Encoding Material Safety Using Control Barrier Functions for Soft Actuator Control

Until recently, the concept of soft robot safety was an informal notion, often attributed solely to the fact that soft robots are less likely to damage their operating environment than rigid robots. As the field moves toward feedback control for practical applications, it becomes increasingly important to define what safety means and to characterize how soft robots can become unsafe. The unifying theme of soft robotics is to achieve useful functionality through deformation. Consequently, limitations in constitutive model accuracy and risks of material failure are inherent to all soft robots and pose a key challenge in designing provably safe controllers. This work introduces a formal definition of material safety based on strain energy functions and provides a controller that enforces it. We characterize safe and unsafe sets of an incompressible hyperelastic material and demonstrate that safety can be enforced using a high-order control barrier function (HOCBF) with quadratic program-based feedback control. As a case study, we consider material safety enforcement on theoretical models of soft actuators with a tubular geometry having inertial effects, first-order viscous effects, and full-state feedback. Simulation results verify that the proposed methodology can enforce the material safety specification.