Simultaneous localization and mapping (SLAM) is a prerequisite for accurate navigation of autonomous vehicles. Although this is often safety-critical, systematic approaches for testing the correctness and accuracy of SLAM algorithms are missing. In this paper, we present an approach for automated and systematic testing of SLAM algorithms. We identify challenging environmental features for SLAM, define coverage criteria that characterize the SLAM problem’s input space, and develop a method for automatically generating high-coverage tests. We demonstrate the effectiveness of our approach with a case study on an existing FastSLAM implementation.