SU991482A1 - Automated control system operator simulator - Google Patents

Description

(54) SIMULATOR OF OPERATORS OF AUTOMATED CONTROL SYSTEMS

The invention relates to the technical means of training and is intended for the day of training and monitoring the degree of preparedness of operators of control systems used in automatic data processing systems and automated control systems, data preparation devices for computers, object tracking systems, etc., and also be used in experimental engineering and psychological studies of the activities of control system operators.

It is known that, in KOTODOM, the response input unit is associated with a response comparison unit and a memory unit connected to a playback unit that is connected to the response comparison unit, the registration unit, the stage counter, the address register, the address comparison unit, the error comparison unit in the learning steps and a tracer; connected to a control unit, an address comparison unit, an oxy6ok comparison unit in training steps, a unit, a response comparison associated with a control unit connected to a registration unit and an address register connected to a memory unit and a unit

Address Comparison, Linkage to Blo-7 Memory and Counter | stages connected to the decoder stages connected to the switch and with the block of error counters on the stages of training associated with the switch connected to the control unit, the output of the registration is connected to the block of error counters on the stages of training. The stage counter and the sampler are combined into a block for determining the stages of the educational process.

The tutorial works in a manner.

After the learner enters the answer, the response input block issues a response code to the response comparison block and sends a tape trigger signal to the memory block. The response comparison unit receives signals from the bits of the playback unit, and signals about the reading of the possible response codes and the distinctive feature code are received from the physical screen determination unit. In the initial state, the device responds only to these codes.

If none of the anticipated codes of possible answers are compared with the response code (the student answers in an unforeseen manner), then the code containing the bit Comparison Comparison is sent, and when reading such a code, the response comparison unit outputs a comparison signal to the control unit. Signal determination unit receives signals from the reproducing unit bits occupied by distinctive features, and decisive signals about the presence of a code type in the comparison comparison unit come from the feature detection unit. The comparison block by the signal from the response comparison block makes up for the presence of the comparison of the response code and the code of the possible answer and issues a signal to write an error to the registration block. The registration unit registers and indicates an error, and also issues an error signal to the error counter block according to the training stages. The stage decoder sends a resolution signal to the input of one of the counters that are in the error block in the learning stages and correspond to the specific training modules. The commutator connects one of the counters of the block of error counters by one of the enabling signals from the decoder of the stages to the comparison of errors by stages of training. The training device allows you to implement adaptive training programs in which the selection of the next frame of each training stage is made depending on the number of students made mistakes during this stage, as well as the transition from stage to stage (from simple to complex, depending on the level of mastering the program This device does not allow the student to correct the actions directly in the learning process, which practically reduces the learning process to the control of his skills. In addition, the learning assessment produces In most practical cases, the operator, having made a mistake, corrects it in a timely manner. In this device, when the operator makes a mistake, the learning process is repeated, which increases the time and reduces the effectiveness of the training. A significant drawback of the device is the greater complexity Formation of the training program, as well as the information presented to the learner, it is necessary to record its possible changes, which limits the possibilities in choosing the material dp training and quick program change. It is also known a device that monitors the activity of an operator of control systems and contains a registration unit, a program setting unit connected to an information presentation unit, a time setting unit and a control unit that is connected to the first inputs through the serially connected time setting unit the comparator unit and through the unit for setting standards to the second inputs of the comparator unit, the third inputs of which are connected to the operator response input unit, and the output - to the counter input; measurement error calculator, total evaluation calculator, match block and total operator error fixation block connected through a match block to the outputs of the comparison block and directly to the first input of the total estimate calculator, the second input of which is connected to the time job block, and the output to the registration block, inputs the measurement error calculator is connected to the program task block and the operator response input block, and the outputs are connected to the fourth input of the comparison block. The device works as follows. The instructor selects and assigns a set of signals presented to the operator by inputting the program number and the training program number into the program setting blocks and standards and entering the signal exposure time into the time setting block. Information is displayed on the information display panel. The results of the operator's response actions, realized with the help of the operator response input block, are fed to the input of the comparator to control the correctness of the sequence of actions by comparing with the reference sequence and to the input of the measurement error calculator, where the amount of error made by students when changing the parameters presented he is signaling. The true values of the measured parameters are fed to the second input of the measurement error calculator from the output of the program task block. If the next action corresponds to the required one, then one of the outputs of the comparison unit will receive a signal arriving at the coincidence unit and used, in addition, to stop the counter, which is triggered by the Start Exposure signal coming from the time setting unit. This makes it possible to determine the reaction time of the operator, which is then compared with the value of the standard in the comparison glare. In the same block, the current measurement error generated by the calculator is compared with the allowable value that comes from the standard reference block.

