Model-free Approach for Sensor Network Localization with Noisy Distance Measurement

A model-free localization method with noisy distance measurement is proposed for estimating a moving robot in 3D space. Considering that the traditional filter-based sensor network localization algorithms can not provide acceptable estimation accuracy in altitude in 3D space, the proposed method utilizes not only current measurements but also previous measurements to localize a robot. This character adds more constraints to localization to avoid local minimum. In addition, different from the traditional filter-based localization methods which need kinetic model for localization, our proposed method is model-free and converts the localization problem to graph optimization problem. The advantage is that we avoid the possible estimation error caused by inaccurate or simplified kinetic model. Considering that the communication limitation in application makes many graph optimization theories such as distributed localization theory and trilateration difficult to be realized, our method proposes to add constrained equation between adjacent positions to solve this problem. Experiments under a variety of scenarios verify the stability of this method and show that the algorithm achieves better localization accuracy than filter-based methods.