First Digital AIP Romania

by Octavian Thor Pleter

Recently I found and scanned some old documents, dated 1991-1992, which prompted memories of my first professional achievements. When I graduated the Faculty of Aeronautics in 1986, I had already discovered my talent for computers and I had built a passion for software, without abandoning my older passion for hardware, evolved from my hobby for electronics. The majority of personal development experts say that we need to do in life something we like, something we are passionate of. My opinion is different. You need to do something that you are talented at, because you can achieve remarkable results with minimum efforts. If you happen to like something where you need to work hard for every result, you will get tired and you will end up by disliking that job. If you have talent, the results appear effortlessly and you will end up liking what you have probably previously ignored.

The communist ideology was completely unadapted to the new computer era. For instance, when we designed a piece of equipment, we had to use only Romanian integrated circuits and eventually a small and limited percentage of components imported from CAER countries (including microprocessors manufactured in the German Democrat Republic, coppied after Intel 8080). Obviously, the results were disappointing. I was lucky to be selected by my faculty professors Cătălin Bungetzianu and Ilie Stiharu to hire me at the first Romanian ILS project, with a prototype installed at the Bacău Airport. I designed some digital logic and instead of using the Zilog Z-80 technology which I was familiar with from what I was doing at home, I was forced to work with Romanian low integration TTL integrated circuits, copied after Texas Instruments. Instead of one microprocessor, several memories and ports, I had to come with two large PCBs full of integrated circuits. Evidently, the Bacău prototype frequently malfunctioned, being much more complex than needed, and being exposed to extreme temperatures and humidity, which the Romanian circuits were not able to take.  Another inadequacy of the communist regime to computer technology was that censorship was highly effective with printed books and journals, with audio (music) and movies (video cassettes), but totally ignored the software, including the electronic documents. That enabled us to be connected to the rest of the world in this field. We had the newest software on the market and we learned to work with them, but we were totally isolated from Western technical books and journals (some clandestine Xerox copies circulated).

In spite of the KGB’s efforts, in 1989 Romania quit communism and got out of the Soviet dominance area, taking a fast course to the West. We cannot imagine now the huge challenges of those times. In 2007 Niculae Droc published a book evoking the beginnings of the Romanian air navigation services in the 1990s. Cosmonaut Dumitru Dorin Prunariu served as the first director of the post-communist (free) Civil Aviation Administration, and Niculae Droc was the visionary leader who created Romatsa, a modern provider of air navigation services. Also, Niculae Droc was the first general director of the Autonomous Management Romatsa, established in 1991.

In his book [1], Droc writes about the initial priorities and among strategic priorities, he mentions the need for a Romanian AIP at Western standards. Thus, the NOTAM Center was created in ROMATSA, “equipment (computers, printers) were purchased, the necessary software was produced (project manager: C. Rusu) and in less than one year, the organization was able to provide a document which could be compared with the best such publications in the world. The superior quality of the AIP allowed us to raise the selling price, in order that a decent profit was ensured to Romatsa.” [1].

In this paper I would like to present and comment on that particular AIP Romania, because it was a tour de force and a great achievent of the Romanian aviation, which had to wait until 2007 to get this short mention. Cristi Rusu was my faculty colleague and friend. He had graduated the navigation section and managed to be hired at the newly established Romatsa. He talked to me about the new AIP project and asked my opinion on how to replace the “manually executed charts” and the “machine written text” multiplied with the “technology of the 1970s” [1]. I had at home an IBM PC-AT 286 compatible computer, put together by myself from various components found with difficulty. Also I had a technology miracle, the new Hewlett-Packard HP 4L laser printer with 300 dpi resolution. It costed me a fortune, $500, but I paid for it partially working in a computer assembly job for a company which was importing components from Germany. As software, I presented to him WordPerfect, which I was using to write articles and books. For aeronautical charts, I presented CorelDraw 2.0. He let me know that they will open a call for tender at Romatsa for the new AIP in August 1991 and recommended that I participate. I did not own a company, I was leader and founder of a non-profit association instead, the Association for Personal Computers, which gathered many remarkable people passionate of this field. Two of the most active members of my association were brothers Alexandru and Radu Corlan. The former was a medicine student and the latter an electronics student. They were the only whom I trusted and at the same time who owned a registered business, called Induss srl. I convinced them to participate at the tender, promissing that I will teach them the aviation part, a field which they were not familiar with. They were computer geniuses instead, and curious to learn new things. Dan Alexandru Corlan is now a prestigious doctor in medicine, founder of Ad Astra Association, and producer in the 1990s of the first Romanian cardio Holter device and some high impact medical software. Radu Corlan emigrated in the US in 1995, where he co-authored a high speed filming technology, the Phantom video cameras. In this position of inventor I saw him in 2012 on the red carpet at the Oscars (Fig. 1). Even since we were working together, he was interested in high speed data acquisitions.