The comparison unit generates all three output signals in response to the operator's action only if the result of this action satisfies three criteria: measurement accuracy, execution time, and compliance of the operator actions with the reference sequence. If these three signals are present, the coincidence unit generates a generic signal that retains the same state of the fixation unit.

If a problem was solved by the operator, he had an error in performing at least one action on one of the criteria, the entire task is considered to be solved incorrectly.

The ratio of correctly solved problems to the total number of problems is taken as an estimate of the activity of the training operator and is given to the registration block.

The device allows you to control the operation of the operator’s training and makes a generalized assessment, according to which you can judge the level of his preparation for process control, a means of detection and tracking, or some other control object 2.

The disadvantage of such a device is the lack of technical means in it, which allows to properly control the work of the operators, since the analysis of the causes of errors is not performed.

In addition, such a device does not provide for an automatic adaptive change of the learning material depending on the degree of its assimilation, which makes it difficult to use the device in the self-learning mode.

The closest to the invention is a device for evaluating the professional suitability of the operator of the control system and comprising a sequentially connected teacher, a program control unit, a real process modeling unit and an operator panel connected to a control driver that is connected to the real process modeling unit and the unit evaluation, coupled with the program control unit and the teacher’s control unit, the control summation unit connected to the control unit program control loku and teacher, additional error fixing unit, connected

with the evaluation unit and with the driver of the control signals, and the block of video control, connected to the driver of the control signals.

The device works in the following way.

The head of the occupation with the absorption of the console instructor dials and introduces a task consisting of individual questions. The number of questions

About the recruited manager,

counted by a question counter. check sum block.

The software control unit, in accordance with the first question, is acting on the modeling unit.

real processes, changing the information of the state of the operator's console, and the evaluation unit, into which the correctness assessment program is introduced

0 response actions of the trained operator, manipulating the governing bodies, laziness. At the same time, the signals from the operator control panels through the driver of control signals

5 are fed to the second input of the block. Modeling real processes, changing the information state of the console of the operator, taking into account the response of the operator / to the input of the block

0 estimates, to the input of the visual control unit, the indicators of the position of the control bodies of which provide the possibility of subjective control of the operator’s actions, and to the inputs of the fixing unit of additional errors. When the operator selects the correct answer, the output signal of the evaluation unit Correctly, through the control panel, the teacher acts on the program control unit, which introduces the next question with a certain time lag. The time interval from the recruitment by the operator to the correct answer to the previous question before entering the next

Question 5 also depends on the requirements of a particular learning situation and the nature of the learning task itself.

