Assessment of Operator-Pilot Training in Conflict Situations Dmytro Kucherov 1 [0000-0002-4334-4175], Olha Sushchenko 1 [0000-0002-8837-1521] Andrei Kozub 2 [0000-0002-6506-2742] and Anton Petrov 1 [0000-0003-3731-4276] 1 National Aviation University, Kyiv, Ukraine 2 National Defence University of Ukraine named after Ivan Cherniakhoskyi, Kyiv, Ukraine d_kucherov@ukr.net Abstract. With the progress in manned and unmanned aircraft technology, the preparation of an operator of aerial vehicles becomes an important task, and its main assignments are training for piloting the aircraft to complex fly conditions and readiness to sharp transition from the remote control to handle control. De- spite the intensive use of aircraft, in particular, unmanned aerial vehicles, for solving military tasks, the scope of application of these devices is gradually ex- panding in the interests of solving peacetime tasks. Apart from delivering moni- toring of territories and objects, the delivery of goods, money, the creation of temporary radio networks for receiving and transmitting information and others are covering. To solve these problems, a qualified training of specialists serving this equipment is necessary. Based on the theoretical and practical training of the operator, criteria for evaluating the operator’s activities are justified and put forward. The evaluation model of the quality training process of specialists in the maintenance and operation of aircraft proposed. Examples of applying this model are given. Keywords: Socio-technical system, Operator-pilot, Assessment Criteria, Quali- ty, Maintenance, Operation. 1 Introduction Socio-technical systems are widespread in many areas of modern industry including aviation. Therefore, studying the problem of training the pilot-operator as an element of a socio-technical system is relevant. The development of such systems and the rapid rate of aviation engineering development take the lead over human qualities. This constrains the growth of operator possibilities for interaction with modern air- craft and leads to conflict situations. Nowadays researches in the area of safe opera- tion of aviation engineering require the creation of new approaches, methodologies, and techniques. This is caused by the following reasons [1]: 1) drastic increase of flight speeds and altitudes; extension of meteorological constraints on carrying out flights; 2) increasing the intensity of air traffic; Copyright © 2020 for this paper by its authors. Use permitted under Creative Commons License Attrib- ution 4.0 International (CC BY 4.0) CMiGIN-2019: International Workshop on Conflict Management in Global Information Networks. 3) complication of the construction as a result of increasing dimensions and au- tomation level; 4) arising additional factors connected with the operation of aging engineering; 5) features of operating conditions; 6) necessity of training and retraining of operator-pilots; 7) computerization of technological processes related to execution and provision of flights; 8) application of “glass cockpits” or cockpits of the high information technolo- gies; the possibility of different types of terrorism. The contemporary approach to providing safe flight is as follows. It is necessary to use the functioning of the human-machine system (“aircraft-operator-pilot”) as a whole. Such an approach provides research on features and undesirable aspects of interacting human-machine system components. The control of the unmanned vehicle can be carried out in manual, automatic and semi-automatic modes, with the human operator playing a significant role in the prep- aration and conduct of the flight. Automation of flight does not at all exclude a person from flight control but changes the direction of his activity. As stated above, aircraft and flight simulators together with an operator-pilot repre- sent closed-loop systems, which belong to the class of human-machine systems. Such a system represents a set of engineering devices and operators, which provides the functioning of these components. There are two types of human-machine systems. Structure schemes of these systems are represented in Figures 1 [2]. Operator Indicators Control units Aircraft Actuators Computer Fig. 1. Structure of the human-machine system of the first type. In the first case, an operator-pilot closes a channel of forming controls in the sys- tem. Otherwise, an operator is a unit of the control system. In the second case, opera- tor functions are checking the operation of the system, and the prevention of acci- dents. Modern human-machine systems include computers, which carry out aircraft con- trol by the