Irfan Hussain | Khalifa University (original) (raw)

Papers by Irfan Hussain

Research Square (Research Square), Feb 12, 2024

Safe and efficient human-robot collaboration (HRC) relies on proactive collision avoidance and ro... more Safe and efficient human-robot collaboration (HRC) relies on proactive collision avoidance and robust detection of unexpected contacts. Although existing learning-based approaches excel in specific areas, they often encounter challenges with robots equipped with variable stiffness actuators-a crucial aspect for ensuring safe and adaptable HRC. To address this, we present MAD-CNN, a novel Modularized Attention-Dilated Convolutional Neural Network designed for highly sensitive and robust collision detection in these adaptable robots. MAD-CNN leverages a dual inductive bias mechanism, combining modularization and dilated convolution, to significantly reduce data requirements. This capability facilitates efficient training even with limited collision datasets, offering a substantial advantage for real-world implementation. Additionally, an attention module prioritizes relevant features, enhancing the network's robustness against variations in stiffness levels-a core challenge in variable stiffness robots. This powerful combination unlocks exceptional performance for MAD-CNN. In extensive evaluations, it achieved zero missed collisions across 516 events, with a minimal detection delay averaging just 12.05 milliseconds and 20% fewer false positives compared to baseline methods. Remarkably, this outstanding performance was attained even when trained on a four-minute dataset focused on the highest stiffness setting, underscoring MAD-CNN's data efficiency and adaptability across diverse stiffness configurations. By enabling accurate and sensitive collision detection across the spectrum of variable stiffness conditions, MAD-CNN paves the way for safer and more reliable HRC in various fields, including manufacturing, healthcare, and assistive robotics.

arXiv (Cornell University), Aug 6, 2022

Frontiers in Bioengineering and Biotechnology

The global increase in the number of stroke patients and limited accessibility to rehabilitation ... more The global increase in the number of stroke patients and limited accessibility to rehabilitation has promoted an increase in the design and development of mobile exoskeletons. Robot-assisted mobile rehabilitation is rapidly emerging as a viable tool as it could provide intensive repetitive movement training and timely standardized delivery of therapy as compared to conventional manual therapy. However, the majority of existing lower limb exoskeletons continue to be heavy and induce unnecessary inertia and inertial vibration on the limb. Cable-driven exoskeletons can overcome these issues with the provision of remote actuation. However, the number of cables and routing can be selected in various ways posing a challenge to designers regarding the optimal design configuration. In this work, a simulation-based generalized framework for modelling and assessment of cable-driven mobile exoskeleton is proposed. The framework can be implemented to identify a ‘suitable’ configuration from sev...

Research Square (Research Square), Apr 9, 2024

The integration of artificial intelligence with sensor technologies has revolutionized precision ... more The integration of artificial intelligence with sensor technologies has revolutionized precision agriculture, offering unprecedented opportunities for enhancing crop management and productivity. This review focuses on the latest advancements in vision-based tactile sensors, a technology at the forefront of this transformation. By combining tactile data with vision-based techniques, these sensors provide a more comprehensive understanding of the agricultural environment. We investigate thoroughly the role of deep learning approaches in refining the functionality of these sensors, highlighting their potential to significantly improve the accuracy and efficiency of agricultural operations. The paper also explores the importance of specialized datasets in training deep neural networks for vision-based tactile applications, assessing the current landscape and identifying gaps in the available data. Through a thorough examination of the current state of the art, this review paper aims to shed light on the potential of AI-driven tactile sensing in precision agriculture and outline future research directions to further advance this field.

Biosystems & biorobotics, 2020

This chapter presents two kinds of electromyographic (EMG) control interfaces for supernumerary r... more This chapter presents two kinds of electromyographic (EMG) control interfaces for supernumerary robotic fingers.

2019 IEEE 9th Annual International Conference on CYBER Technology in Automation, Control, and Intelligent Systems (CYBER)

This paper presents the design, the prototype and the control of SCARA (Selective Compliance Asse... more This paper presents the design, the prototype and the control of SCARA (Selective Compliance Assembly Robot Arm) to perform pick and place tasks for industrial applications. The robot has four degrees of freedom (DoFs) with a capability to carry payloads up to 1 kg with high accuracy, precision, and repeatability. The robot is designed using the CAD tools and its prototype development is carried out by manufacturing its mechanical parts (links, base) and selecting the proper off the shelf electrical (motors, controllers) and transmission components (belts, pulleys). We also present the mathematical formulation consisting of direct kinematics, inverse kinematics, and the dynamical equations of the robot. We also report a closed loop position control based on the Computed Torque Control (CTC) method. The Numerical simulations are performed in order to evaluate the performance of the robot by using the applied control technique. The future work includes the experiments on the hardware using the implemented control technique.

Frontiers in Neurorobotics, 2021

Over the last decade underactuated, adaptive robot grippers and hands have received an increased ... more Over the last decade underactuated, adaptive robot grippers and hands have received an increased interest from the robotics research community. This class of robotic end-effectors can be used in many different fields and scenarios with a very promising application being the development of prosthetic devices. Their suitability for the development of such devices is attributed to the utilization of underactuation that provides increased functionality and dexterity with reduced weight, cost, and control complexity. The most critical components of underactuated, adaptive hands that allow them to perform a broad set of grasp poses are appropriate differential mechanisms that facilitate the actuation of multiple degrees of freedom using a single motor. In this work, we focus on the design, analysis, and experimental validation of a four output geared differential, a series elastic differential, and a whiffletree differential that can incorporate a series of manual and automated locking me...

