How did you get into electronics/engineering and when did you start?

I started at birth. The better question is, “when did your parents know?”

Imagine the pre-internet era of yesteryear. You are proud owner of an analog GRUNDIG phase alternating line television. It is one of those beautiful wood cased devices with rotary capacitors on the front panel for tuning it. You recently purchased it, and are still making payments on it. It is so precious, you asked your wife to knit a silk dustcover for it. Then you come home and catch your single-digit-age child in the act of making holes on your TV with an electric drill. The TV is on. What would be your reaction? A child psychologist would tell you, children do these things because they find adult temperament patterns entertaining and result to guerilla tactics to get it. And recommend the right course of action is not to startle the child in the act, lest you make a bad situation worse; hit the main circuit breaker if it is nearby, if not, approach the kid in a calm, casual manner and negotiate letting go of the power tool.

But no child psychologist can prepare you for a child who is ready to put the drill down himself and argue with you until blue in the face, he was doing it because your TV had design flaws. It is difficult to refute that claim when he demonstrates your TV is now functionally superior (despite aesthetically challenged); the hole is intended for an AM antenna for the receiver he built, for an AM transmitter he also had built, to add remote-control. Would you not agree a hole is a small price to pay if it means you will own the only TV in town with a remote control? You see, being my parent, in no uncertain terms, was an electrifying adventure every day.

When I ask them “when did you know”, my parents say, before I learned to walk or speak, I had this creepy, inexplicable desire to tinker with electromechanical devices. Why would a teething baby insist to suck on a pocket radio while softer, sweeter alternatives are available? Why would a human being whose only method of self propulsion is crawling, will always crawl under cars and giggle at a transmission plate? Why would he lick 9V batteries? Why is it every time you leave him in the car he pulls out your fuses or relays? “That is a sick puppy that baby of yours” was a comment they say to have received often, but not given it serious thought until I developed ability to use tools, and started taking everything they owned apart like there is no tomorrow.

Initially, I was destructive. I would open a device, remove everything that looked electromechanical, sort them, and put them in a cardboard box and hide it under my bed – reassemble the empty shell and leave it where it once functioned. I had a box for DC motors, box for LEDs, box for capacitors, box for antennas… you get the idea. In my eyes they were Legos; I used real components to make my own toys. Had the Lego actually existed in my country back in the day, perhaps things would have been different. But it didn’t. In fact Lego was the last of our worries. The country was fresh out of a civil war. Right-wing/left-wing armed conflicts, often proxy wars between United States supporters and that of the Soviet Union, had fostered a brother-against-brother environment. It was a time when people from your town could throw a grenade at your balcony because you did not share their political views, or open fire at your house when you are asleep because they got the wrong street number to intimidate. I still remember vividly the night my parents’ bedroom was riddled with round after round. You could find shell casings everywhere on the streets – kids my age collected and traded them like baseball cards. It is quite a sight to see 50-caliber ammunition in hands smaller than the casing. To create a pretext for a decisive intervention, the military had allowed the conflicts to escalate (some say they actively adopted a strategy of tension) and then, taken over the government. It was dark days. People around me had different worries than importing toys and entertaining kids. And I, still a child, had to improvise toys to entertain myself. Today, when I look at all the toys you can conveniently buy at the supermarket, I see insults to children’s intelligence, for how sub-optimized and feature-poor and imagination-throttling they are. Even the Lego has lost its charm – today they come with plans and worked-out scenarios, there is little room left for the child to innovate.

Because of my jack-the-ripper attitude with their devices people started hiding them from me, and that is when I figured my supply lines if you will, were threatened. I had just gutted the then-40-year-old tube radio my grandmother had and it was her only living memory of my grandfather whom I never met. She would now not talk to me. And I was not welcome at any of our neighbors anymore either. That is the first time in my life I picked up circuit diagrams and started studying them – see if I can reverse what I was doing. Back in the day most devices came with the circuit diagram attached inside the case, to help the fellow technician diagnose problems and repair them. I knew that because I would rip them out and collect them too. I remember asking my father money to buy a “repair-stick”. I did not know the proper name for soldering-iron, but I knew it was something I needed to repair. I used to sit at the window of repair shops and watch technicians use the soldering iron. One day I found a 15 watt soldering iron in their trash. I took it home, repaired it, and started using it. First time I used it I burned my hand with it; it left a small mark that is still visible today and I wear it proudly.

I do not know if it was for better, or worse, for my parents, when I changed strategies. I wonder how many parents have been given bats in the belfry to enter their elementary-schooler’s room. But if you were my parent, it somewhat went like this: you came to replace a fluorescent light bulb, it glowed in your hand as you entered, and I noted down the Gauss readings that caused this curious effect may be why the other one burned out in the first place, while you are still screaming. First time we moved, a two yard dumpster was needed to haul the scrap electrical parts from my room. Picture that now; a 454 gallon container – something most kids would rather play inside, than own electrical components to fill it. I would demand to be taken to a junkyard for my birthdays. And if you were my parent you would say yes. Because I was handy with tools and I would not hesitate to visit the family car and extract whatever it was I needed. As my parent you had two simple choices: take me to the junkyard today, or take the bus to work tomorrow. It was not uncommon dinner conversation to us, starting with a template “so, your mother found parts of ______ in your room today…”, where you can fill the blank with any of the following phrases: a school-bus, a radar, an ambulance, some military-aircraft, a landmine, an oxy-acetylene-torch, a time-machine (mom used that term once to describe a movie projector), my VCR, my car, neighbor’s car, my boss’ computer, and other curious things, followed by a long silence, and a lot of explaining. My father always used to say “son, I fear the day you will build something using everything in your room”, because he was convinced it would be some kind of doomsday machine.

My obsession to fiddle with electronics was so indomitable, even in the hospital I was tempted to take apart the monitoring equipment. How many kids have heard “…if you do not put that ultrasound machine back together, Koray, your parents will not be able to see your baby brother today”? Repair shops would call my parents to report some 4-feet-tall child of their surname applied for a job. Let me emphasize, for the dear EEWeb readers – if they have not figured out already, I do not come from a country with an OSHA and completely implemented child labor laws. Most businesses readily accept labor from minors. The society praises, if not encourages it. My family was not particularly against it either. But the jobs I tried to sneak in (and sometimes succeeded) involved flyback-transformers, that which crossed the line. My father of all people knew what it meant to be zapped by a lot of volts through the air; he has been working for the electric power company for as long as history itself. It is either his genes to blame for my condition, or watching him play with electricity all day, or the occupational hazards I grew up watching… (amputation by 500kV arc, decapitation by helicopter, are not, strictly speaking, the material for young audiences). But I blame my inability to start any sentence without a what-if. When you start every sentence with the philosophical question of tacit engineering knowledge, it takes you to the frontiers of what is possible, and you get to meet the father of all inventions; doubt. Doubt goes a long way in sciences. We have been taught necessity is the mother of invention. A necessity-minded society trains their children never to keep bread past expiration date. That is why only those who go what-if with the bread get to discover penicillin.

Regardless, every time I did something to stare death in the face behind their back, my parents would have me grounded. If you once saw my room it would immediately occur to you locking me in there was more like a reward, so they figured, I needed to be grounded in the electrical sense. And they conveniently, so it might seem, accomplished that by removing the fuse to my room, so as to shut down my “operations” as my mother used to call, then the things I did. That, and they knew I hated being in the dark. Fear of the dark does things to you. To me, it led to the discovery of inverting amplifiers. Little did my folks know I had developed capability to generate enough electricity, in absolute silence, to run the entire house for about one-half-hour. Because removing the fuse isolated my room from the house, you could leave me without power for days and I could still plot world-domination in my room. I had rigged my doorknob as a touch sensor which shut the system down. So if you were my parent and you walked in to check on me during detention hours, I would be staring back with puppy eyes, through the dark.

I got admitted to a special high-school for gifted students because my “extra-curricular activities” involved gasoline, electron emitters, mosfets, wind tunnels, computer languages, pyrotechnic compounds, and parts of the family car my parents did not know they were missing in ways they did not understand. Fresh out of K12 I owned every electrical diagnostic tool kids my age would not be able to identify and had the technical skills to qualify for an entry level job in electronics. And I proved a very difficult student to satisfy. I constantly interrupted lectures by pointing out inaccuracies in texts and worked-out examples. This is in part due to the real-life problems I regularly had gotten myself into, which never came neatly packaged with the information needed to solve them. That, and I was not conditioned to believe teachers know everything I need to know. First person to throw the towel was our physics teacher, who gave me a copy of the key to physics laboratory so that I could “go and prove it myself” at my own will. This was a first in the history of the school, to allow a student use a laboratory at will (i.e. no supervision) and I could not be happier if I was a dog with two tails because it was also the laboratory that contained the scientific equipment for electrical experiments. I went to that laboratory every recess, and sometimes skipped classes for the sake. During my stay in that school, every blown fuse in the building had something to do with me. But with the projects I did in that lab, at the time I graduated, enough funding was raised for the school used it to have the laboratory renovated and modernized. During my schooling years, as the slimmest and most nerdy kid, I had a significant bully infestation problem. These were not the soft Hollywood movie bullies either; but armed and dangerous kids hell-bent on causing harm to their minors. Little did they know I was befriends with someone who had the power to protect me from all that was evil. His name, Nikolai Tesla, and thanks to him I rarely had a bully visit me (for a physical encounter) twice, and never thrice. Voltage is a universal language – I of all people had figured that out very early. And I knew, people quickly learn, that the business end of a walking Tesla-Coil with wires running through his school uniform is, for all intents and purposes, not where they wanted to be. Despite I was physically defenseless, if you cornered me, you would never forget that day and your grandchildren would keep asking you to tell the story of your first defibrillation experience.

…

I made it to an elite engineering program, the first of five research universities I would later have belonged. Almost by tradition, I received keys to the mechatronics laboratory to use it as both office space and research, hence becoming the only undergraduate student to hold office at the department. I have built my first autonomous robot in that lab, and published my first IEEE paper. The rest, is history, and it is safe to say engineering is an integral of all my activities from my birthday to this day. After I became a professional, I realized I owed it to people whose gadgets I had destroyed. First time I build an MP3 player, I implemented it into the empty shell of radio my grandmother had – so first time in decades, it made another sound again. For every device I destroyed, hundreds I have revived, repurposed, gave them a second chance. My conscience is now clean.

What are your favorite hardware tools that you use?

I will not leave home without a mixed-domain oscilloscope. I do not remember ever having considered any oscilloscope overkill for any purpose. I will happily go pull over a 60 GHz LeCroy mounted on a rolling rack case to see if some AA battery is dead – for the sheer pleasure of turning it on and watching it start to breathe. (Sorry, multimeters). Like a battlefield surgeon running with stethoscope around his neck, it is usual to see me strolling with a probe around mine. It is a tool so favorite, it is the first thing I teach my students taking their first electronics course. My department has one GHz-scope intended for use in VLSI laboratory. It is not in that laboratory, everyone knows where it went. For I am the only person allowed to use it outside the lab.

Multimeter is my second favorite – the more bells and whistles the more I will like it. If it has an OLED screen (or nixie tubes) even better. I like high-speed versions which have a serial port that I can sample their readings, one of my multimeters has this functionality. I own several models from bench types to portables. I also still keep my very first multimeter my mother had bought me as a birthday gift. It still works, despite the first day I had it I had plugged it into the mains and it blew in my face.

I have a more complicated relationship with my logic/protocol analyzers. I like them very much, and I have them from those I cannot comfortably lift, to those that fit in my shirt pocket. But when out the logic analyzer comes, it usually means I am stuck at something that is not behaving like it did during the simulation, and the oscilloscope was not good enough to outfox that bug. Many digital designs can have problems that are typically too difficult to model in simulation, or too time consuming to simulate and often cross multiple clock domains. So looking at a logic analyzer usually brings bad memories to my mind about spending the night at the laboratory, staring at a circuit under magnifying light, plucking my hairs out.

The above two paragraphs probably explain I am a big fan of development boards for FPGA, microcontroller, and microprocessors (and in that order). I own many development boards and I commend on the design practices of Terasic and Mikroelektronika. When I am designing hardware I prefer sockets to soldering until the design is finalized; I do not like development boards where the device is not removable, especially for BGA packages.

I have programmable and universal power supplies from Tektronix and HP, which I depend on a lot, for clean and reliable power. My Agilent arbitrary function generator is also a very handy tool.

For building a complete real-time control system with just one controller board, in rapid control prototyping, and such purposes I prefer dSPACE products and have used them with great success. These allow you to use your own computer, and have Simulink interfaces. I had a chat with Xilinx recently and they are planning to release a similar product line, which I am eagerly waiting to play with.

My Hot Air Rework Station and any soldering equipment connected to it are rarely off (except, when changing tips – much easier when they are cold, but I have done it on hot irons). My very first soldering iron (the one I called repair-stick) was a Weller device. I still use the same brand. But Hakko’s SMD tweezers and wool type tip cleaner cannot be beat. I love both and use them regularly. I usually keep several types of solder at hand, including the Chipquik alloy.

Although I am not a mechanical engineer and do not consider myself an expert in CAD, I am very mechanically oriented and enjoy designing/machining mechanical parts. I own a lot of things that spin and make a buzzing sound and throw sparks when powered. I play with a wide variety of machining tools, and I am fascinated by the CNC lathe-mill. The rotary tool is probably my favorite, but I am not a DREMEL fan, I believe Proxxon makes much better rotary tools.

Beside these toys, I have several toolboxes full of every hand tool imaginable, and always carry a quality suspension tool with me. I have a head mounted LED that focuses and dims, extremely useful for inspections, which I tend to forget I am wearing and it has struck many interesting questions at the supermarket.

I do not know the number of computers I own, operational or otherwise.

What are your favorite software tools that you use?

C++ and RT-Linux are all time favorites whenever I have the time, and I have a tendency to make my own tools if I can help it. When I do not have the time, I use CadSoft EAGLE, MATLAB, MikroC, Simulink, Altera & Xilinx suites (Quartus, ISE, etc), Multisim, MPLAB, Eclipse, GCC, Valgrind, make/cmake, Solidworks, Visual Studio, XPlane, GIMP, and a latex distribution to talk about them. Any computer I use must have a hex editor, a packet sniffer, a terminal, a serial-port monitor, and PuTTY or compatible tool. I do not have particular preference of circuit simulators, but have used National-Instruments’ Electronics Workbench and PSPICE the most. I also really like Chipscope.

What is the hardest/trickiest bug you have ever fixed?

Software wise, it was a bug on the navigation computer of an autopilot meant for helicopter use. When the bug manifested itself, it caused the autopilot application to quit while airborne, either by segmentation fault or a freeze. The code that contained the bug spanned across about 100 files. Therefore it was not so much the bug itself, but the nature of the application that made it extremely difficult to hunt down, because you had to fly to replicate the fault behavior (which you’d agree, is not something you get to do very often). No two runs of the code were ever exactly alike as it depends on live data from aircraft sensors, which meant it did not manifest itself every time and it was difficult to determine what exactly triggered it, you could not view the memory in flight, and it was impossible to execute the code it in reverse. I have tracked it down with Valgrind, after six months of painful debugging, which revealed a pointer that got reassigned by a function every time the autopilot sampled the sensors. What I missed during the design phase, was that the function also incremented the pointer during every call, rather than using the same address – resulting in an erratic memory leak that eventually broke something else in its way.

Hardware wise, it was an inertial navigation unit (also for autopilot) that refused to initialize in the lab. Its error messages were not specific or descriptive; it had a single error bit for initialization success. The device only initialized outside the lab, which meant every time we had to reboot the system, nearly 600 pounds’ worth of of testing equipment would have to be carried outside with it, not to mention power cords, and having to read LCD screens in sunlight, perform the 30-second reboot, and move everything back in. Me and my colleagues had gotten sick and tired of the bug that caused this. I spotted a curious thing that happened when I decided it was time to scope the sensors directly and run an FFT on the readout. It showed that when the device was in the lab it read a 5 Hz vibration, and because it had to be perfectly still while initializing (for calibration) it would not initialize. But what vibrated at 5 Hz such that sensors picked it up but nobody felt it? Turns out my laptop screen was to blame. The latches had gotten loose from heavy use, and every time I typed (which I had to do to start the autopilot) the screen bobbed back and forth with my keystrokes. Hence a big screen, it caused a 5 Hz vibration of the table, that I could not feel, but the autopilot would pick, and interpret, and then bark about.

What is on your bookshelf?

Speaking for the room I am writing this in, and without leaving my desk, C and C++ reference libraries, many C++ books (I even have one in Japanese I received as a gift from a Burmese professor; do not understand the language but I understand the C++ in it), several books on digital image processing and computer vision, books about jitter, noise, and signal integrity, several books about digital/logic design, FPGA, principles of VLSI, several books on electrical circuits (analog mostly), probabilistic robotics, fundamentals of aircraft design, many books on computer architecture and organization, a very thick physics book, a book about artificial intelligence and neuro-fuzzy systems, several books about various microcontroller and microprocessor architectures, a few network books about protocols, standards, interfaces, a book about analog integrated circuit design, CMOS principles, a book about discrete-time signal processing, a few books on operating system principles and supercomputers, a book about real-time systems, digital integrated circuits, defense tech briefs, various issues of Electronics Design magazine, IEEE/ACM/AIAA/COMSOC proceedings of various issues, a stack of various research papers, BYTE magazines as far back as 1979, an encyclopedia of all WWII aircraft, all leaning against a SICK LMS200 laser range finder which prevents them from toppling over and knocking down a 35 year old male human skull. His name is Charlie and he wants to be a robot when he grows up.

I own nearly 3000 books in about every major subject in electrical, computer, and aerospace engineering. Some of them are e-books.

Do you have any tricks up your sleeve?

Yes; I keep a personal pocket-copy of the Murphy’s Laws. I study it regularly – especially before making any serious engineering decisions. I highly recommend this practice. Over the years Murphy has helped me to become a better engineer in every aspect, improved my capability of anticipating errors before they become faults. It is very long, so I am going to quote some of my favorites for my fellow EEWeb readers:

• GOLDEN LAW: Murphy’s Law never fails before all systems that can prevent successful operation of Murphy’s Law, will themselves fail.
• Every semiconductor comes pre-filled with smoke that it needs to operate; it is this magic smoke that makes it all happen. The amount of pre-filled smoke is proportional to the cost of the component. Thanks to advances in VLSI, newer components today have microscopic holes to let the smoke out efficiently without showing any obviously visible sign of failure. Despite that, it is possible to train your nose for the distinct smell of this smoke – oftentimes it is the fastest way to finding the faulty component. Whatever averted your eyes, your nose will catch.
• Big puff of smoke coming out of your design is always made up of several smaller smokes. Which makes it extra important to have a trained nose.
• Overheating components are always faster than you. The only way to beat them is to have some serious cooling strategy operational before the circuit.
• Behind every trivial mini-bug in your circuit, there is a monster bug waiting for its minion to get out of the way. Never underestimate trivial issues.
• When a helicopter throws a blade, it will always find the most expensive equipment nearby, assuming the off chance it missed you. Stay behind the cheap stuff during all testing, and try not to look important.
• The only time a helicopter has too much fuel is when it is on fire. For every other occasion, fill the tank.
• A small object such as a nut or bolt forgotten inside the wind tunnel will never be discovered until it is your turn to operate the wind tunnel and be held responsible for the FODY (foreign-object-damage of the year). The exception to this rule is, of course, if you enter the wind tunnel alone to search for foreign objects, Murphy’s Law will always provide someone to turn the 400 horsepower fan on for you.
• Any guided system will find the most expensive of all targets it was programmed to avoid. If something is capable of self-propulsion do not give it a brain. If you did, do not give it a weapon.
• Whenever you have two wires of same color, the one you believed to be ground the first time, was not ground. Unless you skip your first choice. Always reach for the multimeter.
• The amount of tangle in mutimeter wires is proportional to how quickly it is needed. Make it a habit to store wires neatly.
• Whenever there is two possible ways a connector can go in (i.e. not keyed), and there is 50% chance of getting it right, there is 90% probability first orientation you try will always be the wrong one.
• When there is only one way a connector can go in, it will get stuck, and you will force it the opposite way, which it will then go in, resulting in a short, or bent pins, but it will try to accomplish both if at all possible. If connectors do not want to go in, find other ways than brute force to convince them.
• New systems generate new problems to replace old problems they solve. The amount of problems in any technology is therefore a constant. It is the law of conservation of problems. Trust modern technology only when it is old technology, when the problems are bounded.
• Logic circuits are systematic machines for coming to the wrong conclusion with confidence.
• All great discoveries are made by mistake. Learn to make the right mistakes.
• Every circuit has an invisible emergency-stop-working sensor that measures your desperation. You should not under any circumstances allow any electronic entity feel you are in a hurry.
• The most ominous phrase in engineering is “uh-oh”, which is always late, and the only warning you will ever get.
• The component with the highest failure rate will always be located in the least accessible area of the PCB.
• If you install a 50-cent fuse to protect a $10K component, the component will burn to protect the fuse from harm.
• The probability any design breaks down increases with the importance of expected visit required to fund its development. If it did not break down just before the demonstration, it means during the demonstration, anything that can go wrong will go wrong. But if there is a possibility of several things going wrong, then the one that will cause the most damage will be the one to go wrong. If you perceive that there are only N possible ways in which the procedure can go wrong, and circumvent these, then an N+1 th way, unprepared for, will promptly develop. If there is a worse time for something to go wrong during the event, it will happen only then. If you plan for the case something may go wrong, and it so happens not to go wrong, this will only happen when it would have been ultimately profitable for it to have gone wrong.
• Any wire cut to length will be too short. Any wire cut longer than that, will experience crosstalk and not work, until it has to be cut to length.
• The less intelligent the idea, and the person stating it, the more likely it will be funded.
• A man with one watch is certain about time. A man with two watches is not. Therefore on any real-time design it is best to keep a minimum of three watches and perform majority voting.
• Never say “that can never happen” until you see the demonstration.
• Any given code will expand to fill all the available memory. If it is an embedded system, it will also kill and eat data beyond available memory so it can expand to the bounds of all physical memory.
• Every non trivial circuit has at least one bug. A sufficient condition for circuit triviality is therefore, that it has no bugs. At least one additional bug will be observed after the delivery to customer. The subtlest bugs cause the greatest damage because subtle bug will modify the circuit to mask some other larger bug. A working circuit is the one that has yet-unobserved bugs.
• Every voltage fluctuation will wait until it can assert the worst possible damage.
• Engineering is the only job where success always occurs in private, and failure in full view. If you succeed, you get a sugarcube and a pat on the neck. If you fail, people die. No pressure.
• It will be impossible to fix the N th fault in your design, without breaking the fix on up to N-1 others.
• When working toward the solution of a problem, it always helps if you know the answer. Provided, of course, that you know there is a problem.
• If you are working with tiniest SMD components, those that look like black pepper, with tweezers and magnification equipment on your head, your nose will itch. If you scratch it, you will drop the part. If you do not, you will sneeze. If you do not sneeze, someone else will do it for you. Regardless of the cause, N number of tiny components will fly across the room, out of which the maximum number of them you can possibly find is N-1, unless of course, if the design is due tomorrow, which then it becomes N-2. If you are looking for more than one thing while searching for the parts, you will find the most important one last. After you bought the replacements, you will find the originals.
• Any time you put a component in a “safe place”, it will never be seen again.
• You will find an easier way to design your circuit after you have finished it.
• If your circuit is likely to work AND it is desirable it works, it will not.
• Two wrong connections do not make a right design. It usually takes three or four.
• As soon as you dispose of a random, obsolete, outdated, or unknown component, even one that has gathered dust for years, a pressing need for it will arise.
• The guy you beat out of a prime parking space is the one you have to see for a job interview.
• If you do not need and do not want a particular component, there will always be plenty of it. But if you find something you like, buy a lifetime supply because they will stop making it.
• If you hit two keys on the keyboard, the one you do not want will show up.
• Everything in every electronics laboratory is divided scientifically into three major categories: (1) those that do not work; (2) those that break down; and (3) those that get lost.
• The longer a simulation takes to respond the greater the likelihood it will give the wrong answer.
• The more noise a motor makes, the less power there is available.
• Never say, “I am new at this” to a customer.
• The person who knows “how” will always be the engineer; the person who knows “why” will always be his boss.
• Real machines never come with spares – they come with the exact amount of parts needed. If your repair escapade has produced spares, keep this rule in mind.
• Every fault tolerant system designed to tolerate N faults in fail-operational state, will experience M consecutive faults, where the following formula always holds: M >= N+1
• You can always find what you are not looking for in the supply room.
• That which is attached with only two bolts is directly behind something attached with eight.
• The day you memorize resistor color codes, you will hear about surface mount.

What has been your favorite project?

This is a very difficult question for me; it is like asking parents which one is their favorite child. But if I must pick and choose, it would be the Project-BATTLESPACE, funded by US Air Force Research Laboratory and Air Force Office of Scientific Research. I collaborated with the VRAC independent research laboratory which specializes on virtual environments and pervasive computing; the laboratory that owns the most sophisticated back-projected stereoscopic virtual-reality rooms in the world and operate several synthetic training arenas for the U.S. Army for LVC (Live, Virtual, and Constructive) training through augmented reality. This research enables U.S. soldiers to engage both live and virtual combatants and give them the unfair advantage in training.

Project-BATTLESPACE was a $10 million effort involved immersive command and control of unmanned combat air and ground vehicles (UCAV & UCGV) from an augmented-reality environment, so as to allow one Air Force commander to control multiple vehicles, such as a single pilot fly an entire squadron. This is a critical strategic advantage because piloting a UAV or UCAV is a distressing experience for human pilots. Missions involve very high altitudes and relatively narrow fields of view, this can last for days, and pilots have to be rotated every two hours to prevent many hazardous side effects to their health. Ask any Air Force pilot and they will describe to you, flying one of those aircraft remotely feels like looking at the world through a paper towel tubes for hours at end. Test yourself; look through two paper towels and (carefully) walk around. Try to accomplish some of your daily tasks. Can you drive your car on the highway like this? That is what these pilots go through every day; our limitations as humans are hurting the U.S. Military due to decreased effectiveness in command.

The typical paradigm for UAV or UCAV control is the First-Person-View (FPV) flight. In FPV, the flight-deck experience is brought to a remote pilot via augmenting the real-time visual information with other sensory data. My task was to help flip this paradigm around for navigation by attempting to augment real-time visual information, with higher-abstraction information derived from itself. That is to say by using the UCAV or UCGV sensors as the primary interface context, augment the spatial and temporal context with the myriad of sensory information as it is available. The mathematics behind this undertaking was so intense, a 96-cluster supercomputer was required to calculate it. Due to my expertise in this field, the program director appointed me as chief-engineer, and provided me a team of three aerospace engineers. I successfully led my team to the development of cyberphysical interfaces for the U.S. Air Force Battlespace Simulator, enabling the software to control real life vehicles. It allowed these 96 computers go beyond the simulation, and, (1) control real life UCAV and UCGV platforms, and (2), augment virtual reality with the information gathered from the vehicles.

If you have ever played any real-time tactical strategy game such as Command & Conquer, or World in Conflict, what we accomplished is similar, except, all military units on your screen are also happening in real life. To prove this would work in a real-world U.S. Army LVC training scenario, I also designed and built unarmed man-portable military robots, IUAV and UCGV platforms both, named Michaelangelo and Virgil respectively, to work alongside U.S. warfighters. These robots had many advanced capabilities, including the ability to follow the helmets of U.S. soldiers, talk to them, accept voice commands and more. For instance, they could detect poison or explosive gases, or radioactive emitters, from only a few parts per million in the atmosphere, and immediately move the soldiers away from the threat area. The training scenario they had to play involved an isolated U.S. Military settlement in a primitive suburban setting with desert theme. There are six real soldiers (two with weapons on guard duty; a private first class and a sergeant, a sergeant major, and two command center operators), four virtual enemy soldiers, four virtual environments, three physical locations, two physical robotic combat vehicles (which are my designs), their virtual counterparts, and several other virtual military vehicles. Soldiers, as well as my robots, had to face both real and virtual combatants. Further, soldiers were wearing tactical small-arms-protective vests used by the Marines, which we modified with electronics to cause a harmless and temporary sensation of pain in torso, allowing the wearer to notice they have been shot from a particular direction, experienced pressure, or other ballistic impact. (The vest was designed by one of my former assistants, to whom I am still teaching electronics). Their weapons were real M4 rifles with firing mechanisms removed, replaced with electronics to help calculate bullet trajectories. The helmets they were wearing are standard U.S. Army issue. Following is the scenario U.S. soldiers experienced during the training exercise:

U.S. Army LCV Scenario-1: A military aged male civilian parks a white pickup truck in front of the base and approaches the U.S. troops guarding it. He is warned to stop where he is and show his hands. Despite the clearly stated commands, he acts he is unable to understand English and continues his eccentric, aggressive move towards the base, until the soldiers are agitated and have to take aim at him, and resort to body language to convince him to stop, and get down on the ground. After about 10 minutes of confrontation the compelling civilian cooperates, allows himself to be searched and detained.

Unfortunately the soldiers never notice the other military aged male who crawls out of the pickup truck bed (where the soldiers had no visual) and plants an improvised explosive device (IED) on the side of the road, in front of a casually parked civilian car-bomb on the street. The distracting civilian is on a suicide mission; only obliges for detention after allowing enough time for the IED to be successfully buried. The intent of the perpetrators is to wait for U.S. HUMVEEs to roll out of the base, and detonate the charges remotely when U.S. soldiers drive past them. Enough ordnance was planted in a matter of seconds to utterly destroy two HUMVEEs. This training scenario is, unfortunately, very real. 64% of all U.S. lives lost in Iraq and Afghanistan so far, were lost due to IED explosions that have been treacherously planted like this, exploiting the humanity and rules of engagement of U.S. soldiers. Let us look at the next scenario where my research comes in for the rescue.

U.S. Army LCV Scenario-2: Using the same setup, but different soldiers, and (my) robots assisting them from the ground and air, Scenario-1 plays out. That is to say my physical robots join along the live soldiers but they also appear in the virtual environment, as their sensory perceptions are fed into the system. The sergeant major has the complete digital coverage of the battlefield thanks to the new ability for one commanding officer to command many UCAV/UCGVs with ease. As soon as the white pickup truck pulls over in front of the base, my IUAV, which had been patrolling the area at high altitude, invisible to humans on ground, spots two people leaving the truck and starts tracking them both. The novelty here, is that when you have an intelligent-UAV performing the patrol, unlike humans it can focus on hundreds of moving subjects at once. It will never get tired or distracted, or suffer any of the aforementioned paper-towel tunnel-vision adverse effects. The IUAV reports to the sergeant major a second military aged male is engaging in suspicious activity in front of a parked white van. Sergeant major sends this information to my UCGV. My UCGV immediately creates a threat-zone; a blast perimeter, and instructs any U.S. soldier involved to stay out of it. It then enters the threat zone and begins scanning for explosives. It soon determines the position of the buried IED, replaces its ignition circuit with a U.S. detonator. All explosives disarmed with no collateral damage. All perpetrators caught.

These training scenarios were watched live by 20 U.S. government and industry leaders, including Wright-Patterson Air Force Base, U.S. Missile Defense Agency (MDA), and Lieutenant General Patrick O’Reilly. After the demonstration, MDA director publicly stated his opinion: ``The U.S. Army should hire you’‘, pointing his finger at me. This comment, is coming from a man who plays with U.S. intercontinental ballistic missiles and airborne laser weapons for a day job.

Do you have any note-worthy engineering experiences?

• I invented the IUAV Autopilot and hold an IEEE Best Paper Award on it, my work is also published on AIAA and cited by the leaders in the field including the Editor IEEE-TRO, Authors of Seminal Books such as “Probabilistic Robotics” and “Principles of Robot Motion – Theory, Algorithms, and Implementations”, AIR FORCE INSTITUTE OF TECHNOLOGY, Georgia Institute of Technology, MIT, and Technische Universitat Munchen.
• I have been either directly funded, or contributed in projects that were funded by National Science Foundation (NSF), Rockwell Collins Advanced Technology Center (RCI), Air Force Office of Scientific Research (AFOSR), Air Force Research Laboratory (AFRL), Office of Naval Research (ONR), Lockheed-Martin, U.S. Army RDECOM, Virtual Reality Applications Center (VRAC), Center for Nondestructive Evaluation (CNDE), Space Systems and Controls Laboratory (SSCL – NASA), The Information Infrastructure Institute (iCUBE), Aerospace Robotics Laboratory @ University of Illinois Urbana-Champaign (ARL), and ISU Electrical and Computer Engineering (ECpE).
• Due to exceptional academic performance including academic honors and research performance, I have studied in five research universities without ever having to pay tuition. Two of them are among the top-5 engineering programs in US.
• I provide research jobs for rising young engineers in my field. I also teach, and have trained highly technical people for the U.S. Science and Engineering workforce: Intel, Lockheed-Martin, Boeing, Aerovironment, US Patent Office, National Robotics Engineering Center (Carnegie Mellon), German Aerospace Center (DLR), FCStone (A Fortune-500 company), Mayo-Clinic, US Army 298th Support Maintenance Company, each today employ a U.S. Citizen engineer, who has been my student or research-assistant at some point. One of my assistants recently won an NSF fellowship and admitted to MIT for Ph.D. The list excludes those currently pursuing Ph.D. or formed their own company.

Do you have any experiential stories you would like to share? (Blowing up things, getting shocked, etc.)

I have been shocked by virtually every arbitrary voltage of every possible modulation, from circuits of all complexity. The list includes a lightning bolt. It is a miracle I am here to talk about my dances with death. But I will share with you the most terrifying of all. And no, it is not the lightning strike.

One evening, I was playing electric-guitar, sitting on a 60 Watt amplifier and enjoying the music. It was a beautiful Stratocaster hanging on my neck with thick leather straps, plugged in directly to the amplifier – because it had functionality to send signal into effect processors. I am right handed so naturally my left hand was on the frets. I played for long hours until it was beginning to get dark; too dim to see the tablatures comfortably. My hands had gotten sweaty from the passion. I stopped in the middle of a piece, while holding the second A-minor chord pressed, guitar still humming the chord, and reached for the light switch on a metal-body floor lamp. The switch was in the back side of it so I reached around the tubular body to grab it. The particular lamp in question had been in my house for years and never for once it had hurt anyone. But that evening, as soon as I touched that switch, my right hand clutched the lamp like a shop vise, and my left hand squeezed the guitar frets like a log splitter. It was impossible to let go. Because I had just completed the worst possible circuit for a human being; left hand grounded by guitar strings (through the amplifier), right hand on AC mains voltage (through the lamp), and feet insulated by thick carpet. There I were, someone capable of drawing a circuit of 4000 components and connections by heart, receiving a shock from the simplest of all possible circuits – the AC resistive path. What a pity. I always imagined it would be some sort of death-ray-doomsday-machine that I built, would take me (and preferably digitize me like in TRON but that is a different fantasy).

NIOSH and IEC numbers estimate impedance of a body like mine to be 1200 Ohms under the circumstances, which yields 90 milliamperes passing through my chest, give or take one. There was a 15 ampere fuse on the circuit that supplied the guitar amplifier and the lamp. Considering 60 milliamperes of AC is enough to stop your heart, if left in that condition I was surely going to die. And as an engineer I knew all too well what was happening, which makes it all the more horrifying. But I could not help myself. I was paralyzed by the current. Getting electrocuted, slowly, alone, with nobody to save me. The pain was beyond words. I cannot, for the life of me, liken that one to anything. The closest description might be intense pressure; like there is a bus front tire on your chest squeezing the breath out of you. After some time though you stop feeling any pain. That which is creepy, and a bad sign, because it means your body is accepting defeat, and surrendering. So, if you are getting shocked and the pain stops, it means you are going. Knowing I would never be able to let go of either terminal, and my life quickly fading away, I tried one last guitar move. My legs still worked to a small extent, so I leaned and touched the guitar to the lamp. This grounded the lamp, which tripped the fuse, my hands let go.

I remember waking up in a lot of pain in my joints and my chest which persisted for days and a gallon of pineapple juice was needed to erase the copper taste in my mouth (perhaps from salts forming). After the incident I disassembled and inspected the lamp. A wire insulation had cracked open and contacted the metal body from inside. There was no apparent mechanical reason or signs or aging to provoke it, but it had happened anyway, at a place impossible to see. Perhaps it was there when lamp was purchased, and been there all along, but nobody noticed it because we were always insulated by the thick carpet and touching the lamp had never grounded anybody.

Moral of the story – install GFCI outlets, they are well worth the cost. And always trust Murphy’s Laws.

What are you currently working on?

In collaboration with Rockwell Collins Advanced Technology Center, and through them, Wright-Patterson Air Force Base, I am working on a project named Active Image Navigation Augment (AINA). It is intended to enable U.S. Air Force Unmanned Air Vehicles (UAV) navigate themselves to safety in case GPS coverage is lost or compromised due to spoofing or jamming – transform the UAV as we know it from a remote controlled aircraft into a geo-spatially self-aware flying robot. I was selected to carry out this research due to my proven track record of innovations in making aircraft fly and navigate themselves without human intervention.

They come in all sizes, and the strategic advantage UAV platforms offer for US Military and National interests is undeniable. For the military it is the ultimate intelligence platform, and a force multiplier. For civilian use, UAV could save countless lives in an urban conflagration or natural disaster; search-and-rescue in environments that preclude or discourage direct human involvement (fire, floods, etc), disaster response for nuclear facilities, inspecting bridges, wind turbines, and dams for dangerous cracks and flaws, as well as monitoring the mechanical and structural health of the electric power infrastructure are some of the many applications. However the UAV depends on GPS technology and it is not a question of if, but when, GPS will fail us at a critical time of need. Satellites are ageing, our orbit is a shooting gallery of debris, we are flooding the airwaves with a deluge of radio-conflict, and multi-path errors are already a cause of concern for small UAV platforms. We do not have the same economy when we first launched the satellites, but the future of intelligence, surveillance and reconnaissance missions will depend on our adaptation to GPS-denied environments. Cyber-attacks such as GPS spoofing and jamming are even more immediate threats. Transmitting on the same radio frequencies at a high enough power will deny the service of the radio spectrum to the GPS receiver and it can allow an attacker confuse a UAV. If you have ever flown with any major airline, this also concerns you, because chances are your flight was brought to a safe landing by the Triplex-Autoland system, which too depends on GPS. The grave ramifications of someone with a big antenna, a soldering iron, a degree in EE and malicious intent, can range from the GPS in your automobile directing you to train tracks to armed UAVs, guided munitions and re-entry vehicles getting re-routed.

I am developing an artificial intelligence autopilot for those UAV platforms, capable of sophisticated piloting and navigation in denial of GPS and human interaction. Try this at home today; while looking through two paper towel tubes try to accomplish some of your daily tasks. Gauge how long you can stand it. Imagine yourself driving to work every day in this setting (DO NOT ATTEMPT). How do you feel? That is how it feels to pilot a UAV without GPS. Besides, humans are among the blindest of species due to the chemical reaction times in our eye cells and immense length of our optic nerves. We cannot process images faster than 15 Hz (the pigeon can do it at 75 Hz). We have singular foveae which not only makes our peripheral vision very poor, when threatened it prevents us from focusing on anything but the immediate threat object. Even though we have two eyes, their ocular coupling is outside our control. Human brain is also constantly trying to classify objects and give them a meaning, so it tends to be tolerant of similar-looking or high-entropy objects. Not a reliable navigation strategy. From the air, two similar looking (but geo-spatially different) sections of a river might be perceived as the same section of the river, creating the illusion of re-visiting a landmark. Or, two views of a single forested area may be perceived as two geo-spatially different areas with same plant coverage, creating the illusion this is a new landmark. Occlusions due to weather or other natural phenomena can also deceive human vision in a similar way. Such misconceptions are unacceptable for aircraft navigation. I am trying to address such segmentation issues in a robust way, yielding engineered, discrete, sparse but salient visually augmented landmarks for UAV navigation to optimize aircraft positioning accuracy, without becoming prohibitively taxing on the computational, electrical, or payload resources of the aircraft.

Can you tell us about your other engineering or research projects that you believe have made a difference?

My first published work, while I was still undergraduate, was a remote control which allowed you to control cars on traffic – take over any electromechanical system in the vehicle with the click of a button from virtually anywhere. It was based on GSM network and reverse geocoding, among other curious things. I had a budget of $85 for the entire project which barely covered the cost of electronic parts and other expendables, but I managed to produce a functional prototype and published my design at an international logistics and supply chain conference. The paper later made its way to other scholarly publications including IEEE. At the beginning of 2009, around half of all transit buses in the United States were using such system, and today, virtually every new car – I made my small contribution to this.

In collaboration with ISU Dependable Computing and Networking Laboratory I worked on a research project for autonomous monitoring of the U.S. Power Grid, to protect it from domino blackouts, sabotage and natural disasters, funded in part by National Science Foundation (NSF). At the time the 2003 Northeast blackout had affected about 45 million U.S. Citizens and cost the U.S. Economy $10 Billion in a few days. Because I was living in Ohio at the time, I was one of those people affected, and I watched in horror, how easily the U.S. power grid could be brought down, and how much damage can come out of it. To help mitigate this weakness, I developed an intelligent, small and cost effective computerized eye, thousands of which could be installed on critical spots of the transmission grid. Together, they would communicate with each other wirelessly, and propagate the health state of the grid in real time, at the speed of light. Such technology did not exist in the wireless sensor networking literature. My research appeared on CH13 news, featured on New Scientist magazine, and I was invited to the Midwest-ISO (U.S. operator of over 90.000 miles of power lines) to train power system operators. My IEEE paper on the subject has become one of the most attended distinguished lectures in IEEE Communications Society (COMSOC).

I have several industry projects with Rockwell Collins Incorporated (RCI); the top U.S. Defense Electronics Company which provides tactical defense electronics to the U.S. Department of Defense. RCI brand systems are used by all branches of U.S. Military which cover 70% of all U.S. military airborne systems (including fighters and helicopters) and a vast majority of U.S. Army warfighter systems. Every major airline company flies with their electronics. Simply put, RCI is responsible for some of the the most sophisticated electronics imaginable. We designed a monocular, multi-purpose optical sensor to replace multiple flight instruments on size-weight-and-power (SWaP) challenged Unmanned Air Vehicles (UAVs) to address DARPA nano-air-vehicle challenge. In other words, I created a passive optical range finder; one that could not be detected by the enemy, but allow an aircraft to fly itself. RCI extended the funding, and with it I got to get my first senior research assistant hired (today employed by the National Robotics Institute at Carnegie Mellon University) and together, we investigated ways this system could be integrated into helmets for U.S. Air Force pilots, allowing the aircraft to track the position and orientation of the pilot head inside the cockpit, which could be used to prevent a redout or blackout.

In collaboration with Space Systems and Controls Laboratory (SSCL) I created of one of the most influential aircraft in the U.S. today; the one-of-a-kind Saint- Helicopter, my brainchild, the smallest IUAV helicopter in the world which is fully autonomous, fully self-contained, and featuring on-board monocular simultaneous localization and mapping capability. In other words this machine was capable to draw floor plans of previously unknown buildings and urban areas, in flight, without GPS coverage, autonomously. SSCL is a NASA sponsored independent research laboratory under the Iowa Space Grant Consortium (ISGS), which is further funded by research grants and private donations from Boeing and Lockheed-Martin. It is the laboratory that built and operated the first generation of small spacecraft in Iowa. SSCL projects flew on-board the Space Shuttle Endeavor, and the NASA KC-135A. Saint-Vertigo with the SSCL brand on it was demonstrated to U.S. Air Force, Boeing, RCI, IEEE Robotics and Automation Society, as well as U.S. Army helicopter pilots with flight hours in Vietnam. I have dreamed and created it from scratch, including the airframe, electronics, computer-architectures, control systems, as well as software and algorithms. It required understanding of five different engineering disciplines to invent it. It proved an impacting research platform which allowed development of solutions for bridging the gap in between practical GPS coverage and image navigation. It was well received by the robotics and aerospace society, and the peer-reviewed scientific contributions of this machine are already giving the research in its field a new direction. Saint-Vertigo is a compact, rugged, 3D-agile IUAV transportable in a backpack, very difficult to shoot at, and can fly in congested, isolated, GPS-denied, or hostile areas where fixed-wing aircraft cannot take off, fly through, or land. The strategic advantage this can bring to US Troops and Special Forces in environments where conventional surveillance is not applicable (e.g., below-canopy jungles and riverine environments) aside, my technology is paving the way for GPS-independent navigation systems of the future. The civilian uses of such a spatially aware flying robot, will be in environments that preclude or discourage direct human involvement such as escape from fire and floods, disaster response for nuclear facilities, inspecting bridges, wind turbines, and dams for dangerous cracks and flaws (Minneapolis I-35 Bridge collapse was preventable with such continuous monitoring), search-and-rescue, as well as monitoring the mechanical and structural health of the electric power infrastructure where it can survey an area, process sensor data, identify risk, and help people make safe transit through a dangerous area. This could save countless lives in an urban conflagration or natural disaster. It attracted a $500,000 grant from the Air Force Research Laboratory and $600.000 from Office of Naval Research.

In collaboration with RCI for another project, we have developed a self-calibrating image-navigation system for the Ghostwalker Tactical Vest. This is the same small-arms-protective vest used by the Marines, but equipped with the vision guided autopilot I had previously designed for Saint Vertigo IUAV, using the natural dynamics of the human body as calibration metric. Strategic objective of the system is to allow U.S. Army troops and special forces to be able to map and image-navigate GPS denied environments while walking through them, so they can get out quickly without getting lost, and have a computer calculate safest egress routes for them, with the floorplan stored in the vest. The vest depends on monocular camera. My contributions not only helped this life saving technology become a reality, even if someone else created it, without my research contributions it would not work; the working conditions U.S. forces experience would easily tamper with camera parameters, leading to false navigation solutions, rendering the system ineffective. These life-saving vests are not limited to military; they can also be used by firefighters and other emergency response personnel. During September 11 attacks, stairwell-A remained intact after the second plane hit the South Tower. Only 14 people noticed that and used it to escape. Numerous 911 operators who received calls from the building were not well informed of the situation. They told callers not to descend the tower on their own. If this technology I am developing was in use at the time, it could have directed a lot more people to the intact stairway.

What direction do you see your business heading in the next few years?

More UAVs! Department of Defense UAV Roadmap shows defense spending in UAV development, procurement, and operations have more than doubled in the last decade. The potential for the UAV is immense. In the upcoming decades UAVs of all sizes are expected to play significant role in our daily lives, participating in commercial endeavors and public services as well as the military roles. Federal Aviation Administration (FAA) is currently researching UAV integration into civilian airspace, UPS and FedEx are already looking into technologies that will allow one pilot to fly multiple freight aircraft, and as early as 2015 major airlines may begin to introduce commercial aircraft without pilots (that is to say, your next flight might be piloted by one of my circuits). We are fast approaching an era where autonomous and human directed behavior blend together, revolutionizing pilotage as we know it into a management position.

What challenges do you foresee in our industry?

Just like early aviation pioneers who struggled with the development of aerodynamic control systems of various functions, we too face new roadblocks with the UAV. One is the limitations in current generation of navigation systems. Serious challenges await the ground-based UAV command and control center when a sensory failure (e.g., GPS-denied airspace) forces the pilot (or autopilot) to resort to Visual Flight Rules (VFR) for localization. VFR, like most other flight rules are, is built around human anatomy and implies immersive peripheral vision to the outside of a cockpit. UAV is a poor imitation of this paradigm. A lost UAV is a big trust issue with FAA and they are not prepared to allow one zoom through civilian airspace on its own mind.

What is your teaching philosophy?

“Dubito, ergo cogito; cogito, ergo sum.”

My teaching philosophy is to foster methodical doubt in engineering students, and redirect it to a positive outlet. I believe student curiosity is the fundamental resource for engineering education. My vision of the aspiring engineer of tomorrow is the one believes those who are seeking the truth, but doubts those who find it. It is therefore a tragedy to evade teaching engineering students to doubt and thus depriving them from the key to a sustainable engineering education that will not stop at the diploma. Some students, and I consider myself fortunate to have been one of them, are bitten by the monster curiosity bug, and those that survive the after effects can doubt unassisted. The better share however are not thus fortunate; they either never met the bug or bitten by a baby such that the effects have worn off due to career concerns. It is on our shoulders as the educators to spark their natural inquisitive behavior by a healthy injection of doubt. Either way, my intention is to maximize the total curiosity in the classroom, to make every class interesting, practical, challenging, and most importantly, learner-centric (i.e., effective) so my students look forward to coming to class, and feel bad for missing any.

There was a time when information seemed more finite, and stabilized for long periods before being imparted in an intensive period of a few years. The gains from this period were intended to last a lifetime. What once was is no more. Engineering students of today cannot possibly learn enough information within the course of an engineering program to prepare them for the outside world. In an age information can no longer replace education, what we are planting into our students has become more important that what we are pouring. We have to change the way we structure learning engineering. We have to grow engineering students motivated by their subject, not by grades. It also implies we have to create students of change, as change seems the only thing that will remain constant in engineering.

I am seeing the phrase do not try this at home has become synonymous with the modern day. While I agree some things are better left to professionals, what about things for which the corresponding professionals do not exist? Engineering students are the professionals to be – to whom should we instruct them to leave it to? There is no better way to throttle engineering improvement than suppressing the inner MacGyver. October 9 1903 issue of New York Times had the following headline: “The flying machine which will really fly might be evolved by the combined and continuous efforts of mathematicians and mech