optimal program, prevent accidents, and detect failures with localization of fault place. In other words, they free an operator from many functions. The operator comes into operations in the case of computer failure. Such systems are called auto- mated control systems. So, it is possible to make the following conclusions. On the one hand, the computer carries out complex functions instead of an operator that widens the functional possi- bilities of the system. On the other hand, the increase of computer functional possi- bilities leads to the necessity of their integration, so the relative role of an operator increases. Operator Indicators Control units CS Aircraft Atuators Computer Fig. 2. Structure of the human-machine system of the second type (CS is a control system) The flight control system consists of a receiving and transmitting station, a soft- ware interface located, as a rule, on a tablet computer, an operator and an aircraft itself. The control of the military unmanned apparatus usually involves two operators, one of whom controls the flight of the apparatus (air vehicle operator), and the second one for performing a useful task (mission payload operator). In this regard, the man- agement task is not simple but requires special training of the operator, who must cope with the control tasks in the conditions of natural and deliberate interference, loss of the communication line, loss of direct visibility, delay of transmitted infor- mation, managing of several devices and others. High-quality performance of tasks to a destination requires time to prepare the de- vice for departure in manual or semi-automatic modes, at the same time the prepara- tion of the flight in automatic mode requires much more time-consuming. These cir- cumstances impose some specific requirements for the training of operators who can ensure the performance of the task in adverse flight conditions. 2 Papers Review The important function of the operator-pilot is decision-making, especially in difficult situations. The mathematical model of the decision-making process for human- operator as an element of the socio-technical system is represented in [3]. Determinis- tic and stochastic aspects of this problem are discussed in [4]. An overview of human influence in various areas related to the current and future use of unmanned aerial vehicles is presented in [5]. The study notes that the operator must have specific training, which is not necessarily the same as that of the aviators. However, issues related to the skills, knowledge, and abilities of operators require additional research. The complexity of ensuring the possibility of interaction between the operators of different stations is noted in [6], which is advisable to have for organizing interaction when managing a group of UAVs. The need for precise operator actions to solve use- ful problems is emphasized in [7]. For training operators, of course, appropriate simu- lators should be used. In the development of simulators, an important place is given to an adequate mathematical model of UAVs, one of the versions of which is based on the physical laws of the motion proposed in [8]. Researchers [9] proposed a simulator for training UAV operators of a new type, involving the simulation of the actions of an aircraft. The simulator allows you to simulate the passage of signals, testing on these signals of the airplane in real-time, which allows for the initial assessment of operators. In [10], the error distribution curve of the UAV operator was obtained in the form of the Chi-square Pearson distribution based on the histogram of the error probability density from measurements of the operator's activity, which can be used for the initial assessment of operators. The authors of [11] note the need to develop standards for the selection and training of UAV operators. In [12], it is proposed to evaluate the abilities of the UAV operator according to the profile generated by the special software, which is based on the combination of clus- ter analysis and fuzzy logic. The proposed evaluation metrics make it possible to evaluate the effectiveness of interaction in group management. Features of aircraft control, which define approaches in training of operators for such moving vehicles are represented in [13, 14]. These features are presence of air- borne payload and redundant inertial sensors for UAVs. Training of quadrotor opera- tors also required special approach [15, 16]. A preliminary analysis of the work carried out in the development of effective simulators for the training of operators allows us to draw a preliminary conclusion about the need for research to improve the quality of training of UAV operators. These studies should focus on the development of metrics that allow both to select personnel and improve its training, as well as to serve as the basis for the standards being developed. 