It’s possible that I shall make an ass of myself. But in that case one can always get out of it with a little dialectic. I have, of course, so worded my proposition as to be right either way (K.Marx, Letter to F.Engels on the Indian Mutiny)
Friday, June 20, 2025
Researchers at IIT have demonstrated that a humanoid robot can fly
Researchers in Italy achieved a first flight of iRonCub3.. The robot was able to lift off the floor by approximately 50 cm while maintaining its stability. A research paper has been published in Nature Communications Engineering today
Researchers in Italy at Italian Institute of Technology achieved a first flight of iRonCub3. The robot was able to lift off the floor by approximately 50 cm while maintaining its stability.The result has been possible thanks to thermodynamics and aerodynamics studies, combined with robotics and AI-powered control systems.
Genoa (Italy), 18th June 2025 – The Italian Institute of Technology (IIT) has reached a groundbreaking milestone in humanoid robotics by demonstrating the first flight of iRonCub3, the world’s first jet-powered flying humanoid robot specifically designed to operate in real-world environments. The research team studied the complex aerodynamics of the artificial body and developed an advanced control model for systems composed of several interconnected parts. The overall work on iRonCub3, including real flight tests, took about two years. In the latest experiments, the robot was able to lift off the floor by approximately 50 cm while maintaining its stability. The achievement paves the way for a new generation of flying robots capable of operating in complex environments while maintaining a human-like structure.
The aerodynamics and control studies have been described in a paper published today in Nature Communications Engineering, an open access journal from the Nature Portfolio. The research was carried out by roboticists of IIT in Genoa, Italy, in collaboration with the group of Alex Zanotti at DAER Aerodynamics Laboratory of Polytechnic of Milan – where a comprehensive series of wind tunnel tests were performed – and the group of Gianluca Iaccarino at Stanford University – where deep learning algorithms were used to identify aerodynamic models.
The robot flight demonstration represents the latest milestone of the Artificial and Mechanical Intelligence (AMI) Lab at IIT in Genoa, led by Daniele Pucci. Their research aims to push the boundaries of multi-modal humanoid robotics, combining terrestrial locomotion and aerial mobility to develop robots capable of operating in unstructured and extreme environments.
iRonCub3 is the technological evolution of previous prototypes and is based on the latest generation of the iCub humanoid robot (iCub3), developed to be teleoperated. It integrates four jet engines, two mounted on the arms and two on a jetpack attached to the robot’s back. Modifications to the iCub hardware design were required to support the external engines, such as developing a new titanium spine and adding heat-resistant covers for protection. The robot combined with the jet engines weighs about 70 kg, while the turbines can provide a maximum thrust force of more than 1000 N. This configuration enables the robot to hover and perform controlled flight maneuvers even in the presence of wind disturbances or environmental uncertainties. The exhaust temperature can reach 800 degrees.
“This research is radically different from traditional humanoid robotics and forced us to make a substantial leap forward with respect to the state of the art,” explains Daniele Pucci. “Here, thermodynamics plays a pivotal role — the emission gases from the turbines reach 700°C temperature and flow at nearly the speed of sound. Aerodynamics must be evaluated in real-time, while control systems must handle both slow joint actuators and fast jet turbines. Testing these robots is as fascinating as it is dangerous and there is no room for improvisation.”
The AMI research team focused on the platform’s dynamic balance, which is made particularly complex by the robot’s humanoid morphology. Unlike conventional drones, which have symmetric and compact structures, iRonCub3 has an elongated shape, with masses distributed across movable limbs and a variable center of mass. This required the development of advanced flight balance models that consider the robot’s multibody dynamics and the interaction between jet propulsion and limb movements. Moreover, the movable limbs significantly complicate the aerodynamics, which change with every motion of any of the robot’s limbs.
The researchers at IIT have performed extensive wind tunnel experiments, advanced Computational Fluid Dynamics (CFD) simulations and developed AI-based models capable of estimating aerodynamic forces in real time.
“Our models include neural networks trained on simulated and experimental data and are integrated into the robot’s control architecture to guarantee stable flight” explains Antonello Paolino, first author of the paper and PhD student in a joint program between the IIT and Naples University, who spent a semester as visiting researcher at Stanford University.
As a result, iRonCub3 is equipped with AI-powered control systems that allow it to fly while handling high-speed turbulent airflows, extreme temperatures, and the complex dynamics of multi-body systems.
The advanced aerodynamic modeling developed by IIT demonstrates that it is possible to maintain posture and stability even during non-stationary maneuvers, such as sequential engine ignition or changes in body geometry.
These studies can be transferred to other robots with unconventional morphologies, representing a unique case compared to classical drones, whose balance relies on symmetry and simplified control strategies that often neglect the robot’s own aerodynamics and thermodynamics.
