Video Friday: This Drone Drives and Flies—Seamlessly

Seamless Transitions: The Future of Robotics

Imagine a world where robots can effortlessly navigate between different environments, seamlessly transitioning from air to ground, and from one modality to another. This is the future of robotics, where machines can adapt and learn in real-time, making them increasingly useful in various industries and applications.

Hybrid Designs: The Key to Seamless Transitions

One of the most significant advancements in robotics is the development of hybrid designs that can seamlessly transition between different modes of operation. Duawlfin, a quadrotor drone, is a prime example of this technology. By leveraging its standard quadrotor motors and introducing a differential drivetrain with one-way bearings, Duawlfin eliminates the need for additional actuators or propeller-driven ground propulsion. This allows the drone to transition smoothly between aerial and ground modes, making it ideal for applications like urban logistics and indoor navigation.

Multi-Modal Sensing: The Challenge of Integration

While modern robots are equipped with multiple sensors, including cameras, tactile sensors, and depth sensors, the real challenge lies in integrating these different sensory streams. This is where the concept of multi-modal sensing comes in. Instead of forcing all sensors through a single network, researchers are rethinking how to combine modalities. By training separate expert policies for each modality and learning how to combine their action predictions at the policy level, robots can better navigate complex environments and make more informed decisions.

Collaboration and Adaptation: The Future of Robot Teams

Heterogeneous robot teams operating in realistic settings often must accomplish complex missions requiring collaboration and adaptation to information acquired online. To address these limitations, researchers have developed a framework called SPINE-HT, which grounds the reasoning abilities of Large Language Models (LLMs) in the context of a heterogeneous robot team through a three-stage process. In real-world experiments, SPINE-HT achieved an 87% success rate in missions requiring reasoning about robot capabilities and refining subtasks with online feedback.

Microrobotic Applications: The Potential of Self-Propelling Oil Droplets

Scientists at the Max Planck Institute for Intelligent Systems in Stuttgart, Germany, have developed control strategies for influencing the motion of self-propelling oil droplets. These oil droplets mimic single-celled microorganisms and can autonomously solve a complex maze by following chemical gradients. However, it is very challenging to integrate external perturbation and use these droplets in robotics. To address these challenges, the team developed magnetic droplets that still possess life-like properties and can be controlled by external magnetic fields. In their work, the researchers showed that they are able to guide the droplet's motion and use them in microrobotic applications such as cargo transportation.

Full-Body Teleoperation: The Future of Human-Robot Interaction

Imagine being able to control a robot with your entire body, not just your hands or fingers. This is the future of human-robot interaction, where full-body teleoperation and full-body data acquisition platforms are being developed. Unitree, a company that specializes in robotics and artificial intelligence, has developed a full-body teleoperation platform that allows users to control a robot with their entire body. This technology has the potential to revolutionize various industries, including healthcare, manufacturing, and logistics.

Jumping Robots: The Future of Locomotion

Researchers at the Autonomous Robots Lab at the Norwegian University of Science and Technology have developed a curriculum-based reinforcement learning framework for training precise and high-performance jumping policies for the robot "Olympus". Separate policies are developed for vertical and horizontal jumps, leveraging a simple yet effective strategy. Experimental validation demonstrates horizontal jumps up to 1.25 m with centimeter accuracy and vertical jumps up to 1.0 m. Additionally, the proposed method can be used to learn omnidirectional jumping.

Heavy Lifting: The Future of Robotics

Heavy payloads are no problem for the new KR TITAN ultra, which can move payloads of up to 1500 kg. This makes the heavy lifting extreme in the KUKA portfolio. The KR TITAN ultra is a prime example of the advancements in robotics, where machines can handle heavy loads with ease, making them ideal for various industries and applications.

Conclusion

The future of robotics is exciting and rapidly evolving. With advancements in hybrid designs, multi-modal sensing, collaboration and adaptation, microrobotic applications, full-body teleoperation, jumping robots, and heavy lifting, the possibilities are endless. As researchers and developers continue to push the boundaries of what is possible, we can expect to see even more innovative applications and technologies emerge. The future of robotics is bright, and it's exciting to think about what's next.

Source: https://spectrum.ieee.org/video-friday-multimode-drone