Soft robotics, a rapidly expanding frontier in engineering, is redefining how machines interact with humans and the environment. By incorporating soft or flexible components, these systems promise safer, more intuitive interfaces and open possibilities for environmentally sensitive applications. In agriculture, for instance, a gentle robotic touch can harvest fruit without bruising it. In medicine, adaptable and robust soft components can operate safely within the human body.
Despite the promise, much of soft robotics remains confined to research labs. Efforts are underway to move these technologies into real-world use, tackling challenges across industries. At the 6th IEEE-RAS International Conference on Soft Robotics (RoboSoft), researchers presented innovations that push the boundaries of design and performance.
One breakthrough centers on granular jamming—a method where grains, such as coffee, are packed into a flexible membrane connected to a vacuum line. Under normal pressure, the grains behave like a fluid, conforming around an object. Removing air causes the particles to lock together, creating a firm grip. This technique allows grippers to adapt to diverse shapes and textures.
Researchers have now released the largest dataset of grain types for granular jamming applications. Dr David Howard, AI4Design portfolio lead and head of the Soft Robotics Cluster at CSIRO’s Data61, explained, “We found that quite soft, rubbery grains are the best for gripping. However, say you’ve got a robot out in the field that’s likely to get tossed around a bit. You need to protect the electronics using a skin. We found large, hard grains performed better for shock absorption.” This data enables developers to select grains tailored to specific operational needs.
Performance enhancements have also emerged from an unexpected source—vibrations. The team discovered that sound wave-induced vibrations can improve granular jamming grip strength. “You can tune the behaviour of these kinds of grippers using vibrations, resulting in stronger grips. What’s more, this improvement in performance is affordable; all you need is a single 3D-printed adapter and a $20 amplifier,” Howard noted. Such low-cost upgrades could accelerate adoption in field-ready systems.
Another innovation, the Jamming Donut, addresses a limitation of traditional granular jammers, which require objects to rest on a flat surface. The Jamming Donut combines an inflatable outer ring with an inner granular jammer, enabling secure grasping of objects suspended in air. This hybrid design excels in handling varied shapes and textures, from delicate produce to rigid hardware, and even tasks like turning doorknobs—actions that challenge many conventional grippers.
Beyond mechanical ingenuity, the design process itself is evolving through artificial intelligence. The AI4Design portfolio is developing automated tools for creating complex, integrated robotic systems. By applying these methods to pneumatic grippers, the team has built software that generates fit-for-purpose designs within user-defined material and size constraints. This approach streamlines development, conserving resources while expanding creative possibilities.
Howard described the methodology: “We’re looking at using generative AI to creatively explore the design space, then using physics-based modelling and simulation coupled with real-world experimentation to provide some ground truth data. This can then be used as a tool to do rapid, blanket exploration of these really complicated, interesting design spaces to find intuitive, novel, high performance designs.” Such integration of AI with physical prototyping could transform how engineers approach soft robotics, enabling rapid iteration and deployment.
These advancements—granular jamming optimization, vibration-enhanced gripping, the Jamming Donut, and AI-driven design—represent significant steps toward bringing soft robotics out of the lab and into practical, everyday use. Supported by the Robotics and Autonomous Systems Group and co-funded initiatives like AI for Missions and the Future Digital Manufacturing Fund, the Soft Robotics Cluster is positioning these technologies to address pressing industrial and societal challenges.