Having an accurate, up-to-date estimation of its pose is crucial for any mobile robot. However, different pose estimators can have vastly different update rates and accuracies. They also provide either relative or absolute poses and work best in certain environments. Thus I propose a thesis with the goal to implement a fused pose estimator, which fuses multiple pose estimations with the goal to estimate an accurate pose at frequencies of 10-100Hz in diverse environments. This fused pose estimator will be integrated into the robotic rover Artemis. Further, the thesis will extend Artemis's current set of pose estimators --- consisting of wheel odometry, an IMU and LiDaR map localization --- by integrating a visual odometry component. This will allow the fused pose estimator to also leverage poses acquired using Artemis's cameras. Finally, the thesis shall evaluate the accuracy of the fused pose estimator in real-world experiments compared to ground-truth values.
Development and Analysis of a Fused Pose Estimator and a Visual Odometry Component for a Robotic Rover
In der Regel sind die Vorträge Teil von Lehrveranstaltungsreihen der Universität Bremen und nicht frei zugänglich. Bei Interesse wird um Rücksprache mit dem Sekretariat unter sek-ric(at)dfki.de gebeten.