cs.ROAug 14, 2026

Real-time Estimator of Actuator Control and Health (REACH) on an Eel-Inspired Soft Robot

Authors: Zhangjingyi Jiang, Myungsun Park, Michael T. Tolley, Mark Campbell

Organizations: Department of Mechanical and Aerospace Engineering, Cornell University · Department of Mechanical and Aerospace Engineering, University of California San Diego

Read PDF
Ask AI
Cite

Copy citation

DOINot available for this paper

Abstract

An actuator health estimation algorithm for a soft swimming robot that can perform anguilliform swimming is developed. Due to harsh operational environments of underwater robots, and the common degradation of soft robot materials and actuators, accurate estimation of actuator functionality is necessary for robots to perform their missions as well as return to base in the event of actuator degradation and failure. Termed REACH (Real-time Estimator of Actuator Control and Health), the architecture employs a soft robot model, sigma point filter, and a formal statistical hypothesis test to adequately capture the nonlinearities and changes over time. The performance of REACH using three sensor types (GPS, IMU, and Bend Sensor) with one sensor on each actuator is compared, demonstrating that both bend sensor and IMU are adequate choices. Sensor quantity and placement are evaluated for IMU and bend sensor, showing two sensors are sufficient for IMU, whereas three sensors are needed for bend sensor. Three swimming gaits (linear swimming, wide turning, tight turning) are compared, demonstrating that REACH can successfully predict actuator health for all three gaits, with minimal differences in performance. A filter validation method shows the fault estimation algorithm is statistically consistent in finding the correct degradation. The approach is experimentally evaluated using bend sensor data collected from a fish robot, demonstrating that REACH can successfully estimate actuator health with noisy data and variations in manufacturing.

Figures & tables

Fig. 1 : UCSD Robot Fish [ 12 ]
Fig. 2 : Robot Fish Data Collection Setup
Fig. 3 : Bend Angle Calculation
VariablesValues
Actuator Length Lactuator(m)L_{\text{actuator}}(\text{m})0.0920.092
Head Length Lhead(m)L_{\text{head}}(\text{m})0.040.04
Tail Length Ltail(m)L_{\text{tail}}(\text{m})0.150.15
Connection Length Lconnection(m)L_{\text{connection}}(\text{m})0.020.02
Young’s Modulus of Non-actuators EnonAct(N/m2)E_{\text{nonAct}}(\text{N/m}^{2})2∗1092*10^{9}
Young’s Modulus of Actuators Eactuator(N/m2)E_{\text{actuator}}(\text{N/m}^{2})10610^{6}
TABLE I : Experimental Fish Parameters
Fig. 4 : Filter Validation of Actuator Health Estimation Algorithms on Simulation Data
Fig. 5 : Actuator health estimation using GPS, IMU, and bend sensor of five actuators failures.
Fig. 6 : RMS error and rise time of GPS, IMU, and bend sensors in estimating actuator health for each actuator failure case.
Fig. 7 : Actuator health estimation using IMU, bend sensor, for one sensor, two sensor, three sensor, and four sensor placement cases.
SwimmingCAC_{A}CωC_{\omega}CτOC^{O}_{\tau}
Linear Swimming111100
Low CτOC_{\tau}^{O} Turning ∗0.80.8110.30.3
High CτOC_{\tau}^{O} Turning ∗0.80.8110.80.8
∗ Low/High CτOC_{\tau}^{O} turn is turning
with a large/small radius
TABLE II : Control Parameters - Demo 1
Fig. 8 : RMS error and rise time for control algorithms
T1 ∗T2T3T4T5T6T7T8T9
[1 1 1][1\ 1\ 1][.5 1 1][.5\ 1\ 1][0 1 1][0\ 1\ 1][1 0 0[1\ 0\ 0 ][0 1 0][0\ 1\ 0][0 0 1][0\ 0\ 1][.8 1 1][.8\ 1\ 1][.8 .5 1][.8\ .5\ 1][.5 .8 1][.5\ .8\ 1]
∗ This trial was used to tune stiffness values
TABLE III : Actuator Health for Each Trial
Fig. 9 : Experimental validation of REACH for nine unique actuator health conditions. A1 Red, A2 Green, A3 Blue
More in this topic

Fault Diagnosis

Fault diagnosis identifies, localizes, and explains failures in engineered and computational systems using sensor data, logs, models, and domain knowledge.

See all 148 papers

Explore similar work

CardsList

  1. Observability-Informed Optimal Sensor Placement for Soft Robots

    Sep 27, 2026Samuel Smocot, James Richard Forbes, Audrey SedalSoft Robotics

    Dept. of Mechanical Engineering, McGill University, Montreal, Canada. · Mila – Québec AI Institute, Montreal, Canada.

  2. Learning Control as Enabling Layer for Embodied Intelligence Research explored with Soft Robotic Swimming in diverse Flow Speeds

    Jun 9, 2026Fabian Schwab, Federico Allione, Bingcheng Wang +5Robotic ControlRobust Control

    1Engineering Sciences Department, Swiss Federal Laboratories for Materials Science and Technology, Switzerland · Institute for NeuroInformatics, University of Zurich, Switzerland · Department of Electronic Engineering, University of Rome Tor Vergata, Italy

  3. A Tendon-Driven Robotic Jellyfish with Constrained Soft Actuation and Depth Control via Reinforcement Learning

    Sep 24, 2026Jiarui Peng, Yutong Wu, Zelong Wang +5Underwater RoboticsTendon-Driven Actuation

    Hong Kong Embodied AI Lab, Hong Kong SAR, China · The Chinese University of Hong Kong, Hong Kong SAR, China · The Hong Kong University of Science and Technology, Hong Kong SAR, China