cv

Research Assistant

Korea Advanced Institute of Science and Technology (KAIST)

Developing humanoid whole-body control systems in the Robust Intelligence and Robotics Lab

• Humanoid teleoperation and motion retargeting
• Reinforcement learning-based locomotion
• Vision-Language-Action manipulation

Sole Proprietor

KangZ Robotics

Developed control algorithm solutions for robotic systems, from actuator-level force control to task-level manipulation control

• Real-time haptic-robot synchronization
• Coordinated manipulator tension control

Full-time Researcher

Korea Institute of Machinery and Materials (KIMM)

Developed wheeled-bipedal and dual-arm robots and their locomotion and manipulation control algorithms

• Wheeled-bipedal robot balance and jumping control
• Dual-arm robot development and control
• Novel actuator manufacturing and validation

Graduate Student Researcher

Daegu Gyeongbuk Institute of Science and Technology (DGIST)

Developed quadrupedal robot locomotion and actuator force control algorithms under the supervision of Prof. Sehoon Oh

• Quadrupedal robot high-speed running
• SLIP-based locomotion control
• Series Elastic Actuator force control

Ph.D. in Computer Science

Korea Advanced Institute of Science and Technology (KAIST)

Humanoid Robot, Whole-Body Control, Imitation/Reinforcement Learning, Human-Aware Navigation

• Robust Intelligence and Robotics Lab (RIRO), Prof. Daehyung Park

M.S. in Robotics Engineering

Daegu Gyeongbuk Institute of Science and Technology (DGIST)

Legged Robot Control, Actuator Control

• Motion Control Lab (MCL), Prof. Sehoon Oh
• Thesis: Development and Control of High Driving-Force Quadruped Robot

Visiting Student

University of California, Berkeley

Bachelor of Engineering

Daegu Gyeongbuk Institute of Science and Technology (DGIST)

Robotics, Artificial Intelligence

PC-SureNAV: Preference-Conditioned Superpixel Graph-based Constraint Relaxation for Navigation in Over-constrained Environment

In Preparation

Develops a preference-conditioned navigation method that relaxes constraints over a superpixel graph in over-constrained environments.

Robust Whole-Body Balance Control of Wheeled-Bipedal Robot for Perception-Less Terrain Adaptation

IEEE Access

Presents a robust whole-body controller that enables wheeled-bipedal robots to maintain balance and locomotion in changing environments without terrain perception.

Accurate Jumping Control of Wheeled-Bipedal Robots via Takeoff Velocity Matching and Torque-Optimized Force Trajectory

Robotics and Autonomous Systems (Under Review)

Presents a takeoff-velocity matching and torque-optimized force trajectory controller for accurate commanded-height jumping of a wheeled-bipedal robot.

Iterative Periodic Running Control through Swept Angle Adjustment with Modified SLIP Model

IEEE Robotics and Automation Letters

This paper presents a periodic running control strategy based on a modified Spring-Loaded Inverted Pendulum (SLIP) model to achieve stable running at various velocities. This model simplifies the analytical representation of the stance phase and defines the required swept angle for maintaining periodic motion during the flight phase. Using this model, we develop a feedback control system that ensures the stability of QLSLIP-based periodic locomotion, even in the presence of external disturbances.

Developing a Jumping Algorithm for Overcoming Obstacle in the Wheeled Bipedal Robot

The Korean Society of Manufacturing Technology Engineers (KSMTE) Annual Autumn Conference

Using the dynamics of the SLIP model, the take-off velocity and leg length trajectories needed to achieve the desired height were determined. The robot's kinematics were then controlled to mimic the SLIP model, allowing it to track position and velocity trajectories and jump to the desired height.

Development of Rotating Workspace Ground Contact Force Observer for Legged Robot

2021 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)

Two novel coordinate systems are introduced to describe the joint space motion and the workspace differently, and the ground contact force observer is designed in the novel coordinate systems.

Implementation of Integrated Dual SLIP Dynamics for Sagittal plane motion of Quadruped Robot

International Conference on Control, Automation and Systems (ICCAS)

Proposes a control algorithm that achieves integrated dual-SLIP dynamics while preventing asymmetric behavior between the robot legs.

