Supervised Learning Approaches for Static and Temporal Force Prediction in a Biohybrid Muscle Ring Actuator
Listen
Radio episode about this paper
Transcript
Introduction to the show: ident: Robotics Radio. Generated commentary on the latest robotics and control papers.
Rosa: We're looking at the paper: "Supervised Learning Approaches for Static and Temporal Force Prediction in a Biohybrid Muscle Ring Actuator".
Title and authors: Dev: The authors are listed as Michail Antisthenis Tsompanas, Marco Perez Hernandez, Faisal Abdul-Fattah, Karim Elhakim, Mostafa Ibrahim School of Computing and Creative Technologies at the University of the West of England in Bristol. It's a team from computing and engineering backgrounds.
Rosa: That’s right. They are bringing together computational methods with the biological hardware here. The title tells us exactly what they’re doing: using supervised learning for static and temporal force prediction in biohybrid muscle ring actuators, which is pretty specific.
Taro: What does that actually mean for a robot that might have some unpredictable movement? Does it just give us a better guess of the max force it can push?
Dev: It means they are trying to predict the maximum possible exerted force on the pillars based on things like the muscle sample, electrical stimulation parameters, and baseline behavior before they even apply any stimuli. That’s what that static prediction is for.
Rosa: Exactly. They found that their best static models got an R two score of zero point nine four two five, while the dynamic model achieved an R two score of zero point nine nine five six when predicting the full time series of exerted force in response to electrical stimulation, which is really high accuracy for a dynamic system <ref:2602.16330#pg1>.
Taro: A prediction that works well during steady-state operation but then drops off when there are sudden changes, you know, if something unexpected happens?
Dev: That’s the caveat they found. The results show that forecast accuracy is high and stable during steady-state operation but deteriorates at sudden changes, suggesting a need to incorporate dynamic uncertainty into the digital twin they are building.
The paper's summary: Rosa: So, putting it together, this paper summarizes how they used two complementary modeling strategies to handle the challenges of biological variability and nonlinearity in biohybrid actuation. They looked at static prediction using random forest and neural network models to estimate the maximum possible exerted force based on muscle sample, electrical stimulation parameters, and baseline behavior before stimuli application.
Dev: And then they introduced a dynamic modeling framework using a long short term memory network to construct a digital twin capable of reproducing the full time series of exerted force in response to electrical stimulation. They’re treating it as a univariate time series forecasting problem.
Taro: So, the static models give us an estimate for the maximum force, and the dynamic model gives us what happens over time when we actually stimulate it. How does that help someone who only listens to this show?
Rosa: For someone listening, it means we get a tool to model these muscle actuators. The static part helps optimize performance for targeted applications by estimating the maximum force per experiment based on inputs like the sample number of the muscle ring, frequency, pulse width and waveform type of stimulation signals.
Dev: And the dynamic model is what lets them reproduce the full time series exerted force, which means it’s a foundation for developing robustly adaptive control strategies in future ML-driven biohybrid robotic systems. That’s where the real control potential lies.
The paper's improvements: Taro: I saw some details about how they improved their static model, like adding that extra input feature of the average of the first ten time steps recorded for each experiment. What was the impact of that change?
Rosa: They found that incorporating baseline exerted force prior to stimulation further improved model accuracy by reducing the mean square error by nine point six percent relative to the initial neural network and by twenty point four percent compared to a fine tuned random forest model <ref:2602.16330#pg3>. That’s a solid improvement for handling those inter-sample offsets we talked about earlier, Taro.
Dev: It sounds like that baseline force input was key because they noted there was considerable variation in baseline force across experiments and muscle samples; by supplying this baseline explicitly, the network could better adjust for those offsets. That makes sense from a control perspective.
Taro: So it’s about making the static prediction more robust against those natural differences between different muscle samples or setups in the lab?
Rosa: Precisely. The residual analysis confirmed that all their static models showed a random and homogeneous error distribution, which suggests that while they improved accuracy, there isn't a hidden systematic bias left in their predictions for those maximum force estimates.
