Algorithm Helps Robots Better Mimic Complex and Subtle Movements of Living Creatures
Researchers have long aspired to bring natural wonders like the maneuvers of an octopus arm or efficiency of bacteria into robotics and now scientists from Khalifa University have developed a new model in the field of biomechanics that makes the complex task of robot behavior more precise and simpler in the way robots move and interact with their environment.
The groundbreaking Geometric Variable Strain (GVS) model which transforms the control and movements of hybrid soft-rigid robots — those that combine both soft, flexible parts and rigid components, does not compromise on accuracy despite requiring fewer calculations, making it a powerful tool for the next generation of robotic systems.
This research was in a paper titled ‘Reduced order modeling of hybrid soft-rigid robots using global, local, and state-dependent strain parameterization’ in the, a top 1% journal.
The team includes Khalifa University’s Mechanical and Nuclear Engineering Associate Professor Dr. Federico Renda, and Postdoctoral Fellows Dr. Anup Teejo Mathew, Dr. Daniel Feliu-Talegon, and Dr. Abdulaziz Y. Alkayas, as well as Dr. Frederic Boyer, Faculty, Researcher, , Institut Mines Telecom Atlantique, Nantes, France.
Traditional models often struggle with the complexity of blending both flexibility and rigidity, but the GVS model streamlines the process by using a reduced-order framework, an efficient algorithm, which only needs a minimal amount of data to accurately represent the robot’s behavior. This model is implemented in the SoRoSim Toolbox, an open-source simulator highly regarded in the soft robotics community for its speed, accuracy, and stability, enabling the use of the GVS model in a wide range of modern robotic applications, from human-robot interaction to underwater exploration robots and industrial inspection tools.
Key advantages of the GVS model are its ability to simplify the design and control of robots that need to operate in complex environments. For example, in minimally invasive surgery, continuum manipulators — long, flexible robotic arms — can benefit from the model to achieve more precise movements, reducing the risk to patients. In underwater exploration, robots inspired by the natural propulsion systems of bacteria can navigate with greater efficiency and less noise, making them ideal for sensitive environments, while soft grippers benefit the agricultural and food industries.
In scenarios where soft and rigid components must work together seamlessly including robots used in 3D printing for construction, the GVS model helps tackle operational challenges like vibration and sagging of robots that use steel cables to control heavy loads.
Dr. Federico Renda said: “The GVS model’s ability to simplify the analysis process while maintaining accuracy is beneficial for the design and control of such robots, as it can lead to more efficient algorithms and potentially more responsive and versatile robotic systems. The impact of the model on the future of robotics, especially in the field of soft robotics, could facilitate the development of advanced robotic systems that are capable of more complex and subtle interactions with their environment, similar to living animals, leading to critical innovations.”
Alisha Roy
Science Writer
