综合新闻

硕士生刘备被交通运输领域顶级会议ITSC 2022录用了题为“An Object Detection Method Enhanced by Sparse Point Cloud for Low Illumination in Autonomous Driving”的学术论文。

      论文摘要：Object Detection is an essential task in autonomous driving. The existing object detection methods are susceptible to the variable lighting, which leads to the decrease of detection accuracy. To address the above issue, we introduce LRPN, a robust detection network with multi-modal feature fusion based on prior knowledge of lidar points. LRPN aims to improve the accuracy of object detection in light changing scenes by using fusion method of sparse point cloud and image. We propose two fusion modules: 1) Sparse lidar feature are fused with image in the feature extractor network. 2) Guided anchors are created based on visual projection using lidar points. Our method is applied to Faster RCNN and achieved higher precision under different illumination conditions. Compared with the original Faster RCNN network, the mAP of the method proposed in this paper is raised by 2.2% under good illumination and 3.7% under low illumination. The results show that the introduction of sparse point cloud can significantly improve the detection effect, especially in low illumination scenarios.

硕士生罗钟林被交通运输领域顶级会议ITSC 2022录用了题为“Vehicle-in-the-Loop Intelligent Connected Vehicle Simulation System Based on Vehicle-Road-Cloud Collaboration”的学术论文。

      论文摘要：Autonomous driving test technology is an important guarantee for the large-scale commercialization of autonomous vehicles(AV). The existing test methods are mainly based roads and simulations. Traditional vehicle road test has real traffic environment, but the diversity of test scenarios is limited. It is difficult to customize corner case scenarios safely and efficiently. Simulation testing is flexible and efficient, but the lack of a real traffic flow test environment separates the strong coupling relationship between the vehicle and the environment in practical application scenarios. In view of above, a novel Vehicle-in-the- Loop(ViL) verification method based on vehicle-road-cloud collaboration is proposed in this paper. (1) On the roadside, we propose a road real-time traffic flow element perception method based on monocular camera, and apply the real traffic flow to autonomous driving simulation testing. (2) On cloud platform, we independently develop a simulation platform based on Open- SceneGraph(OSG), which can quickly simulate different weather, lighting and other disturbance factors such as virtual pedestrians and vehicles based on real scenes. (3) On the vehicle, we build a closed-loop test system that combine intelligent connected vehicle(ICV) and mixed environments. This paper takes the autonomous driving obstacle avoidance algorithm in the campus road scene as an example, and completes the system test in the mixed scene. Experiments show that our proposed method can be used as a safer and more efficient test method before autonomous vehicle road test.