At the same time, the output signal evaluation unit acts on the second input of the block for fixing additional errors. This ensures the control of any additional (erroneous) operator's action, after dialing the correct answer, fixing the assessment is correct. Correctly entering the next question. If, after dialing the correct answer, the operator additionally manipulates any control body, then in this case the control signal generating unit generates a pulse arriving at the input of the block of fixations of additional errors. The output signal from this block affects the third input of the evaluation unit whose output signal in this case changes to the opposite (Wrong) and, through the first input of the teacher, influences the program control unit, prevents the next question from being entered. In this case, no control impulse arrives from the output of the program control unit to the control summation unit. If, in the time interval from dialing the correct answer to entering the next question, the operator does not perform additional (erroneous) actions, then after this interval the program control unit enters the next question, and from its output to the input of the office summation unit, the impulse indicates control interval These pilot pulses are counted by the pilot pulse counter of the checksum block. When the number of control pulses becomes equal to the number of questions initially recruited by the supervisor, the output of the control summation unit is a signal. The problem is solved, which is transmitted to the second input of the control panel. The device allows for adaptive learning (entering the next question, depending on the previous answer) of the control system operators. The manager in charge of the required learning situation has the opportunity to change the number of questions related to the problem to be solved by the operator. Visual inspection allows you to evaluate the action of the 3j operator. The known device has the following disadvantages. Firstly, in the event that an operative makes a gross mistake in a short time that does not lead to the development of false skills, the program stops and repeats it, which significantly increases the learning time. In addition, the device has limited accuracy of control of skills, since the evaluation does not take into account the time during which the error has occurred. Secondly, in the initial stages of training, the teacher usually shows the sequence of actions to the operators himself. However, due to psycho-physiological and other factors, he may be mistaken, repeating this or that operation, which also leads to an increase in training time. Thirdly, due to its inertia, the teacher is not able to perform real-time analysis of operator errors. This causes him to stop frequently and repeat the learning process again. The aim of the invention is to expand the didactic capabilities of the device while increasing the efficiency of the learning process and controlling the skills of the operators and expanding the field of application of the device. This goal is achieved by the fact that a simulator containing a series-connected adder, a teacher's console, a program block for simulating real Processes, an operator console, a first pulse shaper, an additional error fixing block, and a second pulse shaper, the first and second outputs of which are connected respectively to the second the teacher’s console inputs and the block for fixing additional errors, the visual control unit, the first input of which is connected to the output of the first imp driver pulses, the first input of the adder is connected to the first output of the teacher’s console, and the second input is connected to the output of the program control unit, a sequential-connected functional converter and a switch are introduced, the second input: which is connected to the third output of the second pulse shaper, the third input is the second output of the teaching console, the fourth input - with the output of the first pulse driver, and the output with the second inputs of the simulation module of real processes and the second pulse driver, the third input of which is connected to the output of the function generator connected to the block vtorym.vhodom visual inspection, functional input transducer connected to the output unit simulation of actual processes. t The functional transducer serves to normalize the information and performs analysis and processing of the information presented to the operator, i.e. performs its normalization in accordance with a predetermined algorithm of the process of training or monitoring the actions of the operator of the control system. As is well known, the training process of the operator of the control system consists in that he is in some optimal (or close to him) way, processing the information presented to him and correctly performing the sequence of actions leading to its change. Those. in the process of learning, the operator must learn to bring the presented information to a given form. For example, when tracking objects, the operator of the tracking system affects the controls so that a special mark matches the object. The operator of the process control system, successively switching on and off certain devices and devices, ultimately results in the incoming information to the required technological process. In this way, . it can be said that, in its immediate work, the operator normalizes the incoming information. . By normalization is meant such processing of incoming information, in which its transformation parameters are determined. For example, when tracking objects, such transformation parameters are defined as an object displacement along the coordinate axes, a change in its scale, an angle of rotation, and the like. In process control systems, for example, transformations of information such as changes in color, temperature, pressure, chemical composition of substances, etc. are determined. Consequently, if we compare the actions of the operator of the control system with the actions of a special information normalization block, we can significantly improve the training efficiency — and the accuracy of control of the operator’s actions, since the functional transducer — a block of information margins — is in fact an ideal operator. Moreover, the introduction of this transducer will significantly reduce costs in the preparation of a program for presenting information, since, for example, in tracking systems together with information about the image of the volume is the background and the neo go to have information on its conversion parameters (coordinates of the object, its scale, rotation angle, etc.) that are required to evaluate the effect of the operator. This is a very time consuming task. In the proposed device, information conversion parameters may not have, since they are automatically determined by the information normalization unit. In addition, the proposed device has another important advantage; The operator panel and the information normalization unit can be connected optically (visually) to the simulation unit of the process, which is sometimes just a necessity. In this case, in order to preserve the working capacity of the training devices at the input of the normalization unit, install an optical-to-electrical signal converter in order to keep it. The switch is designed to distribute the control signals of the former and the information normalization unit (functional transducer). The introduction of the switch has improved the effectiveness of training by correcting the operator’s actions (prompts if he made a rough mistake, which is not a violation of, for example, technical process or the process of tracking an object, the development of false skills. At the same time, the process of learning continues and there is no inconvenience to return to the beginning of the learning phase 1. In addition, The master has the opportunity to carry out a visual control of the actions of the opeato and Fixi, to report any operator's errors / do not stop the process of learning. In a known device, because of the speed of the processes of learning, the teacher is unable to evaluate the operation of the operator objectively and continue the training. As a result This is rated incorrectly, and the learning process at this stage is repeated, which increases the learning time. The training process for instructors in control systems usually begins with the teacher he shows a particular sequence of actions the operator at this stage of training. In the strength of random factors (his psycho-physiological state, etc.), the teacher may even be mistaken in the sequence of actions. In the proposed device, this does not occur, as the teacher performs the role of a functional converter that performs this task ideally. This is also an important feature of the proposed device. FIG. 1 shows the proposed simulator / in FIG. 2 is a converter for normalizing the images in FIG. 3 shows a switch, in FIG. 4 pulse shaper. Schematically shown in FIG. 1, the proposed simulator contains a sequentially connected teacher console 1, a program control unit 2, a real process simulation unit 3, an operator console 4, a pulse generator 5 (control signals) connected to the visual control unit b and an additional error fixing unit 7, the output of which is connected to shaper 8 pulses (. Evaluation of operator actions) connected to the console 1 teacher, and console 1 and

