School of Artificial Intelligence and Data Science

This paper presents the development of a soft actu-ated glove for hand rehabilitation incorporating a twisted string actuator (TSA) mechanism threaded through flexible tubes and ergonomic anchors. The integration of TSA technology offers enhanced dexterity and allows natural hand movements. The flexible structures, coupled with ergonomic anchors, ensure a snug and comfortable fit on the wearer's hand, minimizing discomfort and maximizing usability. The design of the glove opens up possibilities for applications in various fields such as rehabilitation, assistance, and virtual reality interfaces. In this paper, we focus in particular on rehabilitation applications, in which the glove is integrated with a hand-tracking system and a Guided User Interface (GUI) for exercise setting, control, and monitoring.

Identifying products visible in an image of a rack of a supermarket is a challenging and commercially relevant machine vision problem. For identification, the region proposal algorithm generates a number of (mostly overlapped) region proposals around each product on the rack. Each region proposal is then assigned a product class with a certain classification score. Finally, the products are detected using non-maximal suppression (NMS) discovering winners among the region proposals. Greedy-NMS takes classification scores of the proposals as a key factor and thereby often eliminates (geometrically) better-fitted proposals. Graph-based NMS (G-NMS) provides a better alternative performance-wise but an inferior solution timecomplexity-wise (O(N3)) compared to O(N2) time-complexity of greedy-NMS. This paper introduces a adjusted classification score for use in the novel rectified non-maximal suppression (RNMS) setup that runs in O(N2). The efficacy of the proposed adjusted classification score is theoretically characterized in better discriminating overlapped region proposals. Our experiments establish that the performance of the proposed R-NMS is never inferior to G-NMS, and outperforms greedy-NMS while testing on several datasets.

Emanating from the food shortages in the 1960s, the public distribution system (PDS) in India has undergone various transformations through the years to expand food security across different regions of the country. Food grains, procured from the farmers, are processed in mills and transported from the central warehouses to the district warehouses and then finally to the fair price shops in the district. Given the district administration's role in managing public distribution operations by contracting with cooperatives who provide logistics services at the last mile, we develop a light-weight and scalable operations research-based decision support system for food grain distribution from the district warehouses to the fair price shops. Our study focusses on the last-mile distribution in the Ramanathapuram district of Tamil Nadu, India wherein the food grains need to be transported from the eight district warehouses associated with different cooperatives to the fair price shops. Given the current mapping of the district warehouses to the fair price shops, we establish the baseline by running the vehicle routing problem for each district warehouse to arrive at the baseline distances for each district warehouse and the entire district. Subsequently, we perform a remapping of the district warehouses to the fair price shops by running a relaxed version of the assignment problem and then run the vehicle routing problem over these revised clusters. Results indicate that 9% savings in total transportation costs is generated with the remapping procedure. Our light-weight decision support system acts as a valuable policy tool to the district administrators in establishing contracts with the cooperatives for each district warehouse, with implications for scaling it across the country. © 2025 Elsevier Ltd

Decoding of cognitive states aims to identify individuals' brain states and brain fingerprints to predict behavior. Deep learning provides an important platform for analyzing brain signals at different developmental stages to understand brain dynamics. Due to their internal architecture and feature extraction techniques, existing machine-learning and deep-learning approaches are suffering from low classification performance and explainability issues that must be improved. In the current study, we hypothesized that even at the early childhood stage (as early as 3-years), connectivity between brain regions could decode brain states and predict behavioral performance in false-belief tasks. To this end, we proposed an explainable deep learning framework to decode brain states (Theory of Mind and Pain states) and predict individual performance on ToM-related false-belief tasks in a developmental dataset. We proposed an explainable spatiotemporal connectivity-based Graph Convolutional Neural Network (Ex-stGCNN) model for decoding brain states. Here, we consider a developmental dataset, N = 155 (122 children; 3–12 yrs and 33 adults; 18–39 yrs), in which participants watched a short, soundless animated movie, shown to activate Theory-of-Mind (ToM) and pain networs. After scanning, the participants underwent a ToM-related false-belief task, leading to categorization into the pass, fail, and inconsistent groups based on performance. We trained our proposed model using Functional Connectivity (FC) and Inter-Subject Functional Correlations (ISFC) matrices separately. We observed that the stimulus-driven feature set (ISFC) could capture ToM and Pain brain states more accurately with an average accuracy of 94%, whereas it achieved 85% accuracy using FC matrices. We also validated our results using five-fold cross-validation and achieved an average accuracy of 92%. Besides this study, we applied the SHapley Additive exPlanations (SHAP) approach to identify brain fingerprints that contributed the most to predictions. We hypothesized that ToM network brain connectivity could predict individual performance on false-belief tasks. We proposed an Explainable Convolutional Variational Auto-Encoder (Ex-Convolutional VAE) model to predict individual performance on false-belief tasks and trained the model using FC and ISFC matrices separately. ISFC matrices again outperformed the FC matrices in prediction of individual performance. We achieved 93.5% accuracy with an F1-score of 0.94 using ISFC matrices and achieved 90% accuracy with an F1-score of 0.91 using FC matrices. Copyright © 2024 Bhavna, Akhter, Banerjee and Roy.