Fig. 1 – Radu Corlan (second from the right), co-author of high speed filming technology, awarded with Scientific and Engineering Award by the American Film Academy for inovations in the field (2012)

Three companies participated at the call for tender, with an aeronautical chart sample made after a West Germany model supplied by Romatsa (Fig. 2). Induss won, because our chart was produced by a laser printer from Corel at 300 dpi, whereas our competitors placed their bets on plotter and AutoCAD. In those times, the peripherals specialized in graphics were indeed the plotters, and they were able to provide colour prints. We had plotters ourselves, but for aeronautical charts, we considered the plotter weak due to the relatively thick and constant lines. The laser printer could draw very thin lines (1/300″ or under 0.1 mm). There was another fixation that vector graphics required vector peripherals, and the printers were bitmap. However, at high resolutions, with vector graphics, the printed image quality was even better than the plotter. One of the model samples was coloured, and laser printers at the time were black and white only, but we considered the quality more relevant than the colour. The only colour map in the Romanian AIP was the radio navigation chart of Romania on a A2 format. Another difficulty was that some obstacle charts in the AIP were A3 and the only available laser printer at the time was A4 format. We decided to overcome this hurdle by joining together two printed sheets. In 1992 AIP Romania one may notice small discontinuities of the horizontal lines at the junction (Fig. 3). Although the two halves matched perfectly on the screen, the paper was passing through a very hot oven and suffered uneven stretches, and if the lines matched in the upper side, there was a discontinuity in the lower side, and vice versa. To my stupefaction, I found these problems 27 years later in Romania’s AIP at some obstacle charts, including ballpen hadwritten modifications over my old charts. Then I realized that I needed to write about the 1992 digital AIP, with brutal, analog modifications by ballpen, in 2019.

Fig. 2 – Samples of charts provided by Romatsa in 1991 as models for the new AIP, printed with 1200 dpi advanced laser technology, unavailable at the time in Romania

The bid was won by Induss in spite of the powerful lobby of the rivals, who belonged to some Tarom pilots. Not only that our chart sample resembled the models in Fig. 2 the most (except the inferior resolution), and the rival AutoCAD plotted charts looked like cadastre plans, our price mattered. I was eager to get a contract in the field of aviation, I had enough money, and I was not after financial gain. Corlan brothers understood, and our offer was just 776,000 lei (the equivalent of close to $4000 at the moment of signing, and half of the amount at the moment of payment, 6 months later, due to hyperinflation) [2]. It is true that one could have bought a flat in central Bucharest for that money.

On 30^th of October 1991, Romasta signed with Induss probably the most profitable contract in their history, but it is worth mentioning that neither Romatsa was the financial power of later yet. “At the date of incorporating (1^st of February, 1991) ROMATSA had no money in the bank account. (…)  It was borrowing 4 rooms, some desks and chairs from the Civil Aviation Department in their building in Otopeni as headquarters.” [1]. In contrast, we who entered those rooms and sit in those borrowed chairs to apply for tender and negotiate the contract, could consider ourselves rich. Cristi Rusu admitted that our rather modest price was decisive, and that the rivals accused us of dumping. The truth is that I would have worked even for free, I longed to have an aviation project, in my field. I had enough computer projects, but none of navigation. I was working at the faculty a digital avionics project for the IAR-45 remote piloted vehicle for ICA Ghimbav, then at COMOTI I was working a project for the national methane gas distribution grid and a laser anemometry data acquisition project with Sergiu Sidon. Other projects were even more bizarre, for example the first matrimonial agency in Lahovary Square was automated by me in 1990. The agency was very satisfied and, besides the consistent fee, they offered a bonus, to find me a match. Corlan brothers were more pragmatic and tempered my impulse to do voluntary work for Romatsa.