block 2 is connected to an adder 9 serving as a check sum. The block 3 of modeling real processes is connected to the functional converter 10, which is connected to the visual control block b, the switch 11 and the driver 8 connected to the switch 11, the driver 5 is connected through the switch 11 to the block 3 modeling real processes and the driver 8, and the console 1 teacher connected to the switch 11.

As a converter 10, various information processing devices can be used to normalize information depending on the required learning tasks. For example, in the case of training operators in tracking objects that undergo metric transformations (offset by coordinate axes, zoom and rotation), devices for image normalization can be used, containing (Fig. 2) fast Fourier transform unit 12 of the current video signal, unit 13 for determining reference coordinate values, block 14 for recognition, inclusions; node 15 for calculating the function of the similarity of the reference and current image and node, the template storage 16, block 17 for fast Fourier transform lonnogo izobrazheyi unit 18 converting a reference, the control unit 19, the normalization process. The switch 11 can be made according to the scheme shown in FIG. 3 containing keys of the opposite keys 20, 21 and 22, 23 in X and Y and the element OR 24.

It should be emphasized that, as an example, a rather complex device for normalizing images is indicated, showing that almost all the parameters of the image transformation, the displacement along the coordinate axes, the change in the oil ratio, rotation) can be found. In case the object undergoes, for example, only displacements of the m-along the coordinate axes, the device for normalizing the images can be made in a substantially simplified form.

The process of learning and controlling the skills of the operators of the tracking systems in this case is carried out in the following way.

Supervisor with help. The teacher’s console recruits a task consisting of individual questions. The whole process of learning is carried out from simple to complex and is divided into 4 stages: 1 - learning to aim at a stationary object; 2 - learning to follow moving with different

speeds and various tractor objects, 3 - learning to track a moving object, the scale of which varies according to different laws} 4 - learning to follow moving 5 and rotating objects that are subject to zoom.

The number of training cycles at each stage is determined mainly by the complexity and speed of the object's movement. Software control unit 2, in accordance with the first question, affects unit 3 of modeling real processes, for example, a video recorder is used.

Fixed and moving objects are recorded on a magnetic tape in accordance with the learning steps.

0 A television movie recorded on a magnetic tape is played, for example, on the video control device of the console 4 of the operator.

The operator, observed, for example, at the 5th optical sight, and acting on the controls of the console 4, combines a special mark with the center of the object. Signals that characterize the X and Y marQ Ci coordinates from the controls of the operator’s console 4 through the driver and through the switches 20 and 22 of the switch 11 are fed to the input of the pulse former. Evaluations of the operator’s actions, which can be performed, for example, f according to the scheme presented in fig. 4 and containing the node 25 and 26 comparison, comparators 27-30, element OR 31, counter 32 errors, element AND 33, one-shot 34 and 35, element OR 36

0 and element 37. At the same time, the signal characterizing the information of the object to the object is from the video recorder (from block 3, simulation of real processes is fed to the input of converter 10.

5 This signal is subjected to spectral analysis in the Fourier basis in block 12, and the resulting spectrum goes to block 13. In the first stage of normalization, block 19 resolves

One of their reference images, stored in node 16 of block 14, is input to block 17, where it is also subjected to spectral analysis in the Fourier basis. The spectrum of the reference video signal is then passed through block 18 to block 13, where, by calculating the like function, the scale and angle of rotation of the current image are determined. In blocks 12 and 18, the Fourier transform module, which has an invariant transform to the image of the object, is calculated.

