Transparent Pneumatic Tactile Sensors for Soft Biomedical Robotics

Abstract

Palpation is a simple but effective method to distinguish tumors from healthy tissues. The development of miniaturized tactile sensors embedded on endoscopic or robotic devices is key to achieving precise palpation diagnosis and subsequent timely treatment. This paper reports on the fabrication and characterization of a novel tactile sensor with mechanical flexibility and optical transparency that can be easily mounted on soft surgical endoscopes and robotics. By utilizing the pneumatic sensing mechanism, the sensor offers a high sensitivity of 1.25 mbar and negligible hysteresis, enabling the detection of phantom tissues with different stiffnesses ranging from 0 to 2.5 MPa. Our configuration, combining pneumatic sensing and hydraulic actuating, also eliminates electrical wiring from the functional elements located at the robot end-effector, thereby enhancing the system safety. The optical transparency path in the sensors together with its mechanical sensing capability open interesting possibilities in the early detection of solid tumor as well as in the development of all-in-one soft surgical robots that can perform visual/mechanical feedback and optical therapy.

Modified Nonlinear Hysteresis Approach for a Tactile Sensor

“…BPNN was used due to its better generalization and strong nonlinear mapping abilities, which makes it a popular choice in various fields. The BPNN was trained to predict the static resistance value based on the current and historical resistance values (R at and R t−1 , respectively) and the estimated resistance value, which was used as a target [ 41 , 42 , 43 ].…”

“…Uncertainty analysis is used to quantify the uncertainty in the output of a mathematical model by examining the uncertainty in its input [ 41 ]. It aims to understand the extent of uncertainty in the output attributed to each input variable.…”