The field of soft robotics is moving towards the development of more advanced control strategies and modeling techniques to address the challenges of uncertainty and nonlinearity in soft robotic systems. Researchers are exploring the use of stochastic models, such as the Fokker-Planck Equation, to control the probabilistic distribution of soft robots, rather than their state. Additionally, there is a focus on developing more efficient and accurate modeling techniques, such as pseudo-rigid body modeling and actuation-space energy formulation, to enable real-time control and precise force regulation. Bio-inspired designs and novel actuation mechanisms, such as antagonistic fabric-based pneumatic actuators, are also being developed to enhance the capabilities of soft robots. Noteworthy papers include:

• A study on FPE-based Model Predictive Control for a soft robotic finger, which demonstrates the efficacy of this control method in managing uncertainty.
• A novel parameter estimation method for pneumatic soft hand control, which achieves lower error and outperforms traditional PID controllers.
• A pneumatic framework for multi-dimensional haptic property rendering, which enables continuous modulation of object size and stiffness.
• A control strategy leveraging Neural Networks to enhance force-tracking behavior in robotic tasks, which achieves superior performance compared to baseline controllers.