Multi-Sensors Based Simultaneous Mapping and Global Pose Optimization

In recent years, 3D Lidar Based Simultaneous Localization and Mapping (SLAM) has become a hotspot in the research of the autonomous vehicle navigation. In this paper, we proposed a multi-sensors based 3D SLAM system to get more accurate and robust SLAM results in real time. Firstly, the information from different sensors is added into the data-association to reduce the cumulative error of pose. Then, we use the IMU (Inertial Measurement Unit) information complementary filtering to improve the accuracy and efficiency of scan matching and loop closure detection. What’s more, a two-layer loop closure detection algorithm is proposed in our research to improve the real-time performance of the loop closure detection. We have carried a lot of experiments and the experimental results shows that our method is more accurate and better in real-time performance compared with the state-of-the-art algorithm.