Accurate 2D Localization for Mobile Robot by Multi-sensor Fusion

To achieve fully autonomous operation for indoor Mobile Robot. Localization solution has become an active research topic in the past decade. Normally a localization system is required to be reliable, fast, and have high accuracy. But single UWB measurement is very discrete and you can not get the exact orientation of the mobile robot. To overcome these shortcomings, we propose a novel UWB/Gyroscope/Odometer-integrated system, in which UWB measures the distance between the tag and the base station through Time of Flight and then calculating the tag position through the triangulation algorithm. Gyroscope and odometer can get the angular and linear velocity of mobile robot respectively. Which can provide accurate short-term tracking performance. The results of multi-sensor are fused using an Extended Kalman Filter (EKF). The experiment results also show that the algorithm significantly improved accuracy and get reliable yaw angle.