New Robotics and Automation Study Findings Recently Were Reported by Researchers at Daegu Gyeongbuk Institute of Science and Technology (DGIST) (Hysteresis Compensation of Flexible Continuum Manipulator Using Rgbd Sensing and Temporal...).

Researchers at the Daegu Gyeongbuk Institute of Science and Technology (DGIST) in South Korea have developed a data-driven approach using Deep Neural Networks (DNN) to compensate for hysteresis in cable-driven manipulators. Hysteresis, caused by cabling effects such as friction and elongation, poses control difficulties for these manipulators. The researchers collected physical joint configurations using RGBD sensing and fiducial markers to model the hysteresis and found that the Temporal Convolution Network (TCN) demonstrated the highest predictive capability. The proposed control algorithm reduced position and orientation errors by over 60%, suggesting that it could enhance control precision and improve surgical performance. [Extracted from the article]