Miloš Jovanović
Research Subjects
Miloš Jovanović has a total of 48 co-authors. Here are the top co-authors by number of shared publications:
- Dragan Lazić
Tehnički opitni centar Beograd, Srbija
8 shared publications - Slobodan Subotić
Tehnički opitni centar Beograd, Srbija
8 shared publications - Neda Spasojević
Tehnički opitni centar Beograd, Srbija
7 shared publications - Miša Markuš
Tehnički opitni centar Beograd, Srbija
5 shared publications Aleksandar Atanacković
Tehnički opitni centar Beograd, Srbija
3 shared publicationsBeograd Tehnički opitni centar
3 shared publications
Latest Publications
- Open Access
- 19 July 2026
- Published by MDPI AG in Electronics
- Vol. 15 (14) , 3170
Evaluation of Multi-Model Architecture Against Single-Model PPO in the MiniGrid Environment
Reward sparsity is a considerable problem in reinforcement learning when solving complex problems. In an environment where the reward is given only after successfully solving a problem in multiple steps, the signals for optimization can be too weak, preventing the agent from learning. However, this can be solved by breaking the problem into an array of simpler tasks that the agent can learn to perform. This paper presents a possible solution that uses a mixture-of-experts architecture. Separate models are used for planning, routing expert outputs, and solving singular sub-tasks. The solution was implemented in the MiniGrid environment, which allows for the design of arbitrarily complex problems. The results presented in this paper show that a multi-model agent can solve more complex problems comprising multiple steps than a single reinforcement learning agent.
- 19 July 2026
- Published by MDPI AG in Electronics
- Vol. 15 (14) , 3170
Open AccessEvaluation of Multi-Model Architecture Against Single-Model PPO in the MiniGrid Environment
Reward sparsity is a considerable problem in reinforcement learning when solving complex problems. In an environment where the reward is given only after successfully solving a problem in multiple steps, the signals for optimization can be too weak, preventing the agent from learning. However, this can be solved by breaking the problem into an array of simpler tasks that the agent can learn to perform. This paper presents a possible solution that uses a mixture-of-experts architecture. Separate models are used for planning, routing expert outputs, and solving singular sub-tasks. The solution was implemented in the MiniGrid environment, which allows for the design of arbitrarily complex problems. The results presented in this paper show that a multi-model agent can solve more complex problems comprising multiple steps than a single reinforcement learning agent.
- 16 July 2026
- Published by Akademska Misao
- p. 59-64
Comparative Experimental Analysis of Load and Temperature Effects on Precision Resistors under Different Thermal Stabilization Conditions
This paper presents an experimental investigation of self-heating effects in precision resistors under different thermal environments, with a focus on 1 Ω and 10 Ω standards. The study examines resistor behavior under load levels ranging from 2% to 80% of rated power at a controlled ambient temperature of 23 °C. Measurements were performed in both a temperature-controlled oil bath and an air chamber in order to evaluate the influence of heat dissipation conditions. In addition to steady-state analysis, particular attention is given to how resistance and temperature change over time during extended exposure to constant load. Simultaneous monitoring of resistance and internal temperature enabled a detailed assessment of temperature rise (ΔT), resistance variation (ΔR), thermal resistance (RT) and thermal stabilization processes. The results reveal distinct differences between the two environments and highlight the importance of time-dependent effects in understanding self-heating and resistance stability in precision resistors.
- 16 July 2026
- Published by Akademska Misao
- p. 59-64
Comparative Experimental Analysis of Load and Temperature Effects on Precision Resistors under Different Thermal Stabilization Conditions
This paper presents an experimental investigation of self-heating effects in precision resistors under different thermal environments, with a focus on 1 Ω and 10 Ω standards. The study examines resistor behavior under load levels ranging from 2% to 80% of rated power at a controlled ambient temperature of 23 °C. Measurements were performed in both a temperature-controlled oil bath and an air chamber in order to evaluate the influence of heat dissipation conditions. In addition to steady-state analysis, particular attention is given to how resistance and temperature change over time during extended exposure to constant load. Simultaneous monitoring of resistance and internal temperature enabled a detailed assessment of temperature rise (ΔT), resistance variation (ΔR), thermal resistance (RT) and thermal stabilization processes. The results reveal distinct differences between the two environments and highlight the importance of time-dependent effects in understanding self-heating and resistance stability in precision resistors.
