Contact-rich manipulation and force regulation with the observer-based controller

Extend the virtual force-injected dynamic observer and task-space controller for planar pneumatic soft actuators to contact-rich manipulation and force regulation, including the decoupling of environmental contact forces from internal model error and actuation mismatch.

Background

The paper’s observer injects measurement residuals as virtual forces to compensate for parameter mismatch, pneumatic delay, and unmodeled dissipation during free-space tracking. With only position measurements, however, the residual can also include environmental contact effects, so the current implementation cannot uniquely identify contact forces or distinguish them from internal model and actuation errors.

The authors identify contact-rich manipulation and force regulation as outside the demonstrated scope and indicate that addressing them would require additional information, such as pressure or force sensing, known environmental constraints, multiple distributed measurements, or an augmented disturbance/contact estimator. Extending the framework in this direction would enable physics-based control of soft pneumatic actuators during interaction rather than only free-space motion.

References

However, in contact-rich tasks, the same residual may also contain contributions from environmental contact. With only position measurements, the observer cannot uniquely distinguish contact forces from internal model error or actuation mismatch. Therefore, the present implementation should not be interpreted as a contact-force observer. Decoupling these effects would require additional information, such as pressure/force sensing, known environmental/contact constraints, multiple distributed measurements, or an augmented disturbance/contact estimator. Contact-rich manipulation and force regulation are therefore outside the scope of this study and are left for future work.

Task-space model-based control of pneumatic soft actuators  (2608.27186 - Kumar et al., 27 Aug 2026) in Section 4, Discussion, subsection “Observer-Based State Estimation”