The Effect of Features Number on Extended Observation-Cooperative SLAM

This paper investigates the effect of the number of environment features on a cooperative approach of simultaneous localization and mapping (SLAM). The tested cooperative SLAM approach is the Extended Observation-Cooperative SLAM (EO-CSLAM) algorithm which depends on additional, indirect correlated observations of the features (landmarks). The performance gain due to additional correlated observations means that additional features will have similar positive effect. However, as EO-CSLAM adopts extended Kalman filter-simultaneous localization and mapping (EKF-SLAM) solution, the number of environment features will have an important role in the computational burden. Simulation results show that the performance gain provided by EO-CSLAM is more obvious in tha less features cases.