Learning-Based Control Strategies for Soft Robots: Theory, Achievements, and Future Challenges

In the last few decades, soft robotics technologies have challenged conventional approaches by introducing new, compliant bodies to the world of rigid robots. These technologies and systems may enable a wide range of applications, including human–robot interaction and dealing with complex environments. Soft bodies can adapt their shape to contact surfaces, distribute stress over a larger area, and increase the contact surface area, thus reducing impact forces.