Our contractual obligations were tough:

“1.1 The object and the scope of this contract is the realization by the Provider and acquisition by the Customer, within the conditions herein, of a System of Generating and Management of Aeronautical Charts and the 139 charts included in the AIP Romania 1991 document, with the technical specifications described in Annex 1.” [3]

Fig. 3 – The first digital AIP of Romania published in 1992 [4], [5]; (a) Vol. I with the obstacle charts which exceeded the A4 format of the laser printer; (b) Vol. II with the instrument approach charts; (c) detail with the discontinuities of the obstacle charts by joining two A4 halves

The provider was due “to implement the System of Generating and Management of Aeronautical Charts, with the technical and operational characteristics provided in Annex 1; to generate the 139 charts included in AIP Romania, within the specifications in Annex 1 and in compliance to the applicable norms (ICAO Aeronautial Chart Manual 1987 Doc. 8697); to ensure training of the Customer’s staff to use the system provided; to guarantee the correct operation of all provided products, according to Art. 10; to meet the phase deadlines, as per Art. 6.”

With the enthusiasm typical to our young age, we proposed an automated charts generating system from navigation databases. Thus, if for example a radio frequency which intervened in 20 charts was changed, it was modified in the database and the 20 charts were automatically regenerated. This meant a complete digitalization, with a modern concept at that date.

The deadlines were extremely ambitious. The time intervals were given in days: “50 days since signing the contract for the first phase products (9 December 1991); 100 days since signing the contract for the second phase products (28 January 1992); 150 days since signing the contract for the third phase products (18 March 1992);” [3]. To give us credit, I need to mention that the deadlines were met and that the beneficiary was extremely happy. The contract included a software user licence edited by Corlan brothers, who had knowledge of the legal aspects, too.

I need to explain at this point why I did not choose a team from within the Faculty of Aeronautics, which was my employer at the time. I wanted to, but Professor Ilie Stiharu did not show much interest, being absorbed in a automated pilots software which he developed together with Cristi Constantinescu. He was trying to attract me in that project, but I followed Cristi Rusu in his idea that it is high time for our generation to deliver professionally. In 1991 I was an engineer in the research team of the Department of Aeronautics and I was just beginning to work with the students. I involved them not earlier than the third phase of the contract, as I will mention further. Another friend whom I persuaded to participate was Călin Iosof, computer and automated systems engineer, who lived next to me in Rosetti Square, and a former colleague at the Lazar College. Like Corlan brothers, Călin was a computer genius. He built his own computers and in the 1980s he wrote a LISP interpreter in Z-80 Assembly, 10 times faster than any other software interpreter at the time, as well as an unbreakable copy protection encryption software. Unfortunately, his interpreter remained almost unknown due to the copy protection, since Călin asked for a decent price, but not for the poverty in Romania. By contrast, the software I wrote was circulating everywhere. For instance, my Tuttiprint printer driver was used nationwide. Until 1996, as in the whole communist block, in Romania it was not illegal to copy software. The software was unprotected, nobody paid to me anything, but my name was becoming a sort of brand.

The contract execution went very well, as far as I remember (but my memory is highly subjective). Cristi Rusu got to know WordPerfect well and focused on the text and tabular part. I taught Corlan brothers a crash course into ICAO Doc. 8697, reading it for the first time on that occasion. What I did not understand, I asked Cristi Rusu, and he asked Sorin Onițiu. Sorin was a legend, an aviation engineer graduated 1981 who specialized in aeronautical charts. He did all the AIP charts by hand in ink, designed instrument flight procedures, and did all the aeronautical information job. I was to meet him in person much later, but at the time he was the person with the most authority in this field. As proof, Jeppesen, the most prestigious aeronautical information company in the world, made him an offer. The charts that we produced were in fact made after the prototype drawn by hand by Sorin Onițiu. As a matter of fact, I was self taught in this super-specialized field by studying Sorin’s charts. ICAO Doc. 8697 was written in a language which the insiders call Icaoese, and I barely knew English. On top of that, I need examples to understand anything. In AIP Romania 1992 Vol. II MAP section there are still the old charts drawn by hand for the two airports which were considered predominantly military: Craiova și Bacău [5]. Perhaps for military secrecy reasons, we were not informed about these two airports and we were exempt from providing the charts. Another thing that helped me was that I used to fly on flight simulators and experimented with instrument flight procedures. The instrument procedure aeronautical charts (MAP section) are more special in the sense that they must be readable both by humans in the graphic form, and by flight management computers (FMS) in the digitally coded ARINC-424 form. The FMS guides the auto-pilot to fly the trajectories of these procedures, so the charts are not simple charts, they are flight programs which must be compatible with the types of aircraft which execute those programs.