3 Flight Control Problems The most vulnerable for flight control is the UAV control channel. The problem is the limited bandwidth of the radio channel, since the narrowband path is more susceptible to the effects of noise and interference present in the communication channel; in flight, loss of communication is also possible, in places with significant reflections, false signals and commands may be received or erroneous signals and radio control commands may be received. Also, the low data rate is imposed by the complexity of the radio control commands. In connection with the wide use of GPS navigation, for the interception of control, false signals are used that mimic the correct signal and cause the target to incorrectly determine its position and thus allow it to intercept control (spoofing). In addition to radio control problems, it may be difficult to perform a payload. Considering the UAV that monitors (photographs) the ground situation, it should be noted several problems arising from its work, namely: delays and loss of video infor- mation; overlay frames and their passes; camera shake; poor spatial resolution; delays in updating information. 4 Operator Actions Changing of the human ability to operation can be divided into three-time intervals such as an entry in the work, relative stable ability to operation, and disability to work. The first stage is characterized by comparatively low speed and accuracy of actions. The second stage depends on the higher accuracy of operator actions. And the decrease of ability to operate on the third stage is caused by fatigability. The increase of operator reliability is provided by the proper organization of work and rest, and also special training to the regulation of ability to work depending on conditions of work. Also, the operator controlling the flight of the UAV must know the modes and con- trol parameters of the control channel to ensure a stable flight of the device; be able to make adjustments for flight control; know the conditions for switching autopilot from manual and semi-automatic control to automatic; be able to exit the automatic control with the appropriate authorization. The operator responsible for performing payloads must know the equipment used and its capabilities, be able to control the field of view of the video camera, be able to get useful information from the available material, and also work closely with the flight control operator. 5 Operator Evaluation Criterion The estimated quality indicators of the operator should be attributed: - response time t is the time that passes from the moment of perception of infor- mation to the response to it; otherwise, it is the ability to detect, process and respond to a stimulus - the operator’s accuracy  is the degree of correspondence of the read data from the display screen by an operator to the measurement data; - exposure  is time for the operator have to understand the occurred situation; - training , we understand the frequency of errors in the operator activity; - stress tolerance k is the level of resistance to the negative effects of activity; - preparedness  is a set of proficiency, experience, and conditions of mental and physiological faculties that determine its competence to perform certain activities with a necessary quality; - prediction  is the property to see or feel the events of a still unfulfilled, closer fu- ture event. The considered set of parameters allows us to represent them as a vector with the coordinates  = (t, , , , k, , )Т. Then a generalized indicator of the quality operability is introduced to assess the man-operator, it called performance, which can be determined from the set of pro- posed indicators that written as follow M J   c j j , (1) j 1 where ci is the ith weight coefficient, i is the ith normalized parameter values, i.e.  max   j j  , (2)  max   min where j is a parameter, min  i  max, j = 1,…, M, in our case M = 7. The weighting factors are imposed by the limitation M  ci  1 . (3) j 1 The suitability of the operator is estimated by the ratio J  J min , (4) where Jmin is the value by which the decision is made on the admission of the operator to the intended activity. Now the task of evaluating the quality of training is reduced to determining the value Jmin with unknown weights cj. Due to the fact that the introduced indicators make different contributions to the fi- nal criterion, for example, indicators such as reaction time t, accuracy (error)  and exposure  need to be reduced and others to increase, it makes sense to separate these factors and solve the problem of maximizing the form taking into account the equal contribution of each component  1 n1 1 n2  J min  max   c1i   c1i  , (5) ci  n1 i 1 n2 i 1  where n1 + n2 = M, for the considered parameters   (cv , ck , c , c ) ,   ct , c , c  . The calculation procedure is 1. We determine the values of the vector  accepted as permissible for each pa- rameter,   [min, max]. 