Project 01:
Adaptive Human-in-the-Loop Teleoperation via EMG-Informed Impedance Learning
Role: Research Engineer
I developed a real-time human-in-the-loop teleoperation framework for a wearable robot in which joint-level impedance is continuously adapted based on inferred user intent and task phase. The system integrates EMG-informed intent estimation with multimodal sensing from IMU and joint torque to modulate stiffness and damping across interaction phases, replacing command-based control with smooth impedance adaptation. By applying reinforcement learning at the policy-parameter level, the framework personalizes assistance while preserving stability and interpretability. Experimental results showed improved interaction quality, increasing motor support accuracy by 18% and reducing inter-sensor temporal jitter by 22%.
Project 02:
Intent-Aware Shared Control for Dexterous Grasp–Release Manipulation
Role: Research Assistant
I developed an intent-aware shared-control framework for a simulated 7-DOF robotic manipulator that combines real-time neural intent decoding with task-phase–aware torque modulation for dexterous grasp and release. By integrating EEG-based intent estimation with proprioceptive and interaction force feedback, the system dynamically reallocates control authority during contact-rich manipulation. In closed-loop simulation, it achieved 75% intent decoding accuracy with approximately 1.2 seconds latency and improved grasp and release performance by about 30 percent compared to intent-only control.
Project 03:
EMG-Informed Impedance Control for Wearable Upper-Limb Exoskeleton Assistance
Role: Post-Baccalaureate Research Assistant
I led the design and implementation of a 5-DOF wearable upper-limb exoskeleton that delivers intuitive assistance through EMG- and force-modulated impedance control. By adapting joint-level stiffness and damping in real time, the system enables physically transparent human–robot coupling while preserving user agency during contact-rich interaction. Experimental validation showed a 95% improvement in control responsiveness, a 56% reduction in actuation effort, and improved EMG-based intent estimation, supporting ergonomic and reliable long-duration wearable assistance.
Project 04:
EMG-Driven Impedance Control for Post-Stroke Upper-Limb Rehabilitation
Role: Undergraduate Research Assistant and Team Leader
I led the design and development of a 4-DOF EMG-driven upper-limb exoskeleton for post-stroke rehabilitation that adapts impedance control in real time to support safe and intuitive human–robot interaction across multiple therapy modes. By combining EMG- and force-modulated impedance control with a custom low-cost EMG interface, the system regulates joint-level assistance based on user intent and interaction forces. Experimental validation demonstrated 95% joint trajectory fidelity and an 18% reduction in intent-to-torque latency, enabling responsive and stable human-in-the-loop rehabilitation assistance.
Project 05:
Biomechanically Informed Phase-Aware Control of a Muscle-Mimetic Legged Robot
Role: Graduate Research Assistant
I developed a bio-inspired soft legged robot actuated by 30 McKibben-type pneumatic artificial muscles, designed to reproduce muscle-like compliance and nonlinear force generation observed in greyhound locomotion. Using phase-aware coordination of compliant actuation, the control framework integrates phase-synchronized torque control with pressure feedback and IMU-informed joint-phase estimation to regulate force production across the gait cycle. Experimental results showed an 80% increase in locomotion speed, a 25 percent improvement in stability, a 30% gain in force tracking accuracy, and an 18 percent reduction in energy consumption, demonstrating efficient and robust embodied locomotion.
Project 06:
Development of a Wireless EMG Sensor
Role: Undergraduate Research Assistant and Team Leader
I developed a wearable wireless EMG sensor consisting of amplification, filtering, and rectification steps to calculate the Maximum Voluntary Contraction (MVC) value of the biceps and triceps muscles. I transferred the generated signals wirelessly to the graphical user interface via the Bluetooth module of Arduino.
Project 07:
Design of a Passive Micro Flow Sensor System
Role: Undergraduate Research Assistant
I designed an experimental setup in SolidWorks for a microfluidic system with a microflow sensor based on diamagnetic levitation.
Project 08:
Transfer Learning from Real to Imagined Motor Actions in ECoG Data
Role: Team Leader in the Deep Learning Course Project
I led a team of 5 graduate students on the transfer learning from real to imagined motor actions in ECoG data. I trained CNN and LSTM models, achieving 72% accuracy in classifying motor movements.
Project 09:
Classification of motor planning into overt and imagery using an ECoG signal
Role: Team Leader in the Computational Neuroscience Course Project
I directed a team of 5 members on the classification of motor planning into overt and imagery using an ECoG signal. I implemented an SVM classifier to distinguish overt movement and motor imagery.
Project 10:
Representational Mapping Between The Visual Cortex and DCNNs
Role: Team Leader in the Deep Learning Course Project
I led a team of 7 graduate students on the representational mapping between the visual cortex and DCNNs. I trained linear regression models to predict voxel-wise BOLD signals.
Project 11:
Lab-on-card-based sensor for tuberculosis diagnosis using Loop-mediated isothermal amplification
Role: Post-Baccalaureate Research Assistant
I designed the PCB of the low-cost loop-mediated isothermal amplification (LAMP) based Lab-on-card for tuberculosis diagnosis.
Project 12:
Analysis of the Difference in Division Speed of Embryos with Abnormal Chromosomes and Normal Chromosomes
Role: Post-Baccalaureate Research Assistant
I analyzed and compared the division speed data of 11 different cell states of embryos taken from 800 patients with normal and abnormal chromosomes in MATLAB.
Project 13:
Design of a Water Level Controller using PIC16F877A Microcontroller
Role: Team Leader in the Microprocessors and Programming Course Project
I led a team of 2 undergraduate students on the design of a water level controller using the PIC16F877A microcontroller. I drew the circuit diagram of the system and simulated it in the Proteus Design Suite software. I controlled the water level by measuring the water in the reservoir at different levels with the PIC.
Project 14:
Development of a DC to DC Buck Converter
Role: Engineering Intern
I developed a DC to DC Buck (step-down) converter, in which electrical circuit elements and cables are selected in accordance with the desired values. I conducted comprehensive testing of the electronic circuit using advanced diagnostic tools such as a thermal camera and a current probe. I controlled the speed of the DC motor using PWM signals.