Collect useful sentences in Papers (8)

In the last years cooperative robotics has received considerable attention.

Teams of robots are able to overcome
the limitations of single robots and allow to attack more
difficult tasks. 

New challenges appear providing new research
areas. 

As compared to vision-based
SLAM ([3], [18]), stereo matching and reconstruction ([4]),
ground-plane based navigation, or other depth reconstruction techniques, extracting bearings is a simple process,
close to obtaining raw data.

In order to take into account heterogeneous pairs of robots and possible hardware
failures we have implemented our localization algorithm
for two different configurations.

 Many
of the proposed tracking methods do not provide pose
estimation, due to the fact that feature models do not
contain structure information.

cooperative localization has been
an active research area during the last years in the
robotic community. 

There exist different approaches depending on several aspects as the information used
(map availability, static/dynamic features) or the multirobot architecture (communication capabilities, centralized/decentralized).

In
the absence of static references these methods estimate the
relative pose among the robots instead of their location on a  global reference system. However, all these algorithms use
range-bearing sensors that provide an accurate estimation
of their relative positions.

Our method, instead of restricting
the motion of the robots with a known model, uses the
measurements of their displacements to turn the problem
observable. 


Reference:

Montesano, L., Gaspar, J., Santos-Victor, J., & Montano, L. (2005, August). Cooperative localization by fusing vision-based bearing measurements and motion. In Intelligent Robots and Systems, 2005.(IROS 2005). 2005 IEEE/RSJ International Conference on (pp. 2333-2338). IEEE.