Adaptive Localization and Mapping for Planetary Rovers
Javier Hidalgo Carrió
2018. Universität Bremen.
Future rovers will be equipped with substantial onboard autonomy as space agencies and industry proceed with missions studies and technology development in preparation for the next planetary exploration missions. Simultaneous Localization and Mapping (SLAM) is a fundamental part of autonomous capabilities and has close connections to robot perception, planning and control. SLAM positively affects rover operations and mission success. The SLAM community has made great progress in the last decade by enabling real world solutions in terrestrial applications and is nowadays addressing important challenges in robust performance, scalability, high-level understanding, resources awareness and domain adaptation. In this thesis, an adaptive SLAM system is proposed in order to improve rover navigation performance and demand. This research presents a novel localization and mapping solution following a bottom-up approach. It starts with an Attitude and Heading Reference System (AHRS), continues with a 3D odometry dead reckoning solution and builds up to a full graph optimization scheme which uses visual odometry and takes into account rover traction performance, bringing scalability to modern SLAM solutions. A design procedure is presented in order to incorporate inertial sensors into the AHRS. The procedure follows three steps: error characterization, model derivation and filter design. A complete kinematics model of the rover locomotion subsystem is developed in order to improve the wheel odometry solution. Consequently, the parametric model predicts delta poses by solving a system of equations with weighed least squares. In addition, an odometry error model is learned using Gaussian processes (GPs) in order to predict non-systematic errors induced by poor traction of the rover with the terrain. The odometry error model complements the parametric solution by adding an estimation of the error. The gained information serves to adapt the localization and mapping solution to the current navigation demands (domain adaptation). The adaptivity strategy is designed to adjust the visual odometry computational load (active perception) and to influence the optimization back-end by including highly informative keyframes in the graph (adaptive information gain). Following this strategy, the solution is adapted to the navigation demands, providing an adaptive SLAM system driven by the navigation performance and conditions of the interaction with the terrain. The proposed methodology is experimentally verified on a representative planetary rover under realistic field test scenarios. This thesis introduces a modern SLAM system which adapts the estimated pose and map to the predicted error. The system maintains accuracy with fewer nodes, taking the best of both wheel and visual methods in a consistent graph-based smoothing approach.
SLAM, space robotics, localization, mapping, machine learning