Intuitionistic fuzzy (IF) theory has become main approach to representing imprecision and vagueness. The IF divergence measure (IFDivM) based on Jensen–Shannon divergence is perhaps the most widely used measure to compare the similarity of multiple intuitionistic fuzzy sets (IFSs). In the present paper, this IFDivM is examined and applied to multiple examples. It is found that some extant IFDivMs hardly satisfy the axiomatic definition, and in a few cases even unable to show divergence of trivial IFSs. To address these inconsistencies, a new IFDivM based on Jensen–Shannon divergence is proposed, free from these problems. The effectiveness of the proposed IFDivM is tested on several critical cases, and precise analysis of its properties is performed. It is proved that the proposed IFDivM satisfies the axiomatic definition of IFDivMs. To illustrate the practical significance of the IFDivM, a novel intuitionistic fuzzy (IF) TODIM method, based on the proposed IFDivM, is developed, termed as GIF-TODIM method. Unlike the existing IF-TODIM methods, GIF-TODIM does not suffer from the revere ordering inconsistencies. The proposed GIF-TODIM method and the proposed IFDivM are applied to a real-world case study on supplier selection. A detailed comparative analysis is performed taking the TOPSIS method and other IFDivMs as baselines. The role of attitude on the final choice is analyzed in great detail. It is found that the proposed GIF-TODIM method is indeed useful, effective, and superior to the counterpart methods, when it comes to real-world situations. Concomitantly, in the present work, it is also revealed that the TOPSIS method based on the 2-D Hamming distance is a special form of the proposed GIF-TODIM method, when decision-makers have the same attitude towards losses and gains. Thus, an interesting relationship between TOPSIS and TODIM is identified under the intuitionistic fuzzy environment, which is bound to propel significant research in the area of decision making under uncertain conditions. As a whole, the article offers comprehensive analyses of IFDivMs and the TODIM method under the intuitionistic fuzzy environment.

Optimal brain function is shaped by a combination of global information integration, facilitated by long-range connections, and local processing, which relies on short-range connections and underlying biological factors. With aging, anatomical connectivity undergoes significant deterioration, which affects the brain’s overall function. Despite the structural loss, previous research has shown that normative patterns of functions remain intact across the lifespan, defined as the compensatory mechanism of the aging brain. However, the crucial components in guiding the compensatory preservation of the dynamical complexity and the underlying mechanisms remain uncovered. Moreover, it remains largely unknown how the brain readjusts its biological parameters to maintain optimal brain dynamics with age; in this work, we provide a parsimonious mechanism using a whole-brain generative model to uncover the role of sub-communities comprised of short-range and long-range connectivity in driving the dynamic compensation process in the aging brain. We utilize two neuroimaging datasets to demonstrate how short- and long-range white matter tracts affect compensatory mechanisms. We unveil their modulation of intrinsic global scaling parameters, such as global coupling strength and conduction delay, via a personalized large-scale brain model. Our key finding suggests that short-range tracts predominantly amplify global coupling strength with age, potentially representing an epiphenomenon of the compensatory mechanism. This mechanistically explains the significance of short-range connections in compensating for the major loss of long-range connections during aging. This insight could help identify alternative avenues to address aging-related diseases where long-range connections are significantly deteriorated. © The Author(s) 2025.

