Innovative Mechanical Engineering

A REVIEW OF DEEP LEARNING AND SYNTHETIC DATA GENERATION FOR HUMAN POSE ESTIMATION IN AUTONOMOUS ROBOTIC SYSTEMS: METHODS, CHALLENGES, AND FUTURE DIRECTIONS

Wed, 23 Sep 2026 00:00:00 +0800

Reliable perception of human posture and movement is essential for autonomous robotic systems operating in shared or dynamic environments. Human pose estimation provides this capability by detecting body keypoints and reconstructing human motion from visual data, but its practical use in robotics is still affected by challenges such as occlusion, viewpoint changes, limited real-time performance, and insufficient annotated training data. This paper analyzes recent developments in 2D and 3D human pose estimation from the perspective of autonomous robotic perception. The review covers fundamental body representation models, single-person and multi-person estimation strategies, and the main differences between skeleton-based and model-based 3D approaches. Special attention is given to deep learning architectures, including convolutional neural networks, vision transformers, and emerging state-space models, with emphasis on their accuracy, computational complexity, and suitability for real-time robotic applications. In addition, the paper discusses synthetic data generation as a practical solution for improving dataset diversity and annotation quality. Rendering-based simulation, motion-capture data, generative methods, and fusion-based techniques are discussed as tools for producing controllable training samples. A small proof-of-concept example based on a game-based 3D environment is also included to demonstrate how camera and character positions can be extracted and used to automatically calculate camera-to-human distance as a spatial annotation. The main limitation of this approach is the synthetic-to-real domain gap, which can reduce model reliability when synthetic datasets are applied to real-world robotic scenarios. Future research should therefore focus on robot-oriented synthetic datasets, improved synthetic-to-real transfer, generative artificial intelligence, and multimodal sensing to support more robust human perception in autonomous robotic systems.

TRAFFIC AND THE CITY AIR QUALITY

Predrag Živković, Mladen Tomić — Wed, 23 Sep 2026 00:00:00 +0800

The air quality is very important for the quality of living in modern cities. It is known that the traffic is the single largest pollutant in Europe. Therefore, proper treatment of this component of pollution is very important for precise estimation of pollution levels. In this paper, specific approach for measurement of roadside CO2 concentration, as the tracer gas, is analysed. TESTO 454 gas analyzer has been used for CO2 concentration measurements. The wind speed and direction have been measured on several locations in the City of Niš, considering that pollution distribution is globally determined by wind characteristics, and for wind measurements cup anemometer was used. For wind data collecting, specific measuring system was developed, a well as the software needed. Traffic intensity has been measured on six major crossroads in the City. The data obtained were used to estimate the traffic generated CO2 emission on the major streets in Niš. Given methodology shows possibilities for further implementation of other pollutant components measurement, which allows tracking of the main pollution pathways.

MAGNETOHYDRODYNAMIC CONVECTIVE DARCY–FORCHHEIMER FLOW OF A PENTA-HYBRID NANOFLUID IN A VERTICAL CHANNEL IN THE PRESENCE OF THERMAL RADIATION

Wed, 23 Sep 2026 00:00:00 +0800

Nanofluids, i.e., commercial fluids in which nanoparticles are dispersed, have attracted the attention of many researchers for use in efficient thermal management systems. This study considers fluid in which five different types of nanoparticles are dispersed, known as a penta-hybrid nanofluid. Darcy–Forchheimer flow in a vertical channel with walls at different constant temperatures is investigated. A homogeneous magnetic field acts perpendicular to the channel walls and thermal radiation is also present. The differential equations describing this flow were solved using the homotopy perturbation method. Relations for the dimensionless fluid velocity, dimensionless temperature, shear stresses and Nusselt numbers at the channel walls were determined as functions of the characteristic parameters introduced for this problem, as well as the physical properties of the base fluid and nanoparticles. For specific parameter values, the dimensionless velocity and temperature distributions are presented in the form of graphs, while the shear stresses and Nusselt numbers are given in a table. The results obtained were discussed and corresponding conclusions were drawn.

DATA-DRIVEN MODELLING OF BUILDING THERMAL DYNAMICS IN A DISTRICT HEATING SYSTEM USING ARX AND RC MODELS

Milica Tasic, Ivan Ćirić, Emina Petrović, Miloš Simonović, Marko Ignjatović — Wed, 23 Sep 2026 00:00:00 +0800

This paper presents a comparative data-driven modelling framework for identifying building thermal dynamics using Autoregressive with exogenous inputs (ARX) and Resistance-Capacitance (RC) models. The proposed methodology is developed using measured operational data from a district heating substation and automatically performs data preprocessing, identification of indoor temperature dynamics, parameter estimation, model validation, and performance evaluation.
Two ARX structures, including single-lag and multi-lag formulations, are identified and compared with a physics-inspired RC model incorporating an equivalent thermal mass state obtained through a first-order low-pass filter. The RC model parameter governing thermal inertia is optimised by minimising the prediction error on an independent test dataset, enabling automatic estimation of the equivalent thermal time constant. In addition to conventional statistical indicators, including RMSE, MAE, and the coefficient of determination (R²), residual uncertainty, thermal reserve, and comfort reliability maps are introduced to provide further insight into model behaviour and predictive capability.
The proposed methodology was validated using a dataset of 12,898 SCADA samples collected at 15-minute intervals. The multi-lag ARX model achieved the highest prediction accuracy, with RMSE = 0.0271 °C and R² = 0.9981. The optimised RC model achieved RMSE = 0.0407 °C and R² = 0.9958 with the optimal value of α = 0.08, which corresponds to an equivalent thermal mass time constant of approximately 3 h. The results show that the ARX model provides higher prediction accuracy, while the RC model offers better physical interpretation of building thermal dynamics. The proposed framework provides a basis for predictive control of buildings.

APPLICATIONS OF TENSOR CALCULUS IN POINT MECHANICS, ELASTICITY THEORY, AND FRACTURE MECHANICS

Katica (Stevanovic) Hedrih — Wed, 23 Sep 2026 00:00:00 +0800

This paper presents fundamental applications of tensor calculus in point mechanics, elasticity theory, and fracture mechanics. The inertial force is formulated in vector form, and its contravariant and physical coordinates are derived in generalized curvilinear coordinate systems. The stress tensor is presented in vector and tensor forms, together with its physical, covariant, and contravariant components. The Lagrange strain tensor is also discussed in curvilinear orthogonal coordinate systems.
Special attention is devoted to the kinematics of deformable bodies. The vector of the total reduced transformation of an oriented infinitesimal line element is defined, and the corresponding vector of proper rotation is introduced together with its physical, covariant, and contravariant tensor coordinates.
The paper further outlines several basic concepts of fracture mechanics. The potential energy density, the Eshelby energy tensor, and the J and M integrals are presented in tensor form. Their physical interpretation is discussed, emphasizing the significance of the J integral as a generalized force associated with crack propagation.
Key words: Tensor calculus, Covariant coordinates, Contravariant coordinates, Material point, Deformable body, Stress tensor, Strain tensor, Rotation angle, Fracture mechanics, J integral.