A RTLS/DR based localization system architecture for indoor mobile robots

Abstract

In this paper, aiming for mobile robot localization in indoor environments, a loosely coupled architecture, which combines indoor Real-time Locating System (RTLS) and dead reckoning (DR) system, is proposed. Within this architecture, the DR is simple and cost effective but subjected to accumulative error; the indoor RTLS can provide absolute, consistent localization estimates, therefore is complementary to DR sensors. The most outstanding feature of this architecture is that it can easily integrate any type of existing RTLS and DR sensors. Another advantage lies in the fact that this architecture adopts a linear model, thus is exempted from the risk of linearization of non-linear models. In order to verify the proposed architecture, an example localization system consisting of an ultra-wideband (UWB) RTLS system and wheel encoders based DR, on a real robot, is implemented. Experiments have been conducted on this example system and the results have demonstrated its performance.