Serhii Bahdasariants

Serhii Bahdasariants

Auto-generated profile, based on publicly available information

Serhii Bahdasariants

Carnegie Mellon University

Research Focus & Metrics

Academic Career Length

4Years (2023-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
15 April 2024 in IEEE Sensors Letters
Verified
Journal Article
Closed access
SBSerhii Bahdasariants
Serhii Bahdasariants
MYMatthew G. Yough
Matthew G. Yough
VGValeriya Gritsenko
Valeriya Gritsenko
This letter presents a novel method for solving the inverse kinematic problem of capturing human reaching movements using a dynamic biomechanical model. The model consists of rigid segments connected by joints and actuated by markers. The method was validated against a rotation matrix-based method using motion capture data recorded during reaching movements performed by healthy human volunteers. The results show that the proposed method achieved low errors in joint angles and compensated for noise in motion capture data. The angles were comparable to those calculated using the standard marker-based method. The proposed bioinspired method can be used in real-time medical applications for processing noisy marker data with occlusions.

13 December 2023 in PLOS ONE
Verified
Journal Article
Open access
PDF file
MMMykhailo Manukian
Mykhailo Manukian
SBSerhii Bahdasariants
Serhii Bahdasariants
SYSergiy Yakovenko
Sergiy Yakovenko
Simulating human body dynamics requires detailed and accurate mathematical models. When solved inversely, these models provide a comprehensive description of force generation that considers subject morphology and can be applied to control real-time assistive technology, for example, orthosis or muscle/nerve stimulation. Yet, model complexity hinders the speed of its computations and may require approximations as a mitigation strategy. Here, we use machine learning algorithms to provide a method for accurate physics simulations and subject-specific parameterization. Several types of artificial neural networks (ANNs) with varied architecture were tasked to generate the inverse dynamic transformation of realistic arm and hand movement (23 degrees of freedom). Using a physical model, we generated representative limb movements with bell-shaped end-point velocity trajectories within the physiological workspace. This dataset was used to develop ANN transformations with low torque errors (less than 0.1 Nm). Multiple ANN implementations using kinematic sequences solved accurately and robustly the high-dimensional kinematic Jacobian and inverse dynamics of arm and hand. These results provide further support for the use of ANN architectures that use temporal trajectories of time-delayed values to make accurate predictions of limb dynamics.

1 June 2026 in bioRxiv
Verified
Preprint
Open access
PDF file
SBSerhii Bahdasariants
Serhii Bahdasariants
SYSergiy Yakovenko
Sergiy Yakovenko
Seamless interaction between humans and machines requires interfaces that remain robust to the variability inherent in biological signals and physical environments. Advanced human-machine interfaces (HMIs) increasingly rely on machine learning to predict or control limb dynamics. These systems must learn input-to-output mappings between control variables and limb state, such as the mapping from muscle forces or joint torques acting about segmented arm joints to limb posture over time. Such statistical input-to-output transformations can result in numerical instability of predicted musculoskeletal kinematics and dynamics. Achieving the robustness of biological motor control requires solving both forward and inverse dynamics problems; however, these problems are computationally asymmetric because they entail opposing operations–integration and differentiation. Since we have previously shown that neural networks solve the inverse dynamics problem when trained to map kinematic to dynamic signals during reaching, we hypothesized that representing separately the approximation of equations of motion (EOM) and their temporal numerical integration may capture the relevant computational structure of the forward dynamics problem. We tested this hypothesis by comparing a conventional direct-mapping recurrent neural network (RNN) with a two-stage model, the artificial physics engine (APE). When predicting the state of a two-segment system under external perturbations not encountered during training, the direct-mapping, monolithic model produced large prediction errors inconsistent with the expected interaction torque, whereas the APE maintained low error and remained stable under novel initial conditions and perturbations. Mapping system dynamics in the terms of the EOM improves robustness against intrinsic and extrinsic sources of variability by imposing a causal, physics-based structure on HMI design.