Radu Corlan took the aerodrome charts and his first work was the Bucharest Otopeni airport (Fig. 4). I took the obstacle charts, which were bigger, and Alexandru Corlan with Călin Iosof started the database software for procedures charts.

Fig. 4 – The first chart realized by Radu Corlan was the aerodrome chart of the Bucharest Otopeni airport in AGA section; the contract signed with Romatsa and the Induss offer are nearby

My PC AT-286 computer had CGA graphics, this is monochrome at 640×200 resolution and I was using an analogue TV set as monitor. It was extremely challenging, because I only saw a small bit of the chart I was working at. I belive Radu had colour VGA, that is 640×480, but I had to finish the contract to get the money for a VGA card and monitor. In fact all the money I made I invested in hardware and peripherals.

The contour of Romania represented a challenge. The digitizers which allowed to digitize points were not accurate and I had to enlarge maps with the Romanian borders to scan them (Fig. 5). It took me one whole week to get that file with accurate contour points for Romania, which was later used on a wide scale. At sea, the Romanian airspace extended more than territorial waters and I needed to calculate there. Along the AIP charts, there was the Radio Navigation Chart of Romania, A2 format, where the contour imperfections would have mattered.

Fig. 5 – The Romania’s contour was digitized at a resolution larger than needed for AIP charts, because the A2 Radio Navigation Chart of Romania followed; Cristi Rusu applied an OK with a red pencil when the chart was accepted at the NOTAM Department

The charts when ready were delivered to Cristi Rusu, who took them at the office and brought them back either with observations, or with an OK written with a red pencil (Fig. 5).

The MAP section instrument procedure charts came later because the software to generate them out of the navigation database took long. However, after the software successfully passed the tests, all charts got generated at once. Two problems persist in my memory with these charts. The first is a bug which gave us headaches for a while. The charts were generated and then compared to those drawn by Sorin Onițiu by hand. One of them included a VOR beacon which the software refused to represent. On Sorin’s chart the VOR was present, on the generated chart all the other details were in place, except the VOR. Of course, Cristi Rusu did not give his red pencil OK and Alexandru Corlan assured us that the software has no personality to discriminate a VOR from all others. Finally, redoing calculations, we realized that indeed, the VOR was misplaced in the original chart, and in fact its coordinates placed it outside the boundaries of that chart. Thus, by that time we wasted, we paid the small guilty satisfaction to find an error in the charts which circulated until then in AIP Romania. It was for the first time when the map objects were rigorously represented to match the reality in the field and the navigation database. The second incident came from the fact that the software automatically placed the labels of the radio aids where it found free space. For the viewer, these automated placements seemed bizarre, sometimes far from the radio aid itself. A human designer would have squeezed other elements and would have made room for the labels where the viewer expected to see them, but the automated generator was a robot who placed all the chart elements, leaving labels at the end, placing them where it found free space. Cristi’s red pencil kept correcting the placements, but we knew that what looks simple to a human is complicated for a robot and vice versa. Here, Alexandru and Călin came with a formidable solution. They wrote a LISP algorithm from the class of Artificial Intelligence (how we called it at the time, as a dream compared to today’s reality). It redid the placement of all chart elements which were not fixed by geographical coordinates to find a suitable place for each label, according to “human” rules (Fig. 6). A kind of Elisa test for placing the labels. After this algorithm was employed, OKs started to flow. There were pilots who did not even realize that these were automated generated charts. They knew Sorin Onițiu was away, at Jeppesen, in Germany, and they wondered how he continued to design charts for AIP Romania.

Fig. 6 – ILS instrument approach chart where the labels of the radio aids are placed according to “human” rules (somehow in a line but not strictly aligned and eventually covering unessential elements, replacing the white space placements, which looked artificial)

A problem which we could not solve was the obstacle chart of Caransebeș Airport, which has a mountain on the runway axis. I innovated from the ICAO norms by a local change of vertical scale (Fig. 7), but I have never come to terms with this solution, because I was haunted by the idea what would happen if a pilot who has never encountered such a situation would wrongly interpret the height of the mountain. Evidently I got approval of the NOTAM Department. The innovation would not have been needed if we could use a larger printer.