2. We calculate the values of the coefficients ci by the method of hierarchy analy- sis, involving the preparation of matrix A of pairwise comparisons of size pp, where p is the number of criteria. Set the diagonal elements of the matrix equal to 1, set the elements aij following the importance of the parameter so that the important parameter takes a number in the range from 1 to 9 so that the most important takes the value 9, and the least important 1. The element aji = 1 / aij. 3. We verify the fulfillment of the consistency condition for the matrix of pairwise comparisons by calculating the consistency coefficient CR p( pmax  p) CR  . (6) 1.98( p  1)( p  2) If CR  0.1, the level of inconsistency is considered acceptable, otherwise matrix A is reviewed. 4. Determine the value Jmin by (5). 5. We measure the parameters of the vector  for a particular human operator, ap- ply (2), (1) and (4). Corollary 1. The quantity J satisfies the condition 0 < J  Jmax. Corollary 2. The value Jmax = 1. 6 Case Study The interaction of operator-pilot and machine can be illustrated by the scheme repre- sented in Fig. 3 [2]. Receiving information about the system state, the operator pro- cesses this information and forms necessary influences on control units. The operator can influence on control units of the system engineering components employing effectors. Basic human effectors are muscles of hands and legs. Control can have implemented also by the voice and changing electric potentials of a separate body point (body currents of control of manipulators for checking the physiological activity of the operator), effects of physiological activity (pulse, breath frequency). Technical Effectors Receptors Operator components Fig. 3. Structural scheme of interaction of the operator with the technical components of the control system The operator can influence on the technical part of the control system in two ways. For the first way, the operator is the main source of energy necessary for control. For the second way, the operator can determine the only time of implementation of the definite control unit. It should be noted that the greatest effect of aircraft control can be achieved in the case if the operator has the so-called “sense of handle” [2]. Then the operator-pilot can correlate efforts applied to the control handle with the executed manoeuvre. The operator-pilot carries out various functions in control systems. These functions are driving the system into action, coordination functioning different components of the control system, and also receiving, storing and converting information. So, developing control systems it is necessary to take into consideration both tech- nical requirements and psychological, physiological, physiological and anthropomet- ric characteristics of the operator-pilot [17]. Indicators are of great importance in changing information between the operator- pilot and engineering facilities. This process includes optical, acoustic, mechanical and other effects. It should be noted that the system of analyzers of the operator has many channels and great possibilities for signal receiving. During using indicators, the sight analyzers of the operator-pilot have the greatest loading. Sound signalling is used with restrictions. Other analyzers are used very rarely. So, to organize the optimal interconnection between the operator-pilot and indica- tors, it is necessary to take into consideration, on the one hand, basic features of hu- man sight system first of all visual acuity, and on the another hand, – indicator param- eters related with human engineering-physiological features including symbol dimen- sions, contrast range, dearness, alphabet type and frequency of information repetition. During mounting indicators and control units it is necessary to carry out some principles such as priority, grouping, interaction. The priority principle lies in the location of the most important indicators and control units in optimal zones of the working place of the operator-pilot. Grouping is a combination of indicators and con- trol units into logical blocks. Carrying out the principle of interaction