- Open Access
- 1 May 2026
- Published by MDPI AG in Electronics
- Vol. 15 (9) , 1921
Benefits of Using LLM for Long-Term Planning with Distilled Subtask Model Compared to End-to-End Reinforcement Learning in the MiniGrid Simulator
Policy learning under delayed reward conditions remains a significant challenge for end-to-end reinforcement learning (RL) agents. The difficulty increases for problems that require long-term planning and the execution of multiple dependent subtasks. As a result, solutions based on a single monolithic policy often suffer from unstable training. One possible solution to this problem could be to delegate the long-term planning to a separate model. This paper presents an implementation comprising two models: a large language model (LLM) for long-term planning and an execution model that solves subtasks. The execution model was trained via distillation from multiple teacher models trained with RL on individual tasks. The results presented in this paper demonstrate the benefits of this approach. By delegating long-term planning to the LLM, the agent can solve more complex problems than end-to-end agents trained with the proximal policy optimization (PPO) algorithm.
- 1 May 2026
- Published by MDPI AG in Electronics
- Vol. 15 (9) , 1921
Open AccessBenefits of Using LLM for Long-Term Planning with Distilled Subtask Model Compared to End-to-End Reinforcement Learning in the MiniGrid Simulator
Policy learning under delayed reward conditions remains a significant challenge for end-to-end reinforcement learning (RL) agents. The difficulty increases for problems that require long-term planning and the execution of multiple dependent subtasks. As a result, solutions based on a single monolithic policy often suffer from unstable training. One possible solution to this problem could be to delegate the long-term planning to a separate model. This paper presents an implementation comprising two models: a large language model (LLM) for long-term planning and an execution model that solves subtasks. The execution model was trained via distillation from multiple teacher models trained with RL on individual tasks. The results presented in this paper demonstrate the benefits of this approach. By delegating long-term planning to the LLM, the agent can solve more complex problems than end-to-end agents trained with the proximal policy optimization (PPO) algorithm.
- 1 January 2026
- Published by Akademska Misao
- p. 108-111
Jedna metoda etaloniranja stroboskopa upotrebom fotodetektora sa posebnim osvrtom na merne mogucnosti
U radu je predstavljena jedna metoda etaloniranja stroboskopa upotrebom silicijumskog fotodetektora. Predstavljanji su osnovni fizički i matematički rada storposkopa, njihova primena, osnovne karakteristike i ograničenja. Poseban osvrt u radu, dat je na budžet merne nesigurnosti, glavne elemente budžeta merne nesigurnosti, kao i proračun ukupne merne nesigurnosti tokom etaloniranja.
- 1 January 2026
- Published by Akademska Misao
- p. 108-111
Jedna metoda etaloniranja stroboskopa upotrebom fotodetektora sa posebnim osvrtom na merne mogucnosti
U radu je predstavljena jedna metoda etaloniranja stroboskopa upotrebom silicijumskog fotodetektora. Predstavljanji su osnovni fizički i matematički rada storposkopa, njihova primena, osnovne karakteristike i ograničenja. Poseban osvrt u radu, dat je na budžet merne nesigurnosti, glavne elemente budžeta merne nesigurnosti, kao i proračun ukupne merne nesigurnosti tokom etaloniranja.
- 1 January 2026
- Published by Akademska Misao
- p. 99-103
Uticaj oslabljivača na merenje harmonijskih izobličenja signala
Ovaj rad razmatra uticaj oslabljivača na merenje harmonijskih izobličenja signala, koji se radi prilagođenja nivoa vezuje na ulaz analizatora spektra. Imajući u vidu frekvencijsku zavisnost slabljenja oslabljivača, izražena je sumnja da njegov uticaj neće biti isti na osnovnoj frekvenciji i njenim harmonicima. Kako u literaturi nije razmatran ovaj aspekt, cilj istraživanja je bio da se utvrdi uticaj oslabljivača na merenje harmonijskih izobličenja. Analiza je obuhvatila tri varijante osnovnog metoda baziranog na merenju harmonijskih izobličenja bez primene oslabljivača i sa njim. Istraživanje je sprovedeno u ograničenim uslovima koji se tiču odabira pogodnih oslabljivača, frekvencije nosioca, merne nesigurnosti itd. Eksperimentalni rezultati su pokazali da frekvencijska karakteristika oslabljivača ima uticaj na merenje harmonijskih izobličenja signala, pri čemu su dati uslovi pod kojima se njegov uticaj može zanemariti.