Using block 19 in block 18, for each cycle of calculating the function of the similarity, the spectrum of the standard is transformed in accordance with the specified values of the parameters of rotation and scaling. The calculation of the similar function is performed in the following sequence. In the first cycle, it calculates the measure of similarity for the position of the spectrum of the standard c. parameters X O and Y (X is the scale parameter, Y is the angle of rotation of the image of the object). In the second cycle, the image is rotated by. y is the angle y, and the similarity measure is again calculated, the third is by the angle Y Y., and all values of the rotation parameter are sorted out. Then the image scale is changed to D,., And again the measure of similarity of the spectrum of the current and reference images is calculated with the values O O, U, ... The parameters of the current image J are determined by the maximum of similarity calculated in block 13, Gadget. which arrive at block 18. At the same time, block 19 permits the passage of the video signal from node 16 to block 18, where it is converted by the received D. Y. After that, the video signal goes to node 15 of block 14, where it is compared with the current image, which has already been subjected to offset only. Such a procedure can be carried out consistently for a whole series of standards. Thus, in the converter 10, in this case, the transformation parameters of the zalona determine its coordinates X and Y, as well as the zoom parameters jL g | rotation The received signals, characterizing the target coordinates (if the operator has to learn how to de-scale and rotate, then rotation and zoom parameters), are compared with nodes 25 and 26. At the output of nodes 25 and 26, the comparators 27-30 are set to the dead zone, These are the comparators 27 and 30 for a rough mistake, they have a large insensitivity zone. Than comparators 28 and 29, defining a given error. Signals from comparators 28 and 29, which determine the tracking error, through element 31 arrive at counter 32. -Evaluation is correctly formed by counter 32 and element 33 in the event that the operator held the mark on the object for a period of time, and made no more than once the number of errors solved, as well as in case he didn’t allow any additional actions after issuing the Capture command. If the operator has made a gross error, then within a short period of time (during this time, for example, tracking failure in real conditions does not occur, the comparators 27 and 30 are gross errors, the signals from which through element 36 and element 3 come to counter 32 block its operation, and the element OR 24. At the same time, signals from the operator’s controls are corrected With or sent to D with signals from the converter 10. The time that can be performed by the operator is determined by the time No one-shotrs 34 and 35, started from comparators 27 and 30. By pressing,, the Capture button on console 4 (not shown), the operator stops sending control signals and then withstands the required 1 set time. If during this time the operator does not perform additional actions, -To the block 7 for fixing additional errors allows the signal to pass correctly through the AND 33 element, the control panel 1 to the program control block 2, which introduces the next training program. In this case, a video monitoring device can be used as a visual control unit 6. During the initial training, the role of the teacher is performed by the converter 10, which determines the parameters of the transformation (its coordinates, change of scale, angle of rotation) and their normalization. In this case, in the KGM switch 11, under the action of a signal coming to the element OR of console 1, switch: the keys from 20-23 and the signal passes from the converter 10 to the block 3 of simulation of the real Processor are allowed. Thus, the feedback loop is required for the tracking process. Otherwise, the operation of the proposed device is similar to the operation of the known device. The use of the device for image normalization as a converter 10 has significantly reduced the cost of preparing the training program (TV movie), since together with the information about the object, it is no longer necessary to record the parameters of its transformation X and Y coordinates, zoom and rotation . In addition, due to the reduction in training time, the capacity of the proposed device is increased. . The efficiency of training is also enhanced by using a television film with real objects and backgrounds as information presented to an operator. If the proposed training device is used as a training device for operators whose tasks include color recognition (such tasks are often found in medicine, metallurgy, metrology, and other areas of technology, a device for normalization of color can be used as a transformer 10 The proposed simulator works as follows, Ru The master of jobs with the help of the console 1, containing, for example, the control keyboard and the register unit, types and introduces a task that consists of individual questions. The number of questions typed by the manager from the maximum number of questions related to this educational task It is determined depending on the requirements of a particular training situation, for example, the operator’s training stage, the complexity of the task, the psycho-physiological state of the operator and his level of training, as well as the conditions of interaction with other Streamers during conducting complex occupations, etc. The number of questions dialed by the job manager is counted by a question counter (not shown) of the adder 9. The program control unit 2, containing, for example, the control node and the decoders of control commands according to the first question. It acts on the block 3 of modeling real processes, containing, for example, simulators of the operator's visual environment and changing the information state of the console 4. The student, the minipulator with control elements, performs a sequence of actions. In this case, the signals from the controls of the console 4 through. the former 5 is fed to the first input of the switch 11 and the first input of the block, b of the visual control, the position indicators of the controls of which provide the possibility of subjective control of the actions of the operation and the inputs of the block 7 of fixing additional errors. At the same time, the signal characterizing the information state of the operator's console from the block 3 of modeling real processes is fed to the converter 10, which processes it in accordance with a given learning algorithm and determines the parameters of its conversion (change. The converter 10 performs the sequence of actions that need to be performed by the trained operator The output signals from the switch 11 are fed to the second input of the block 3, which changes in this case the information state of the console 4, taking into account the operator’s actions, and the input of the driver 8, where the signals characterizing the operator’s actions are compared with the signals generated by the converter 10. In case the operator makes a mistake, the signal is generated incorrectly on the first output of the driver 8, which is received through the console 1 in software control block 2. Block 2 repeats the question in this case, if in the process of performing the sequence of actions the operator makes a grave error that does not lead to a violation of the tactics or training strategy C, the operator does not produce false. skills); the spruce 8 generates an Error signal, which is sent from its second output to the second input of the switch 11. The operator control signals that arrive at the switch 11 in this case are corrected (over-. change) by the signals of the converter 10, which are fed to the input KOM-I of the mutator 11. At the same time, at the first output of the driver 8, the signal is formed correctly, and the learning process continues. With the help of block b of visual control, the teacher has the ability to visually evaluate the actions of the operator, observe the moments of correction and correctly determine the degree of operator training. In addition, the operator himself sees his mistake and quickly fixes it. In case the operator performs the sequence of actions correctly, the output signal of the correct imaging unit 8 through the console 1 acts on the program control unit 2, which introduces the next question with a certain delay. The length of time from the recruitment by the operator to the correct answer to the question before entering the next question also depends on the requirements of the particular learning situation and the nature of the learning task itself. At the same time, the output signal of the imaging unit 8 acts on the second input of the block 7 for fixing additional errors. This thereby makes it possible to control any additional (operator’s erroneous action. Thus, after performing a given sequence of actions, the assessment is fixed and the following question is entered correctly. If, after performing the correct action, the operator additionally manipulates any control, then the driver 5 generates a signal at the input of the block 7 fixing additional

