Optimal filtering in autonomous driving systems
组织者
演讲者
李升波
时间
2024年05月08日 14:30 至 15:30
地点
理科楼A-304
摘要
In autonomous vehicles, optimal filtering plays a crucial role in environment perception. Traditional pipeline mainly relies on a modular design, where raw sensor measurements are processed sequentially to estimate the states of the ego vehicle and surrounding vehicles. This approach may lead to the loss of critical environmental information. This talk will introduce an alternative strategy that directly maps sensor measurements to traffic states, bypassing the intermediate modular processing steps. By doing so, we aim to enhance the accuracy and efficiency of environment perception in autonomous vehicles. This talk will share some preliminary investigation into this topic, highlighting the benefits and challenges associated with sensor-to-state mapping.