3 July 2023 in PLOS ONE
Verified
Journal Article
Open access
PDF file
SBSerhii Bahdasariants
Serhii Bahdasariants
ABAna Maria Forti Barela
Ana Maria Forti Barela
VGValeriya Gritsenko
Valeriya Gritsenko
OBOdair Bacca
Odair Bacca
JBJosé Angelo Barela
José Angelo Barela
SYSergiy Yakovenko
Sergiy Yakovenko
The nervous system predicts and executes complex motion of body segments actuated by the coordinated action of muscles. When a stroke or other traumatic injury disrupts neural processing, the impeded behavior has not only kinematic but also kinetic attributes that require interpretation. Biomechanical models could allow medical specialists to observe these dynamic variables and instantaneously diagnose mobility issues that may otherwise remain unnoticed. However, the real-time and subject-specific dynamic computations necessitate the optimization these simulations. In this study, we explored the effects of intrinsic viscoelasticity, choice of numerical integration method, and decrease in sampling frequency on the accuracy and stability of the simulation. The bipedal model with 17 rotational degrees of freedom (DOF)—describing hip, knee, ankle, and standing foot contact—was instrumented with viscoelastic elements with a resting length in the middle of the DOF range of motion. The accumulation of numerical errors was evaluated in dynamic simulations using swing-phase experimental kinematics. The relationship between viscoelasticity, sampling rates, and the integrator type was evaluated. The optimal selection of these three factors resulted in an accurate reconstruction of joint kinematics (err < 1%) and kinetics (err < 5%) with increased simulation time steps. Notably, joint viscoelasticity reduced the integration errors of explicit methods and had minimal to no additional benefit for implicit methods. Gained insights have the potential to improve diagnostic tools and accurize real-time feedback simulations used in the functional recovery of neuromuscular diseases and intuitive control of modern prosthetic solutions.

11 October 2023 in TechRxiv
Verified
Preprint
Open access
PDF file
MYMatthew Yough
Matthew Yough
SBSerhii Bahdasariants
Serhii Bahdasariants
VGValeriya Gritsenko
Valeriya Gritsenko
This paper presents a novel method for solving the inverse kinematic problem of capturing human reaching movements using a dynamic biomechanical model. The model consists of rigid segments connected by joints and actuated by markers. The method was validated against a rotation matrix-based method using motion capture data recorded during reaching movements performed by healthy human volunteers. The results showed that the proposed method achieved low errors in joint angles. The angles were comparable to those calculated using the standard marker-based method. The proposed bioinspired method can be used in real-time medical applications, such as rehabilitation, assessment, and prosthesis/orthosis control.

8 June 2026 in bioRxiv
Verified
Preprint
Open access
PDF file
SBSerhii Bahdasariants
Serhii Bahdasariants
LPLauren Parola
Lauren Parola
KKKriti Kacker
Kriti Kacker
AFAriel K. Feldman
Ariel K. Feldman
ZZZachary Zdobinski
Zachary Zdobinski
DWDouglas J. Weber
Douglas J. Weber
The movement of articulated bodies, whether biological or robotic, involves two interrelated computational challenges. Inverse dynamics determines the forces or torques required to achieve a specified movement, while forward dynamics computes the resulting accelerations from given forces or torques. Although both computations are governed by the same equations of motion, they are typically implemented as separate inverse and forward mappings in both model-based and data-driven approaches. This division can obscure the underlying structure common to both problems. Here, we introduce the Appellian Neural Network (ANNet), a physics-informed neural network that unifies both computations through a shared learned representation by modeling a single scalar quantity from classical mechanics, specifically Appell’s acceleration energy. Inverse dynamics is obtained by differentiating the learned energy function with respect to acceleration to recover joint torques. Forward dynamics is computed without retraining by embedding the same learned energy landscape into an optimization objective, where the unregularized minimum satisfies the Gibbs–Appell equation. The resulting accelerations are then integrated forward in time. ANNet is evaluated using a double pendulum paradigm. For trajectories not encountered during training, both inverse and forward simulations demonstrate accuracy and real-time capability. Both computations propagate input noise similarly to their analytical counterparts, and adaptation of the inverse dynamics to a modified pendulum generalizes to forward simulation without additional forward-specific training. These results establish a first-principles approach for employing a single learned representation to support both inverse and forward dynamic computations.