Vladimir Sibinović
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Vladimir Sibinović has a total of 15 co-authors. Here are the top co-authors by number of shared publications:
- Anđela Jovanović
University of Ni,Faculty of Electronic Engineering,Department of Automatic Control,Ni,Serbia
7 shared publications - Vladimir Mitić
University of Nis
7 shared publications - Boban Veselić
University of Ni,Faculty of Electronic Engineering,Department of Automatic Control,Ni,Serbia
5 shared publications - Mirko Raković
University of Novi Sad
3 shared publications - Milutin Nikolić
University of Novi Sad
2 shared publications - Aleksandar Milenković
University of Niš, Faculty of Electronic Engineering, Department of Computer Science, Republic of Serbia
1 shared publication
Latest Publications
- 1 June 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-6
Development of a Mobile Platform for Maze-solving
- 1 June 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-6
Development of a Mobile Platform for Maze-solving
- 13 October 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-4
Comparison of Different TinyML Implementations for an Industrial Control System
Requirements regarding security, trustworthiness and low latency in modern industrial control and robotic systems have led to implemented machine learning solutions considerably shifting from cloud computing to edge computing. In this context, the emerging concepts of Edge Artificial Intelligence and Tiny Machine Learning are becoming essential for implementing machine learning based industrial control solutions using resource-constrained devices. In this paper Tiny Machine Learning is used to solve the industrial control problem of apple classification using two small edge devices. Several versions are compared while presented experimental results and comparisons with our previously published results demonstrate that Tiny Machine Learning can provide a viable solution even for complex industrial grade control problems in the food industry.
- 13 October 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-4
Comparison of Different TinyML Implementations for an Industrial Control System
Requirements regarding security, trustworthiness and low latency in modern industrial control and robotic systems have led to implemented machine learning solutions considerably shifting from cloud computing to edge computing. In this context, the emerging concepts of Edge Artificial Intelligence and Tiny Machine Learning are becoming essential for implementing machine learning based industrial control solutions using resource-constrained devices. In this paper Tiny Machine Learning is used to solve the industrial control problem of apple classification using two small edge devices. Several versions are compared while presented experimental results and comparisons with our previously published results demonstrate that Tiny Machine Learning can provide a viable solution even for complex industrial grade control problems in the food industry.
- 13 October 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-4
Experimental Evaluation of Neural Network Complexity in Constrained Hardware Environments
This paper focuses on the challenge of deploying a neural network to be used for precise localization capabilities on the constrained hardware by exploring a novel strategy, transforming raw magnetic field sensor readings into accurate real-time magnet position estimations within a defined workspace. The collected dataset of magnetic field readings, corresponding to known magnet position, was used to train different configurations of feed forward neural networks. These configurations vary in key architectural parameters, including the number of hidden layers and the number of neurons per layer. The trained networks are evaluated for their total memory footprint and execution time on an 8-bit microcontroller. This methodology, demonstrated through the localization problem, is applicable to other machine learning application, including data classification and can be adapted for different hardware constrained devices.
- 13 October 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-4
Experimental Evaluation of Neural Network Complexity in Constrained Hardware Environments
This paper focuses on the challenge of deploying a neural network to be used for precise localization capabilities on the constrained hardware by exploring a novel strategy, transforming raw magnetic field sensor readings into accurate real-time magnet position estimations within a defined workspace. The collected dataset of magnetic field readings, corresponding to known magnet position, was used to train different configurations of feed forward neural networks. These configurations vary in key architectural parameters, including the number of hidden layers and the number of neurons per layer. The trained networks are evaluated for their total memory footprint and execution time on an 8-bit microcontroller. This methodology, demonstrated through the localization problem, is applicable to other machine learning application, including data classification and can be adapted for different hardware constrained devices.
- 2 September 2025
- Published by Springer Nature
- p. 568-576
Estimating 2D Position from Magnetic Sensor Readings Using Artificial Neural Network
- 2 September 2025
- Published by Springer Nature
- p. 568-576
Estimating 2D Position from Magnetic Sensor Readings Using Artificial Neural Network
- 9 June 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-6
Robust Trajectory Tracking of Differential Drive Platforms Using Discrete-Time Sliding Mode Control
This paper proposes a control structure suitable for trajectory tracking tasks performed by a differential drive mobile platform, operating under external disturbances. To address these challenges, a discrete-time sliding mode controller (DSMC) is implemented within the inner velocity control loops, offering robust behavior. In addition to enhanced robustness, the proposed controller effectively eliminates wheel slippage. The performance of the proposed system is evaluated through simulations and compared to the control system having conventional digital PI controllers in the velocity control subsystems.