Conclusion: Dev: So to wrap up on this paper, the implication is that machine learning tools are expected to have a key role in future designing processes and optimization of biohybrid machines because deterministic modeling is difficult due to high biological variability and nonlinear dynamics.
Rosa: They demonstrated that ML models can effectively predict these outputs through both static and dynamic training on experimental data, with the dynamic prediction model reproducing the full time series of force achieved by the actuator with an r two score of zero point nine nine five six <ref:2602.16330#pg1>.
Taro: I think what this really changes for us is that we now have a way to use these models to guide how we design and optimize these systems, even if they aren't perfect in every scenario.
Dev: Yeah, the dynamic model provides a foundation for developing robustly adaptive control strategies in future biohybrid robotic systems. We can use that time series prediction as a baseline for whatever control loop we build on top of it.
Rosa: That’s the big picture. We learned how to use supervised learning to handle these complex biological dynamics, and we got high fidelity results for both static and dynamic force prediction in this paper, "Supervised Learning Approaches for Static and Temporal Force Prediction in a Biohybrid Muscle Ring Actuator."
Michail-Antisthenis Tsompanas, Marco Perez Hernandez, Faisal Abdul-Fattah, Karim Elhakim, Mostafa Ibrahim, Judith Fuentes, Florencia Lezcano, Riccardo Collu, Massimo Barbaro, Stefano Lai, Samuel Sanchez
Unconventional Computing Laboratory & School of Computing & Creative Technologies, University of the West of England, Bristol Institute for Bioengineering of Catalonia (IBEC), The Barcelona Institute for Science and Technology (BIST), Department of Electrical and Electronic Engineering, University of Cagliari, Institució Catalana de Recerca i Estudis Avancats (ICREA)
cs.RO, cs.ET
Submitted: 2026-02-18
Updated: 2026-11-17
DOI: 10.1002/aisy.70637
License: http://creativecommons.org/licenses/by-nc-sa/4.0/
Importance score: 83/100
The gist: Skeletal muscle-based biohybrid actuators anchored on flexible polymer pillars were studied under static and dynamic machine learning prediction models.
Key concepts
- Static Prediction Models
- These models estimate the maximum possible force a muscle ring can exert before stimulation by using inputs like muscle sample number, stimulation frequency, and waveform type. They help optimize actuator performance for specific force requirements.
- Dynamic Modeling Framework (LSTM)
- This framework uses a Long Short-Term Memory network to create a digital twin of the actuator. It learns to replicate the entire sequence of force measurements over time in response to electrical stimulation, which is crucial for understanding temporal behavior.
- R2 Score
- The R2 score measures how well the model's predictions match the actual experimental data. A high R2 score indicates that the machine learning model is highly accurate at predicting both static force limits and dynamic force changes.
- Digital Twin
- In this context, a digital twin is a virtual replica of the physical biohybrid muscle actuator. The LSTM model acts as this twin, allowing researchers to simulate and predict how the physical device will behave under different stimulation conditions.