A temporally stable functional brain network pattern among coordinated brain regions is fundamental to stimulus selectivity and functional specificity during the critical period of brain development. Brain networks that are recruited in time to process internal states of others’ bodies (like hunger and pain) versus internal mental states (like beliefs, desires, and emotions) of others’ minds allow us to ask whether a quantitative characterization of the stability of these networks carries meaning during early development and constrain cognition in a specific way. Previous research provides critical insight into the early development of the theory-of-mind (ToM) network and its segregation from the Pain network throughout normal development using functional connectivity. However, a quantitative characterization of the temporal stability of ToM networks from early childhood to adulthood remains unexplored. In this work, reusing a large sample of children (n = 122, 3–12 years) and adults (n = 33) dataset that is available on the OpenfMRI database under the accession number ds000228, we addressed this question based on their fMRI data during a short and engaging naturalistic movie-watching task. The movie highlights the characters’ bodily sensations (often pain) and mental states (beliefs, desires, emotions), and is a feasible experiment for young children. Our results tracked the change in temporal stability using an unsupervised characterization of ToM and Pain networks DFC patterns using Angular and Mahalanobis distances between dominant dynamic functional connectivity subspaces. Our findings reveal that both ToM and Pain networks exhibit lower temporal stability as early as 3-years and gradually stabilize by 5-years, which continues till adolescence and late adulthood (often sharing similarity with adult DFC stability patterns). Furthermore, we find that the temporal stability of ToM brain networks is associated with the performance of participants in the false belief task to access mentalization at an early age. Interestingly, higher temporal stability is associated with the pass group, and similarly, moderate and low temporal stability are associated with the inconsistent group and the fail group. Our findings open an avenue for applying the temporal stability of large-scale functional brain networks during cortical development to act as a biomarker for multiple developmental disorders concerning impairment and discontinuity in the neural basis of social cognition.

In this paper, a new type of robotic finger is introduced that uses a twisted string actuator (TSA) and modular passive return rotational (PPR) joints. The design is intended to be simple, compact, lightweight, and energy-efficient while producing high grasping force with a relatively small motor. The PPR joints are based on the beam-buckling principle and are designed to match the non-linear TSA force profile, resulting in high grasping force throughout the finger's full flexion motion and passive finger extension. To evaluate the performance of the robotic finger, we fabricated a prototype and conducted experiments to assess its object grasping cycle, passive finger extension, grasping force, stable grasping condition, shape adaptability, and energy consumption. The finger weighs 170 grams and achieved a high force throughout the flexion motion, producing a maximum grasping force of 43.3 N at full flexion using a stall torque of 32 mNm. The modularity of the PPR joint allows for scalability and adaptability to handle different objects. We also demonstrated the finger's potential as a wearable sixth robotic finger (SRF), evaluating its object grasping competency, shape adaptability, and wearability. The finger was able to grasp various objects with a maximum payload of 1.0 kg and a hanging payload of up to 5 kg. Overall, the proposed robotic finger has the potential to be used as an SRF to compensate for arm disorders’ grasping capability.

Through mechanization, automation, and intensification, there has been a substantial increase in agricultural production over time. The efficiency, reliability, and precision of agricultural equipment have improved significantly with automation, leading to a reduced dependency on human intervention. The surge in the adoption of agricultural robotics research and technologies is a response to the growing recognition that robots offer an effective solution to address the shortage of skilled workers in crop production. This paper aims to present a systematic overview of recent advancements in precision delivery technology within agricultural robotics, with a primary focus on the following aspects: 1) precision agriculture market; 2) design and development of spray robot technologies, encompassing both terrestrial and aerial platforms; 3) spray technologies and their application mechanisms; 4) various spraying techniques tailored to specific pests and vegetation; and 5) evolution of sensor technologies for precision spraying. Additionally, this article explores the current state of the art in robotic technologies employed in precision agriculture.

In the real world, it is common to observe both randomness and fuzziness concurrently. Accordingly, an event/system/decision has multiple possibilities. Thus, in order to compare multiple uncertain options, a measure of overall uncertainty is often desired. This is the topic of this paper. A versatile and intuitive framework that could quantify concurrent probabilistic, fuzzy, and prob-fuzzy uncertainties is proposed. Based on the same, new entropy functions are developed to quantify the prob-fuzzy uncertainty. The proposed entropy functions are inspired from the popular entropy functions such as Shannon's entropy. The properties of the entropy functions are rigorously analysed. A real case-study in the agriculture and environment domain is included to demonstrate the usefulness of the work.