Rodney Brooks had it right in his 1991 paper “Intelligence without Reason.”  His approach to Artificial Intelligence is based an emergence of behaviors not explicitly programmed into a system.  Instead, a hierarchy of discrete behaviors is organized in such a way that higher-level behaviors subsume the resources required by lower-priority behaviors when appropriate conditions are true.  The rapid switching between behaviors causes an emergence of behaviors that transcend the collective behaviors that are explicitly programmed.

When I first applied this subsumption architecture to my robot it behaved unpredictably at the micro level but predictably at the macro level.  I built a maze in my living room from 1 x 8 boards lying on their side and released my robot at one end.  The goal was for my robot “Beto” to find the exit on the other side of the 12′ x 12′ labyrinth of wooden walls.  The behaviors I programmed were simple.  From highest priority to lowest, the behaviors were:

1. When bumper switch is touched on right, stop, turn left.
2. When bumper switch is touched on left or in front, turn right.
3. When IR sensor sees something on right, turn left.
4. When IR sensor sees something in front or left, turn right.
5. Unconditionally drive forward while arcing to the right.

Each behavior was responsible for one single, simple thing.  They each ran as discrete processes and each monitored its inputs.  When their respective behavior condition was true, each took control of the actuators and performed its action.  Higher priority behaviors could subsume actuator resources from lower-priority behaviors.

The result was fascinating.  I had deliberately designed a long narrow corridor in the maze in order to try to confuse the robot.  The robot drove right down the middle of the corridor, slowed before reaching the end wall, stopped for about a second, turned 180 degrees and proceeded out the way it came from.  None of those behaviors were programmed into the robot but the rapid switching between the simple few behaviors caused this complex behavior to emerge.

I spent some time decomposing this kind of emergent behavior and was never able to completely and confidently explain every nuance; however, it was obvious that the emergent behaviors came from the hard-coded behaviors operating at a few milliseconds at a time–switching tens or hundreds of times a second, impacting motor speeds, voltage levels and performing logic.  With these simple few behaviors, a lot of the analysis was speculation and I quickly concluded that in order to create more organic-behaving robots, I had to just let go.

The robot successfully navigated his way through a different maze layout every time–validating another of Dr. Brooks’s tenets that robots should be able to react to a dynamic and changing world environment.

In Mark Buchanan’s article, “Law and Disorder,” Buchanan shares a case in General Motors when in 1992 the company was struggling to optimize schedule of the robots that automatically painted trucks coming off the assembly line.  GM’s Dick Morley suggested that the robots should be left to determine their own painting schedules.

Morley put out some simple rules for each machine where each would “bid” for new jobs with an unconditional desire to stay busy.  “The results were remarkable, if a little weird.  The system saved General Motors more than $1 million each year in paint alone.  Yet the line ran to a schedule that no one could predict, made up on the fly by the machines themselves as they responded to emerging needs.”

Trying to decompose any emergent behavior down to its hard-coded constituent behaviors is possible but the answer is inherently elusive because it’s always moving.  It’s like trying to reverse engineer the behavior of a rand() function.

Steven Wolfram is yet another from this behavior-based camp.  In his book, “A New Kind of Science,” he argues that the rules in nature aren’t necessarily limited to traditional mathematics.  Instead, he suggests that more general rules in nature can be embodied in computer programs.

In each of these the proposal was deemed “controversial.”  Einstein scoffed at the the theory of spooky and unpredictable quantum behaviors.  He declared that “God does not play dice with the universe.”  It seems more and more plausible now that the tiny strings vibrating in the elements of the atom’s nucleus are the basic behaviors from which everything has emerged.

As robotics and Artificial Intelligence enter the dawn of the Singularity, the complexities manifesting from the core threads of behaviors will become as unpredictable as humans.  All we need to do is to let go.

References:

• Intelligence Without Reason
• A New Kind of Science
• Why complex systems do better without us, Mark Buchanan, NewScientist, August 6, 2008