Four-legged robots are becoming more common in jobs that are difficult or dangerous for people, including search-and-rescue missions, inspections of industrial sites and carrying equipment across rough ground.
Unlike wheeled robots, however, these machines use much more energy because every step requires their motors to repeatedly lift and move their legs.
Now, researchers have developed a new type of 3D-printed robot foot that helps these machines walk more efficiently while using less battery power.
The study was published in the International Journal of Precision Engineering and Manufacturing–Green Technology.
The research was carried out by Dr. Keun Park and Dr. Jung-Yup Kim from Seoul National University of Science and Technology in South Korea.
Their goal was to find a simple way to reduce the amount of energy needed for walking without making the robot more complicated.
Many engineers have tried adding springs to robot legs to capture energy when a foot hits the ground and release it during the next step.
While this idea works well in some situations, it is less effective when robots walk slowly because much of the stored energy is lost instead of helping the robot move forward. This can also make the robot less stable.
Instead of adding springs to the legs, the researchers focused on the robot’s feet. They created special lightweight feet using a 3D-printing technique.
The feet contain tiny repeating structures filled with small open spaces, making them both light and flexible while still being strong. These porous designs can compress when the robot steps down and then spring back, returning some of the stored energy during the next movement.
The team tested three different foot designs, known as primitive, gyroid and diamond structures. After measuring how well each one absorbed and returned energy, they found that the diamond-shaped design provided the best combination of flexibility, strength and energy recovery.
Creating better feet alone was not enough. The robot also needed to learn how to take advantage of them. To do this, the researchers used a type of artificial intelligence called deep reinforcement learning.
Instead of following a fixed walking pattern, the robot practiced different ways of walking and gradually learned which movements used the least energy while working with the flexible feet.
The AI system learned to match the robot’s steps with the natural compression and rebound of the new feet. This allowed the motors to do less work because part of the movement came from the energy stored in the feet themselves.
The researchers tested the new system on a commercially available four-legged robot. Compared with standard solid feet, the new porous feet reduced battery power use by between 1.4% and 6.2% at walking speeds ranging from 0.4 to 1.0 meters per second. Just as importantly, the robot remained stable throughout the tests, showing that the AI controller was able to make effective use of the energy stored in the feet without causing unnecessary movements.
Although the energy savings may seem modest, even small improvements can make a significant difference for robots that operate for long periods or carry heavy loads. Longer battery life means fewer charging stops and more time spent completing useful work.
The researchers believe this combination of smart 3D-printed materials and artificial intelligence could help create quieter, more energy-efficient four-legged robots for many real-world tasks.
In the future, robots used for inspections, warehouse logistics, delivery services and emergency response could all benefit from this simple but effective design.
