iRonCub3 is equipped with jet engines, weighs 70 kg, generates over 1,000 N of thrust, and has exhaust temperatures exceeding 600 degrees Celsius.
Robot developers at the Italian Institute of Technology (IIT) have introduced a testing and preliminary validation area for the world’s first humanoid robot equipped with jet engines, as reported by Interesting Engineering on August 19. Named iRonCub, this robot features four compact jet engines, granting it the ability to fly and perform advanced tasks. The research team believes that the flying capability will benefit the robot in various applications, including disaster relief.
The iRonCub robot can fly due to its four jet engines. (Photo: IIT).
The testing began in 2021, and the IIT research team faced several challenges, such as preventing the robot from catching fire or even exploding due to exhaust from the engines. They developed prototypes of iRonCub, based on the iCub v2.5 and v3.0 platforms. iCub is a humanoid robot created by IIT for research purposes, designed to aid in the development and testing of AI algorithms.
iCub can rotate freely up to 53 degrees and is covered in a full-body skin, equipped with force/torsion sensors, cameras, microphones, gyroscopes, accelerometers, and encoders at each joint. With iRonCub, both versions are equipped with four jet engines—two in the arms and two in the jetpack attached to the robot’s back. Current testing focuses on the iRonCub version 3. The research team has made significant improvements to the hardware design of iCub to accommodate external engines, creating a titanium spine and adding heat-resistant protection. iRonCub3 weighs approximately 70 kg, with turbines generating a maximum thrust of over 1,000 N and exhaust temperatures above 600 degrees Celsius.
Researchers are currently testing iRonCub3 in flight and control areas, marking a significant advancement compared to iRonCub2. iRonCub3 features several upgrades over its predecessor. Built on the iCub3 platform, this version eliminates the ligaments and integrates torsion sensors into the jetpack. Additionally, new electronic devices have been designed, including a next-generation control and planning system that operates at a higher frequency. According to the research team, these improvements enhance the robot’s capabilities and performance.
A key challenge for aerial humanoid robots is planning flight and navigation routes, including the transition process between two activities. A momentum-based route planning algorithm developed in Python uses a multiple shooting method to address the issue. This algorithm has been tested through simulations and will soon be trialed on the actual robot.
For flight control, the algorithm is designed to manage the robot’s direction and position, applying conditional optimization. The research team states that the mechanism can be adjusted for various numbers of jet turbines. They noted that the project is significantly more complex than traditional humanoid robots. Thermodynamics plays a crucial role, with exhaust from the turbines reaching hundreds of degrees Celsius and approaching the speed of sound. The aerodynamics of the multi-component system requires a neural network for real-time assessment.