- 9 June 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-6
Robust Trajectory Tracking of Differential Drive Platforms Using Discrete-Time Sliding Mode Control
This paper proposes a control structure suitable for trajectory tracking tasks performed by a differential drive mobile platform, operating under external disturbances. To address these challenges, a discrete-time sliding mode controller (DSMC) is implemented within the inner velocity control loops, offering robust behavior. In addition to enhanced robustness, the proposed controller effectively eliminates wheel slippage. The performance of the proposed system is evaluated through simulations and compared to the control system having conventional digital PI controllers in the velocity control subsystems.
- Open Access
- 1 January 2025
- Published by National Library of Serbia in Facta universitatis - series: Electronics and Energetics
- Vol. 38 (3) , 563-580
Emerging concept of personalized medicine vending machine based on the era of COVID-19 pandemic
This paper presents the new concept of Medicine Vending Machine (MVM), called Personalized Medicine Vending Machines (PMVM), motivated by the need to increase social distancing during the COVID-19 pandemic. The conceptual model, the scenarios of using PMVM, and all the necessary preconditions are provided. This paper proposes modifying the existing MVM concept, which can provide better social distancing in times of pandemics and epidemics, emphasizing personalization and picking up prescribed therapy. The medical information system MEDIS.NET has been used as a keystone in the development of PMVM. Pharmacies and patients use prescription data stored in patient electronic health records in MEDIS.NET to provide all conditions needed for the operation of PMVM. Real data from the Pharmacy Institution Nis and Health Center Nis are used to define the scenarios for the use of PMVM. The proposed PMVM would be a new service for smart cities that can contribute to the significant increase of the efficiency of smart cities during the combat against COVID-19 and any similar pandemics in the future.
- 1 January 2025
- Published by National Library of Serbia in Facta universitatis - series: Electronics and Energetics
- Vol. 38 (3) , 563-580
Open AccessEmerging concept of personalized medicine vending machine based on the era of COVID-19 pandemic
This paper presents the new concept of Medicine Vending Machine (MVM), called Personalized Medicine Vending Machines (PMVM), motivated by the need to increase social distancing during the COVID-19 pandemic. The conceptual model, the scenarios of using PMVM, and all the necessary preconditions are provided. This paper proposes modifying the existing MVM concept, which can provide better social distancing in times of pandemics and epidemics, emphasizing personalization and picking up prescribed therapy. The medical information system MEDIS.NET has been used as a keystone in the development of PMVM. Pharmacies and patients use prescription data stored in patient electronic health records in MEDIS.NET to provide all conditions needed for the operation of PMVM. Real data from the Pharmacy Institution Nis and Health Center Nis are used to define the scenarios for the use of PMVM. The proposed PMVM would be a new service for smart cities that can contribute to the significant increase of the efficiency of smart cities during the combat against COVID-19 and any similar pandemics in the future.
- 3 June 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-6
Trajectory Following in Differential Drive Mobile Platform
This paper considers a trajectory tracking problem in motion control of a mobile platform. The platform is designed to have differential drive, Arduino microcontroller, Bluetooth connectivity and is battery powered, thus representing a complete mechatronic system. The task at this stage was to provide the platform motion along simple desired trajectories, such as straight lines and curves. During its motion, the platform draws its trajectory on the surface, upon which motion accuracy can be evaluated. Odometry approach is used in solving this task where each wheel motor is driven within a cascaded control structure with conventional digital controllers. Experimental results of several tests are presented and analyzed.
- 3 June 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-6
Trajectory Following in Differential Drive Mobile Platform
This paper considers a trajectory tracking problem in motion control of a mobile platform. The platform is designed to have differential drive, Arduino microcontroller, Bluetooth connectivity and is battery powered, thus representing a complete mechatronic system. The task at this stage was to provide the platform motion along simple desired trajectories, such as straight lines and curves. During its motion, the platform draws its trajectory on the surface, upon which motion accuracy can be evaluated. Odometry approach is used in solving this task where each wheel motor is driven within a cascaded control structure with conventional digital controllers. Experimental results of several tests are presented and analyzed.
