Shyamanta M. Hazarika

Ami Chopra, Supriya Bordoloi, Shyamanta M. Hazarika

Graph attention networks (GATs) provide one of the best frameworks for learning node representations in relational data; but, existing variants such as Graph Attention Network (GAT) mainly operate on static graphs and rely on implicit temporal aggregation when applied to sequential data. In this paper, we introduce Electroencephalography-temporal Graph Attention Network (EEG-tGAT), a temporally augmented formulation of GATv2 that is tailored for affordance classification from interaction sequences. The proposed model incorporates temporal attention to modulate the contribution of different time segments and temporal dropout to regularize learning across temporally correlated observations. The design reflects the assumption that temporal dimensions in affordance data are not semantically uniform and that discriminative information may be unevenly distributed across time. Experimental results on affordance datasets show that EEG-tGAT achieves improved classification performance compared to GATv2. The observed gains helps to conclude that explicitly encoding temporal importance and enforcing temporal robustness introduce inductive biases that are much better aligned with the structure of affordance-driven interaction data. These findings show us that modest architectural changes to graph attention models can help one obtain consistent benefits when temporal relationships play a nontrivial role in the task.

Hirakjyoti Basumatary, Daksh Adhar, Atharva Shrawge et al.

Achieving human-level dexterity in robotic grasping remains a challenging endeavor. Robotic hands frequently encounter slippage and deformation during object manipulation, issues rarely encountered by humans due to their sensory receptors, experiential learning, and motor memory. The emulation of the human grasping reflex within robotic hands is referred to as the ``bionic reflex". Past endeavors in the realm of bionic reflex control predominantly relied on model-based and supervised learning approaches, necessitating human intervention during thresholding and labeling tasks. In this study, we introduce an innovative bionic reflex control pipeline, leveraging reinforcement learning (RL); thereby eliminating the need for human intervention during control design. Our proposed bionic reflex controller has been designed and tested on an anthropomorphic hand, manipulating deformable objects in the PyBullet physics simulator, incorporating domain randomization (DR) for enhanced Sim2Real transferability. Our findings underscore the promise of RL as a potent tool for advancing bionic reflex control within anthropomorphic robotic hands. We anticipate that this autonomous, RL-based bionic reflex controller will catalyze the development of dependable and highly efficient robotic and prosthetic hands, revolutionizing human-robot interaction and assistive technologies.

Hirakjyoti Basumatary, Daksh Adhar, Riddhiman Shaw et al.

Grasp force synthesis is a non-convex optimization problem involving constraints that are bilinear. Traditional approaches to this problem involve general-purpose gradient-based nonlinear optimization and semi-definite programming. With a view towards dealing with postural synergies and non-smooth but convex positive semidefinite constraints, we look beyond gradient-based optimization. The focus of this paper is to undertake a grasp analysis of biomimetic grasping in multi-fingered robotic hands as a bilinear matrix inequality (BMI) problem. Our analysis is to solve it using a deep learning approach to make the algorithm efficiently generate force closure grasps with optimal grasp quality on untrained/unseen objects.

Narkhede Kunal Sanjay, Shyamanta M. Hazarika

This paper presents a bionic reflex control strategy for a kinematically constrained robotic finger. Here, the bionic reflex is achieved through a force tracking impedance control strategy. The dynamic model of the finger is reduced subject to kinematic constraints. Thereafter, an impedance control strategy that allows exact tracking of forces is discussed. Simulation results for a single finger holding a rectangular object against a flat surface are presented. Bionic reflex response time is of the order of milliseconds.

Rupam Bhattacharyya, Adity Saikia, Shyamanta M. Hazarika

Retention of residual skills for persons who partially lose their cognitive or physical ability is of utmost importance. Research is focused on developing systems that provide need-based assistance for retention of such residual skills. This paper describes a novel cognitive collaborative control architecture C3A, designed to address the challenges of developing need- based assistance for wheelchair navigation. Organization of C3A is detailed and results from simulation of the proposed architecture is presented. For simulation of our proposed architecture, we have used ROS (Robot Operating System) as a control framework and a 3D robotic simulator called USARSim (Unified System for Automation and Robot Simulation).

Alexy Bhowmick, Shyamanta M. Hazarika

We present a comprehensive review of the most effective content-based e-mail spam filtering techniques. We focus primarily on Machine Learning-based spam filters and their variants, and report on a broad review ranging from surveying the relevant ideas, efforts, effectiveness, and the current progress. The initial exposition of the background examines the basics of e-mail spam filtering, the evolving nature of spam, spammers playing cat-and-mouse with e-mail service providers (ESPs), and the Machine Learning front in fighting spam. We conclude by measuring the impact of Machine Learning-based filters and explore the promising offshoots of latest developments.