- 1 January 2026
- Published by Akademska Misao
- p. 99-103
Uticaj oslabljivača na merenje harmonijskih izobličenja signala
Ovaj rad razmatra uticaj oslabljivača na merenje harmonijskih izobličenja signala, koji se radi prilagođenja nivoa vezuje na ulaz analizatora spektra. Imajući u vidu frekvencijsku zavisnost slabljenja oslabljivača, izražena je sumnja da njegov uticaj neće biti isti na osnovnoj frekvenciji i njenim harmonicima. Kako u literaturi nije razmatran ovaj aspekt, cilj istraživanja je bio da se utvrdi uticaj oslabljivača na merenje harmonijskih izobličenja. Analiza je obuhvatila tri varijante osnovnog metoda baziranog na merenju harmonijskih izobličenja bez primene oslabljivača i sa njim. Istraživanje je sprovedeno u ograničenim uslovima koji se tiču odabira pogodnih oslabljivača, frekvencije nosioca, merne nesigurnosti itd. Eksperimentalni rezultati su pokazali da frekvencijska karakteristika oslabljivača ima uticaj na merenje harmonijskih izobličenja signala, pri čemu su dati uslovi pod kojima se njegov uticaj može zanemariti.
- 1 January 2026
- Published by Akademska Misao
- p. 94-98
Određivanje mase nestandardnih tegova
Tema ovog rada je određivanje mase nestandardnih tegova, odnosno tegova koji nisu urađeni prema Međunarodnoj preporuci OIML. U radu je opisan postupak određivanje mase tegova sa proračunom merne nesigurnosti.
- 1 January 2026
- Published by Akademska Misao
- p. 94-98
Određivanje mase nestandardnih tegova
Tema ovog rada je određivanje mase nestandardnih tegova, odnosno tegova koji nisu urađeni prema Međunarodnoj preporuci OIML. U radu je opisan postupak određivanje mase tegova sa proračunom merne nesigurnosti.
- Open Access
- 6 August 2025
- Published by MDPI AG in Electronics
- Vol. 14 (15) , 3127
Optimization of Delay Time in ZigBee Sensor Networks for Smart Home Systems Using a Smart-Adaptive Communication Distribution Algorithm
As smart homes and Internet of Things (IoT) ecosystems continue to expand, the need for energy-efficient and low-latency communication has become increasingly critical. One of the key challenges in these systems is minimizing delay time while ensuring reliable and efficient communication between devices. This study focuses on optimizing delay time in ZigBee sensor networks used in smart-home systems. A Smart–Adaptive Communication Distribution Algorithm is proposed, which dynamically adjusts the communication between network nodes based on real-time network conditions. Experimental measurements were conducted under various scenarios to evaluate the performance of the proposed algorithm, with a particular focus on reducing delay and enhancing overall network efficiency. The results demonstrate that the proposed algorithm significantly reduces delay times compared to traditional methods, making it a promising solution for delay-sensitive IoT applications. Furthermore, the findings highlight the importance of adaptive communication strategies in improving the performance of ZigBee-based sensor networks.
- 6 August 2025
- Published by MDPI AG in Electronics
- Vol. 14 (15) , 3127
Open AccessOptimization of Delay Time in ZigBee Sensor Networks for Smart Home Systems Using a Smart-Adaptive Communication Distribution Algorithm
As smart homes and Internet of Things (IoT) ecosystems continue to expand, the need for energy-efficient and low-latency communication has become increasingly critical. One of the key challenges in these systems is minimizing delay time while ensuring reliable and efficient communication between devices. This study focuses on optimizing delay time in ZigBee sensor networks used in smart-home systems. A Smart–Adaptive Communication Distribution Algorithm is proposed, which dynamically adjusts the communication between network nodes based on real-time network conditions. Experimental measurements were conducted under various scenarios to evaluate the performance of the proposed algorithm, with a particular focus on reducing delay and enhancing overall network efficiency. The results demonstrate that the proposed algorithm significantly reduces delay times compared to traditional methods, making it a promising solution for delay-sensitive IoT applications. Furthermore, the findings highlight the importance of adaptive communication strategies in improving the performance of ZigBee-based sensor networks.
