ML-Driven Stingray Robots Enhance Marine Exploration
Teaching Underwater Stingray Robots to Swim Faster and with Greater Precision Using Machine Learning
Tags: Singapore University of Technology and Design, Singapore, Science & Exploration, Electronics & Robotics
SUTD researchers have used machine learning to model the swimming dynamics of stingray-like robots, enabling faster, more precise movement in underwater environments. Their Deep Neural Network (DNN) model predicts optimal flapping motions to achieve target propulsive forces, simplifying complex modeling challenges inherent in soft robotics. Applications include marine exploration and inspection, where precise control is essential. Conducted in collaboration with A*STAR, the research demonstrates that DNNs can effectively mimic the complex physical properties of soft-bodied robots. This innovation could advance autonomous underwater robotics by allowing real-time adaptation to dynamic marine conditions.
IP Type or Form Factor: Software & Algorithm
TRL: 4 - minimum viable product built in lab
Industry or Tech Area: Marine Exploration & Conservation; Robotics
