Mrinmoy Sarkar

Xuyang Yan, Shabnam Nazmi, Biniam Gebru et al.

In this paper, we proposed a new clustering-based active learning framework, namely Active Learning using a Clustering-based Sampling (ALCS), to address the shortage of labeled data. ALCS employs a density-based clustering approach to explore the cluster structure from the data without requiring exhaustive parameter tuning. A bi-cluster boundary-based sample query procedure is introduced to improve the learning performance for classifying highly overlapped classes. Additionally, we developed an effective diversity exploration strategy to address the redundancy among queried samples. Our experimental results justified the efficacy of the ALCS approach.

Justn williams, Kishor Datta Gupta, Roy George et al.

Vision-language-action (VLA) models have shown strong semantic grounding and task generalization in manipulation, but aerial deployment remains difficult because drones require low-latency closed-loop guidance under strict onboard compute and communication constraints. We present LiteVLA-H, a compact 256M-parameter VLA system designed for dual-rate operation on an NVIDIA Jetson AGX Orin: a fast outer-loop guidance mode for short action-token outputs and a slower semantic mode for scene understanding, hazard description, and operator-facing narration. The central empirical observation is that, in this compact edge regime, end-to-end latency is dominated by multimodal pre-fill rather than by the marginal cost of decoding a few extra tokens. This motivates a scheduler that issues reactive action tokens at 50.65,ms (19.74,Hz) while still supporting sentence-level semantic outputs at 149.90--164.57\ms (6.08--6.67,Hz) on the same embedded platform. To specialize the model without collapsing its descriptive competence, we use a knowledge-preserving fine-tuning recipe that mixes reactive flight data, aerial semantic data, and generic caption/VQA supervision. Beyond reporting current latency measurements, we position the system against recent state-of-the-art architectures, including AnywhereVLA, FutureVLA, and ReMem-VLA, showing that the measured action branch reaches a higher edge inference rate under our deployment conditions while retaining periodic semantic awareness.

Mrinmoy Sarkar

In this project, a state-of-the-art deep convolution neural network (DCNN) is presented to segment seismic images for salt detection below the earth's surface. Detection of salt location is very important for starting mining. Hence, a seismic image is used to detect the exact salt location under the earth's surface. However, precisely detecting the exact location of salt deposits is difficult. Therefore, professional seismic imaging still requires expert human interpretation of salt bodies. This leads to very subjective, highly variable renderings. Hence, to create the most accurate seismic images and 3D renderings, we need a robust algorithm that automatically and accurately identifies if a surface target is a salt or not. Since the performance of DCNN is well-known and well-established for object recognition in images, DCNN is a very good choice for this particular problem and being successfully applied to a dataset of seismic images in which each pixel is labeled as salt or not. The result of this algorithm is promising.

Justin Williams, Kishor Datta Gupta, Roy George et al.

Vision-Language-Action (VLA) models provide a unified framework for perception, language conditioning, and action generation, but many existing systems remain difficult to deploy in embedded robotic settings because of their computational requirements and inference latency. In this paper, we present LiteVLA-Edge, a deployment-oriented VLA pipeline for fully on-device inference on Jetson Orin-class hardware. Our approach combines supervised image-to-action fine-tuning in FP32 with post-training 4-bit GGUF quantization and GPU-accelerated inference through the \texttt{llama.cpp} runtime. Under our deployment configuration, LiteVLA-Edge achieves a mean end-to-end latency of 150.5\,ms (approximately 6.6\,Hz) while operating entirely offline within a ROS~2-integrated perception--reasoning--action pipeline. Rather than introducing a new policy objective, our contribution is a practical systems path for executing compact multimodal control models locally on embedded hardware while preserving modular interfaces between perception, reasoning, and actuation. These results establish timing feasibility for reactive language-conditioned control and provide a reproducible baseline for future task-level evaluation of on-device VLAs in robotics.

Justin Williams, Kishor Datta Gupta, Roy George et al.

The deployment of artificial intelligence models at the edge is increasingly critical for autonomous robots operating in GPS-denied environments where local, resource-efficient reasoning is essential. This work demonstrates the feasibility of deploying small Vision-Language Models (VLMs) on mobile robots to achieve real-time scene understanding and reasoning under strict computational constraints. Unlike prior approaches that separate perception from mobility, the proposed framework enables simultaneous movement and reasoning in dynamic environments using only on-board hardware. The system integrates a compact VLM with multimodal perception to perform contextual interpretation directly on embedded hardware, eliminating reliance on cloud connectivity. Experimental validation highlights the balance between computational efficiency, task accuracy, and system responsiveness. Implementation on a mobile robot confirms one of the first successful deployments of small VLMs for concurrent reasoning and mobility at the edge. This work establishes a foundation for scalable, assured autonomy in applications such as service robotics, disaster response, and defense operations.

