Authors
Result type
conference paper
Description
LIDAR (2D) has been widely used for mapping and
navigation in mobile robotics. However, its usage is limited to
simple environments. This problem can be solved by adding more
sensors and processing these data together. This paper explores a
method of how measurements from a stereo camera and LIDAR
complement each other in dynamical mapping. An occupancy
grid map from LIDAR data is used as a prerequisite and we
describe how it could be extended by a 2D grid map from
a stereo camera. This approach is based on the ground plane
estimation in disparity map acquired from the stereo vision. For
ground plane detection, RANSAC and Least Squares methods
are used. After obstacles determination, 2D occupancy map is
generated for future fusion with complementary 2D LIDAR map.
Experimental results obtained from Willow Garage Stereo and
Hokuyo UTM-30LX Laser are good enough to determine that
this method is a benefit, although my implementation is still a
prototype. In this paper, we present the applied methods, analyze
the results and discuss the modifications and possible extensions
to get better results.
navigation in mobile robotics. However, its usage is limited to
simple environments. This problem can be solved by adding more
sensors and processing these data together. This paper explores a
method of how measurements from a stereo camera and LIDAR
complement each other in dynamical mapping. An occupancy
grid map from LIDAR data is used as a prerequisite and we
describe how it could be extended by a 2D grid map from
a stereo camera. This approach is based on the ground plane
estimation in disparity map acquired from the stereo vision. For
ground plane detection, RANSAC and Least Squares methods
are used. After obstacles determination, 2D occupancy map is
generated for future fusion with complementary 2D LIDAR map.
Experimental results obtained from Willow Garage Stereo and
Hokuyo UTM-30LX Laser are good enough to determine that
this method is a benefit, although my implementation is still a
prototype. In this paper, we present the applied methods, analyze
the results and discuss the modifications and possible extensions
to get better results.