Terminology
Summary
Skeletal muscle-based biohybrid actuators anchored on flexible polymer pillars were studied under static and dynamic machine learning prediction models. p. 0 To address these issues, this study
investigates the application of supervised learning, a form of machine learning, to model and predict
the behavior of biohybrid machines (BHMs), focusing on a muscle ring anchored on flexible polymer
pillars. First, static prediction models (i. However, their intrinsic biological variability and nonlinearity pose
significant challenges for controllability and predictability. To address these issues, this study
investigates the application of supervised learning, a form of machine learning, to model and predict
the behavior of biohybrid machines (BHMs), focusing on a muscle ring anchored on flexible polymer
pillars. A PREPRINT - FEBRUARY 19, 2026
demonstrate high predictive accuracy. The best performance of the static models is characterized
by R2 of 0. The best performance of the static models is characterized
by R2 of 0. 9425, whereas the dynamic model achieves R2 of 0.Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Modeling Approaches
The research proposes two complementary modeling strategies to address the challenges of biological variability and nonlinearity in biohybrid actuation. The first approach focuses on static prediction, employing random forest (RF) and neural network (NN) models to estimate the maximum possible exerted force based on muscle sample, electrical stimulation parameters, and baseline behavior before stimuli application. p. 1 To address the key challenges of variability and nonlinearity in biohybrid actuation, we
explore two complementary modeling strategies. The first approach focuses on static prediction, employing at first
random forest (RF) and then feedforward neural network (NN) models to estimate the maximum possible exerted
force on the pillars based on the muscle sample, electrical stimulation parameters, and baseline behavior before the
application of stimuli. The first approach focuses on static prediction, employing at first
random forest (RF) and then feedforward neural network (NN) models to estimate the maximum possible exerted
force on the pillars based on the muscle sample, electrical stimulation parameters, and baseline behavior before the
application of stimuli. The second approach introduces a dynamic modeling framework, utilizing a recurrent NN,
specifically a Long Short-Term Memory (LSTM) architecture, to construct a digital twin of the biohybrid mechanism
2Where the paper says it
Where the paper says it
Static Model Performance
The static prediction models were trained on a dataset of 161 data points, with the inputs being the sample number of the muscle ring, frequency, pulse width, and waveform type of stimulation signals. The Random Forest Regressor (RFR) achieved an R2 score of 0. p. 2 The performance of the Random Forest Regressor (RFR) was optimized through hyperparameter tuning, via randomized
search cross validation. The RFR model achieved its best performance with the following hyperparameters: number of
estimators equal to 300, maximum depth of individual trees equal to 10, minimum samples split equal to 2 and minimum
samples leaf equal to 1. The RFR model achieved its best performance with the following hyperparameters: number of
estimators equal to 300, maximum depth of individual trees equal to 10, minimum samples split equal to 2 and minimum
samples leaf equal to 1. The optimized model fitted the dataset with a mean square error (MSE) of 6. The NN achieved an MSE of 5. 31 × 10−9
and an R2
score of 0. The RFR model achieved its best performance with the following hyperparameters: number of
estimators equal to 300, maximum depth of individual trees equal to 10, minimum samples split equal to 2 and minimum
samples leaf equal to 1. The optimized model fitted the dataset with a mean square error (MSE) of 6. To further enhance the NN model performance and mitigate the effect of high biological variability, an additional
input feature was introduced. Namely, the average of the first 10 time steps (equivalent to the first 0. The best performance of the static models is characterized
by R2 of 0. 9425, whereas the dynamic model achieves R2 of 0. Incorporating the
baseline exerted force prior to stimulation further improved model accuracy, as expected. It is noteworthy here that the
