Sergiy Yakovenko

A phase decoder for reconstructing missing muscle activity

This paper introduces a phase decoder that reconstructs missing muscle activity from sparse EMG signals, achieving accurate phase estimation and muscle activation reconstruction in real-time.

Neural networks learn forward dynamics when freed from numerical integration

The paper shows that a two-stage model (APE) outperforms a direct-mapping neural network in predicting limb dynamics under novel conditions by explicitly modeling equations of motion and numerical integration.

Evidence of sensory error threshold in triggering locomotor adaptations in humans

The paper investigates the error threshold in triggering locomotor adaptations in humans by manipulating kinematic and kinetic constraints and finding that asymmetric loading, not kinematic asymmetry alone, induces persistent adaptation.

Damage explains function in spiking neural networks representing central pattern generator

The paper develops a spiking neural network model of the central pattern generator to study distributed computations and shows that lesioning different neural elements affects CPG function differently.

Artificial physics engine for real-time inverse dynamics of arm and hand movement

This paper develops an artificial physics engine using neural networks to simulate inverse dynamics of arm and hand movement with low torque errors.

Does joint impedance improve dynamic leg simulations with explicit and implicit solvers?

This study evaluates how joint impedance affects the accuracy and stability of dynamic leg simulations using explicit and implicit solvers, finding that viscoelasticity reduces errors in explicit methods.

Intralimb locomotor coordination in rats walking on asymmetric pegway

The paper characterizes spatiotemporal muscle activation patterns in rat forelimbs during asymmetric stepping, finding stereotypical and lateralized patterns similar to those in cats.

Neuromechanical Coupling is Reflected in the Spatial Organization of the Spinal Motoneuron Pools

This study shows that the spatial organization of spinal motoneuron pools reflects the functional relationships between muscles they innervate, supporting neuromechanical tuning for upper limb movements.

Solving musculoskeletal biomechanics with machine learning

This paper applies machine learning to model musculoskeletal transformations in the human arm and hand, showing that both gradient boosting and neural networks can accurately predict muscle length and moment arm relationships.

The primary afferent activity cannot capture the dynamical features of muscle activity during reaching movements.

This study shows that primary afferent feedback cannot replicate the dynamic muscle activity patterns during reaching movements, suggesting that task-dependent muscle co-activation and nonlinear feedback are key to controlling limb dynamics.

Functional and Structural Moment Arm Validation for Musculoskeletal Models: A Study of the Human Forearm and Hand

The paper validates and scales musculoskeletal models of the human forearm and hand using posture-dependent moment arms, isometric force measurements, and error detection to improve accuracy in torque prediction.

Model of a bilateral Brown-type central pattern generator for symmetric and asymmetric locomotion

The paper provides an analytical solution to a Brown-type CPG model, demonstrating its efficiency and ability to control symmetric and asymmetric locomotion through parametric sensitivity analysis.