Abenezer Girma, Abdollah Homaifar, M Nabil Mahmoud et al.

Standard Convolutional Neural Network (CNN) designs rarely focus on the importance of explicitly capturing diverse features to enhance the network's performance. Instead, most existing methods follow an indirect approach of increasing or tuning the networks' depth and width, which in many cases significantly increases the computational cost. Inspired by a biological visual system, we propose a Diverse and Adaptive Attention Convolutional Network (DA$^{2}$-Net), which enables any feed-forward CNNs to explicitly capture diverse features and adaptively select and emphasize the most informative features to efficiently boost the network's performance. DA$^{2}$-Net incurs negligible computational overhead and it is designed to be easily integrated with any CNN architecture. We extensively evaluated DA$^{2}$-Net on benchmark datasets, including CIFAR100, SVHN, and ImageNet, with various CNN architectures. The experimental results show DA$^{2}$-Net provides a significant performance improvement with very minimal computational overhead.

Xuyang Yan, Mrinmoy Sarkar, Biniam Gebru et al.

Feature selection methods are widely used to address the high computational overheads and curse of dimensionality in classifying high-dimensional data. Most conventional feature selection methods focus on handling homogeneous features, while real-world datasets usually have a mixture of continuous and discrete features. Some recent mixed-type feature selection studies only select features with high relevance to class labels and ignore the redundancy among features. The determination of an appropriate feature subset is also a challenge. In this paper, a supervised feature selection method using density-based feature clustering (SFSDFC) is proposed to obtain an appropriate final feature subset for mixed-type data. SFSDFC decomposes the feature space into a set of disjoint feature clusters using a novel density-based clustering method. Then, an effective feature selection strategy is employed to obtain a subset of important features with minimal redundancy from those feature clusters. Extensive experiments as well as comparison studies with five state-of-the-art methods are conducted on SFSDFC using thirteen real-world benchmark datasets and results justify the efficacy of the SFSDFC method.

Mrinmoy Sarkar, Xuyang Yan, Abenezer Girma et al.

In this paper, we developed a generalized simulation framework for the evaluation of electric vertical takeoff and landing vehicles (eVTOLs) in the context of Unmanned Aircraft Systems (UAS) Traffic Management (UTM) and under the concept of Urban Air Mobility (UAM). Unlike most existing studies, the proposed framework combines the utilization of UTM and eVTOLs to develop a realistic UAM testing platform. For this purpose, we first enhanced an existing UTM simulator to simulate the real-world UAM environment. Then, instead of using a simplified eVOTL model, a realistic eVTOL design tool, namely SUAVE, is employed and an dilation sub-module is introduced to bridge the gap between the UTM simulator and SUAVE eVTOL performance evaluation tool to elaborate the complete mission profile. Based on the developed simulation framework, experiments are conducted and the results are presented to analyze the performance of eVTOLs in the UAM environment.

Xuyang Yan, Abdollah Homaifar, Mrinmoy Sarkar et al.

The non-stationary nature of data streams strongly challenges traditional machine learning techniques. Although some solutions have been proposed to extend traditional machine learning techniques for handling data streams, these approaches either require an initial label set or rely on specialized design parameters. The overlap among classes and the labeling of data streams constitute other major challenges for classifying data streams. In this paper, we proposed a clustering-based data stream classification framework to handle non-stationary data streams without utilizing an initial label set. A density-based stream clustering procedure is used to capture novel concepts with a dynamic threshold and an effective active label querying strategy is introduced to continuously learn the new concepts from the data streams. The sub-cluster structure of each cluster is explored to handle the overlap among classes. Experimental results and quantitative comparison studies reveal that the proposed method provides statistically better or comparable performance than the existing methods.

Dhiman Chowdhury, Mrinmoy Sarkar

Electrical network frequency (ENF) is the signature of a power distribution grid which represents the nominal frequency (50 or 60 Hz) of a power system network. Due to load variations in a power grid, ENF sequences experience fluctuations. These ENF variations are inherently located in a multimedia signal which is recorded close to the grid or directly from the mains power line. Therefore, a multimedia recording can be localized by analyzing the ENF sequences of that signal in absence of the concurrent power signal. In this paper, a novel approach to analyze location forensics using ENF sequences extracted from a number of power and audio recordings is proposed. The digital recordings are collected from different grid locations around the world. Potential feature components are determined from the ENF sequences. Then, a multi-class support vector machine (SVM) classification model is developed to validate the location authenticity of the recordings. The performance assessments affirm the efficacy of the presented work.