No abstract

LiDAR-based Multiple Object Detection and Tracking (MODT) is one of the essential tasks in autonomous driving. Since MODT is directly related to the safety of an autonomous vehicle, it is critical to provide reliable information about the surrounding objects. For that reason, we propose a semantic point cloud-based adaptive MODT system for autonomous driving. Semantic point clouds emerge with advances in deep learning-based Point Cloud Semantic Segmentation (PCSS), which assigns semantic information to each point in the point cloud of LiDAR. This semantic information provides several advantages to the MODT system. First, any point corresponding to any static object can be filtered. Because the class information assigned to each point can be directly utilized, filtering is possible without any modeling. Second, the class information of an object can be inferred without any special classification process because the class information is provided from the semantic point cloud. Finally, the clustering and tracking module can consider unique dimensional and dynamic characteristics based on class information. We utilize the Carla simulator and KITTI dataset to verify our method by comparing several existing algorithms. In conclusion, the performance of the proposed algorithm is improved by about 176% on average compared to the existing algorithm. INDEX TERMSSemantic point cloud, point cloud semantic segmentation, multiple object detection and tracking (MODT), class-adaptive tracking, autonomous vehicle, LiDAR

No abstract

The irradiation angle of the headlights is regulated by the United Nations Economic Commission for Europe (UNECE) for safe driving by preventing glare from oncoming drivers. Automatic headlamp leveling is a system that automatically controls the angle of the headlamp horizontally to meet safety regulations. To control the irradiation angle, the automatic headlamp leveling system must estimate the roadto-headlamp pitch angle. This paper proposes an accurate and robust road-to-headlamp pitch angle estimation system using a low-cost MEMS (Micro Electro Mechanical Systems) IMU (Inertial Measurement Unit) sensor. The proposed system sequentially estimates the headlamp pitch angle, the road slope angle, and the road-to-headlamp pitch angle. First, headlamp pitch angle is estimated using the gravity from IMU. Second, the road slope angle is estimated based on vehicle prior knowledge that the vehicle is moving in the direction of the road. Finally, road-to-headlamp pitch angle is estimated by subtracting road slope angle from headlamp pitch angle. However, the measurement data of a MEMS IMU sensor contains various types of noise, such as bias and white noise. For that reason, reducing the effects of MEMS IMU noise is highly important when estimating the road-to-headlamp pitch angle. Therefore, the proposed system focuses on minimizing the effects of the low-cost MEMS IMU noise. In addition, the proposed method is verified by simulations and experiments. The simulation analyzes the effect of the MEMS IMU sensor noise on the estimation accuracy and precision. Finally, the proposed algorithm shows more robust and accurate performance compared with other methods on the test and common road verifications.INDEX TERMS Automatic headlamp leveling, inertial measurement unit (IMU), road to headlamp pitch angle estimation, road slope estimation.