Gastrointestinal tract diseases affect millions worldwide, and traditional drug delivery methods suffer from low targeting efficiency and potential side effects due to systemic drug distribution. Magnetic soft robots have emerged as a revolutionary solution for minimally invasive medical operations, thanks to their small size, untethered drive, and agile motion. However, existing magnetic soft robots face critical limitations in multi-angle folding, real-time reconfigurable magnetization, and adaptability to the irregular, confined gastrointestinal cavity environment, hindering their clinical application for targeted drug delivery.
To address these challenges, a collaborative research team consisting of scholars from China University of Mining and Technology, Soochow University, RWTH Aachen University, and the University of Oxford developed a magnetic soft sheet robot based on magnetorheological fluids. The robot adopts a four-layer fully soft sheet structure, composed of upper and lower linear low-density polyethylene surface layers, a magnetorheological fluids core layer, and a polyamide nylon mesh support layer. With a compact size (30 mm in length, 10 mm in width, 1.5 mm in thickness) and light weight (0.55 g), the robot is non-magnetized in a zero magnetic field, eliminating unintended magnetic interference in the human body.
The core innovation of this robot lies in its real-time reconfigurable magnetization and reversible folding performance. The internal magnetorheological fluids can form magnetic particle chains along the magnetic field direction in milliseconds under external magnetic, and the magnetization direction can be dynamically adjusted with the spatial magnetic field. Driven by a 5-degree-of-freedom magnetic field platform, the robot achieves precise multi-angle folding, reducing its surface area to one-third of the original to navigate narrow intestinal tracts, and can be unfolded to a large surface area for stable movement in the stomach cavity—realizing flexible adaptation to the varying spatial sizes of the gastrointestinal tract.
The research team fabricated five prototypes of the soft sheet robot with different magnetorheological fluid densities (3.0 g/mL to 4.2 g/mL) and conducted comprehensive motion performance tests across diverse scenarios. The robot demonstrated stable flip, steering, and folding motions on smooth surfaces, flexible fluff surfaces, and slope surfaces, as well as in underwater environments. Even under load (carrying biodegradable hydrogel drugs with a mass of 0.15 g, approximately 30% of the robot’s own mass), it maintained reliable movement performance. Critical validation was conducted through ex vivo porcine stomach experiments simulating the human gastrointestinal environment. In 10 repeated tests, the robot reached any preset lesion site in the porcine stomach within an average of 5 minutes and stably attached to the drug release position, with the loaded hydrogel drugs dissolving within 30 minutes to achieve localized targeted therapy. Additionally, ultrasonic detection technology (Voluson E10) successfully monitored the robot’s real-time motion in the closed gastric cavity, confirming its traceability and controllability in enclosed biological spaces—providing a technical guarantee for clinical application monitoring.
Biocompatibility tests further verified the robot’s safety for human body application: immersed in simulated gastric juice (pH 1.2) and intestinal juice (pH 6.8) at 37°C for 24 hours, the robot showed no surface rupture, volume expansion, or shape deformation. Detection of heavy metals and harmful substances in the extract solutions found no exceeding of safety limits, and no bacterial colony growth was observed in microbial culture tests, confirming the robot’s biocompatibility and non-toxicity in the gastrointestinal tract environment.
The research team noted that this magnetic soft sheet robot breaks through the technical bottlenecks of traditional magnetic soft robots in folding adaptability and magnetization reconfigurability. Its advantages of untethered drive, fully soft structure, and high targeting accuracy make it an ideal noninvasive medical device for gastrointestinal tract targeted drug delivery. Future research will focus on optimizing the robot’s driving and control capabilities to adapt to the acidic gastric environment, gastrointestinal peristalsis, and fluid disturbances, as well as improving the coordination between magnetic field control and ultrasonic detection to achieve more precise clinical operation.
About China University of Mining and Technology (CUMT)
China University of Mining and Technology, located in Xuzhou, Jiangsu Province, is a national key university with distinctive strengths in engineering, particularly in mechanical and electrical engineering, materials science, and biomedical engineering. The university is committed to interdisciplinary research and innovation, bridging engineering technology with medical science to develop cutting-edge biomedical devices and technologies. With advanced research laboratories and a strong collaborative network with domestic and international universities, CUMT continues to make significant contributions to the development of noninvasive medical technology and intelligent robotics.
About Author
Xinhua Liu is a researcher at the School of Mechanical and Electrical Engineering, China University of Mining and Technology, focusing on the research and development of magnetic robots, magnetorheological materials, and biomedical engineering applications. His research interests include the design of intelligent soft robotic systems, magnetic field control technology, and targeted drug delivery devices. Prof. Liu has published over 150 papers in important domestic and international journals and has obtained more than 60 national invention patents.
Funding information
This research was supported by the National Natural Science Foundation of China (52405081, 52575080), the Natural Science Foundation of Jiangsu Province (BK20231066), and the Joint Funds of the National Natural Science Foundation of China (U24A20116).
Media contact:
Xinhua Liu
School of Mechanical and Electrical Engineering, China University of Mining and Technology
Email: liuxinhua@cumt.edu.cn
Ting Zhang
School of Mechanical and Electrical Engineering, Soochow University
Email: tzhang@suda.edu.cn
END