means the reali- zation of the proper and constant interconnections between every control unit and an appropriate indicator. One of the most complex problems of the human-machine system design is a divi- sion of functions between the operator-pilot and computer. Nowadays the heuristic methods based on the experience of human-machine systems operation. One of the methods is based on the creation of lists of human-machine system functions, where every function is decided on the possibility to be carried out by the operator or com- puter. A decision obtained by this method is conditional and does not guarantee an absence of errors. Another method is based on the analysis of functions of the human- machine system. These functions are supplied with some parameters, for example, frequency, speed, stability, accuracy, meaning estimated by the method of expert assessments. Then, redundancy or lack of the operator or computer loading, are de- termined based on analytical dependencies. But using expert assessments makes this method approximate and subjective. The method based on the principle of the need to influence the operator-pilot in all cases, when the transformation of information is subjected to essential changes, has no such disadvantages. Moreover, the latter meth- od takes into consideration the connections between system functions. This is the advantage of the method in comparison with methods, based on studying a list of system functions only. In contrast to the above-stated methods, the proposed method of operator training is adapted to conflict situations. The initial data taken as the norm are given in Table 1. Table 1. Normal data. Parameter t    k   Value 0.5 0.5 0.5 2 2 2 2 ci 0.1976 0.4905 0.3119 0.3873 0.1397 0.2748 0.1981 Using the method of analytical hierarchy process [18-20], weighting coefficients were found that satisfy the condition for consistency of the initial matrices of pairwise comparisons. Then, the following (5), the value of Imin = 0.3355. Let check the weights for other, obviously worse values of the considered parameters [20-22], for example, for data in Table.2. Table 2. The second set of data. Parameter t    k   Value 0.8 0.8 0.8 1.5 1.5 1.5 1.5 ci 0.1976 0.4905 0.3119 0.3873 0.1397 0.2748 0.1981 We get the value I = 0.111, which does not satisfy the condition (4). Now we will consider the best indicators of Table 3 against the data of Table 1. Table 3. The third set of data. Parameter t    k   Value 0.3 0.3 0.3 2.5 2.5 2.5 2.5 ci 0.1976 0.4905 0.3119 0.3873 0.1397 0.2748 0.1981 We obtain the value I = 0.526, which completely satisfies condition (4). 7 Conclusions Since the scope of application of UAVs has recently been expanding, the problem of training operators to ensure the preparation, operation, and execution of tasks for the purpose is becoming acute. The paper proposes an approach in which the generalized quality indicator is a linear function of partial heterogeneous quality criteria for per- forming typical functions with unknown weight coefficients. The described method is a modification of the method of analytical hierarchy process based on identifying single-type indicators, the efficiency of which is confirmed by numerical examples. The considered approach can be used in the development of simulators for training operators or evaluating gamers. References 1. Transport ergatic system on aviation. http://avia.pro/blog/ergaticheskaya-sistema-na- transporte-aviaciya. 2. Tishchenko, N.M.: Introduction in Design of Control Systems. Energoatomizdat, Moscow (1986). 3. Rizun, N., Shmelova, T.: Strategic Imperatives and Core Competencies in the Era of Ro- botics and Artificial Intelligence. Chapter 9. Decision-Making Models of the Human- Operator as an Element of the Socio-Technical Systems. International Publisher of Pro- gressive Information Science and Technology Research, USA, Pennsylvania. (2016). 4. Shmelova, T., Sikirda, Y., Scarponi, C., Chialastri, A.: Deterministic and Stochastic Mod- els of Decision Making in Air Navigation Socio-Technical System. In: Proceedings of the 14th International Conference on ICT in Education, Research and Industrial Applications. Integration, Harmonization and Knowledge Transfer; Part III: 4th International Workshop on Theory of Reliability and Markov Modelling for Information Technologies (TheRMIT 2018), vol. II, pp. 649-656, Kyiv, Ukraine (2018). http://ceur-ws.org/Vol- 2104/paper_221.pdf 5. McCarley, J.S., Wickens, C.D.: Human Factors Implications of UAVs in the National Air- space. https://www.researchgate.net/publication/228358350_Human_factors_implications_of_U AVs_in_the_national_airspace, last accessed 2019/03/13 6. Taylor, R.M.: Human Automation Integration for Supervisory Control of UAVs. In Virtual Media for Military Applications, 12-1 – 12-10 (2006). 