- 3 June 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-6
Neural Network Regression Analysis of Magnetic Sensor Data for Spatial Magnet Positioning
This paper deals with the magnet position estimation based on the measurements from multiple Hall sensors. Estimation of the position is obtained from trained artificial neural networks. Data are gathered using a custom sensor board that has eight three-axis Hall sensors on it, and correlation between magnetic field and position was determined to be able to estimate position in X and Y directions. First measurements were made to determine the optimal height of the magnet for later data gathering. The outcome of the research was evaluated by comparing the ground truth with the estimated values of the best-trained network to determine the error of estimation. Precise measurement of the position of the magnet is intended to be used for the development of a tactile sensor that estimates the contact force through the displacement of the magnet in the deformable layer.
- 3 June 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-6
Neural Network Regression Analysis of Magnetic Sensor Data for Spatial Magnet Positioning
This paper deals with the magnet position estimation based on the measurements from multiple Hall sensors. Estimation of the position is obtained from trained artificial neural networks. Data are gathered using a custom sensor board that has eight three-axis Hall sensors on it, and correlation between magnetic field and position was determined to be able to estimate position in X and Y directions. First measurements were made to determine the optimal height of the magnet for later data gathering. The outcome of the research was evaluated by comparing the ground truth with the estimated values of the best-trained network to determine the error of estimation. Precise measurement of the position of the magnet is intended to be used for the development of a tactile sensor that estimates the contact force through the displacement of the magnet in the deformable layer.
- 3 June 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-5
Modeling, Identification and Compensation of Friction in Solo12 Quadruped Robot
The accuracy of the dynamic model of a robot plays an important role for robots performing highly dynamic tasks. Quadruped robots are often performing highly dynamic tasks and in order to be able to do that, fine and precise control of all joints is necessary. Friction compensation can play an important role in accomplishing this. A friction identification procedure for quadruped legged robots is presented in this paper. The procedure consists of a predefined joint motion that minimizes the joint coupling and collects the necessary data for friction estimation. Using the identified friction terms, a PD torque controller with friction compensation was implemented. The identification procedure and the torque controller are tested on a Solo12 quadruped robot using MuJoCo simulation software and on the real robot.
- 3 June 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-5
Modeling, Identification and Compensation of Friction in Solo12 Quadruped Robot
The accuracy of the dynamic model of a robot plays an important role for robots performing highly dynamic tasks. Quadruped robots are often performing highly dynamic tasks and in order to be able to do that, fine and precise control of all joints is necessary. Friction compensation can play an important role in accomplishing this. A friction identification procedure for quadruped legged robots is presented in this paper. The procedure consists of a predefined joint motion that minimizes the joint coupling and collects the necessary data for friction estimation. Using the identified friction terms, a PD torque controller with friction compensation was implemented. The identification procedure and the torque controller are tested on a Solo12 quadruped robot using MuJoCo simulation software and on the real robot.
- 1 January 2024
- Published by University Library in Kragujevac
- p. 27-30
Modeling, Control and Simulation of Differential Drive Mobile Platform
Differential drive mobile platforms are common in the robotic world, for teaching, research, and tackling different applications. These platforms are very agile and can perform movement, trajectory following, and positioning tasks. To test different control algorithms for a mobile platform, its model must be used to simulate the real-world platform. This allows students and researchers to investigate different control strategies and fine-tune control algorithms for specific tasks. This paper presents a simulation model of the mobile platform, which was developed with the aim of matching the physically realized platform as much as possible. The created model can be intuitively used for control design and simulation.
- 1 January 2024
- Published by University Library in Kragujevac
- p. 27-30
Modeling, Control and Simulation of Differential Drive Mobile Platform
Differential drive mobile platforms are common in the robotic world, for teaching, research, and tackling different applications. These platforms are very agile and can perform movement, trajectory following, and positioning tasks. To test different control algorithms for a mobile platform, its model must be used to simulate the real-world platform. This allows students and researchers to investigate different control strategies and fine-tune control algorithms for specific tasks. This paper presents a simulation model of the mobile platform, which was developed with the aim of matching the physically realized platform as much as possible. The created model can be intuitively used for control design and simulation.