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

arXiv:2608.14865 · cs.RO · Submitted 2026-08-14 · Read on arXiv

Listen

Transcript

Introduction to the show: ident: Robotics Radio. Generated commentary on the latest robotics and control papers.

Rosa: Today's paper: "Real-time Estimator of Actuator Control and Health (REACH) on an Eel-Inspired Soft Robot".

Dev: The gist 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;

Rosa: First, who's behind it and why it matters.

Title and authors: Rosa: Moving on to this paper, we have "Real-time Estimator of Actuator Control and Health (REACH) on an Eel-Inspired Soft Robot." The authors are Zhangjingyi Jiang, Myungsun Park, Michael T. Tolley, and Mark Campbell. This paper is about using a sigma point filter and a statistical test to estimate actuator health in soft swimming robots.

Dev: It’s interesting because they didn't just build the estimation part; they compared three sensor types—GPS, IMU, and Bend Sensor—to see which one works best for predicting that health. This comparison is key to understanding the practical application outside of a perfect simulation setup.

Taro: I wonder what this means for autonomous systems operating in real-world conditions where you don't have perfect control over the environment or sensor noise, right?

Rosa: That’s exactly what we’re thinking about, Taro. The paper shows that both the bend sensor and the IMU are adequate choices for health estimation when used on all five actuators of this fish robot.

Dev: But they also found some important details about how much data you actually need, which is something engineers always care about regarding loop rates and computational load.

The paper's summary: Rosa: So, the core of REACH is using that sigma point filter to predict the state, and then using a specific filter validation method to make sure the statistical results are actually significant before you trust them.

Dev: They use a test statistic called the average normalized innovation squared, lambda KF k (N), calculated over N time steps, which helps ensure the filter is giving statistically sound results based on comparing it to two-sided threshold statistics bL and bU.

Taro: That sounds like they’re making sure the estimation doesn't just look good in one spot but holds up under rigorous statistical scrutiny, which is crucial when you’re relying on this for real-world decisions.

Rosa: Right. It shows that the filter gives statistically significant results by comparing that average normalized innovation squared lambda KF k (N) to those threshold statistics bL and bU.

Dev: They also used simulation data from the Anguilliform Swimming Soft Robot Simulation Platform, or ASSRSimP, as the input model for their estimation process.

The paper's improvements: Rosa: One of the main improvements they highlight is that their comparison shows a distinction in sensor performance when it comes to localizing the fault. They found that the bend sensor has a lower RMS error for most actuator failure cases and a lower rise time for actuator five failure compared to the IMU.

Dev: That’s significant because it means if an actuator fails, you can pinpoint where that degradation is happening much more quickly using the bend sensor data than with just an IMU reading.

Taro: So, if we’re in a situation where the robot suddenly starts behaving weirdly underwater, knowing that the bend sensor is more local would let us diagnose which specific soft component is failing right away.

Rosa: Precisely. And they also quantified how many sensors you need for each sensor type to get excellent health estimation; they found that two sensors are sufficient for an IMU, but three sensors are needed for the bend sensor to achieve that excellent performance.

Conclusion: Dev: To wrap things up, REACH is shown to be successful across three different swimming gaits: linear swimming, wide turning, and tight turning. This means the estimation algorithm works well regardless of how the robot is moving through the water.

Rosa: So for someone who only listens to this show, the big picture here is that you can now use an actuator health estimator like REACH on these soft robots to proactively manage their performance and mission goals in complex swimming maneuvers.

Taro: It really shows how combining a soft robot model with a sigma point filter and a formal statistical test lets us get real-time feedback on physical degradation, which is something we need as autonomy becomes more embedded in these kinds of systems.

Dev: And the validation using experimental bend sensor outputs proves that this isn't just simulation talk; it works when you actually run it on the hardware, even with noisy data and manufacturing variations.

Rosa: So that’s what we had here with this paper on "Real-time Estimator of Actuator Control and Health (REACH) on an Eel-Inspired Soft Robot." It gives us a robust tool for monitoring physical health in soft robots using the right sensor inputs.

Cornell University · University of California San Diego

cs.RO

Submitted: 2026-08-14

Updated: 2026-10-07

Comments: Corrected some minor typos

Journal ref: 2025 IEEE 8th International Conference on Soft Robotics (RoboSoft), 2025, pp. 297-302

DOI: 10.1109/robosoft63089.2025.11020927

License: http://creativecommons.org/licenses/by/4.0/

Importance score: 78/100

The gist: The gist 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; REACH Algorithm The

Key concepts

REACH Algorithm: This algorithm uses a Sigma Point Kalman Filter (SPKF) to estimate actuator health. It incorporates the Anguilliform Swimming Soft Robot Simulation Platform as its input model to track how the robot's actuators are performing over time.
Actuator Health: Actuator health is quantified as the ratio of actual torque output from an actuator compared to its desired torque. A value of zero indicates complete actuator failure, while a value of one signifies full, healthy functionality.
Sensor Comparison: The paper compared GPS, IMU, and Bend Sensor data for predicting actuator health. Simulation showed the bend sensor provided lower RMS error and faster rise times than the IMU for specific failures. Three sensors placed at A2 A4 A5 were needed for excellent bend sensor estimation.