mistakes. The output signal from this block acts on the third input of the imaging unit 8, the output signal of which in this case changes to a temperamentally wrong and, through the first input of the console 1 on the program control unit 2, prevents the next question from being entered. In this case, from the output of block 2 to the adder 9, the control pulse is not received. If in the interval, the time from dialing the correct answer to entering the next question, the operator does not produce additional (erroneous actions, then after this interval, block 2 enters the next question, and from its output to the input of the adder 9 a control pulse arrives, indicating the end of the interval control. These control pulses are counted by a counter pulse counter (not shown) of the adder 9.

When the number of control pulses becomes equal to the number of questions initially collected by the supervisor, the output of adder 9 turns on a signal. The problem is solved, which is transmitted to the input of console 1. At the initial stages of training control system operators, the teacher usually shows the sequence of necessary actions. However, by virtue of random factors, it may occur.

In the proposed device, the role of the teacher is performed by the converter 10. The training in this case is carried out as follows.

The instructor, acting on the controls of the console 1, delivers. the signal to the input of the switch 11. In this case, the signals produced by the converter 10 are fed through the switch 11 to the input of the block 3 of simulation of real processes, changing. In this case, the information state of the console 4, taking into account the actions of the converter 10. Thus, the teacher at the first stages of training, shows the sequence of necessary actions to a cf with the converter 10.

The proposed device, ocj ecTB, adaptive training and control of the skills of operators of control systems, improves the accuracy of control of training, reduces training time and

possesses and. an extended functional, H {возмож 1 possibilities, which ultimately makes it possible to increase the efficiency of the learning process.

Claims (3)

1. Author's certificate of the USSR 746690; cl. G 09 B 7/02, 1979. 2. The author sytelestvo.SSSR 693426. cl. G 09 B 7/00, 1978. 3. The author's certificate of the USSR 693425. cl. From 09 to 7/00. 1978 (prototype). Faye. 2