We are pleased to announce that three papers from our laboratory have been accepted to ICRA 2025. One of the accepted papers is “Application of Koopman Direct Encoding-Based Model Predictive Control to Nonlinear Electromechanical Systems”.

This study extends the Koopman Direct Encoding (KDE) framework to nonautonomous and real-world systems by introducing a KDE-based Model Predictive Control (KDE-MPC) method. The proposed controller achieved more stable and smoother position tracking than traditional PID control in nonlinear electromechanical systems with contact–noncontact transitions, demonstrating its practicality for complex robotic control.

Next accepted paper is “Spatial Sensitivity Equalization of ERT-Based Robotic Skin Through Gauge Factor Distribution Optimization”.

This study proposes a topology optimization-based method to equalize the spatial sensitivity of Electrical Resistance Tomography (ERT) sensors) for more accurate tactile sensing. By modulating sensor conductivity and fabricating optimized conductive patterns via screen printing, the method significantly reduced sensitivity disparities and improved contact force reconstruction in both simulation and real-world tests.

Last one is “Wearable Soft Sensing Band with Stretchable Sensors for Torque Estimation and Hand Gesture Recognition”

This study introduces a wearable soft sensing band that estimates muscle activity through volume changes without the need for pressure or adhesive electrodes. The device achieved torque estimation errors below 11.5% and 92.9% gesture classification accuracy, demonstrating a lightweight and flexible alternative to sEMG for applications in healthcare, rehabilitation, and robotics.