International trading and robots
Interview to Odyssey Robotics in Encín Golf Madrid
(Translated by Irene Robles, translator and science fiction author)
The human has always been an animal tied to technology and, precisely because of this, the appearance of the new technologies is opening a new definition of human being.
The commerce and international logistics, which are so tied to man and his technology, have suffered a great advance in the past years in fields so different like telematic managment of documentation and communication between different parts of the supply chain, or the new control and navigation systems of the different means of transport. However, there is still much path to walk for the electric vehicles and the autonomous vehicles, the distribution by drone or factory ships travelling around the world or installed at free zones.
Thanks to the increase of computing power of computers, nowadays it is possible to manufacture robots which are able to gather millions of data ordered scientifically. Maybe the general questioning about the capability of robots to extract conclusions or take decisions is not related with a robot processing the information in the same fashion that a human brain could do, there is rather the possibility that a robot can or can not have creative intelligence (if we understand that as the capacity to generate new operatives not previously programmed or imagined or even existant by now) or strategic intelligence (if we understand that as the capability to establish a path with a clear goal and develop it around initial criteria, but modifiying when foreseed or not events came into play). In other words, the key point for the development of the artificial intelligence is guesing if a robot could be able to take decisions on the fly, taking into account unexpected events, or the other hand, its acting will be always tied to previous programming, case in which it will be important to know under which algorithm is working a vehicle or machine when it moves and if threatens the life of hundreds or thousands of people, or even the value of the transportated goods. For vehicles that carry people or goods is important to know, to clarify, in case of accident, which was the priorty followed or the program underlying the decision tree of the self-guided method. Insurance companies will also adapt their policies depending how the algorithm has been set in one way or the other, as well as governments will create laws to protect the welfare state.
On the other side, there is talks about robots that work as avatars of their creators offering them a physical alter ego in another place. This could threaten the way in which internacional commecial meetings are developed today. It makes even wondering about public life further than an old or handicaped body… It scares to think that our presence could be replaced by a virtual image of ourselfs and that an android robot asummes our identity or the set of characteristics that define our being. In this case, our body would lose its importance and everything would be set aside the brain activity. In the end, companies would be formed by brains in diferente places, agglutinating some corporative culture and specific know how, or maybe companies would be global with one global and shared knowledge.
In a population that desires to live forever, that wants to keep life quality at all cost, a robot could help to keep a life standard or can leave the physical human life to a machine that plays the interaction with the environment and others. Said that, Matrix unveils itself as a film more and more near to the reality that could take place. And, obviously, it gives some fear. The prostetics, nanotechnology, laser, are tools to enlarge life, to make our organic parts regenerate, create or replace tissues. What would happen when we like to elongate our brain life? Will prostetics also be created? Grafts? Transplants? Or simply like an egyptian mommification to store the value of a body, will be mommified or freezed, or copied the most valued brains?
Japan is preparing robotic interpreters to attend visitors at the Olympic Games. Not necesary androids, but they are able to answer questions with more advanced algorithm than an actual call center can provide. Maybe in brief time there would be androids like the ones that work at the Osaka University, that can live together with families. Maybe helping old people with autism or demence. And here comes the key: could they take decisions for us? Up to which point do we have to delegate this last signature?
Commercially speaking, could a robot sign a contract in our name? Could it sell a product or service to a company, once it learns the catalog, price tables, discounts and profit margins alongside with technical characteristics and usability? If it understands the client needs, couldnt it provide the best solution in record time? There is software that suggest what, to who and how to sell, as there is software about how to invest on trade markets, after the analisis of thousands of parameters of different companies.
To speak about this questions and artificial intelligence on the whole, I had the pleasure to interview Javier Martínez García, CEO at Odyssey Robotics, a company that designs and builds guiadance systems for next-generation autonomous vehicles, solving specific needs and who aims to be, without doubt, a great promise in technologies like machine vision. Javier has just won the third edition of the program Santader YUZZ UMH (Miguel Hernández University in Elche, Alicante) for young enterpreneurs and will travel soon to Silicon Valley to develop its projects more widely.
Cristina Peña Andrés: Javier, what is your view about the new definition of human being introduced by new technologies?
Javier Martínez García: Our evolution has specialized us into thinking, we have greater capabilities at this than any other known species. If we consider the application of the new technnologies, all of them are focused into giving more resources to our mind, so we can focuse it limited resources at tasks that really scratch a little more of this evolutive path.
CPA: Do you think that a robot could really develop strategic or creative intelligence?
JMG: If a robot wins me at chess no matter how much I play, does it have greater strategic intelligence? If a robot compares a enormous set of medical data and cames up with a drug combination that reduces to almost zero the side effects while maximizes the healing action, has it creative intelligence? The answer to both questions is yes, and this systems are somewhat common in our day to day lives. The problem is the doubt to see as intelligence something that can be explained like a mechanism, because is hard for us to give any intelligence to a complex gearbox. Our intuitive conception about intelligence says that it should have similarities with us.
