Imagine a small robot traveling along a coastline. First, it crawls across dry ground. A few moments later, it encounters a step or steep bank. Then, without changing its body or attaching new parts, it enters the water and starts swimming.
That sounds like a difficult challenge for a robot—but researchers at New York University’s Tandon School of Engineering have developed a machine designed to do exactly this.
Called WorMa, the worm-like robot can move across different environments, including flat surfaces, slopes, steps and open water. Instead of using complicated mechanical systems to adapt to each terrain, WorMa changes how it moves by redistributing water inside its body.
The research could eventually help create robots for environmental monitoring, coastline inspection and infrastructure surveys, where machines may need to travel through several types of terrain during a single mission.
The Challenge of Amphibious Robots
Robots that move like snakes or worms are naturally useful for exploring difficult environments. Their flexible bodies allow them to squeeze through spaces and travel over uneven surfaces.
However, moving from one environment to another is not simple.
A robot that works well on solid ground may struggle in water. A swimming robot may not have enough traction to climb a slope. Steps and curbs create another problem because the robot needs to generate enough force at the right part of its body to lift itself upward.
Many amphibious robots deal with these challenges by changing their movement patterns, using additional appendages or mechanically changing their shape.
While these solutions can work, they also make robots more complicated. More mechanical parts can mean additional weight, energy requirements and potential points of failure.
The team behind WorMa took a different approach.
Rather than changing the robot's basic movement, the researchers changed where its mass is located.
Meet WorMa
WorMa was developed by Nana Obayashi, an Assistant Professor at NYU Tandon, and Ph.D. student Daniil Filimonov, who works in Obayashi's Prema Lab.
The name WorMa combines the words "worm" and "mass," reflecting the main idea behind the robot.
WorMa is about 50 centimeters (20 inches) long and weighs slightly more than 1 kilogram (2.2 pounds). It has five links connected by four joints, allowing its body to move with an undulating, wave-like motion.
The robot also carries an unusual internal system.
It has a latex balloon water tank in its head and another in its tail. A pump positioned in the middle of the robot can move roughly 300 grams (10.6 ounces) of water between these two tanks.
That amount represents about 28% of the robot's total mass.
By pumping the water toward its head or tail, WorMa can change its center of mass.
This simple change has a major effect on how the robot interacts with its surroundings.
More Weight in the Head Helps It Climb
When WorMa encounters an inclined surface, it needs enough traction to prevent itself from sliding backward.
The researchers found that moving the water toward the robot's head increases the force pressing its front contact points against the surface.
That gives the robot greater traction.
The effect became especially important on steep slopes. During testing, the head-heavy configuration was the only configuration that allowed WorMa to climb the steepest incline tested—19.5 degrees. With other weight distributions, the robot slid backward.
The head-heavy configuration also made the robot more energy-efficient during climbing. Its cost of transport was reduced by at least 33% compared with the other tested configurations.
So, rather than adding special climbing equipment, the robot simply moved water inside its body.
In Water, the Robot Does the Opposite
Swimming presents a completely different challenge.
When WorMa enters open water, placing more mass toward its tail becomes advantageous.
With its center of mass shifted toward the rear, the robot was able to swim up to 26% faster than when its mass was concentrated toward the head.
It was also 52% more efficient in that configuration.
This demonstrates the central idea behind WorMa: the same robot can behave differently in different environments simply by changing its internal mass distribution.
The robot does not need to replace its body or attach a completely different swimming mechanism.
How WorMa Climbs a Step
Steps were even more challenging.
A fixed head-heavy configuration could not reliably get the robot over a step, while a fixed tail-heavy configuration also failed.
The researchers therefore developed a sequence that uses both configurations.
First, WorMa approaches the step with water concentrated toward its head. This gives the front of the robot enough traction.
Then, the robot moves the water toward its tail. This shifts its center of mass backward and allows the head and neck to rise, while the tail pushes the robot closer to the step.
Once the head catches onto the edge of the step, the water is pumped back toward the head.
The robot can then continue moving forward.
Using this strategy, WorMa successfully crossed steps as high as 15 centimeters (6 inches).
Getting Out of Water Also Requires Weight Shifting
Interestingly, entering and leaving water can require different strategies.
When WorMa needs to climb out of the water and onto a slope, the researchers found that concentrating its mass toward the head was necessary.
With the weight positioned elsewhere, the robot did not generate enough force at its front to pull itself onto the slope.
This shows how important the center of mass can be for a small robot moving between very different environments.
One Robot, Multiple Terrains
The researchers eventually combined these strategies into a single test course.
The course included flat ground, a step, an inclined surface and open water.
Instead of stopping to change its body or equipment, WorMa adapted by moving water between its head and tail.
The robot could therefore use different mass distributions for different sections of the course.
According to the researchers, this represents the first demonstration of an undulatory robot making amphibious terrain transitions that include both entering and exiting water.
Why This Technology Matters
The significance of WorMa is not simply that it can crawl and swim.
Its larger potential comes from its simple method of adaptation.
A robot designed for environmental monitoring might encounter dry land, rocks, curbs, slopes, mud and water during the same mission. Carrying specialized equipment for every environment could make such a robot larger and more complicated.
WorMa suggests another possibility: instead of adding more hardware, a robot could adapt by changing how its existing mass is distributed.
This could be useful for applications such as coastline monitoring, wetland exploration and infrastructure inspection.
For example, a future version might travel along a shoreline, climb over an obstacle and enter shallow water without requiring a mechanical transformation between each stage.
What Comes Next?
WorMa is still a research platform, and the current system has room for improvement.
The researchers suggest that faster pumps and sensor-driven control systems could make future versions more responsive.
Instead of being manually programmed for each terrain, sensors could potentially detect changes in the environment and automatically adjust the robot's center of mass.
That could allow the robot to decide when it needs more traction for climbing or when it should shift its weight to improve swimming performance.
The work is part of a broader research direction from Obayashi's group, which also includes ScaFi, a fish-inspired robot designed to be built at different sizes for environments ranging from shallow streams to open water.
Together, these projects explore an important question in robotics: Can one robot adapt to many environments without becoming mechanically complicated?
WorMa offers an intriguing answer. Sometimes, the key to adapting a robot to its surroundings may not be changing its shape or adding new limbs—it may simply be moving what is already inside its body.
Reference: Daniil Filimonov et al., “WorMa: An Undulatory Robot With Center of Mass Regulation via Internal Fluid Redistribution for Amphibious Locomotion,” Advanced Robotics Research (2026), DOI: 10.1002/adrr.70163.
Comments
Post a Comment