In-Situ Sensing and Dynamics Predictions for Electrothermally-Actuated Soft Robot Limbs
This work addresses the problem of practical deployment for untethered soft robots by enabling compact designs with reliable motion predictions, though it is incremental in improving sensing and modeling for a specific domain.
The paper tackles the challenge of sensing and modeling actuator states in electrothermally-actuated soft robots by proposing a framework that uses in-situ temperature and deflection sensors to train an LSTM neural network, achieving motion predictions over 10 minutes with errors matching sensor accuracy.
Untethered soft robots that locomote using electrothermally-responsive materials like shape memory alloy (SMA) face challenging design constraints for sensing actuator states. At the same time, modeling of actuator behaviors faces steep challenges, even with available sensor data, due to complex electrical-thermal-mechanical interactions and hysteresis. This article proposes a framework for in-situ sensing and dynamics modeling of actuator states, particularly temperature of SMA wires, which is used to predict robot motions. A planar soft limb is developed, actuated by a pair of SMA coils, that includes compact and robust sensors for temperature and angular deflection. Data from these sensors are used to train a neural network based on the long short-term memory (LSTM) architecture to model both unidirectional (single SMA) and bidirectional (both SMAs) motion. Predictions from the model demonstrate that data from the temperature sensor, combined with control inputs, allow for dynamics predictions over extraordinarily long open-loop timescales (10 minutes) with little drift. Prediction errors are on the order of the soft deflection sensor's accuracy. This architecture allows for compact designs of electrothermally-actuated soft robots that include sensing sufficient for motion predictions, helping to bring these robots into practical application.