Abstract

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

Highlights

  • S OFT robotic sensors have immense potential to revolutionize the field of health-monitoring, human motion detection, human-machine interfaces, and soft robotics, owing to their high conformability [1]

  • The main idea behind this letter is to compensate for the time-variant hidden states of a soft-bodied system using these redundant and disjoint sensors to obtain the true strain state in a static manner

  • Theoretical underpinnings of the concept is demonstrated through information theoretics and learning-based methods are used to validate the theory

Read more

Summary

INTRODUCTION

S OFT robotic sensors have immense potential to revolutionize the field of health-monitoring, human motion detection, human-machine interfaces, and soft robotics, owing to their high conformability [1]. Depending on the strain responsive mechanism, this can be because of the rearrangement of conductive particles [13], damage, geometric effects induced by the surrounding visco-elastic matrix and/or de-lamination due to impedance mismatch [14] Solving these problems from the material side is still an open challenge which is hampered by the unavailability of benchmarking tools for comparing different technologies. Recent advancements in machine learning can be an alternate solution to handling the nonlinear time-variant dynamics of these soft sensors [15], [16] This can be done by explicitly providing the past sensor information [17], by using recurrent neural networks that can retain information about the past data [18]– [20], or by using adaptive algorithms [21]. We present simple statistical techniques that can be used for quantifying the quality of a soft strain sensor and its combinations with any complex modelling techniques

THEORY
MATERIAL CHARACTERIZATION
Multi-Material Sensor Matrix
Static Strain Estimation
Case Study
CONCLUSION
Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call