Transparent Torque Sensor-less Impedance Rendering for Low-cost Direct Drive Motor

2020 IEEE 16th International Workshop on Advanced Motion Control (AMC)

Proposes transparent sensor-less impedance control for a low-cost direct-drive motor by measuring, analyzing, and compensating for torque ripple.

Ripple Minimization for Harmonic-geared Series Elastic Actuator under Force Control

2020 IEEE 16th International Workshop on Advanced Motion Control (AMC)

Develops an accurate force control algorithm that reduces the effect of torque ripple in Series Elastic Actuators (SEA).

Torque Ripple Estimation, Characterization and Compensation for High Torque BLDC Motor based on Multi-sine Function

19th International Conference on Control, Automation and Systems (ICCAS)

Estimates and compensates for periodic torque ripple in a high-torque BLDC motor using a multi-sine function.

Humanoid Whole-Body Teleoperation Pipeline

Developed an Isaac Sim-based teleoperation and human-to-robot motion-retargeting pipeline for whole-body humanoid control and demonstration data collection

• Motion retargeting
• Whole-body control
• Isaac Sim and Isaac Lab
• Real-robot deployment

IROS 2026 Humanoid IKEA Assembly Challenge

Developing an autonomous furniture-assembly pipeline that integrates locomotion, precise base positioning, and Vision-Language-Action manipulation policies

• Reinforcement learning
• Humanoid locomotion
• Precise positioning
• Vision-Language-Action models

Real-Time Haptic-Robot Synchronization Algorithm

Implemented haptic force feedback and synchronized a manipulator trajectory with a haptic device using inverse kinematics

• Geomagic Touch force rendering
• Inverse kinematics
• ROS-based real-time experiments
• MATLAB simulation

Coordinated Manipulator Control for Desired Tension

Developed coordinated manipulator and gripper force control to achieve a desired cable tension

• Tension control
• Inverse kinematics
• Gripper force control
• MATLAB simulation

Wheeled-Leg Robot Development and Locomotion Control

Designed a topology-optimized robot leg and developed balance, navigation, and accurate jumping control algorithms for a wheeled-bipedal robot

• LQR balance control
• SLIP-based jumping
• Torque-optimized force trajectory
• ROS-based experiments

Dual Arm Robot Development and Control

Developed a dual-arm robot and implemented inverse-kinematics control to grasp cups and pour water using integrated grippers

• Actuator validation
• Inverse kinematics
• ROS-based real-time experiments
• Robot World 2023 demonstration

Quadrupedal Robot High-Speed Running Locomotion Algorithm

Developed hopping and gait-pattern generation algorithms and proposed high-speed running control using estimated ground reaction forces

• Sensorless force estimation
• LabVIEW real-time experiments
• MATLAB Simscape simulation

Development of High-Speed Running Multi-Legged Robot Using Back-Driveable Actuator System

Designed a quadruped robot platform with a waist actuator and developed SLIP-based locomotion and Series Elastic Actuator force control

• SLIP-based running
• SolidWorks robot design
• LabVIEW real-time experiments
• MATLAB Simscape simulation

Variable Exercise Machine with a Tension Control Module

Designed a high-torque actuator module and developed force control that adjusts exercise load in response to user intent

• High-torque actuator development
• Force control
• Human-robot interaction

Interworking Modular Exo-Suit Technology for Strength Support

Developed an auxiliary force controller that improves force accuracy for a Series Elastic Actuator integrated into an exosuit

• Impedance control
• Harmonic-drive ripple compensation
• Cogging-torque compensation

Probabilistic Inference of Emotion in Text Using Deep Learning

Developed a recurrent neural network for seven-class emotion analysis of text

• Recurrent Neural Network
• Caffe framework

Beginning and Significance of Scientific Skepticism

Analyzed cases of pseudoscience and proposed methodologies for distinguishing pseudoscience from scientific claims

• Research
• Technical writing

Full Scholarship

Daegu Gyeongbuk Institute of Science and Technology (DGIST)

Received a national scholarship covering full tuition and a stipend.

DGIST Best Research Award

Daegu Gyeongbuk Institute of Science and Technology (DGIST)

Won the Best Topic Research Award in the Undergraduate Group Research Program.