In the same way that terrestrial life evolved from ocean swimmers to land walkers, soft robots are progressing, too, thanks to recent Cornell research in battery development and design.
A modular worm robot built by the Organic Robotics Lab and a jellyfish that was a collaboration with the Archer Group, both in Cornell Engineering, demonstrate the benefits of “embodied energy” — an approach that incorporates power sources into the body of a machine, to reduce its weight and cost.
The worm and jellyfish are direct descendants of an aqueous soft robot, inspired by a lionfish and unveiled in 2019, that could store energy and power its applications via a circulating hydraulic fluid — i.e., “robot blood.” Similar blood sustains the new species, but with an improved design for greater battery capacity and power density.
Here is an exclusive Tech Briefs interview, edited for length and clarity, with Lead Author Chongchan Kim, Postdoctoral Researcher.
Tech Briefs: What was the biggest technical challenge you faced while developing these robots and this RFB with a tendon?
Kim: The biggest challenge in the process of robot development was ensuring the reliability of its internal battery system. As the overall system grew more complex, it became crucial for every component to operate without failure. In particular, the robot is equipped with 16 battery cells that must continuously power it, even as the robot moves and deforms — and the batteries themselves were required to deform in tandem with these movements. To maintain stability under these conditions, we adopted a more systematic design and manufacturing approach.
Tech Briefs: What was the catalyst for this project?
Kim: This research began from the need for agricultural robots designed for underground exploration. Since movement underground demands high power and continuous monitoring requires large capacity, we believed that our battery would serve as an excellent example of these capabilities.
Tech Briefs: Can you explain in simple terms how it works?
Kim: The modular worm is composed of four actuation segments (which we call pods) and one control unit located at the end of the body. Each pod is an individual actuator that can contract and expand in length by about 10 percent via a tendon-driven mechanism and also functions as a battery supplying 1.3 V and 3.4 Wh of energy. The four pods are connected in series both mechanically and electrically, delivering a total voltage of 5.2 V to power the control and locomotion of the robot. The modular design of the pods allows the robot to be tailored to the power requirements of the system, while the independent motion of each pod enables a versatile crawl gait pattern, thereby achieving more efficient movement.
We have integrated a zinc iodide redox flow battery into each pod. The battery electrolyte not only stores energy but also serves as the fluid for a hydrostatic skeleton that supports the body of the robot. The incompressibility of the electrolyte ensures that the body maintains a constant volume even under pressure, providing structural support, and also induces radial expansion during longitudinal contraction, which drives the movement of the robot.
We demonstrated the capabilities of the robot by navigating a narrow, hexagonally shaped pipe. The robot bent its battery-integrated body to pass through curved sections. It traveled 3.8 m in 1 hour and 20 minutes using only 4 percent of the total energy capacity.
Tech Briefs: The article I read says, “A crucial part of the design was the decision by postdoctoral researcher Chongchan Kim, the study’s lead author, to use a dry-adhesion method to automatically bond Nafion separators to the worm’s silicone-urethane body as it was printed.” My question is: What made you come to that decision?
Kim: An RFB that stores energy in the form of a liquid within the electrolyte is a promising energy source that allows soft robots to retain compliance and functionality even when batteries with high capacity are embedded. However, the sealing of the ion exchange membrane (which prevents the mixing of the electrolytes in the cathode side and the anode side) has complicated integration into soft robots. Typically, sealing of ion exchange membranes is achieved through rigid constraints imposed by mechanical force, resulting in a loss of compliance. A method that permits deformation while still achieving hermetic sealing was needed, and our discovery shows that strong adhesion between the ion exchange membrane and the soft battery cell housing allows battery cells to maintain perfect sealing even under various deformation conditions. Additionally, dry adhesion improves manufacturing consistency and enables more reliable operation.
Tech Briefs: The article also says, “In the future, Shepherd anticipates making high-capacity, embodied-energy robots with the same type of lithium-polymer batteries, but that also have skeletons and can walk … The result will be something that is more ‘like us,’ he said. ‘An imperfect organism. But still doing pretty good.’” My question is: Do you have plans for further research/work to do this?
Kim: Our group does not conduct research using lithium batteries currently, but we see the rise of embodied energy robots as an inevitable trend. Just as fat in living organisms not only stores energy but also helps maintain body temperature, protects organs from mechanical shock, and provides structural support for the skeleton, distributed energy components throughout the robot's body will enhance system efficiency through multifunctional applications.
Tech Briefs: If not, what are your next steps?
Kim: Our technology is not yet powerful enough to drive robots that can walk or run. Additionally, there are still many practical challenges in design and manufacturing methods that require significant improvements. However, we are tackling these challenges one by one. Just as living organisms evolved — progressing from aquatic to terrestrial environments and from soft-bodied forms to structures with skeletal support — our approach will similarly advance in stages. New efforts and innovations are continuously underway to drive this evolution forward.
Tech Briefs: Do you have any updates you can share?
Kim: We are focusing on developing a fully liquid redox flow battery, in which all active redox species are dissolved in the electrolyte. This approach completely resolves the coupling issues between electrodes and capacity that were a persistent problem in the previous zinc iodide RFB system, enabling a system with higher capacity and enhanced stability.