IEEE Access

The field of wearable devices and exoskeletons for rehabilitation and assistance is one of the ma... more The field of wearable devices and exoskeletons for rehabilitation and assistance is one of the main fields where variable stiffness actuators (VSAs) are continuously incorporated. This is due to the need for such devices to adapt to erratic environments and mimic human motions. Recently, a passive revolute joint design with controllable variable stiffness was proposed by the authors. However, the previous design was only introduced as a proof of concept, with no further investigation of the design stiffness change capabilities for application utilization. In this work, we introduce an extended analysis and comprehensive parametric study of the system's main design parameters. Based on the performed analysis, we propose an optimization framework for this design concept that allows implementation in knee exoskeletons. The main design parameters that affect the stiffness performance are defined. The proposed method allows the identification of these design parameters, which enables the redesign of the system according to the predefined user requirements, such as size, desired stiffness range, and stiffness change rate. A bench-top experimental setup is constructed to validate the obtained optimal design parameters. Results show the ability to achieve the desired stiffness performance for the intended application.

IEEE Robotics and Automation Letters

The International Journal of Robotics Research

This paper presents the design, analysis, fabrication, experimental characterization, and evaluat... more This paper presents the design, analysis, fabrication, experimental characterization, and evaluation of two prototypes of robotic extra fingers that can be used as grasp compensatory devices for a hemiparetic upper limb. The devices are the results of experimental sessions with chronic stroke patients and consultations with clinical experts. Both devices share a common principle of work, which consists in opposing the device to the paretic hand or wrist so to restrain the motion of an object. They can be used by chronic stroke patients to compensate for grasping in several activities of daily living (ADLs) with a particular focus on bimanual tasks. The robotic extra fingers are designed to be extremely portable and wearable. They can be wrapped as bracelets when not being used, to further reduce the encumbrance. Both devices are intrinsically compliant and driven by a single actuator through a tendon system. The motion of the robotic devices can be controlled using an electromyograp...

IEEE ... International Conference on Rehabilitation Robotics : [proceedings], Jul 1, 2017

The Supernumerary robotic limbs are a recently introduced class of wearable robots that, differen... more The Supernumerary robotic limbs are a recently introduced class of wearable robots that, differently from traditional prostheses and exoskeletons, aim at adding extra effectors (i.e., arms, legs, or fingers) to the human user, rather than substituting or enhancing the natural ones. However, it is still undefined whether the use of supernumerary robotic limbs could specifically lead to neural modifications in brain dynamics. The illusion of owning the part of body has been already proven in many experimental observations, such as those relying on multisensory integration (e.g., rubber hand illusion), prosthesis and even on virtual reality. In this paper we present a description of a novel magnetic compatible supernumerary robotic finger together with preliminary observations from two functional magnetic resonance imaging (fMRI) experiments, in which brain activity was measured before and after a period of training with the robotic device, and during the use of the novel MRI-compatibl...

Robotics and Autonomous Systems, 2017

Biosystems & Biorobotics, 2016

2013 13th International Conference on Control, Automation and Systems (ICCAS 2013), 2013

ABSTRACT

Abstract—Above Elbow Prosthesis is one of the most commonly amputated or missing limbs. The resea... more Abstract—Above Elbow Prosthesis is one of the most commonly amputated or missing limbs. The research is done for modelling techniques of upper limb prosthesis and design of high torque, light weight and compact in size elbow actuator. The purposed actuator consists of ...

Frontiers in Neurorobotics, 2016

This paper presents design, analysis, fabrication, experimental characterization and evaluation o... more This paper presents design, analysis, fabrication, experimental characterization and evaluation of two prototypes of robotic extra fingers that can be used as grasp compensatory devices for hemiparetic upper limb. The devices are the results of experimental sessions with chronic stroke patients and consultations with clinical experts. Both devices share a common principle of work which consists in opposing to the paretic hand/wrist so to restrain the motion of an object. They can be used by chronic stroke patients to compensate for grasping in several Activities of Daily Living (ADL) with a particular focus on bimanual tasks. The robotic extra fingers are designed to be extremely portable and wearable. They can be wrapped as bracelets when not being used, to further reduce the encumbrance. Both devices are intrinsically-compliant and driven by a single actuator through a tendon system. The motion of the robotic devices can be controlled using an Electromyography (EMG) based interface embedded in a cap. The interface allows the user to control the device motion by contracting the frontalis muscle. The performance characteristics of the devices have been measured through experimental set up and the shape adaptability has been confirmed by grasping various objects with different shapes. We tested the devices through qualitative experiments based on ADL involving five chronic stroke patients. The prototypes successfully enabled the patients to complete various bi-manual tasks. Results show that the proposed robotic devices improve the autonomy of patients in ADL and allow them to complete tasks which were previously impossible to perform.