Autonomous mobile robot perception and navigation in indoor
and outdoor environments. Multiple sensors (bump sensors, infra red
proximity sensors, sonar range sensors, cameras, odometry sensors (via
encoders) etc.) are used to acquire perceptions of the current state
of the robot's operating environment. These perceptions are used to
build and dynamically maintain a map. The map is then used to perform
autonomous collision-free navigation.
Multisensor fusion and integration of sensing data received
from various sensors. Bayesian networks are used to perform sensor
fusion and the unified information is stored in a modified occupancy
grid. The modifed occupancy grid serves as the map for the control
and navigation algorithms.
Potential field control theory is used to overlay the
occupancy grid (map) with the potential gradients. The control
algorithm follows the potential gradient to the goal, while performing
obstacle avoidance.
Back to Main Page This page is maintained by Amit Singhal.
Last update: 11/17/97.