Project aim: Investigate digital fabrication technology to create a robot skin and seek out structural optimization strategy. 

Project aim: Integrating a soft skin sensor on a soft pneumatic actuator. 

Collaboration with MPI-IS

This project developed an application-specific integrated chip (ASIC) to miniaturize ERT sensors. 

Paper: Intelligent Service Robotics

Collaboration with IMS Chips

This project attempts to use sim-to-real transfer learning with a finite element multiphysics model. This new calibration method has the possibility to improve the sensing performance of ERT-based tactile sensors.

Papers: Transactions on Automation Science and Engineering 2022, ICRA2020

Dataset: Experiment Data and Python Codes

Collaboration with KAIST and MPI-IS

This project created a robotic skin using hydrogel-elastomer hybrids. Both deep pressure and light touch were detected from this skin using resistance tomography and passive acoustic tomography. As a proof-of-concept, damage repairment and sensorized prosthetics are demonstrated.

Papers: Science Robotics 2022

Videos: Multi-modal sensing, Damage repair, Sensorized Prosthetics

Collaboration with KAIST, MIT

Robot therapy is a promising intervention for children with autism spectrum disorder (ASD). This project designed a soft tactile sensing suit on a commercially available humanoid robot. 

Papers: Frontiers in Robotics and AI, Paladyn Journal of Behavioral Robotics 2021, HRI2021

Videos: DIY Guide, Experiment Video

Dataset: Experiment Data, Sensor Pattern

Collaboration with MPI-IS

Using barometric chips is a low-cost way to create a tactile sensor. This project created a 3-D fingertip tactile sensor using 16 barometric chips.

Patents: US patent filed

Collaboration with MPI-IS

The resistance tomography is applied to a special piezoresistive laminate. As a result, lateral strain distribution and contract pressure distribution can be simultaneously estimated from a flexible laminate. 

Patents: US patent filed

This project creates textile-based tactile skin that can cover the entire body of a robot. A reconstruction method called electrical resistance tomography (ERT) is used to achieve large-area coverage, ease of manufacturability, and robustness. A low-cost, large-area sensor has been demonstrated and comprehensively compared with biological systems.

Papers: ICRA 2020

Videos: Demonstration

Collaboration with KAIST

This project creates textile-based tactile skin that can cover the entire body of a robot. A reconstruction method called electrical resistance tomography (ERT) is used to achieve large-area coverage, ease of manufacturability, and robustness. A low-cost, large-area sensor has been demonstrated and comprehensively compared with biological systems.

Papers: AsiaHaptics 2019, ICRA2019, Smart Materials and Structures 2021

Videos: Sensor Calibration, Internal Electrode Effect

Collaboration with KAIST

This project aimed to build soft tactile sensors using soft nanocomposite (MWCNT/Silicone). The sensor can detect multi-point, multi-directional strains using anisotropic electrical resistance tomography (aniso-ERT). The strain sensors using the proposed approach can be made in various three-dimensional shapes with low manufacturing cost. 

Papers: Scientific Report 2017, HRI2017

Patents: KR-10-2016-0006287,  KR-10-2017-0033907

Videos: Tactile Pad, Soft Tactile Interface

This project aimed to build a printable skin adhesive stretch sensor to estimate rotation angles of multi-axis joint for biomedical engineering applications, such as gait analysis, gesture recognition, and motion monitoring. As a printable piezoresistive material, silicone rubber mixed with multiwall carbon nanotube composites were fabricated to make a highly stretchable (up to 120%) strain sensors. 

Papers: Smater Materials and Structures 2018, ICRA2016

Videos: Demonstration 

An active upper limb support system was developed in this project. The system has an actuator to lift the user's upper limb. The user's motion intention was detected by two signals: surface electromyography (sEMG) signals and a strain gauge. 

Papers: Journal of Medical Robotics Research 2017, URAI2014, SICE2014

Patents: KR-10-1223209-0000

This project investigated a simulation model of flexible needle-soft tissue interaction. A multi-layer soft tissue is considered in this model to estimate the needle deflection accurately. The study was validated with a porcine tissue with optical cameras. 

Papers: Medical & Biological Engineering & Computing 2014

This project developed a robotic palpation system to mechanically examine prostate cancer. The system has a 3DOF positioner with a 2DOF palpation tip equipped with a custom force-torque sensor. The system was validated with ten in-vitro prostate specimens in collaboration with Yonsei Severance Hospital  

Papers: Journal of Medical Robotics and Computer Assisted Surgery 2014, Journal of ICROS 2010, Biorob2012, ICRA2011, Biorob2010

Patents: KR-10-1223209-0000

This project developed a 3DOF haptic device and finite element (FE)-based soft tissue simulator. The device is made of three BLDC motors with a cable-driven mechanism to reduce the backlash effect. The simulator computes reaction force using a hyperelastic tissue model and a collision detection algorithm. 

Patents: KR-10-1230950-0000