Inferring 3D Forearm Trajectories Using Self-derived sEMG Motion Features

Surface electromyography (sEMG) provides a non-invasive method to record muscle activation patterns. This technology is vital for estimating human motion in wearable robotics and human-machine interaction. Traditionally, researchers mapped these signals to intermediate kinematic variables. However, a new study introduces sEMG motion features that allow for direct inference of 3D forearm trajectories.

Directly inferring 3D trajectories is historically challenging because it usually requires additional sensing modalities. This novel method avoids external sensors by deriving extensible features directly from the sEMG signal itself. Consequently, the system becomes more streamlined and practical for real-world clinical use.

Improving Wearable Robotics with sEMG Motion Features

To validate the proposed framework, researchers performed an in-vivo experiment with ten subjects performing nine typical forearm motions. The results were impressive. Specifically, the self-derived features improved motion estimation performance by an average of 10% to 15%. This improvement occurred across multiple evaluation metrics, demonstrating a robust relationship between the derived features and actual physical movement.

Furthermore, these findings assist in an in-depth understanding of forearm motor control. Because the framework is extensible, developers can integrate more critical features in the future. Therefore, this advancement directly improves human motion trajectory inference for robotics and prosthetic applications.

Frequently Asked Questions

How do sEMG motion features improve trajectory accuracy?

They provide a direct link between muscle activation patterns and 3D movement. This reduces the error often introduced by mapping through intermediate variables like joint torque or angles.

Can this technology be used in prosthetic control?

Yes. By improving 3D trajectory inference, these features enable smoother and more intuitive control of prosthetic limbs, making movements feel more natural for the user.

Disclaimer: This content is for informational and educational purposes only. It does not constitute professional medical advice, diagnosis, or treatment. Always seek the advice of your physician or other qualified healthcare provider with any questions you may have regarding a medical condition. Refer to the latest local and national guidelines for clinical practice.

References

Lan B et al. Self-derived Motion Features from sEMG for Inferring 3D Forearm Trajectories. IEEE Trans Biomed Eng. 2026 Apr 22. doi: 10.1109/TBME.2026.3686312. PMID: 42019063.

Grimmelsmann N et al. sEMG-based prediction of human forearm movements utilizing a biomechanical model based on individual anatomical/ physiological measures and a reduced set of optimization parameters. PMC. 2023 Aug 03.

Review of sEMG for Exoskeleton Robots: Motion Intention Recognition Techniques and Applications. MDPI. 2025 Apr 13.