[spb_text_block pb_margin_bottom=”no” pb_border_bottom=”no” width=”1/1″ el_position=”first last”]
Projection mapping has been used for lots of applications in many industries, but I never would have thought of this one. Some mad scientists over at MIT have used projection mapping to visualize the thoughts of robots.
MIT’s Aerospace Controls Lab invents algorithms for robots to navigate unknown terrain or communicate and collaborate in roboty tasks. The trouble is that it is difficult to visualize why an algorithm may be working. What is the robot thinking? Why did it send that message or choose to move over there?
[/spb_text_block] [spb_single_image image=”1556″ image_size=”full” frame=”noframe” intro_animation=”none” full_width=”no” lightbox=”yes” link_target=”_self” width=”1/1″ el_position=”first last”] [spb_text_block pb_margin_bottom=”no” pb_border_bottom=”no” width=”1/1″ el_position=”first last”]
“Normally, a robot may make some decision, but you can’t quite tell what’s going on in its mind — why it’s choosing a particular path,” Agha-mohammadi says. “But if you can see the robot’s plan projected on the ground, you can connect what it perceives with what it does to make sense of its actions.”
[/spb_text_block] [spb_single_image image=”1558″ image_size=”full” frame=”noframe” intro_animation=”none” full_width=”no” lightbox=”yes” link_target=”_self” width=”1/1″ el_position=”first last”] [spb_text_block pb_margin_bottom=”no” pb_border_bottom=”no” width=”1/1″ el_position=”first last”]
To visualize these hidden decisions, the team created a real-time projection mapped playground for their robots. They used a system of 18 motion-capture camers (looks like either NaturalPoint or Vicon) to precisely track the position of the robot in the room. Then overhead projectors display the information onto the floor of the room.
You may have noticed a similar setup (Motion-tracking + projectors) in the super popular Omote video, along with the Miley Cyrus world tour by VYV.
[/spb_text_block] [spb_single_image image=”1163″ image_size=”full” frame=”noframe” intro_animation=”none” full_width=”no” lightbox=”yes” image_link=”http://projection-mapping.org/omote/” link_target=”_self” width=”1/2″ el_position=”first”] [spb_single_image image=”884″ image_size=”full” frame=”noframe” intro_animation=”none” full_width=”no” lightbox=”yes” image_link=”http://projection-mapping.org/miley-cyrus/” link_target=”_self” width=”1/2″ el_position=”last”] [spb_text_block pb_margin_bottom=”no” pb_border_bottom=”no” width=”1/1″ el_position=”first last”]
A cute benefit of their system, the projected setup enables the researchers to simulate outdoor flying drone scenarios without worrying about the FAA. For instance, projecting a forest landscape as a backdrop to investigating algorithms for fighting forest fires.
[/spb_text_block] [spb_text_block pb_margin_bottom=”no” pb_border_bottom=”no” width=”1/1″ el_position=”first last”]
This project reminds us of the Romo Cart project by Ken Kawamoto which we posted on last month. What other simulation/training environments do you want to see projection mapping help with? Hit up the comments!
[/spb_text_block] [spb_text_block pb_margin_bottom=”no” pb_border_bottom=”no” width=”1/1″ el_position=”first last”]
Via: [sf_button colour=”accent” type=”standard” size=”standard” link=”https://newsoffice.mit.edu/2014/system-shows-robot-intentions-1029″ target=”_blank” icon=”ss-oven” dropshadow=”no” extraclass=””]MIT News[/sf_button]
[/spb_text_block] [spb_single_image image=”1557″ image_size=”full” frame=”noframe” intro_animation=”none” full_width=”no” lightbox=”yes” link_target=”_self” width=”1/1″ el_position=”first last”]