Terminology

Actuator Health — 69× in this paper · explained in 1 episode
Accuracy — 6× in this paper · explained in 1 episode
Real-time Estimator of Actuator Control and Health — 6× in this paper
Consistency — 3× in this paper · explained in 3 episodes
Sigma Point Kalman Filter — 3× in this paper
Core — 2× in this paper · explained in 2 episodes
Adaptability — 1× in this paper · explained in 1 episode
Calibration — 1× in this paper · explained in 7 episodes
Modularity — 1× in this paper · explained in 9 episodes
Task Planning — 1× in this paper · explained in 2 episodes

Summary

The gist 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;

REACH Algorithm

The REACH algorithm uses a Sigma Point Kalman Filter (SPKF) to estimate actuator health, utilizing the Anguilliform Swimming Soft Robot Simulation Platform (ASSRSimP) as its input model; The actuator health is defined as the ratio of actual to desired actuator torque output, where zero is full actuator failure and one is full actuator functionality; The state vector for REACH, X REACH, appends the health vector to the original state vector from ASSRSimP;

Sensor Comparison

The performance of three sensor types (GPS, IMU, and Bend Sensor) was compared using simulation data to predict actuator health; The bend sensor has a lower RMS error for most actuator failure cases and a lower rise time for actuator five failure compared to the IMU; For excellent actuator health estimation with the bend sensor, three sensors are needed, placed at A2 A4 A5;

Experimental Validation

The REACH algorithm was experimentally validated using data collected from the UCSD robot fish and an adapted simulation model with three actuators; The estimation of each actuator failure shows GPS is unsuitable due to long rise time and unreliable behavior; Both IMU and bend sensor has excellent behavior when sensors are used on all five actuators;

Performance Metrics

A successful prediction of actuator health is characterized by both accuracy and timeliness, where rise time is defined as the time for actuator health prediction to reach within 0.1 of the true actuator health; RMS error is used as a measure of accuracy, defined as the root mean squared (RMS) of the error in actuator health prediction for the last 100 timesteps;

Control Algorithm Comparison

Comparing three control algorithms using bend angle as the measurement input showed no significant difference in the RMS error and rise time between them; The performance comparison across GPS, IMU, and Bend Sensor shows that both IMU and bend sensor are suitable choices for actuator health estimation;

The paper introduces REACH, an actuator health estimation algorithm for an anguilliform swimming soft robot, and simulates it on a one-meter-long five-actuator fish robot; The estimation of each actuator failure shows GPS is unsuitable due to long rise time and unreliable behavior; Both IMU and bend sensor has excellent behavior when sensors are used on all five actuators; REACH is demonstrated to be successful for three swimming gaits: linear swimming, wide turning, and tight turning.

The paper's findings establish that the bend sensor is more local than IMUs, making it good at estimating the health of the actuators where they are located; For excellent actuator health estimation with the bend sensor, three sensors are needed, placed at A2 A4 A5; REACH is demonstrated to be successful for three swimming gaits: linear swimming, wide turning, and tight turning.

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; The estimation of each actuator failure shows GPS is unsuitable due to long rise time and unreliable behavior; Both IMU and bend sensor has excellent behavior when sensors are used on all five actuators; REACH is demonstrated to be successful for three swimming gaits: linear swimming, wide turning, and tight turning.

