All the best sci-fi films have them, and they may become our future automated space explorers. Currently, one of the biggest drawbacks for using robots in space is that they depend on human input (i.e. commands need to be sent for every robotic arm motion and every rover wheel rotation). This means that, especially with missions operating far from Earth (such as the Phoenix Mars Lander and Mars Expedition Rovers), very simple and mundane tasks can take hours or even days to complete. One of the main reasons supporting manned exploration of space is that very complex science can be carried out very rapidly (after all, astronauts are human and many robotic operations that take weeks can be completed in seconds). But say if our robotic explorers had a high degree of automation? Say if they could sever the requirement for human input and carry out tasks with intelligent reasoning? As robotic and computer technology increases in sophistication, one Caltech scientist believes space exploration by artificial intelligence is closer than we think…
I remember watching the start of Star Wars: The Empire Strikes Back thinking it was so unfair that Darth Vader and his ilk had access to intelligent space exploration droids that could fly around the galaxy, land on alien worlds and automatically seek out the rebels on Hoth (directing the battle fleet to the icy moon, creating one of the most famous and atmospheric sci-fi battle sequences in movie history. In my opinion at least). But say if we were able to build such “droids” (in fact, droid is a good description of these space explorers, defined as ‘self-aware robots’) that could be sent out into space to explore and report back to mission control without depending on instruction from Earth?
Wolfgang Fink, physicist and researcher at Caltech, believes robotic exploration of space will always take the lead, and even reverse the need for manned missions. “Robotic exploration probably will always be the trail blazer for human exploration of far space,” he says in an interview with Sharon Gaudin. “We haven’t yet landed a human being on Mars but we have a robot there now. In that sense, it’s much easier to send a robotic explorer. When you can take the human out of the loop, that is becoming very exciting.”
While Fink is encouraged by the progress made by missions such as Phoenix and its robotic arm, he is keen to emphasize that the link between human and robot needs to be removed, thus allowing robots to make their own decisions on what science needs to be carried out. In reference to Phoenix’s robotic arm he said, “The arms are the tools, but it’s about the intent to move the arms. That’s what we’re after. To [have the robot] know that something there is interesting and that’s where it needs to go and then to go get a sample from it. That’s what we’ve after. You want to get rid of the joystick, in other words. You want the system to take control of itself and then basically use its own tools to explore.”
The key attribute robots need to possess is the ability to recognize something of interest, such as a rock or crater, something that a human mind would see as a scientific opportunity. At Caltech, Fink and others are working on programs that use images for robots to distinguish colours, textures, shapes and obstacles. Once artificial intelligence has the ability to do this, if the programming is complex enough, the robot can notice something that is out of place, or a region worth investigating (such as a strangely coloured patch of Mars regolith that a Mars robot will decide to dig into).
As you’d expect, software is being tested and Caltech scientists are beginning to try it out on a rover’s navigation functions. However, the robotic decision-making is very basic presently, but NASA has taken a keen interest in Fink’s work. For example, in 2017 NASA intends to send a robotic mission to Titan, one of Saturn’s moons. In all likelihood the moon will be explored by a balloon-type vehicle. However, it would be impractical for such a vehicle to depend on commands being sent from Earth (as it would take more than an hour for communications to transmit over that distance), so there would need to be a certain degree of automation built into the craft so fast decisions can be made in a dynamic environment such as Titan’s atmosphere.
Although this is all interesting and necessary, there will still be a basic human desire to explore space via manned missions, although a certain degree of self-awareness may be required of our robotic explorers as they carry out reconnaissance trips before we make the trip…
Source: PC World