Viacheslav Koshman

This study presents a methodology for identifying the most informative frequencies and channels in electromyography (EMG) data to evaluate muscle recovery using Decision Tree classifiers. EMG signals, recorded from the vastus lateralis muscle during squat exercises, were analyzed across varying rest intervals to assess optimal recovery periods. By employing single Decision Tree classifiers, the study enhances interpretability, offering insights into feature importance — essential for applications in medical and sports settings where transparency is critical. The experimental protocol utilized a grid search for hyperparameter tuning and cross-validation to address class imbalance, ultimately achieving a reliable classification of rest intervals based on power spectral density features. The results indicate that a limited subset of highly informative features provides sufficient accuracy, suggesting that streamlined, interpretable models are effective for the evaluation of muscle recovery. This approach can guide future research in developing compact, robust models adapted to EMG-based diagnostics.

The development of exoskeleton technologies has garnered increasing interest from industrial companies, particularly in the context of active exoskeletons that are powered by an external energy source. A key component in these systems is the actuation device, which plays a pivotal role in their overall performance. Among the various actuation mechanisms, the twisted string actuator (TSA) has emerged as a promising candidate for wearable robotic systems. The TSA mimics the behavior of human muscles but offers higher efficiency, making it an attractive solution for exoskeleton applications.
This study introduces a novel lever-based transmission mechanism designed to adapt TSAs to human joints that require torque, rather than a linear force. To support this approach, we developed a mathematical model that accurately describes the dynamics of the proposed mechanism. A series of experiments were conducted to validate the model, confirming its reliability. In addition to the theoretical work, we integrated the lever-based TSA into an exoskeleton and tested its effectiveness in reducing muscle loads. The experiments focused on squatting exercises, where significant reductions in muscle activity were observed, demonstrating the exoskeleton’s potential for easing physical strain on users. The accuracy of the lever-based TSA model was further confirmed by the experiments in predicting the generated force and the actuation angles achieved. The estimated error between the sensor data and the models predictions were 1.64 MAE (degrees) for the revolute joint angle and 1.79 MAE (Newtons) for the tension force of the string.
Moreover, the results showed that the lever-based TSA provided the necessary torques for the hip joint, aligning well with the natural movement of the human body. This makes it easier to control and adapt the system in practical exoskeleton applications, enhancing its usability and effectiveness.

The use of exoskeletons in manufacturing, construction, and healthcare shows potential for enhancing performance and reducing injury risk. This study evaluates exoskeleton efficacy during heavy lifting using an advanced system integrating electromyography (EMG) and electrocardiography (ECG). Real-time monitoring with baseline correction and filtering ensured precise data. EMG analyses using Root Mean Square (RMS) and Integral methods revealed reduced muscle activation and cumulative exertion during exoskeleton-assisted tasks. ECG data indicated lower cardiovascular strain. Testing with a hip exoskeleton confirmed its ability to decrease physical load, emphasizing the value of integrated physiological monitoring for comprehensive exoskeleton performance assessment and future research directions.