Computational and experimental evaluation demonstrates accurate dynamic state estimation in soft continuum robots using base force-torque sensing, indicating a pathway toward real-time...
Key Points
To develop a dual boundary observer capable of reconstructing infinite-dimensional dynamic states in soft continuum robots using only base wrench measurements.
Formulated a boundary state observer using Cosserat rod theory and energy dissipation principles driven by base force/torque sensing.
Conducted a Lyapunov-based stability analysis to evaluate state convergence and characterize optimal observer gain tuning.
Validated the observer framework through dynamic numerical simulations and hardware experiments on a tendon-driven continuum robot.
Demonstrated accurate full-state estimation during rapid dynamic motions without requiring external motion tracking systems.
Uncovered a convex relationship between observer gains and convergence rate, enabling systematic optimization of parameter tuning.
Confirmed observer robustness against external perturbation forces, sensor noise, and modeling uncertainties under real-time computation.