Autonomous driving (auto-driving) aims at reducing car accidents, traffic congestion, and greenhouse gas emissions by automating the transportation process. The potential impact of the cross-disciplinary technology has attracted heavy R&D investments not only by leading car manufactures (e.g., Toyota, Tesla, BMW) but also Internet companies (e.g., Google, Apple, Baidu). As one of the key operations in auto-driving, the vehicular sensing acts as a basic and important technique for auto-vehicles that senses all vehicles (including line-of-sight (LoS) and non-LoS (NLoS) vehicles) via accurately detecting their positions, driving-directions, sizes, velocities, trajectories, etc. The information then serves as inputs for computing and control tasks, such as driving-path planning, navigation, and accidence avoidance. However, notice that the state-of-art vehicular sensing techniques, such as LIDAR, Camera, and RADAR, are incapable of detecting hidden vehicles without LoS. Our initial work on this areas focuses on sensing hidden vehicles by exploiting asynchronous multi-path vehicle-to-vehicle (V2V) transmission via jointly leveraging signal processing and optimization theory [VS1].