View
219
Download
0
Tags:
Embed Size (px)
Citation preview
Autonomous Robots
By: Samantha CatanzaroJessie Mitchell-JemisonMatthew KlingensmithSharon KongAnthony Santos
Introduction
Autonomous robots: intelligent machines capable of performing tasks in the world by themselves, without explicit human control over their movements (Bekey, Autonomous Robots)
Simply, machines that move intelligently, performing actions without direct human control
Examples: humanoids, rovers, cars, rescue, vacuum cleaners, line-following robots, etc.
Technology Three basic components:
Sensors – send input to computers
Computers – interpret input Motors – take signals from
computer and generate actions. Sensors:
RADAR, LIDAR, video, or GPS Most effective robots combine
these sensors into one model. Some sensors, like Velodyne
LIDARs, generate many data points that must be statistically analyzed to remove redundant or meaningless data.
Three types of data analysis Model-based sensor-based statistical combination of models
and sensors.
History of Autonomous Robots
1962: First
industrial arm robot
created for GM
(Unimate)
1966: Standford
creates Shakey,
first robot to know
and respond
to its own actions.
1974: Silver Arm autonomous robot
1st used to assemble
small parts using touch
sensors.
1977: Voyager space probes launch; most
autonomous space probes at
time.
1977 1st Autonomous
Vehicles develop
ed in Japan
Autonomous robotics has improved as computers, sensors, and general technology have advanced.
1979: New
vision system
designed by Hans Moravec (CMU) for greater
autonomy.
1980s: Commerci
al autonomous robots created
for hospitals, etc. using floor plans
to plan routes
1980s: DARPA funds ALV,
autonomous vehicle that uses
laser radar and computer
vision.
1995: Both CMU and
Dickmanns (Germany)
create driverless cars that
travel long distances with little
aid of human drivers.
96-2001: UAVS
become more
autonomous; driverless
cars use new method
to ‘understand’ environment
instead of using radar
(ARGO).
2002 1st
DARPA
Grand Challenge
Pros Automation of tasks Increases efficiency Little or no human input necessary Reduces cost of labor Keeps humans out of hazardous situations
Cons
Deterministic program may have unpredictable results on error
Should still have human observation
High initial costs for development
Group Opinion
Autonomous robots have the potential to be beneficial to society.
They should not replace humans to the extent that humans depend on robots for daily activities, but they should be used in dangerous situations (ex. rescue missions).
Therefore, we support the research of autonomous robots.