NN with additional input feature of the unstimulated force reduced MSE by 9. It is noteworthy here that the
NN with additional input feature of the unstimulated force reduced MSE by 9. 6% relative to the initial NN model and
by 20.Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Dynamic Model Accuracy
The dynamic prediction model utilized an LSTM network trained on 122,176 training windows of 10 consecutive force measurements. After training for 30 epochs, the resulting MSE was 0. p. 3 The data were split into 70% for training, 15% for validation and 15% for testing. After training
the LSTM for 30 epochs, the resulting MSE was 0. After training
the LSTM for 30 epochs, the resulting MSE was 0. 0013, while the R2
score was equal to 0.Where the paper says it
Where the paper says it
Limitations and Future Work
The present study has several limitations, including a small dataset size for static models which prevents generalization to different configurations Furthermore, the dynamic model was implemented in only an open-loop context; a closed-loop feedback implementation would contribute a more robust methodology Future work will focus on including experiments with different configurations and larger numbers to the training dataset The extension of dynamic ML models to include multiple inputs, such as the time series of the stimulation electrical signal, will be investigated to better anticipate sudden changes in system state The inclusion of closed-loop ML methodologies will be studied to enable self-adaptive control strategies for BHMs that can be directly deployed in hardware
Experimental Setup
The experimental setup involved C2C12 mouse myoblasts cultured in a custom hydrogel molded into PDMS-based circular rings anchored between two posts separated by 9 mm Electrical stimulation was provided using handmade electrodes placed 13 mm apart on opposite sides of the Petri dish cover Three different stimulation strategies were applied to verify tetanic contractions and modulate actuator contraction in a controllable manner The overall dataset consists of 123,786 time steps where force is measured with a measurement interval of 0.04s
Conclusion
ML tools are expected to have a key role in future designing processes, optimization and controllability of BHMs due to the difficulty in deterministic modeling caused by high biological variability and nonlinear dynamics This work demonstrated that ML models can effectively predict BHMs outputs through static and dynamic training on experimental data The dynamic prediction model reproduced the full time series of force achieved by the actuator with an R2 score of 0.9956 This work demonstrates that ML models can be utilized to effectively predict BHMs outputs through static and dynamic training on experimental data The static prediction models achieved increasing accuracy in predicting the maximum force per experiment based on inputs like the sample number of the muscle ring, frequency, pulse width and waveform type of stimulation signals The inclusion of baseline exerted force before electrical stimulus enhanced NN model performance by reducing MSE by 9. p. 4 The static prediction models achieved increasing accuracy in predicting the maximum force per
experiment based on inputs like the sample number of the muscle ring, the frequency, pulse width and waveform type
of stimulation signals. Whereas, the inclusion of baseline exerted force before the application of electrical stimulus
enhanced the performance of a NN model by reducing the MSE by 9. Whereas, the inclusion of baseline exerted force before the application of electrical stimulus
enhanced the performance of a NN model by reducing the MSE by 9. 6% relative to the initial NN configuration and by
20.Where the paper says it
Where the paper says it
REFERENCES
[1] C. Appiah, C. Arndt, K. Siemsen, A. Heitmann, A. Staubitz, and C. Selhuber-Unkel. Living materials herald a new era in soft robotics
[2] A. Bartolucci, J. Fuentes, D. p. 0 tsompanas@uwe. uk
Marco Perez Hernandez, Faisal Abdul-Fattah, Karim Elhakim, Mostafa Ibrahim
School of Computing & Creative Technologies,
College of Arts, Technology and Environment,
University of the West of England,
Bristol, BS16 1QY, United Kingdom
Judith Fuentes, Florencia Lezcano
Institute for Bioengineering of Catalonia (IBEC),
The Barcelona Institute for Science and Technology (BIST),
Barcelona, Spain
Riccardo Collu, Massimo Barbaro, Stefano Lai
