Embedded Force Sensor with Deep Transformation Calibration for Interventional Soft Robots

A calibration method for gelatin-graphite-based soft sensors is proposed. This approach uses convolutional deep learning approaches that account for a sensor's non-linear behaviour and reduce noise amplification. This technique offers a smaller minimum detectable force than other approaches and is particularly useful in sensitive surgical scenarios. The best calibration (CQT) scheme provides high performance, with a Mean Absolute Error of ≤11.2 mN, and accurate force estimation, especially for forces below 400 mN of amplitude. This sensing principle and calibration method can revolutionize surgical procedures and capitalize on the benefits of soft robotics, potentially enhancing precision and reducing surgical trauma.