The final design of iRonCub3 is the result of an advanced co-design process, specifically developed to integrate artificial intelligence and multi-physics into the design of flying robots. These techniques, which are innovative in the field of robotics, allow for the simultaneous optimization of both body shape and control strategies, considering the complex interactions between aerodynamics, thermodynamics, and multibody dynamics.
Co-design was used to determine the optimal placement of the jet turbines to maximize control and stability during flight. Advanced design techniques were also employed to manage the heat dissipation generated by the engines, thus ensuring the structural integrity of the robot even under extreme operating conditions.
The robot has been completely re-engineered to withstand the harsh conditions associated with aerial locomotion, introducing major improvements focused on precision actuation, enhanced thrust control via integrated sensors, and advanced planners for coordinated takeoff and landing.
Throughout the design process, numerous iterative adjustments were made based on the results of advanced simulations and experimental testing, leading to the robot’s current configuration. This approach has allowed the team to overcome the limitations of traditional methodologies and represents a step forward in the automatic and integrated design of complex robotic systems.
The first flight tests of iRonCub3 have been conducted in IIT’s small flight-testing area, where the robot was able to lift off the floor by approximately 50 cm. In the coming months, prototype testing will continue and will be further enhanced thanks to a collaboration with Genoa Airport (Aeroporto di Genova), which will provide a dedicated area that will be set up and equipped by the Italian Institute of Technology in compliance with all required safety regulations. The area will host future experimental campaigns.
Applications of flying humanoid robots like iRonCub3 are envisioned in a variety of future scenarios, such as search-and-rescue operations in disaster-struck areas, inspection of hazardous or inaccessible environments, and exploration missions where both manipulation capabilities and aerial mobility are essential.
iRonCub3 is the technological evolution of previous prototypes and is based on the latest generation of the iCub humanoid robot (iCub3), developed to be teleoperated. It integrates four jet engines, two mounted on the arms and two on a jetpack attached to the robot’s back
The robot flight demonstration represents the latest milestone of the Artificial and Mechanical Intelligence (AMI) Lab at IIT in Genoa, led by Daniele Pucci. Their research aims to push the boundaries of multi-modal humanoid robotics, combining terrestrial locomotion and aerial mobility to develop robots capable of operating in unstructured and extreme environments.
QUT robotics researchers have developed a new robot navigation system that mimics neural processes of the human brain and uses less than 10 per cent of the energy required by traditional systems.
In a study published in the journal Science Robotics, the researchers detail a new system which they call LENS – Locational Encoding with Neuromorphic Systems.
LENS uses brain-inspired computing to set a new, low-energy benchmark for robotic place recognition.
“To run these neuromorphic systems, we designed specialised algorithms that learn more like humans do, processing information in the form of electrical spikes, similar to the signals used by real neurons,” Dr Hines said.
“Energy constraints are a major challenge in real-world robotics, especially in fields like search and rescue, space exploration and underwater navigation.
“By using neuromorphic computing, our system reduces the energy requirements of visual localisation by up to 99 per cent, allowing robots to operate longer and cover greater distances on limited power supplies.
“We have known neuromorphic systems could be more efficient, but they’re often too complex and hard to use in the real world – we developed a new system that we think will change how they are used with robots.”
In the study, the researchers developed LENS, a system that was able to recognise locations along an 8km journey but using only 180KB of storage – almost 300 times less than other systems.
LENS combines a brain-like spiking neural network with a special camera that only reacts to movement and a low-power chip, all on one small robot.
“This system demonstrates how neuromorphic computing can achieve real-time, energy-efficient location tracking on robots, opening up new possibilities for low-power navigation technology,” Dr Hines said.
“Lower energy consumption can allow remotely operated robots to explore for longer and further.
“Our system enables robots to localise themselves using only visual information, in a way that is both fast and energy efficient.”
Dr Fischer, ARC DECRA Fellow, said the key innovation in the LENS system was a new algorithm that exploited two types of promising bio-inspired hardware: sensing, via a special type of camera known as an “event camera”, and computing, via a neuromorphic chip.
“Rather than capturing a full image of the scene that takes in every detail in each frame, an event camera continuously senses changes and movement every microsecond,” Dr Fischer said.
“The camera detects changes in brightness at each pixel, closely replicating how our eyes and brain process visual information.
“Knowing where you are, also known as visual place recognition, is essential for both humans and robots.
“While people use visual cues effortlessly, it’s a challenging task for machines.”
Professor Michael Milford, director of the QUT Centre for Robotics, said the study was representative of a key theme of research conducted by the centre’s researchers.
“Impactful robotics and tech means both pioneering ground-breaking research, but also doing all the translational work to ensure it meets end user expectations and requirements,” Professor Milford said.
“You can’t just do one or the other.
“This study is a great example of working towards energy-efficient robotic systems that provide end-users with the performance and endurance they require for those robots to be useful in their application domains.”
No comments:
Post a Comment