Department of Electrical and Electronic Engineering,
University of Cagliari,
Cagliari, Italy
Samuel Sánchez
Instituciò Catalana de Recerca i Estudis Avancats (ICREA) and
Institute for Bioengineering of Catalonia (IBEC),
The Barcelona Institute for Science and Technology (BIST),
Barcelona, Spain
Andrew Adamatzky
Unconventional Computing Laboratory,
College of Arts, Technology and Environment,
University of the West of England,
Bristol, BS16 1QY, United Kingdom
February 19, 2026
ABSTRACT
Skeletal muscle-based biohybrid actuators have proved to be a promising component in soft robotics,
offering efficient movement. tsompanas@uwe. uk
Marco Perez Hernandez, Faisal Abdul-Fattah, Karim Elhakim, Mostafa Ibrahim
School of Computing & Creative Technologies,
College of Arts, Technology and Environment,
University of the West of England,
Bristol, BS16 1QY, United Kingdom
Judith Fuentes, Florencia Lezcano
Institute for Bioengineering of Catalonia (IBEC),
The Barcelona Institute for Science and Technology (BIST),
Barcelona, Spain
Riccardo Collu, Massimo Barbaro, Stefano Lai
Department of Electrical and Electronic Engineering,
University of Cagliari,
Cagliari, Italy
Samuel Sánchez
Instituciò Catalana de Recerca i Estudis Avancats (ICREA) and
Institute for Bioengineering of Catalonia (IBEC),
The Barcelona Institute for Science and Technology (BIST),
Barcelona, Spain
Andrew Adamatzky
Unconventional Computing Laboratory,
College of Arts, Technology and Environment,
University of the West of England,
Bristol, BS16 1QY, United Kingdom
February 19, 2026
ABSTRACT
Skeletal muscle-based biohybrid actuators have proved to be a promising component in soft robotics,
offering efficient movement. tsompanas@uwe. uk
Marco Perez Hernandez, Faisal Abdul-Fattah, Karim Elhakim, Mostafa Ibrahim
School of Computing & Creative Technologies,
College of Arts, Technology and Environment,
University of the West of England,
Bristol, BS16 1QY, United Kingdom
Judith Fuentes, Florencia Lezcano
Institute for Bioengineering of Catalonia (IBEC),
The Barcelona Institute for Science and Technology (BIST),
Barcelona, Spain
Riccardo Collu, Massimo Barbaro, Stefano Lai
Department of Electrical and Electronic Engineering,
University of Cagliari,
Cagliari, Italy
Samuel Sánchez
Instituciò Catalana de Recerca i Estudis Avancats (ICREA) and
Institute for Bioengineering of Catalonia (IBEC),
The Barcelona Institute for Science and Technology (BIST),
Barcelona, Spain
Andrew Adamatzky
Unconventional Computing Laboratory,
College of Arts, Technology and Environment,
University of the West of England,
Bristol, BS16 1QY, United Kingdom
February 19, 2026
ABSTRACT
Skeletal muscle-based biohybrid actuators have proved to be a promising component in soft robotics,
offering efficient movement.Where the paper says it
Where the paper says it
Where the paper says it
[3] L. Breiman. Random forests
[4] R. W. Carlsen and M. Sitti. Bio-hybrid cell-based actuators for microsystems
[5] Y. Chi, Y. Zhao, Y. Hong, Y. Li, and J. Yin. A perspective on miniature soft robotics: Actuation, fabrication, control, and applications
[6] M. Cianchetti, C. Laschi, A. Menciassi, and P. Dario. Biomedical applications of soft robotics
[7] R. Collu, J. Fuentes, F. p. 0 tsompanas@uwe. uk
Marco Perez Hernandez, Faisal Abdul-Fattah, Karim Elhakim, Mostafa Ibrahim
School of Computing & Creative Technologies,
College of Arts, Technology and Environment,
University of the West of England,
Bristol, BS16 1QY, United Kingdom
Judith Fuentes, Florencia Lezcano
Institute for Bioengineering of Catalonia (IBEC),
The Barcelona Institute for Science and Technology (BIST),
Barcelona, Spain
Riccardo Collu, Massimo Barbaro, Stefano Lai
Department of Electrical and Electronic Engineering,
University of Cagliari,
Cagliari, Italy
Samuel Sánchez
Instituciò Catalana de Recerca i Estudis Avancats (ICREA) and
Institute for Bioengineering of Catalonia (IBEC),
The Barcelona Institute for Science and Technology (BIST),
Barcelona, Spain
Andrew Adamatzky
Unconventional Computing Laboratory,
College of Arts, Technology and Environment,
University of the West of England,
Bristol, BS16 1QY, United Kingdom
February 19, 2026
ABSTRACT
Skeletal muscle-based biohybrid actuators have proved to be a promising component in soft robotics,
offering efficient movement.Where the paper says it
[8] M. Filippi, A. Balciunaite, A. Georgopoulou, P. Paniagua, F. Drescher, M. Nie Sensor-embedded muscle for closed-loop controllable actuation in proprioceptive biohybrid robots
