The X1, the world’s first multirobot system, integrates a humanoid robot with a transforming drone capable of launching from its back, driving, and flying. This innovative system is designed to tackle emergency scenarios by deploying autonomous agents quickly.
The new multimodal system is a product of a three-year collaboration between Caltech’s Center for Autonomous Systems and Technologies (CAST) and the Technology Innovation Institute (TII) in Abu Dhabi, United Arab Emirates. It showcases innovative projects possible through combined global expertise in autonomous systems, artificial intelligence, robotics, and propulsion.
Aaron Ames, director and Booth-Kresa Leadership Chair of CAST, and Bren Professor at Caltech, highlighted the system’s advantage. He noted that while robots can fly, drive, or walk, combining these locomotion modalities into one package maximises benefits and mitigates individual limitations.
A team recently demonstrated X1's capabilities on Caltech’s campus, simulating an emergency requiring rapid autonomous agent deployment. For the test, an off-the-shelf Unitree G1 humanoid was modified to carry M4, Caltech’s multimodal robot that can fly and drive, as a backpack.
The demonstration began with the humanoid walking from Gates–Thomas Laboratory through Sherman Fairchild Library to an elevated spot for safe deployment. The humanoid bent forward, allowing M4 to launch in drone mode before landing and transforming into driving mode for efficient ground travel.
M4 then encountered Turtle Pond, switching back to drone mode to fly over the obstacle, reaching the “emergency” site near Caltech Hall. A second M4 and the humanoid eventually joined the first responder.
Mory Gharib, Ph.D., Hans W. Liepmann Professor at Caltech and CAST’s founding director, explained the challenge of integrating different robots into a single, multi-functional system. He said the collaboration provided the ideal solution.
Gharib’s group, which originally built the M4 robot, specialises in flying and driving robots, as well as advanced control systems. Ames’ lab contributes expertise in locomotion and developing algorithms for safe humanoid robot use. TII provides extensive knowledge in autonomy and sensing for robotic systems in urban environments.
A Northeastern University team led by Alireza Ramezani, an associate professor, assists with morphing robot design. Ramezani praised the collaboration, highlighting how researchers with diverse skill sets addressed challenging robotics problems, from perception and sensor data fusion to locomotion modelling, controls, and hardware design.
In July 2025, TII engineers visited Caltech, where partners built a new M4 version utilising Saluki, a secure flight controller and computer technology from TII for onboard computing. Future work aims to equip the entire system with sensors, model-based algorithms, and machine learning-driven autonomy for real-time navigation and adaptation.
Claudio Tortorici, director of TII, stated, “We install different kinds of sensors—lidar, cameras, range finders—and we combine all these data to understand where the robot is, and the robot understands where it is in order to go from one point to another.” He affirmed TII’s contribution to autonomous movement.
Ames clarified that the humanoid robot’s actions in the demo were more advanced than simple walking. While many humanoid robots use human movement data, Ames’ group builds mathematical models describing physics for broader applications.
These models, fused with machine learning, provide robots with general abilities to navigate various situations. Ames noted, “The robot learns to walk as the physics dictate,” enabling X1 to walk on different terrains, stairs, and while carrying items like M4.
An overarching goal of the collaboration is to make such autonomous systems safer and more reliable. Tortorici believes society is increasingly accepting robots, but reliability is crucial for their widespread integration.
Ames confirmed the ongoing work on safety-critical control, ensuring system trustworthiness and security. He added that multiple projects extend beyond this one, addressing various facets of autonomy to collectively advance the field.
The X1 is the world’s first multirobot system, combining a humanoid robot and a transforming drone.
The system can launch a drone from its back, enabling it to drive, fly, and navigate various terrains for emergency response.
It is a result of a three-year collaboration between Caltech’s CAST and the Technology Innovation Institute (TII) in Abu Dhabi.
Source: TECHXPLORE
