Have you ever wonder what robots dream of? The idea that robots can dream might sound implausible to some. Robots are machines.
They do not have feelings emotions and they certainly cannot dream, many people would undoubtedly argue.
If even artificial intelligence is familiar with the concept of self-awareness, then perhaps it is possible to create robots that can also dream?
In Arthur C. Clarke's classical space adventure 2010: Space Odyssey Two, Sal-900 asks: "Will I dream?" Dr. Chandra replies "Of course you will.
All intelligent beings dream. Nobody knows why... Perhaps you will dream about HAL - as I often do."
According to J. Bongard, Associate Professor at the Department of Computer Science, University Vermont, "robots equipped with an algorithm that infers their
own physical structure from stored sensory data, dreams of their prior actions, so to speak- perform better in a simple forward locomotion task than robots
whose decisions are not dream-inspired.
Can robots dream?
Furthermore, robots that use these self-models to plan future actions can recover autonomously from injuries, by adapting their gait to compensate for the
In a way, one could say that robots, not totally unlike people do get "tired" at times and have a need for "rest".
Dr. Adami, Professor of Applied Life Sciences at the Keck Graduate Institute believes that a robot might benefit from some "down time" just like people do.
Dr. Adami speculates that if robots were given an alternate state, one in which the robot stopped exploring and instead focused on a problem or obstacle,
it could provide benefits for them just like it provides benefits for human beings.
This raises the question: "How would dream-inspired algorithms work in terra incognita?"
A robot would spend the day exploring part of the landscape, and perhaps be stymied by an obstacle.
At night, the robot would replay its actions and infer a model of the environment. Armed with this model, it could think of-that is, synthesize- actions that
would allow it to overcome the obstacle, perhaps trying out those in particular that would best allow it to understand the nature of the obstacle. Informally,
robot would dream up strategies for success- just as the robot constructed by Bongard dreams about its own shape and form-and approach the morning with fresh ideas.
Although such an algorithm would require far more complex simulations than those giving rise to self-models in the work of Bongard, robots relying on this
kind of navigation could play an interesting role in our quest to understand the nature of consciousness.
Perhas robots dream about landscapes or relaxing activities..?
For example, we ought to be able to record the changes in the robot's artificial brain as it establishes its beliefs and models about the world and itself,
and from those infer not only its cognitive algorithms, but also witness the emergence of a personality.
Thus, perhaps the discipline of experimental robot psychology is not too far off in the future. And even though the robots studied by Bongard seem to prefer
to dream about themselves rather than electric sheep, they just may have unwittingly helped us understand what dreams are for.
So, robots can certainly dream, but their dreams will be slightly unlike ours and who knows may androids do dream of electric sheep after all.