Data‐Driven Navigation of Ferromagnetic Soft Continuum Robots Based on Machine Learning

Abstract

Ferromagnetic soft continuum robots (FSCRs) have great potential in biomedical applications due to their miniaturization and remote control capabilities. However, to direct the FSCR accurately and effectively, it is critical to realize inverse kinematics control in navigation, which is difficult for existing mechanical models. Herein, with the path segmentation strategy, an automatic method to navigate the FSCR in different paths based on machine learning is developed. A data‐driven artificial neural network (ANN) model to guide the steering of the magnetically responsive tip is presented. Using parametric simulations as the training data, the ANN model shows good generalization performance to predict control parameters. Moreover, the basic framework of the learning model remains effective when the FSCR materials change, which shows high scalability and is important for adapting to various environments. The study presents a promising strategy for guiding FSCRs in the narrow and tortuous vasculature, which is essential for many biomedical operations.

Hard magnetics and soft materials—a synergy

“…Data-driven modeling techniques leverage available experimental or observational data to develop models that capture the behavior of hMSMs [349]. These approaches often utilize machine learning [292] with neural network architectures [291,350,351] to extract patterns, correlations, and intrinsic relationships present in the collected (or simulated) data. By assimilating empirical data, these models enhance the understanding of the complex magneto-mechanical behaviors exhibited by hMSMs.…”

“…The neural network utilized to mimic the FE architecture is a fully interconnected structure, featuring an output regression layer. Following this approach, Ni et al [292] developed an ANN architecture for ferromagnetic soft continuum robots prepared by reinforcing NdFeB particles into PDMS. The model was trained using parametric simulations generated using the custom development functions of ABAQUS-python software.…”

“…Lastly, detailed visualization portrays the navigation process undertaken by ferromagnetic soft continuum robots. Reproduced from [292]. CC BY 4.0.…”

Continuum Robots and Magnetic Soft Robots: From Models to Interdisciplinary Challenges for Medical Applications

“…Three-dimensional operations and complex magnetic fields pose challenges for data collection, necessitating specialized sensors or computer vision techniques. Data modeling [ 204 , 205 , 206 ] and hybrid modeling offers multiple options for magnetic soft robots, in contrast to the mature technologies of continuum robots. Researchers should draw on continuum robot strategies, emphasizing the integration of precise models, advanced algorithms, and sensing technologies while focusing on interdisciplinary biocompatibility studies in biological environments.…”