic nuclei (AGNs) remains a time-consuming and challenging task, as it requires follow up spectroscopic observations and detailed analysis. Hence, there exists an urgent requirement for alternative redshift estimation techniques. The use of machine learning (ML) for this purpose has been growing over the last few years, primarily due to the availability of large-scale galactic surveys. However, due to observational errors, a significant fraction of these data sets often have missing entries, rendering that fraction unusable for ML regression applications. In this study, we demonstrate the performance of an imputation technique called Multivariate Imputation by Chained Equations (MICE), which rectifies the issue of missing data entries by imputing them using the available information in the catalog. We use the Fermi-LAT Fourth Data Release Catalog (4LAC) and impute 24% of the catalog. Subsequently, we follow the methodology described in Dainotti et al. (2021) and create an ML model for estimating the redshift of 4LAC AGNs. We present results which highlight positive impact of MICE imputation technique on the machine learning models performance and obtained redshift estimation accuracy.

astro-ph.IM↗

arXiv2022

Predicting the redshift of gamma-ray loud AGNs using Supervised Machine Learning: Part 2

Measuring the redshift of active galactic nuclei (AGNs) requires the use of time-consuming and expensive spectroscopic analysis. However, obtaining redshift measurements of AGNs is crucial as it can enable AGN population studies, provide insight into the star formation rate, the luminosity function, and the density rate evolution. Hence, there is a requirement for alternative redshift measurement techniques. In this project, we aim to use the Fermi gamma-ray space telescope's 4LAC Data Release (DR2) catalog to train a machine learning model capable of predicting the redshift reliably. In addition, this project aims at improving and extending with the new 4LAC Catalog the predictive capabilities of the machine learning (ML) methodology published in Dainotti et al. (2021). Furthermore, we implement feature engineering to expand the parameter space and a bias correction technique to our final results. This study uses additional machine learning techniques inside the ensemble method, the SuperLearner, previously used in Dainotti et al.(2021). Additionally, we also test a novel ML model called Sorted L-One Penalized Estimation (SLOPE). Using these methods we provide a catalog of estimated redshift values for those AGNs that do not have a spectroscopic redshift measurement. These estimates can serve as a redshift reference for the community to verify as updated Fermi catalogs are released with more redshift measurements.

astro-ph.HE↗