Using Redundant and Disjoint Time-Variant Soft Robotic Sensors for Accurate Static State Estimation

Soft robotic sensors have been limited in their applications due to highly nonlinear time variant behavior. Current studies are either looking into techniques improve the mechano-electrical properties of these or modelling algorithms that account for history each sensor. Here, we present a method combining multi-material soft strain obtain equivalent higher quality sensors; better than individual sensors. The core idea behind this work is use combination redundant and disjoint compensate time-variant hidden states soft-bodied system, finally true state static manner using learning-based approach. We provide methods develop variable metrics estimate dissimilarity efficacy sensor combinations, which can double down as benchmarking tool proposed approach experimentally validated on pneumatic actuator with embedded Our results show data from sufficient accurately system.