[9] M. Guix, R. Mestre, T. Patiño, M. De Corato, J. Fuentes Biohybrid soft robots with self-stimulating skeletons
[10] D. R. Higueras-Ruiz, K. Nishikawa, H. Feigenbaum, and M. Shafer. What is an artificial muscle? p. 0 tsompanas@uwe. uk
Marco Perez Hernandez, Faisal Abdul-Fattah, Karim Elhakim, Mostafa Ibrahim
School of Computing & Creative Technologies,
College of Arts, Technology and Environment,
University of the West of England,
Bristol, BS16 1QY, United Kingdom
Judith Fuentes, Florencia Lezcano
Institute for Bioengineering of Catalonia (IBEC),
The Barcelona Institute for Science and Technology (BIST),
Barcelona, Spain
Riccardo Collu, Massimo Barbaro, Stefano Lai
Department of Electrical and Electronic Engineering,
University of Cagliari,
Cagliari, Italy
Samuel Sánchez
Instituciò Catalana de Recerca i Estudis Avancats (ICREA) and
Institute for Bioengineering of Catalonia (IBEC),
The Barcelona Institute for Science and Technology (BIST),
Barcelona, Spain
Andrew Adamatzky
Unconventional Computing Laboratory,
College of Arts, Technology and Environment,
University of the West of England,
Bristol, BS16 1QY, United Kingdom
February 19, 2026
ABSTRACT
Skeletal muscle-based biohybrid actuators have proved to be a promising component in soft robotics,
offering efficient movement.Where the paper says it
[11] E. B. Joyee and Y. Pan. Additive manufacturing of multi-material soft robot for on-demand drug delivery applications
[12] K.-i. Kabumoto, T. Hoshino, Y. Akiyama, and K. Morishima. Voluntary movement controlled by the surface emg signal for tissue-engineered skeletal muscle on a gripping tool
[13] R. Kinjo, Y. Morimoto, B. Jo, and S. Takeuchi. Biohybrid bipedal robot powered by skeletal muscle tissue
[14] S. Lai, J. Fuentes, M. p. 0 tsompanas@uwe. uk
Marco Perez Hernandez, Faisal Abdul-Fattah, Karim Elhakim, Mostafa Ibrahim
School of Computing & Creative Technologies,
College of Arts, Technology and Environment,
University of the West of England,
Bristol, BS16 1QY, United Kingdom
Judith Fuentes, Florencia Lezcano
Institute for Bioengineering of Catalonia (IBEC),
The Barcelona Institute for Science and Technology (BIST),
Barcelona, Spain
Riccardo Collu, Massimo Barbaro, Stefano Lai
Department of Electrical and Electronic Engineering,
University of Cagliari,
Cagliari, Italy
Samuel Sánchez
Instituciò Catalana de Recerca i Estudis Avancats (ICREA) and
Institute for Bioengineering of Catalonia (IBEC),
The Barcelona Institute for Science and Technology (BIST),
Barcelona, Spain
Andrew Adamatzky
Unconventional Computing Laboratory,
College of Arts, Technology and Environment,
University of the West of England,
Bristol, BS16 1QY, United Kingdom
February 19, 2026
ABSTRACT
Skeletal muscle-based biohybrid actuators have proved to be a promising component in soft robotics,
offering efficient movement.Where the paper says it
[15] T. Li and S. Takeuchi. Advancing biohybrid robotics: Innovations in contraction models, control techniques, and applications
[16] U. Mahmood, G. Casula, J. Llanos, I. Niosiline, C. Salvatori Integration of organic field-effect transistor-based strain sensors to soft robotic devices and systems
[17] D. R. Merrill, M. Bikson, and J. G. Jefferys. Electrical stimulation of excitable tissue: design of efficacious and safe protocols
[18] P. Ramdya and A. J. Ijspeert. The neuromechanics of animal locomotion: From biology to robotics and back
[19] L. Ricotti and A. Menciassi. Bio-hybrid muscle cell-based actuators
[20] L. Ricotti, B. Trimmer, A. W. Feinberg, R. Raman, K. K. Parker Biohybrid actuators for robotics: A review of devices actuated by living cells
[21] A. Sarker, T. Ul Islam, and M. R. Islam. A review on recent trends of bioinspired soft robotics: Actuators, control methods, materials selection, sensors, challenges, and future prospects
[22] J. Wang and A. Chortos. Control strategies for soft robot systems
[23] V. A. Webster-Wood, M. Guix, N. W. Xu, B. Behkam, H. Sato Biohybrid robots: recent progress, challenges, and perspectives
[24] A Short text and graphic for the Table of Contents. Skeletal muscle-based biohybrid actuators anchored on flexible polymer pillars were studied under static and dynamic machine learning prediction models
[25] Skeletal muscle-based biohybrid actuators anchored on flexible polymer pillars were studied under static and dynamic machine learning prediction models.