Fig. 7 – Caransebeș airport obstacle chart where we needed to innovate from ICAO norms (upper right)

When the Faculty of Electronics left Polizu campus, moving to Leu in 1990, the Rooms A-146/L114 in Polizu were given to us and there I had the mission to organize a computers laboratory. I installed CorelDraw and showed to the avionics students how we made the aeronautical charts for Romatsa. I thought it made a useful teaching exercise. It never crossed my mind that the students could help us because we worked against the clock with the deliveries and I was under stress with the large radio navigation map of Romania. The A2 format was too big and even if we split it in A4 bits, it was a colour chart, so only to be drawn on a plotter, and Corel was not even suitable to generate plotter files. A solution was in the waiting. I gave the students the chart symbols database and the contour of Romania, I asked them to learn CorelDraw, a possibly useful tool. It was not for a mark. After one week, there was this student Răzvan Mărgăuan who called me at one of the stations, with another student, Radu Cioponea standing next to him. I will never forget my amazement that I was in front of the colour radio navigation chart of Romania, in all its complexity. I could understand how enormous amount of work was in that jewel of a chart, but I could not understand how can you do it in only one week and learning Corel in the same time. That chart was later delivered to the customer Romatsa (Fig. 8). Later, Răzvan Mărgăuan and Radu Cioponea were offered a job by Romatsa imediatelly after graduation. Otherwise, general director Niculae Droc insisted to be present in person at their diploma exam, as an unprecedented and unique event in the history of Faculty of Aeronautics (Aerospace Engineering). Răzvan in a team with Radu and another colleague made under my supervision a professional air traffic control simulator as a diploma project, presented at the diploma exam. The simulator was later implemented at Romatsa.

According to the contract we wrote manuals on charts design [6] and trained Romatsa personnel, and the contract ended successfully and in time. Cristi Rusu gained a merited position in Romatsa. He got interest from many countries for the new Induss system, but his boss, Jenică Poenaru was not interested to give a competitive edge to others. Sadly the people we trained left the country with Cristi Rusu among them, Cristi emigrating to Canada.

I do not know how long did Romatsa fully use the system. The specialists left in the NOTAM Department avoided us, probably fearing the dumping prices of such brilliant solutions.

Although it was expensive, I got an AIP Romania 1992 [4], [5] as courtesy of Romatsa. My AIP copy was maintained until 1993 (that is the reason for some pages dated 1993). Appreciation and other honours did not occur. I am glad that at least Cristi Rusu was mentioned in the history of Niculae Droc. In the end, it was not this first digital AIP of Romania (forgotten today) the most valuable thing. It was great that I managed to attract some exceptional students to this fascinating field of air navigation and that today they are accomplished experts at an international level. Răzvan Mărgăuan is Head Technical Systems at EUROCONTROL, Maastricht Upper Area Control and Radu Cioponea is Safety Aviation Expert at EUROCONTROL in Brussels. Also, I enjoy that I suceeded in attracting to aviation (albeit shortly) exceptional people like Radu and Dan Alexandru Corlan, or like Călin Iosof. Personally, this project provided professional self confidence. It was so hard that, by comparisson, the challenges that followed seemed a piece of cake.

Fig. 8 – The radio navigation chart of Romania, 2019 Edition, a much more recent version of the 1992 original, which sadly was lost

References

[1] Niculae Droc (2007) Scurtă istorie a serviciilor de trafic aerian, (A Short History of Air Traffic Services, in Romanian), https://www.scribd.com/doc/247930266/Istoria-serviciilor-de-trafic-aerian

[2] Induss (1991) Deviz de lucrări la licitația Romatsa pentru AIP Romania (Estimate of work for the Romatsa tender for AIP Romania, in Romanian)

[3] Contract Romatsa – Induss (1991) pentru realizarea unui Sistem de Creare și Gestiune a Hărților Aeriene și a celor 139 de hărți cuprinse în documentația AIP România 1991 (Contract Romatsa – Induss 1991 for the development of a System of Generation and Management of Aeronautical Charts and of the 139 charts included in the AIP Romanian 1991, in Romanian)

[4] Departamentul Aviației Civile, Romatsa (2012) AIP Romania, Vol. I, GEN, AGA, COM, MET, RAC, FAL, SAR

[5] Departamentul Aviației Civile, Romatsa (2012) AIP Romania, Vol. II, MAP

[6] Induss (1992) Manual de desen pentru hărțile AGA-AOC în CorelDraw 2.0 (Manual to design AGA-AOC charts in CorelDraw 2.0, in Romanian)