The field of soft robotics is moving towards more flexible and adaptable control methods, shifting away from rigid-body control logic and embracing control compliance. This approach enables robustness, flexibility, and cross-task generalization, and is inspired by human motor control. Researchers are also exploring innovative methods for shape reconstruction, such as vision-based approaches that leverage the robot's natural surface appearance. Additionally, there is a growing focus on human-robot interaction, particularly in the context of unmanned surface vehicles, where usability challenges and operator difficulties are being addressed through user-centered design. Notable papers include:

• AFT, which proposes a vision-based, markerless, and training-free framework for soft robot shape reconstruction.
• Learning Visually Interpretable Oscillator Networks for Soft Continuum Robots from Video, which introduces a plug-and-play module for autoencoder-based latent dynamics learning that generates pixel-accurate attention maps.
• Toward generic control for soft robotic systems, which proposes a generic control framework grounded in control compliance.