Blog Post #386:  The Beaver Robot did it!!!

Blog Post #386:  The Beaver Robot did it!!!

Taken from:  http://www.buffalo.edu/news/releases/2018/06/017.html

Artificial intelligence is stretching out in almost all directions.  Maybe one of the unusual applications is making a robot that does machine learning that is like a beaver.  Seemingly beavers build their dams without a blueprint. (They also don’t go to Home Depot or Lowes to get their wood products).  The beavers seemingly have the main concept of stopping the flow or water. Each pond is different – different rocks, terrain, and even water flow.  The beaver has to adjust to the terrain to make the dam stop the water. And to continue to monitor the environment around them in case of a dam break.  The article says that there is a randomness in nature that the beaver has to adjust to when building a dam.

Likewise trying to develop artificial intelligence applications to work through random environment constraints could be important.  One first concept might occur in search and rescue operations – like after a tornado. After a tornado, the rescue team might need help in finding people buried under debris.  The randomness of what is left after a tornado calls for human rescuers to analyze the situation and to find a path around holes in the ground (maybe even caved-in basements), downed trees, appliances thrown around, and unconscious people.  Human rescuers would use their vision to find a path to when the unconscious person might be buried under the debris. But that might be tricky or even dangerous for a human such as falling down a well or their weight disrupting a precarious balance of boards and bricks.

The solution is to develop an artificial intelligence robot that can work its way around the random destruction.  

The article describes it this way: “Using off-the-shelf components, [the researchers] outfitted a mini-rover vehicle with a camera, custom software and a robotic arm to lift and deposit objects.

“They then created uneven terrain — randomly placed rocks, bricks and broken bits of concrete — to simulate an environment after a disaster such as a tornado or earthquake. The team also places hand-sized bean bags of different sizes around the simulated disaster area.

“Researchers then activate the robot, which uses the algorithms to continuously monitor and scan its environment. It picks up bean bags and deposits them in holes and gaps in between the rock, brick and concrete. Eventually the bags form a ramp, which allows the robot to overcome the obstacles and reach its target location, a flat platform.”

I’m impressed.  Programming a robotic device to adjust to obstacles and to find a path through and around them could be important in search and rescue operations.  Maybe I can program such a robot to find my car keys when I have misplaced them!!

What do you think?

Bruce

 

Posted by Bruce White

Leave a Reply