- 4 August 2025
- Published by Akademska Misao
- p. 125-129
Temperature Effects on Precision Resistors: A Review of Methods for Determining Temperature and Load Coefficients, Their Advantages, Limitations and Results
The electrical resistance of precision resistors is affected by temperature due to two main factors: ambient temperature changes and self-heating from power dissipation. The first factor is quantified by the temperature coefficient of resistance (TCR), which indicates how much the resistance varies with temperature. The second factor, known as the load coefficient or load effect, measures the change in resistance caused by the resistor heating up as it dissipates power due to the current flowing through it. This review examines various methods used to determine temperature and load coefficients, highlighting their advantages, limitations, and the results observed in different studies. Furthermore, the review discusses key findings from past research, comparing the reliability and consistency of different approaches.
- 4 August 2025
- Published by Akademska Misao
- p. 125-129
Temperature Effects on Precision Resistors: A Review of Methods for Determining Temperature and Load Coefficients, Their Advantages, Limitations and Results
The electrical resistance of precision resistors is affected by temperature due to two main factors: ambient temperature changes and self-heating from power dissipation. The first factor is quantified by the temperature coefficient of resistance (TCR), which indicates how much the resistance varies with temperature. The second factor, known as the load coefficient or load effect, measures the change in resistance caused by the resistor heating up as it dissipates power due to the current flowing through it. This review examines various methods used to determine temperature and load coefficients, highlighting their advantages, limitations, and the results observed in different studies. Furthermore, the review discusses key findings from past research, comparing the reliability and consistency of different approaches.
- 1 January 2025
- Published by Akademska Misao
- p. 144-147
Etaloniranje infracrvenih beskontaktnih merila temperature koršćenjem IC kalibratora sa posebnim osvrtom na ukupnu mernu nesigurnost
U radu je predstavljena metoda etaloniranja infracrvenih beskontaktnih merila temperature. Predstavljanji su osnovni fizički i matematički postulati implementirani kod beskontaktnim merila temperature, njihova primena, osnovne karakteristike i ograničenja. Poseban osvrt u radu, dat je na budžet merne nesigurnosti, glavne elemente budžeta merne nesigurnosti, kao i proračun ukupne merne nesigurnosti tokom etaloniranja.
- 1 January 2025
- Published by Akademska Misao
- p. 144-147
Etaloniranje infracrvenih beskontaktnih merila temperature koršćenjem IC kalibratora sa posebnim osvrtom na ukupnu mernu nesigurnost
U radu je predstavljena metoda etaloniranja infracrvenih beskontaktnih merila temperature. Predstavljanji su osnovni fizički i matematički postulati implementirani kod beskontaktnim merila temperature, njihova primena, osnovne karakteristike i ograničenja. Poseban osvrt u radu, dat je na budžet merne nesigurnosti, glavne elemente budžeta merne nesigurnosti, kao i proračun ukupne merne nesigurnosti tokom etaloniranja.
- 1 January 2025
- Published by Akademska Misao
- p. 131-134
Etaloniranje CCLD predpojačavača naelektrisanja
U radu je prikazano etaloniranje CCLD predpojačavača naelektrisanja, tip B&K 2647, tj. predpojačavača sa strujnom pobudom. Metoda se realizuje u Tehničkom opitnom centru primenom etalona za vibracije B&K 3629W.
- 1 January 2025
- Published by Akademska Misao
- p. 131-134
Etaloniranje CCLD predpojačavača naelektrisanja
U radu je prikazano etaloniranje CCLD predpojačavača naelektrisanja, tip B&K 2647, tj. predpojačavača sa strujnom pobudom. Metoda se realizuje u Tehničkom opitnom centru primenom etalona za vibracije B&K 3629W.