CPA: Which programming is implicit inside a self-guided vehicle?
JMG: The fundamental idea for every self-guided vehicle is to avoid creating dangerous situations. The control signals that command the machine have hierarchical priorities assigned for that pourpose: if my command as a crane is to lower the load but there is something below it, this presence signal has higher priority and cancels the descend action. But that not only applies to completely autonomous vehicles, it can be found, for example, at cars and trucks with frontal collision detection system. They constantly measure the distance with the vehicle in front, it takes into account the speed and actitudes of the last seconds and, if the front vehicle brakes, the system cancels the orders of throttle and activates other functions to action the brakes and stop our vehicle safely.
Aside from security systems, the program that rules the functioning of an autonomous vehicle has the task of knowing where it is and which is its actitude (position, direction, speed, load state…), so it can acomplish the given instructions. There is no much more inside this machines, really.
CPA: Which kind of relationship could a robot and the human brain have in a future? Could the neural map of the human brain be reproduced? Could a software be insert in? In some ways, nowadays software complets our memory, our agendas, our cameras, our way of communicate…
JMG: If we understand that our brain is, somehow, similar to a computer, then its possible to wonder if it couldn’t be programed like one and there is really no argument to think that is impossible. The problem to bring this to the real world is practical and it can be divided into three parts. On the one hand, a neuron is really small, tiny. Imagine trying to do eye surgery to an ant using a barrow bar, thats the size scale between a hypodermic nedle and a neuron. On the other side, a neuron is alive and to know its state we have to interfer with its inner workings. So, when we try to measure a neuron, we are distorting its operation and we are also shortening its lifespan. Finally, even if all this gets solved, there is still to find a way to know how the neurons are connected together. A neuron can have connection to a hundred of others and there are around 20 billion neurons in our brain. In conclusion, its a problem about the lack of technology.
CPA: What is your consideration about android shaped robots assumming human identities? Could they displace us?
JMG: When you meet someone, you generate a big amount of conclusions in a fraction of a second: if it is someone you know, you can even guess how is the day going for him. Nowadays there are androids that can create a bit of confussion at this initial seconds and make us feel that we are in front of a human. But something strange awakens our brain and, after a more powerfull analysis, we know that what we have in front is an android. The activity in this field is high and the competition to create higly credible human androids is set. My opinion is that we are far from stumbling with a robot and not identify it in a few seconds.
Speaking about the possibility of a displacement, there is a more realistic possibility. We have shaped the world for us, and a human shaped robot can use the same tools and vehicles without modiffication and they automatically asume some of our works. There was a competition, celebrated until 2015, the DARPA Robotics Challenge, whose goal is exactly that: android robots that can use manual tools and do other actions like opening and closing vales, doors, climbing stairs… Definitely, working inside a industrial environment.
CPA: Do you think that the commercial job can be at risk by the higher capabilities of robots to work with client databases, realtime prescriptions and bussisness decisions?
JMG: I think that the risk comes from not asuming that a new and more capable tool has born and we have to learn and master it. If it is useful, it could not only be applied on the more demanding task, but could be applied to multiply our output as this tools can be used at the same time and to increase power, while our brain cannot.
CPA: Javier, do you think that the last decision can be in the hands of a robot?
JMG: An example of decision automatism can be the timer of a bomb, where seconds accumulate in a counter and when it reaches the predetermined moment, triggers the explosion. This timer has detonated the bomb, but it did it because an external condition, fixed by someone. This way, although the timer triggered the explosion, the decision came from out of the system by someone who set it and he was human. This argument applies to all the machines known because we are the ones giving them a pourpose to acomplish. To what extend a very long chain of events relate a decision with its consequence is something that, in my opinion, lies in the field of philosophy and has no objective solution.
But this does not mean no one is working in systems that are able to define their own objectives, because there are researches which aim to understand how a machine could, with curiosity for its environment as a base, establish its own goals and objectives by itself. In that case, we could start thinking again about who or what is deciding the actions of the machine, being this same question aplicable to us.
CPA: Suggest me an interesting lecture about the last trends that are being worked on and are now reality.
JMG: Undoubtedly I recommend you the post “Neuralink and the Brain’s Magical Future”, from the site “Wait But Why”. Is not a reality, but it is a fragment of the future, it is about the origin of our brain, the neurons and the Neuralink company, founded by Elon Musk. Their objective is to build advanced brain-machine interfaces so an artificial intelligence can be joined with our brain to complete us. Its author, Tim Urban, puts a great effort at unwinding every concept for making it easier to understand.
Find more about Odyssey Robotics at www.odysseyrobotics.com