Development of the simultaneous mapping system for the robotic fish

This paper describes a mapping system for robotic fish in partially structured environments. Several infrared sensors are used to obtain information about the structures in such environments. An algorithm has been developed to extract object features, together with their uncertainty, from infrared sensors. The obtained information is incorporated into a feature based mapping algorithm running an extended Kalman filter. Simultaneously, robotic fish's position is provided to the feature extraction algorithm to correct the disturbance that the robotic fish motion produces in the mapping process. Moreover, the following procedures which are the build and maintaining the sequence of local maps as well as posteriorly recovering the full global map has been adapted for the presented application. Traditional robots have rigid underlying structures that limit their ability to interact with their environment. However, the robotic fish has the flexible body and the efficient in using power thus the robotic fish is a good selection for simultaneous mapping.