Academic mobility is a valuable and indispensable mechanism in each country's higher education quality assurance system. It also contributes to building and improving the capacities of individual universities because it enables the introduction of positive and proven teaching practices, modern teaching and research methods, and operational and administrative processes. The added value of mobility is the establishment of personal contacts and professional networking of teaching, scientific, artistic and non-teaching staff of the University of Sarajevo (from now on: "UNSA"), which has a long-term effect on the development of teaching, scientific and technological capacity. Studying through mobility gets an international note - students understand their work in a global context, their CV is enriched with new experiences, their knowledge of a foreign language (s) is improved, and after graduation, they become a competitive workforce. Teaching / scientific, artistic and non-teaching staff gain international experience through mobility, improve themselves, participate in active professional development, and increase the quality of the working environment. In the last 15 years, the University of Sarajevo has had over 2,500 outgoing mobility while hosting more than 1,500 international students, teaching / scientific, artistic and non-teaching staff. The mobility of students and staff significantly improves both the quality and the professional standard of UNSA activities. It indirectly contributes to the development and transformation of society as a whole. Through the integration of UNSA, many processes are uniform and simplified, but still, the heterogeneity and specificity of each organizational unit are present. In addition to all the benefits of student mobility, the most significant administrative challenge is the equivalence and recognition of ECTS credits earned at foreign institutions upon return to home institutions. Student services are faced with many formalities around keeping registers and other administrative forms that need to be tailored to exchange students. The current legal regulations in higher education do not value participation in mobility programs, which is undoubtedly not motivating for students and staff to get involved in them. On the contrary, the current vague legal framework and administrative barriers are significantly demotivating for students as they lose a semester or school year due to participation in mobility programs. It is necessary to develop more adequate regulations that will encourage more frequent, productive, more straightforward and better implementation of academic mobility for students. This could be achieved through participation in programs intended for international students (one-year specialist study, master's degree, dual education, summer schools, etc.). It is necessary to develop mechanisms for adequate evaluation of mobility for teaching, artistic and scientific staff of UNSA, which is particularly important for advancement to higher academic titles. It is also crucial to promote the exchange of administrative staff, which is often unjustifiably neglected, to improve the administrative processes at UNSA and its (sub) organizational units. To achieve this, we should be focused on developing cooperation with potential partners interested in administrative / non-teaching staff mobility.

The paper extends the concept of universal motion controller (UMC), by introducing an adaptive mechanism in the original form. The adaptive universal motion controller (AUMC) allows superior position tracking in free motion by allowing better utilization of available control resources. AUMC, as well as UMC, allows concurrent position and force control with a single control structure. Thus, it can be used for trajectory tracking in free motion and also for the interaction force control. This control strategy is of essential importance for the growing field of human-robot interaction (HRI) applications.

Occupancy detection is one of the key elements in improving the energy performance of buildings. Due to their nature, occupancy detection models could be trained on old building data and adapted to new buildings for faster onboarding. We explore and analyse the transfer learning framework applied to occupancy detection. We use a combination of Long-short Term Memory neural network and convolutional neural network architectures and test the transfer learning framework on three datasets. The results show that the transferred models perform better than non-transferred models in almost all metric and dataset combinations.

This paper presents unified force and position control based on sliding mode control (SMC) for a series elastic actuator (SEA). Compliant motion of robotic systems is crucial when dealing with unstructured environments as in the case of physical human-robot interaction. Therefore, not only traditional mechanical systems with stiff joints but also mechanically compliant systems such as SEAs have been actively studied. In order to accomplish versatile tasks, the strategy enabling both position control and force control is favorable. In this paper, the controller synthesizing position and force controllers on the basis of SMC for the control problem of SEAs is proposed by extending our previous work. Simulation results demonstrate the feasibility of the proposed method.

Owing to the increasing engagement of service robots in everyday life, significant requirements are imposed on their control systems to ensure safe interaction between robots and humans. The stiffness of the motion executed by the service robots is not high, as with industrial robots, but has to be variable depending on the defined task. Therefore, a service robot needs to have soft actuation, delivering “human-like” motion dependant on the interaction force between the robot and its environment. Such an operation requires switching from the trajectory tracking (position control) mode to the interaction (force control) mode, and vice versa. Conventional control methods, based on hybrid position/force control, or switching between a position and force controller, may fail short in these cases. Thus, we have previously proposed a new control method, denoted as universal motion controller, that merges the position and force control into a single control structure. The control method is elaborated in this article, and its experimental validation is presented for the first time for multi-degree-of-freedom systems.

This paper presents a combination of two methods that can be effectively combined for control of electrical machines. The first method enables real-time identification of electrical and mechanical parameters based on differential geometry and geometric algebra. The second method enables robust control of electrical machines, even when the knowledge about parameters is incomplete. In combination, the two methods open a path for successful control of electrical machines with unknown and/or time varying parameters.

Ambient conditions, especially temperature and humidity, have a huge impact on the performance of an air quality sensor. In this paper, four correction models were built to compensate the impact of ambient conditions. Linear regression and machine learning algorithms were used for building the models. Correction models were trained by using three types of measurement data. Raw measurement data was used in the first case. Secondly, measurement data was corrected and a significant improvement was shown. Lastly, measurements of various ambient conditions were used as well. Using corrected and extended measurement data brought a great improvement in accuracy of the models. A neural network correction model proved to be the most efficient in all cases. Compensating the impact of ambient conditions on the performance of an air quality sensor by using correction models was efficient and this method could be used in the air quality monitoring applications. This is of particular importance for usage of low-cost sensors in the air quality monitoring.

The broader use of devices powered by rechargeable batteries, especially constrained embedded devices, makes the efficient Battery Management System (BMS) increasingly more important. The estimation accuracy of the amount of remaining charge in the battery is critical as it affects the device’s operation and reliability. For that reason, the estimation of state-of-charge (SoC) is considered one of the main functionalities of a BMS. However, SoC estimation remains a complex task that depends on a range of internal and external factors. Most traditional SoC estimation methods are either computationally complex, require special laboratory equipment or additional configuration efforts. In addition, most methods require continuous measurement of battery parameters, which, in turn, renders these methods not applicable to the class of constrained embedded devices. This paper aims to extend the Coulomb counting method to the class of duty-cycled energy-constrained devices by designing an algorithm that combines voltage-based evaluation and pre-recorded task power profiles to estimate the SoC. In addition, a setup for identifying the battery parameters and algorithm validation setup were also developed and described in the paper.

The paper discusses a control strategy that merges position and force control into a single control structure. The structure, denoted as the universal motion controller in our previous work, can be utilized to build a smart actuating system that runs a mechanical system with $n$ degrees of freedom. A smart actuating system has an integrated controller and it can be used in plug-and-play fashion for different trajectory tracking and force control tasks, defined either in configuration space, or in the task space. The only input of the actuating system is the attraction force in configuration space. Based on the attraction force, the smart actuating system is capable of imposing input forces to the mechanical system that will ensure execution of a specified task.