Lizards’ astounding crash-landing ability validated with Soft Robot perching

Aerial soft robotic physical models with an active tail reflex provide experimental validation field discovery of Geckos' superpowers in the rainforest canopy are not entirely down to its unparalleled feet. Versatile tails play just as much a pivotal role.

Like Comment
Read the paper

We position our interdisciplinary research at the interface of engineering, materials science, and biomechanics, specializing on soft active materials, biomimetic mechanisms, and robotics inspired by original biomechanical discovery. We explore the behavioral and morphological adaptations of natural systems to recreate robust multimodal locomotion at our Lab at the Max-Planck Institute for Intelligent Systems.

Figure. 1. Left: Illustration of Fall Arresting Response (FAR). Credit: Peter Bräm, and Andre Wee. Right: Robot lander and a gecko, resting after landing on a tree trunk. Credit: Ardian Jusufi.

The Asian flat-tailed gecko (Hemidactylus platyurus) is found in the Indomalayan lowland tropical rainforests, where it lives in the trees. Although it was thought to be non-volant, we recorded it can glide considerable distances from tree to tree to evade predators. When the trees are spaced close together, as geckos take to the air, jumping and landing takes place at the blink of an eye. Moving through the air at high speeds of ~120 body lengths per second at 6m/s (or ~21 Km/h) for 'directed aerial descent' the geckos crash-land unbraked into the neighboring tree. Remarkably, following a dedicated landing maneuver the lizards are still able to stick to the vertical surface without becoming dislodged.

Movie 1: Summary video containing supporting information movies from the paper, modified. Narrated by Ardian Jusufi. 

We conducted experiments in the Indonesian-Malaysian rainforest to study the mechanism behind this remarkable landing ability. We performed field research climbing seven-meter-high ladders at high ambient temperatures of up to 38 degrees Celsius and 85 % humidity. Thanks to high-speed cameras, we were able to film the flight and perching behavior of geckos in their natural environment, tracking how they use their tails not only to control gliding flight, but also for perching on a challenging vertical target. 

The lizards aim at a target, take off and land at lightning speed and are capable of landing in a confined space. 

As can be seen in the video below, the gecko tries to cushion the impact by bending its trunk backwards by more than 100 degrees. During this bending, the front feet lose their grip and only the hind legs remain attached. This backward bending of the trunk dissipates energy by pressing the tail against the tree trunk. Animals that lost their tails could not sufficiently cushion the energy and fell off the trunk. The tail thus acts like a fifth leg, helping the gecko stabilize itself after impact.

But how can one conclusively show that the tail has this stabilizing effect? Going back to the drawing board in the lab we developed a perching soft robot to better comprehend the pressures acting on the animal. This robotic physical model of the gecko's landing behavior serves for experimental validation of the dynamics on impact. To emulate the lizard's compliant body, we built a soft robotic platform with a soft torso, allowing the tail to be detached and reattached. When the front foot of the robot comes into contact with a surface, it bends its tail, similar to the climbing gecko reflexes previously found in the rainforest. A microswitch on the robot's underbelly provided tail actuation input, and gave feedback when the bottom of the head made contact with the wall. The information is processed by a microcontroller on the "shoulder" to immediately activate a motor in the "pelvis" that pulls on a "tendon" to push the tail against the wall to stabilize the robot. 

The soft robotic lizard was catapulted at the same velocity that the geckos collided with the tree, onto a felt-lined wall to which the robot's Velcro-covered foot could stick. After collision, we assessed the force exerted on the robot's feet. The longer the robot's tail, the less force was required to lift its hind feet off the ground. The less force applied, the easier it was for the robot to maintain its grip. However, for a robot lacking a tail, the pressures acting on the hind feet become excessive, causing the robot to lose its grip, bounce off, and fall.

Fig 2. Ardian Jusufi with Soft Robotic Lander (image credit: Ardian Jusufi Lab). 
Thus, this experiment supported our idea that the gecko can stabilize itself on a vertical surface only after striking it with its tail at great speed.

Ultimately, we would like to transfer the insights gained from the gliding flight of the gecko to robotics. We have already succeeded in doing this with climbing robots, which have a biologically inspired mechanical tail that, like geckos, gives them more stability when climbing.

This research was done by Rob Siddall, Greg Byrnes, Bob Full, and Ardian Jusufi at the Max-Planck Institute for Intelligent Systems Stuttgart, Siena College, and UC Berkeley.

