Magnetic Fields Drive Smarter, Intelligent Soft Robots

National University of Singapore College of Design and Engineering

Soft robots are prized for their agility and gentle touch, which makes them ideal for traversing delicate or enclosed spaces to perform various tasks, from cultivating baby corals in laboratories to inspecting industrial pipes in chemical plants. However, achieving embodied intelligence in such systems, where sensing, movement and power supply work together in a untethered configuration, remains a challenge. Flexible materials can deform and adapt, but their power sources are unable to do so. Conventional batteries often stiffen the robot's body, drain quickly, or degrade under strain, all of which leave soft robots tethered or with a short lifespan.

Assistant Professor Wu Changsheng and his team across the Department of Materials Science and Engineering and the Department of Electrical and Computer Engineering, College of Design and Engineering, National University of Singapore, found a way to turn that limitation into an advantage. Their study, published in Science Advances , demonstrates that the same magnetic fields used to control soft robots can also enhance the performance of the batteries inside them.

"Magnetic fields are typically used to stimulate motion in soft robots, in a process called 'actuation,' but we realised they could also stabilise electrochemical reactions inside flexible batteries," explained Asst Prof Wu. "Allowing actuation and energy management to share the same physical principle enables us to make the robot truly self-contained and efficient."

The team designed flexible zinc-manganese dioxide (Zn-MnO₂) batteries encapsulated in soft silicone and vertically stacked within a robot body inspired by the manta ray. Importantly, unlike traditional lateral arrangements, this vertical integration maximises space and keeps the robot pliable.

"We took a page from the manta ray because its body embodies exactly what we want to achieve — a natural integration between movement, sensing and energy use," said Asst Prof Wu. "Its form allows synergistic coordination of multifunctionalities in an efficient and compact way, making it a perfect biological model for embodied intelligence."

Through tests, the researchers revealed that the magnetic field produced by the robot's own ferromagnetic actuators stabilised the internal chemistry of the flexible batteries, reducing the risk of dendrite growth — needle-like metal deposits that can cause short circuits — and maintaining energy output under repeated stress and bending. Under magnetic enhancement, the batteries retained 57.3% of their capacity after 200 cycles, nearly double that of unenhanced samples.

"After further investigation, we found out how this enhancement works. The Lorentz force generated by the magnetic field acts on moving ions in the electrolyte and redirects zinc ion trajectories during plating, creating a more uniform ion flux that promotes even zinc deposition at the anode and effectively suppresses dendrite growth. Simultaneously, the magnetic field aligned electron spins within the manganese oxide lattice, reinforcing atomic bonds and preventing crystal degradation during charge and discharge," added Xiao Xiao, a PhD student in Dr. Wu's team and a co-first author of the publication. "This dual magneto-electrochemical stabilisation, achieved in a fully flexible format, is an exciting step towards durable onboard power systems for soft robots operating in challenging, dynamic environments."

Embodied intelligence in action

To demonstrate the concept, the team built a magnetically actuated manta ray robot equipped with the flexible batteries, soft magnetic elastomer actuators and a lightweight hybrid circuit for sensing and wireless communication. The robot's fins flap in response to external magnetic fields generated by a coil or electromagnet array, enabling it to stabilize the locomotion and adapt to different water surfaces.

As expected, the same magnetic fields that drive and steer the robot also enhance its energy stability — confirming the researchers' vision to merge motion control with power management. The robot can perform basic swimming manoeuvres, such as linear propulsion, 90-degree turns, and complex trajectories, all while transmitting real-time data to a computer that visualises its movements in a digital twin environment.

Within this setup, the robot exhibited autonomous decision-making abilities. For instance, when it encountered an obstacle, onboard inertial sensors detected sudden changes in acceleration, which in turn prompted the control system to adjust its orientation and reroute. The robot successfully navigated narrow passages through posture adjustments and executed U-turns when facing impassable obstacles. During perturbation tests, the feedback algorithm rapidly corrected deviations in yaw, pitch, and roll angles caused by external forces like waves or physical contact, maintaining stable trajectory control. The integrated temperature sensors enabled environmental monitoring, mapping thermal gradients in aquatic environments.

"Integrating actuation, sensing and power vertically across the robot's body allows us to maximise functional area without compromising the 'softness' of the robot," said Asst Prof Wu. "This design enables the robot to move, feel and respond to its environment in real time."

Looking ahead, the researchers plan to expand the robot's sensing capabilities by incorporating miniaturised sensors such as ultrasonic sensors for surrounding perception or chemical detectors for water-quality monitoring. They are also exploring how magnetic enhancement could improve other battery chemistries, such as lithium-ion, or other battery forms, such as wearable battery fibres, to enhance energy density and operational endurance.

"Our vision is to enable soft robots that can think and act autonomously in complex or inaccessible spaces — whether inspecting pipelines, monitoring marine habitats or supporting medical interventions in the operating theatre," added Asst Prof Wu. "Through some creative and critical thinking about how energy and intelligence are built into the body, we can bring soft robotics much closer to nature's own designs — like the graceful giant of the sea, the manta ray."

Asst Prof Wu's work was carried out in collaboration with researchers from Tsinghua University, the University of California, Los Angeles (UCLA) and Dartmouth College.

/Public Release. This material from the originating organization/author(s) might be of the point-in-time nature, and edited for clarity, style and length. Mirage.News does not take institutional positions or sides, and all views, positions, and conclusions expressed herein are solely those of the author(s).View in full here.