New High-Performance Workstation for Rehabilitation and Bioinspired Robotics
Our Laboratory for Bioinspired Robotics has recently acquired a new high-performance computing workstation to support ongoing research in rehabilitation robotics, bioinspired robotics, electromyography (EMG), data-driven modeling, and advanced control.
The workstation will provide the computational backbone for connecting, synchronizing, and processing data from our multichannel EMG systems and motion-tracking equipment. It will also support large-scale numerical simulations, machine-learning workflows, development of Koopman-based models and controllers, and future real-time experiments with soft robotic rehabilitation devices.
The system is equipped with:
- AMD Ryzen Threadripper PRO 9975WX – 32 cores / 64 threads
- 256 GB DDR5 ECC memory
- NVIDIA RTX PRO 4500 Blackwell with 32 GB ECC GPU memory
- 6 TB NVMe storage
- Dual 10 Gb Ethernet
- Linux and ROS 2 research environment
The workstation will be particularly important for combining multichannel sensing, motion analysis, soft robotics, numerical simulation, and data-driven modeling and control within a common experimental and computational platform.
The equipment was acquired with financial support from the Croatian Science Foundation (HRZZ) and the European Union through the University of Rijeka (UNIRI). The HRZZ-supported part of the equipment was acquired within the project ASoRo – Advanced Soft Robots: Data-Driven Development, Modeling and Control (UIP-2025-02-1541), while the UNIRI-supported part was acquired within the project KMD-EMG – Koopman-driven real-time sEMG signal decomposition for robotic rehabilitation (uniri-iz-25-116), funded by the European Union – NextGenerationEU.
The new infrastructure significantly expands our capabilities for computationally demanding research and provides a platform for the next stages of our work on adaptive and intelligent rehabilitation robotic systems.
For additional technical information: