We often see the term “video game” associated with people having a good time and/or being entertained. That’s pretty natural, as such games were originally designed to entertain and be a lot of fun. However, these days video games are being used in many different ways. For instance, they can be used to help disabled people or victims of strokes to regain their movements and improve their lifestyle.
It is a truly revolutionary concept indeed and Unity is helping to make it real. Kleber de Oliveira Andrade is one of the “revolutionaries” behind this new era, besides being one of the owners of the Ponto V web site, he is a teacher and coordinator at Faculdade Dom Bosco (a university in the city of Piracicaba, Brazil) and is undergoing doctorate with the guidance of professor Glauco Augusto de Paula Caurin (EESC/ USP). More importantly though: he is the head of a team of researchers that is developing a one- Brazilian solution that shall assist people with limited bodily movements. We spoke about his career and the exciting technological future that’s ahead of us.
Unt3Dmag: It is a pleasure talking to you, Kleber. Tell us a little bit about your career and how you first started working with Unity.
Kleber Andrade: Well, I started developing video games back in 1999. I was merely 13 then. In 2008 I got a degree on Genetic Algorithms and Games by using XNA and C#. I took part of the SCGames in Florianópolis back in 2009, someone there showed me a game they had made by using Unity. In that very same year I started my master’s degree. I continued studying XNA even further, while starting with Unity as well. However, the latter seemed pretty complicated to me, so I stayed with XNA for programming. I had a blog and by that time I became an associate at Ponto V. I published all my XNA tutorials in Ponto V. It was then around 2011 – when I started my PhD, actually – I realized XNA would not last much longer, so I gave it up for good and started using Unity full time. Nowadays I use it for all my projects and classes.
It seems to me that Ponto V is a very collaborative effort from you guys!
It definitely is! It all came about when Vinicius Godoy first started his blog, which already bore the name Ponto V. Then Bruno Sanches joined him. Bruno was one of the guys that created Erinia. Finally, I joined them both. Now the three of us are the main guys behind it, and we rely on a wide range of other people to help us. Anyone can collaborate with Ponto V, but all submitted materials must meet our approval first.
So you have been programming and developing games for a long time now…
Yep. Like I said, I started making games back in 1999, and before that I used to draw a lot. I got into college in 2005 to study Computer Science and graduated in 2008. By 2010 I was already teaching classes… Actually, since 2009 I give lectures on game development and teach C++ with SDL. During this time I have used a lot of different engines and experienced programming with many different languages, including odd ones, such as RPG Maker [laughs].
How did you turn from a boy who used to like to draw to a full grown adult that studies complex stuff such as Genetic Algorithms?
Good question! My older brothers purchased an Atari 2600 console when I was 3, so I started playing it at a very young age. By the time I was 5 I started drawing and my main inspiration for my illustrations were the cartoons, movies and video games I enjoyed as a child. I began using my computer to play DOS-based video games when I was 10 and when I was 11 I came by some magazines with articles about programming. Finally, when I was 13, I discovered that all those games I loved were done thanks to programming, so now I wanted to develop them myself. When in college I acquired a keen interest on Artificial Intelligence, so I started studying Genetic Algorithms and stuff like that.
During my teenage years I used to buy those computer mags that printed pages and more pages with endless codes. I would copy them for hours, as clueless as one could be about the meaning of all that, and was never successful in compiling a single darned game!
Exactly! But come on, that was fun… Anyway, in 2008 I took part of the very first Campus Party and won the first prize in Artificial Intelligence for Games.
You career is pretty academic, if I may say so. In this sense, how hard is it to use Unity for things other than games?
Yeah, I guess we can say my life is pretty academic nowadays, you’re right about that. I am a college teacher working for a fair number of academic institutions in Brazil. I teach robotics and AI, and my students range from people who are studying to be game developers to people who will graduate as engineers. However, games are what we still do, basically – in the broader sense of the word. But they are “serious” games. They are games aimed at helping people with health issues, games used for rehabilitating people who suffered strokes, for example. Unity is very helpful when it comes to this because it is really quick and easy to prototype and test stuff. Besides that, Unity is quite open, so I can do whatever I need to it through C# – for example, networking, programming specific AI techniques, accessing databases, and so on.
That is truly fascinating! I would like to talk more deeply about these things, if you will. For example, rehabilitation of people who suffered strokes. How is this done?
This is something very motivating for me and for so many reasons. First, because it allows me to combine what I love to do with helping people with an issue that is socially significant and not just a health issue. Second, because it concerns my doctor’s degree as well. In 2010 I met Gisele Ito, an occupational therapist who needed our team at the university to develop some kind of mechatronic system for wrist rehabilitation. Then Ricardo Cesar Joaquim, Bruno Jardim and myself – along with professor Adriano Siqueira and professor Glauco Caurin – started working on it. First thing we did was hook up some engines to move images in Microsoft XNA. It seemed pretty cool and promising straight away. From then on we started studying the subject much more seriously. It is a field not as new as we first thought – the idea first came up in the US back in 1989: robots, games, rehabilitation. We started working in these three fronts, and assembled a sizable group of researchers in our lab: mathematicians, physicists, engineers, therapists, computer experts and so on. If you take a look at the world’s population you’ll realize it is getting older everywhere, so we can expect strokes to happen more frequently. They tend to affect older people – so the more older people we have living amongst us the greater the likelihood of somebody having a stroke.
The math is simple. In this scenario we can end up having a shortage of therapists to care for the debilitating patients, as a consequence death rates are likely to increase amongst them. Strokes are the second most frequent cause of death worldwide, so rehabilitation is of the upmost importance for the patients. They can regain control or partial control of their bodily movements and it greatly enhances their social lives. Games can be a motivational tool, something that encourages the patient to make the movement. Robots are used as joysticks that are capable of reading the movement, they can help in the act of performing the movement as well. This is made possible by the neuroplasticity of the brain, which allows nerve cells to reorganize, therefore the person can learn to use other areas of the brain and perform the movements he once previously lost.
That is simply amazing! When will the general public be able to benefit from that?
Well, a couple of robots are already available, but they are pretty expensive. You can check some of them out at here or here. These are not ours though; in our case, we are developing a solution that is one-hundred percent Brazilian. It is also top- quality and involves a great deal of multi-disciplinary research that we want to make more affordable. The project goes through several phases, the first one was research, the second was structural development and the third was the development of a pilot game (The Catcher). Currently we are being evaluated by the Committee of Ethics so that we meet the conditions to start testing our solution with healthy individuals. After that we can test it with disabled people. Finally, if all works fine, we shall register the patent and then establish a company to start selling. I guess this can all be done in a year and a half to two year’s time.
So this technology is currently available and is already helping some people – provided they can afford it. On the other hand, you guys are also coming up with a Brazilian solution whose low cost can help the masses too, right?
I wouldn’t use the words “low cost”. When we say “low cost” people tend to think “low quality”, and that is not the case here. Some hospitals and institutions in Brazil already have the robots I mentioned earlier – Lucy Montoro is one of them. In our case, we are developing a high-quality system with engines and controllers that are undoubtedly cutting-edge, all of the technology involved is cream of the crop but we can make it more affordable because we reduced the whole system and opted to attack the point in case directly.
For instance, most of these robots are designed to have more than one degree of motion, in our case, we designed our robot to have a single degree of motion. Therefore we shall have eight robots in the future, and each one of them will be used for a single purpose, be it feet, ankles, elbows, shoulders, arms or wrists. It is gonna be costly nonetheless, with a retail price of R$ 50 thousand, I would say (16 thousand dollars approximately). But it is a better price if we compare it to the 200 thousand dollars that you must pay to get the other robots. It is basically twelve times cheaper.
That makes a huge difference indeed. Now I want to change the subject for a while. Like you said earlier, you are a college teacher and many of your students look forward to becoming video game developers. What’s the proficiency level of the developers coming out of the Brazilian universities nowadays?
Hmmmm… That’s difficult to say. I will try to answer, even though it varies a lot. Some of my students have entered college with a great deal of development knowledge they’d already acquired. Others get there with no knowledge whatsoever and some of them leave knowing a lot, some of them leave with limited knowledge nonetheless. The problem is that the courses are generally very generic. They have to deal with arts, audio, programming, design… This lack of focus tend to get in the way of the students. But I would say that most of our students are good and they like what they do. The thing is, though: making a game is complex in so many aspects… Sometimes a bad game is much more successful than a good one. Sometimes it would take too much time and money to do what you first envisioned, so you wrap the game up before it is actually finished. The Ninjin game was developed by two Fatec-Americana students of mine, and it was a huge success. However, I must say that the students do leave college with a very superficial education. This is a very broad field of knowledge, and it is up to you to keep up with everything you need to learn in order to succeed. It is your personal effort that counts, there is always something new to learn – there is and there will always be. After fifteen years I can say I am still learning how to make games [laughs].
Going back to AI, is it an area that is also still progressing when it comes to video game development?
Yes, it is. We can thank computational excellence for that, but then there is the other side of the coin: sometimes we do have algorithms that are pretty efficient, but no computational prowess to process them. Anyway, the coolest ones in my opinion are the adaptive algorithms, algorithms that “learn” through reinforcement. People need to try to map the player in order to verify if he is still motivated to keep playing. After all, if he loses motivation he stops playing…
Sure, keeping motivation is perhaps a greater challenge than developing the game itself…
That is a fact! The game can be gorgeous, with great mechanics and all. However, if it doesn’t motivate the player, he won’t play it. Plain and simple.
Now tell me: is it true you developed a system that makes parallel parking in cars an automatic feat?
Yep, it was my master’s degree thesis. I had to defend it by using a 2D simulator, because the car could not be manufactured
in time. I basically programmed two algorithms, one based upon deterministic finite automatons and the other based upon neural networks. It was an amazing challenge, and when the car is finally ready I can’t wait to test it properly. In the meantime I am putting the final touches in the 3D simulator for more tests.
I guess you can grasp how important for humanity this feat can be [laughs].
Sure. For people who can read Portuguese (or can use Google translate — laughs), here it is. I still want to make it better, my idea is to work more deeply on this, although it is hard to find time for that…
It may be the end of a worldwide reign of ruined and crashed bumpers [laughs]…
And there will be no more waiting until that guy finally manages to park his car. Plus, any person will be able to park in any given parking space, no matter how difficult the maneuver is.
According to my calculations, in this case “any person” means 5 billion human beings…
Probably [laughs]. The problem is that everything demands time, a lot of time, to be developed…
Amongst all the areas we have talked about in this interview, which is the one whose near future is the brightest and/or the most amazing?
Near future? Hmmmm, that’s a tough one… We do have a lot of research going on nowadays. For me, near future is a time period of up to twenty years… I guess autonomous vehicles is a good guess. Google is already trying to sell its first driverless cars. There is a lot of research being made on robotic arms for object manipulation. They will be used in a wide range of things, especially in the Orthotics field, and will be very beneficial in that area. There is also a lot going on when it comes to combining rehabilitation and robotics, these will also advance us pretty quickly. So I guess these are the three areas with the greatest potential nowadays: autonomous cars, rehabilitation through robotics, and robotic manipulation.
I think there will be many surprises in all these fields, and I hope that a one-hundred percent Brazilian I mentioned is one of them.
That would be wonderful not only for us, but also for the rest of the world I am sure. Kleber, I would like to thank you very much for your time. This interview was very revealing for me, and I think our readers will feel likewise about it. Would you like to leave some message for the developers that are reading us now?
Well, I guess I should tell them to keep on doing whatever they enjoy doing, but having in mind our primary needs as human beings as well. One day – who knows – you may be that one person with a certain condition, a person that so badly needs new and relevant things to be developed. Keep that in mind. Furthermore, there is more info about our rehab work in the page of the game. Thank you very much.