[26] Skeletal muscle-based biohybrid actuators anchored on flexible polymer pillars were studied under static and dynamic machine learning prediction models. p. 0 To address these issues, this study
investigates the application of supervised learning, a form of machine learning, to model and predict
the behavior of biohybrid machines (BHMs), focusing on a muscle ring anchored on flexible polymer
pillars. First, static prediction models (i.Where the paper says it
Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.
Improvements for AI systems
-
Static prediction models can enable optimization of muscle actuator performance for targeted applications and required force outcomes by estimating
the maximum possible exerted force on the pillars based on the muscle sample, electrical stimulation parameters, and baseline behavior before the application of stimuli.
-
Dynamic modeling frameworks based on Long Short-Term Memory networks can serve as a digital twin to
replicate the time series of exerted forces observed in response to electrical stimulation,
providing a foundation for developingrobustly adaptive control strategies in future biohybrid robotic systems.
-
The integration of baseline force as an auxiliary input feature into the neural network model enhances performance, as it is noted that this step
motivated by the observed considerable variation in baseline force across experiments and muscle samples; by supplying this baseline explicitly, the network could better adjust for inter-sample offsets.
-
The dynamic prediction model exhibits high fidelity to temporal dynamics, achieving an R2 score of 0.9956, which suggests it can accurately forecast
the full time series exerted force in response to electrical stimulation,
though its performance deteriorates at sudden changes.
Abstract
Skeletal muscle-based biohybrid actuators have proved to be a promising component in soft robotics, offering efficient movement. However, their intrinsic biological variability and nonlinearity pose significant challenges for controllability and predictability. To address these issues, this study investigates the application of supervised learning, a form of machine learning, to model and predict the behavior of biohybrid machines (BHMs), focusing on a muscle ring anchored on flexible polymer pillars. First, static prediction models (i.e., random forest and neural network regressors) are trained to estimate the maximum exerted force achieved from input variables such as muscle sample, electrical stimulation parameters, and baseline exerted force. Second, a dynamic modeling framework, based on Long Short-Term Memory networks, is developed to replicate the time series of exerted forces observed. Both modeling approaches demonstrate high predictive accuracy. The best performance of the static models is characterized by a mean R2 of 0.9346, whereas the dynamic model achieves a mean R2 of 0.9944 over five training instances. The static models can enable optimization of muscle actuator performance for targeted applications and required force outcomes, while the dynamic model provides a foundation for developing robustly adaptive control strategies in future biohybrid robotic systems.
Related papers
- Rethinking Actions: Continuity-Aware Priors for World-Model Planning
- MOSAIC: Modular Obstacle and Spatial-Aware Framework for Manipulation In Cluttered Scenes
- SAGAS: Semantic-Aware Graph-Assisted Stitching for Offline Temporal Logic Planning
- DexSeed: Expanding Human Hand-Object Interactions into Robot Demonstrations for Dexterous Policy Learning
- LaST0.5: A Generalist World Action Model with Interaction-Centric Latent Reasoning
- CIALLO: Learning Shared Intent Representations for Cross-Embodiment Manipulation