SonarBased RealWorld Mapping and Navigation

1
Sonar-Based Real-World Mapping and Navigation

• by
• ALBERTO ELFES
• Presenter
• Uday Rajanna

2

• Involves a sonar based mapping and navigation
  system developed for an autonomous mobile robot
  operating in unknown or unstructured
  environments.
• An active range finding device based on an
  ultrasonic range transducer to build two
  dimensional maps of the robots surroundings.
• Each sonar distance reading provides information
  concerning empty and occupied volumes in a cone
  in front of the sensor.
• These empty and occupied readings are
  interpreted as empty and occupied probability
  distribution functions
• Range measurements from multiple points of view
  taken from multiple sensors are integrated into
  the sonar map. These measurements are
  constantly updated with the previous map to
  produce new maps, thereby increasing accuracy of
  detected probabilities of empty and occupied
  volumes.

3
(No Transcript)
4
(No Transcript)
5

• R -gt range measurement returned by the sonar
  sensor
• e -gt maximum sonar measurement error
• ? -gt sensor bandwidth
• O -gt solid angle subtending the main lobe of the
  sensitivity function.
• S -gt position of the sonar sensor
• d -gt distance from P to S
• ? -gt angle between the main axis of the beam and
  P as seen from S
• Probably Empty Region -gt pE
• Somewhere Occupied Region -gt pO

6
(No Transcript)
7
(No Transcript)
8
(No Transcript)
9
(No Transcript)
10

• Dolphin Architecture
• Multiple axis of representation of sonar mapping
  information
• (a) The Abstraction Axis
• Probabilistic representation, Geometric
  Level, Symbolic Level
• (b) The Geographical Axis
• View, Local Map, Global Map
• (c) The Resolution Axis
• Sensor-level local map and generate
  progression of maps with increasingly
  less detail

11
(No Transcript)
12
(No Transcript)
13
(No Transcript)
14
(No Transcript)
15

• Conclusions
• A Sonar based mapping and navigation system that
  proves to be able to operate in unknown and
  unstructured environments.
• This method provides a way of explicitly
  describing unknown, empty, and occupied areas.
  This can be combined with other robust techniques
  to perform effective navigation.
• This method combines well with the Matching
  technique described in the paper to determine
  the robots estimate of its position and
  orientation and also as a way to detect
  landmarks.