Detecting High-Level Features by Multisensor Fusion
Robust sensing of the environment of a mobile robot is an important task both in localizing the robot and in building a complete and reliable map of such an environment. One of the fundamental ideas to achieve this robustness is the use of redundancy, that is, to combine environmental information obtained by several sensors. Such an approach provides more reliable and accurate information about what the sensory system of the robot really observes. Credibility of the observed features is therefore enhanced. Dealing with redundancy requires both the availability of a robust modeling tool to represent uncertain geometric information and a multisensor fusion mechanism capable of handling information obtained by different sensors.
KeywordsReference Frame Mobile Robot High Level Feature High Angular Resolution Perturbation Vector
Unable to display preview. Download preview PDF.