The paper can be found here:

Ardian Jusufi

Max Planck Independent Group Leader, Max Planck Institute for Intelligent Systems


Dr. Ardian Jusufi positions his highly interdisciplinary research at the interface of engineering, materials science, and biomechanics. He specializes on soft active materials, biomimetics, and robotics inspired by original biomechanical discovery. Ardian is a pioneer in the biorobotic investigation of robust locomotion of robots underpinned by his discoveries of tails as control appendages enabling rapid disturbance rejection and multi-modal transitions. His research has been featured in leading journals including PNAS and Nature (both cover articles), as well as Soft Robotics, Current Biology, and Advanced Intelligent Systems.


With a goal to achieve the dream of life-like movement, he integrates soft actuators and flexible sensors made of hyperelastic silicone elastomers (containing liquid metal) and embeds them into a diving, soft robotic fish, for instance. Their integration enables new skills in swimming and arboreal robots to increase their ability to overcome obstacles. In contrast to robots with mostly hard components, the musculoskeletal system is capable of adapting to multiple dynamic perturbations simultaneously. Life-like movement not only expands upon robot capability, it also enables original discovery in experimental biomechanics. His research integrates robotics, ‘smart’ materials, with comparative biomechanics. In this spirit, Dr. Ardian Jusufi is a leader in the field of biorobotics enabled by biomimetic materials and mechanisms.


What is the research focus and the team’s vision?

As the physicist Richard Feynman famously said: “What I cannot create, I do not understand”. 

The past decade has presented a dramatic expansion in the development of mobile robots and the application of robotic systems to practical tasks. Despite the proliferation of computation and sensing at the small scale, robots still remain largely unable to access all but the most structured environments, and unable to reach the performances of natural systems.


The capability gap between nature and human-made devices is most apparent in locomotion. Biological organisms traverse various cluttered terrains (e.g. granular media) effortlessly, whereas even the most advanced robots get stuck. Biomechanical system robustness emerges from the interplay of materials and mechanics based solutions barely dreamt of by robotics engineers. Although novel computational solutions and biologically inspired mechanisms have advanced robot locomotion, most platforms are made of predominantly rigid parts, offering precision and control. By contrast, biological tissues consist in large part of water, making compliance ubiquitous in nature. Morphing structures such as fins, tails, and wings allow for shape changes on the body, thus enabling unparalleled transmedium movement with multi-functional appendages. Soft active materials are now required to emulate and decipher the general principles of natural motion systems – Experimental robotics can advance motion science by breathing new life into physical models. Embodimentprovides physical resilience by exploiting the morphological intelligence of the body to simplify control. 


Experimental validation of gliding and climbing is measured in field research. However, the softest model systems are found under water. To this end, Dr. Jusufi’s research has shown that soft fluidic actuators made of silicone elastomers that enable shape changes on the body, facilitating body stiffness modulation during undulatory swimming in flow tanks. Electroactive polymers are also used in a fluid movement context. Fluidic soft sensors, in turn, provide proprioception for closed loop control. 


Robophysical modelling thus unlocks nature’s secrets to maximum mobility andmanipulationReverse-engineering nature’s patents offers design blueprints for resilient biomimetic devices that expand the reach of human experience with potential for diverse industry applications.


Recent Publications:


Curriculum Vitae:

Dr. Ardian Jusufi graduated from the University of California at Berkeley, where most of his formal training took place. He moved to the University of Cambridge where he was a Queens' College Postdoctoral Research Associate. Subsequently, Ardian was a research scientist at Harvard University, under the supervision of Prof. Wood of the Harvard Microrobotics Laboratory at the School of Engineering and Applied Sciences and the Wyss Institute for Biologically-Inspired Engineering. He was a lecturer in Sydney before founding the Max Planck Research Group in 2018. His research group is an integral part, in fact one of the first research groups of the Cyber Valley ecosystem.  Ardian has been a Max Planck Research Group Leader [~Associate Professor] at the MPI for Intelligent Systems (formerly Metals Research) for over three years.


Awards & Honors (Selected) 


Outstanding Teaching Award for G.S.Instructors – U.C.Berkeley.


William V. Power Award – U.C. Berkeley.


Queens’ College Postdoctoral Research Associate, Cambridge University.                                                                           

Best Poster Presentation – 8thInternational Symposium Adaptive Motion in Animals and Machines, Japan.


Best Presentation Competition, Runner-up, Soc. Int. Comp. B. Annual main conference.


Invited Speaker RSS 2019.

Invited Symposium Speaker SICB 2011 and 2021.

Invited Speaker ICRA 2022