As quirky as they sound, jumping robots are more than just a gimmick. They’re useful in disaster rescue, planetary exploration, and surveillance. Now, researchers might have designed the optimal jumper robot. This small, inconspicuous device can jump a whopping 120 meters in the air.

To make a jumping bot

Jumping robots usually rely on a classic physics mechanism: the spring. Springs store elastic energy and then abruptly release it.

Spring-driven jumping robots use motors to gather that elastic energy and then release it to propel themselves into the air. This method, known as power amplification, allows springs to generate more power and do more work than standalone motors.

This field of research is surprisingly broad. The technology is generally inspired by nature, particularly jumping insects like fleas and grasshoppers. The first applications were space-based. Researchers were looking to explore harsh extraterrestrial environments with them. Such robots have also found applications in hazardous environments such as disaster rescue and surveillance, but the main challenges still lie in hardcore physics.

Co-author Dr. John Lo, Research Associate in Space Robotics at The University of Manchester, explains:

Robots are traditionally designed to move by rolling on wheels or using legs to walk, but jumping provides an effective way of traveling around locations where the terrain is very uneven, or where there are a lot of obstacles, such as inside caves, through forests, over boulders, or even the surface of other planets in space.

“While jumping robots already exist, there are several big challenges in the design of these jumping machines, the main one being to jump high enough to overcome large and complicated obstacles. Our design would dramatically improve the energy efficiency and performance of spring-driven jumping robots.”

The big challenge, however, is timing the jump.

Premature takeoff

To be as efficient as possible, the robot has to jump at the exact right time. Robots often fail to reach their maximum jump height due to premature take-off, where the robot disengages from the ground before the spring energy is fully released. This results in an incomplete transfer of stored elastic energy to gravitational potential energy, significantly reducing jump height.

Co-author Dr. Ben Parslew, Senior Lecturer in Aerospace Engineering, says was not clear at all how to address this problem.

“There were so many questions to answer and decisions to make about the shape of the robot, such as should it have legs to push off the ground like a kangaroo, or should it be more like an engineered piston with a giant spring? Should it be a simple symmetrical shape like a diamond, or should it be something more curved and organic?”

In the end, the focus was on removing any unneeded movement while maintaining the necessary strength and stiffness. They landed somewhere in the middle: a robot that’s small enough to be light and agile, but large enough to carry a motor.

“Our structural redesigns redistribute the robot’s component mass towards the top and taper the structure towards the bottom. Lighter legs, in the shape of a prism and using springs that only stretch are all properties that we have shown to improve the performance and most importantly, the energy efficiency of the jumping robot.”

The ideal design

The study classified take-offs into three categories: idealized take-off, premature take-off, and delayed take-off. Idealized take-off occurs when all stored elastic potential energy is fully released at take-off. Premature take-off happens when the spring still stores elastic potential energy at take-off, while delayed take-off occurs when the spring length exceeds its natural length at take-off, storing excess energy. Both premature and delayed take-offs are undesirable as they lower energy conversion efficiency.

Researchers used two simple dynamic models to illustrate these take-off categories: a prismatic multibody system driven by a translational spring and a rotational rigid body driven by a rotational spring.

The prismatic model demonstrated delayed take-off due to the presence of an unsprung mass, which resulted in a reduced center of mass velocity. The rotational model showed that centripetal forces acting on rotational masses could cause premature take-off, further reducing energy efficiency.

However, the rotational model proved superior in the end. By concentrating mass at the body and minimizing mass in the legs and feet, this type of robot can reduce inertial effects that lead to premature take-off and inefficient energy conversion and jump hundreds of time its size.

Despite this optimized design, however, the researchers still have some hurdles against them. They’ve found the way to use the robot’s strength, but controlling its direction is still a challenge. The next goal is to perform more controlled jumps and also harness the energy to ensure several jumps per charge.

Journal Reference: John Lo et al, Characterising the take-off dynamics and energy efficiency in spring-driven jumping robots, Mechanism and Machine Theory (2024). DOI: 10.1016/j.mechmachtheory.2024.105688