7. Chen, J.Y.C.: Effects of operator spatial ability on UAV-guided ground navigation. In: Proceedings of 2010 5th ACM/IEEE International Conference on Human-Robot Interac- tion (HRI), 2-5 March 2010, Osaka, Japan, 139 – 140 (2010). doi: 10.1109/HRI.2010.5453227 8. Kucherov, D.P., Kozub, A.M.: Model of UAV as agent of multi-agent system. In: Proceed- ings of Dependable Systems, Services, and Technologies (DESSERT-2018), 24-27 May 2018, Kyiv, Ukraine, 358 – 362 (2018). doi: 10.1109/DESSERT.2018.8409156 9. Wu, J., Wang, W., Zhang, J., Wang B.: Research of a kind of new UAV training simulator based on equipment simulation. In: Proceedings of 2011 International Conference on Elec- tronic & Mechanical Engineering and Information Technology, 12-14 August 2011, Har- bin, China, 4812 – 4815 (2011). doi: 10.1109/EMEIT.2011.6024116 10. Kozhokhina, O.V., Gribov, V.M., Blahaia, L.V.: Processing statistical data about UAV operator errors. In: Proceedings of IEEE 3rd International Conference on Actual Problems of Unmanned Aerial Vehicles Developments (APUAVD), 13-15 October 2015, Kyiv, Ukraine, 124 – 127 (2015). doi: 10.1109/APUAVD.2015.7346578 11. McCarley, J.S., Wickens C.D.: Human factors concerns in UAV flight. https://www.researchgate.net/publication/241595724_HUMAN_FACTORS_CONCERNS _ IN_UAV_ FLIGHT, last accessed 2019/03/13. 12. Rodríguez-Fernández, V., Menéndez, H.D., Camacho, D.: Automatic profile generation for UAV operators using a simulation-based training environment. Progress in Artificial Intelligence. 5 (1), 37 – 46 (2016). doi: 10.1007/s13748-015-0072-y 13. Sushchenko, O.A., Tunik, A.A.: Robust stabilization of UAV observation equipment. In Proccedings of IEEE 2nd International Conference on Actual Problems of Unmanned Aer- ial Vehicles Developments (APUAVD), 15-17 October, 2013, Kyiv, Ukraine, 176 – 180 (2013). doi: 10.1109/APUAVD.2013.6705318 14. Sushchenko, O.A., Bezkorovainyi, Y.N., Novytska, N.D., Shafran, K.: Design of robust controller for UAV with redundant inertial sensors. In Proceedings of IEEE 5th Interna- tional Conference on Methods and Systems of Navigation and Motion Control (MSNMC), 16-18 October 2018, Kyiv, Ukraine, 127-131, (2018). doi: 10.1109/NSMNC.2018.8576290 15. Kucherov, D., Sushchenko, O., Rasstrygin, A., Zhdanov, S., Kozub, A.: Synthesis of the switching control law for a quadrotor autopilot, International Journal of Engineering and Technology (UAE), vol. 7, issue 4, pp. 3065-3069 (2018). doi: 10.14419/ijet.v7i4.16368 16. Kucherov, D., Kozub, A., Rasstrygin, A.: Setting the PID controller for controlling quad- rotor flight: a gradient approach. In Proceedings of IEEE 5th International Conference on Methods and Systems of Navigation and Motion Control (MSNMC), 16-18 October 2018, Kyiv, Ukraine, 90-93, (2018). doi: 10.1109/NSMNC.2018.8576294. 17. Avetistova A.A., “Psychological features of players in computer games”, Psychology. Journal of the Higher School of Economics, vol. 8, issue 4, pp.35 – 58. (2011) https://psy- journal.hse.ru/data/2013/10/30/1283367862/Avetisova_8-04pp35-58.pdf 18. Saaty, T.L.: Relative Measurement and Its Generalization in Decision Making Why Pair- wise Comparisons are Central in Mathematics for the Measurement of Intangible Factors the Analytic Hierarchy. Network Process. Rev. R. Acad. Cien. Serie A. Mat., 102 (2), 251 – 318 (2008). https://link.springer.com/article/10.1007/BF03191825 19. Mazin Al Hadidi, Jamil S. Al-Azzeh, R. Odarchenko, S. Gnatyuk, A. Abakumova, Adap- tive Regulation of Radiated Power Radio Transmitting Devices in Modern Cellular Net- work Depending on Climatic Conditions, Contemporary Engineering Sciences, Vol. 9, № 10, рр. 473-485, 2016. 20. Mazin Al Hadidi, J. Samih Al-Azzeh, O. Tkalich, R. Odarchenko, S. Gnatyuk, Yu. Khokhlachova. ZigBee, Bluetooth and Wi-Fi Complex Wireless Networks Perfor- mance Increasing, International Journal on Communications Antenna and Propagation, Vol. 7, № 1, рр. 48-56, 2017. 21. O. Solomentsev, M. Zaliskyi, R. Odarchenko, S. Gnatyuk, Research of energy characteris- tics of QAM modulation techniques for modern broadband radio systems, Proceedings of the 2016 IEEE International Conference on Electronics and Information Technology (EIT), Odesa, Ukraine, May 23-27, 2016, рр. 14-20. 22. R. Odarchenko, S. Gnatyuk, T. Zhmurko, O. Tkalich, Improved Method of Routing in UAV Network, Proceedings of the 2015 IEEE 3rd International Conference on «Actual Problems of Unmanned Aerial Vehicles Developments (APUAVD), Kyiv, Ukraine, Octo- ber 13-15, Vol. 1, 2015, рр. 294-297.