Soft robotics is a field of engineering that aims to create robots that can mimic the flexibility, adaptability and resilience of natural organisms. One of the main challenges in soft robotics is to design robots that can perform complex tasks in diverse environments, such as grasping delicate objects, navigating rough terrains or interacting with humans.
To inspire and guide soft robotics designers, a team of researchers from the University of California, Berkeley, has developed an open-source platform that simulates the behaviour and morphology of various wildlife species. The platform, called Wild-Soft, allows users to explore how different animals move, sense and interact with their surroundings, and to apply these insights to their own robotic designs.
Wild-Soft is based on a physics engine that can model the dynamics of soft bodies, such as muscles, tendons and skin. The platform also incorporates realistic sensory inputs, such as vision, touch and proprioception, as well as neural networks that control the animal's actions and reactions. Users can choose from a library of pre-built animal models, such as snakes, fish, birds and mammals, or create their own custom models using a graphical user interface.
The platform enables users to test their robotic designs in various scenarios, such as obstacle courses, mazes or predator-prey interactions. Users can also compare the performance of different designs using metrics such as speed, energy efficiency and robustness. The platform also provides feedback and suggestions on how to improve the design based on the principles of biomimicry.
Wild-Soft is intended to be a collaborative and accessible tool for the soft robotics community. The platform is free and open-source and can run on any computer with a web browser. The researchers hope that Wild-Soft will foster creativity and innovation in soft robotics, and help bridge the gap between biology and engineering.