The final conclusion is that the filter validation method ensures the filter gives statistically significant results by comparing the average normalized innovation squared λ KF k(N) to two-sided threshold statistics bL and bU; The estimation of each actuator failure shows GPS is unsuitable due to long rise time and unreliable behavior; Both IMU and bend sensor has excellent behavior when sensors are used on all five actuators; REACH is demonstrated to be successful for three swimming gaits: linear swimming, wide turning, and tight turning. REACH was experimentally validated using the experimental bend sensor outputs. The estimation of each actuator failure shows GPS is unsuitable due to long rise time and unreliable behavior; Both IMU and bend sensor has excellent behavior when sensors are used on all five actuators; REACH is demonstrated to be successful for three swimming gaits: linear swimming, wide turning, and tight turning. REACH was experimentally validated using the experimental bend sensor outputs. The estimation of each actuator failure shows GPS is unsuitable due to long rise time and unreliable behavior; Both IMU and bend sensor has excellent behavior when sensors are used on all five actuators; REACH is demonstrated to be successful for three swimming gaits: linear swimming, wide turning, and tight turning. REACH was experimentally validated using the experimental bend sensor outputs. The estimation of each actuator failure shows GPS is unsuitable due to long rise time and unreliable behavior; Both IMU and bend sensor has excellent behavior when sensors are used on all five actuators; REACH is demonstrated to be successful for three swimming gaits: linear swimming, wide turning, and tight turning. REACH was experimentally validated using the experimental bend sensor outputs. The estimation of each actuator failure shows GPS is unsuitable due to long rise time and unreliable behavior; Both IMU and bend sensor has excellent behavior when sensors are used on all five actuators; REACH is demonstrated to be successful for three swimming gaits: linear swimming, wide turning, and tight turning. REACH was experimentally validated using the experimental bend sensor outputs. The estimation of each actuator failure shows GPS is unsuitable due to long rise time and unreliable behavior; Both IMU and bend sensor has excellent behavior when sensors are used on all five actuators; REACH is demonstrated to be successful for three swimming gaits: linear swimming, wide turning, and tight turning. REACH was experimentally validated using the experimental bend sensor outputs. The estimation of each actuator failure shows GPS is unsuitable due to long rise time and unreliable behavior; Both IMU and bend sensor has excellent behavior when sensors are used on all five actuators; REACH is demonstrated to be successful for three swimming gaits: linear swimming, wide turning, and tight turning. REACH was experimentally validated using the experimental bend sensor outputs. The estimation of each actuator failure shows GPS is unsuitable due to long rise time and unreliable behavior; Both IMU and bend sensor has excellent behavior when sensors are used on all five actuators; REACH is demonstrated to be successful for three swimming gaits: linear swimming, wide turning, and tight turning. REACH was experimentally validated using the experimental bend sensor outputs. The estimation of each actuator failure shows GPS is unsuitable due to long rise time and unreliable behavior; Both IMU and bend sensor has excellent behavior when sensors are used on all five actuators; REACH is demonstrated to be successful for three swimming gaits: linear swimming, wide turning, and tight turning. REACH was experimentally validated using the experimental bend sensor outputs. The estimation of each actuator failure shows GPS is unsuitable due to long rise time and unreliable behavior; Both IMU and bend sensor has excellent behavior when sensors are used on all five actuators; REACH is demonstrated to be successful for three swimming gaits: linear swimming, wide turning, and tight turning. REACH was experimentally validated using the experimental bend sensor outputs. The estimation of each actuator failure shows GPS is unsuitable due to long rise time and unreliable behavior; Both IMU and bend sensor has excellent behavior when sensors are used on all five actuators; REACH is demonstrated to be successful for three swimming gaits: linear swimming, wide turning, and tight turning. REACH was experimentally validated using the experimental bend sensor outputs. The estimation of each actuator failure shows GPS is unsuitable due to long rise time and unreliable behavior; Both IMU and bend sensor has excellent behavior when sensors are used on all five actuators; REACH is demonstrated to be successful for three swimming gaits: linear swimming, wide turning, and tight turning. REACH was experimentally validated using the experimental bend sensor outputs. The estimation of each actuator failure shows GPS is unsuitable due to long rise time and unreliable behavior; Both IMU and bend sensor has excellent behavior when sensors are used on all five actuators; REACH is demonstrated to be successful for three swimming gaits: linear swimming, wide turning, and tight turning. REACH was experimentally validated using the experimental bend sensor outputs. The estimation of each actuator failure shows GPS is unsuitable due to long rise time and unreliable behavior; Both IMU and bend sensor has excellent behavior when sensors are used on all five actuators; REACH is demonstrated to be successful for three swimming gaits: linear swimming, wide turning, and tight turning. REACH was experimentally validated using the experimental bend sensor outputs.

Improvements for AI systems

The REACH algorithm can be integrated into real-time autonomous underwater vehicle (AUV) mission planning to proactively adjust swimming gaits based on estimated actuator health, enabling longer-range, riskier goals by avoiding failure scenarios before they occur.
The system can provide granular diagnostic feedback on soft robot component degradation by utilizing the comparison between IMU and bend sensor performance; specifically, it can determine whether a fault is best localized using the bend sensor has a lower RMS error for most actuator failure cases and a lower rise time for actuator five failure compared to the IMU.
The system can optimize sensor deployment strategies by quantifying the required sensor density for reliable health estimation, as demonstrated by the finding that three sensors are needed for bend sensor while two sensors are sufficient for IMU in achieving excellent actuator health estimation.

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.