Szabolcs Vajna

Tamás Matuszka, Iván Barton, Ádám Butykai et al.

Autonomous driving is a popular research area within the computer vision research community. Since autonomous vehicles are highly safety-critical, ensuring robustness is essential for real-world deployment. While several public multimodal datasets are accessible, they mainly comprise two sensor modalities (camera, LiDAR) which are not well suited for adverse weather. In addition, they lack far-range annotations, making it harder to train neural networks that are the base of a highway assistant function of an autonomous vehicle. Therefore, we introduce a multimodal dataset for robust autonomous driving with long-range perception. The dataset consists of 176 scenes with synchronized and calibrated LiDAR, camera, and radar sensors covering a 360-degree field of view. The collected data was captured in highway, urban, and suburban areas during daytime, night, and rain and is annotated with 3D bounding boxes with consistent identifiers across frames. Furthermore, we trained unimodal and multimodal baseline models for 3D object detection. Data are available at \url{https://github.com/aimotive/aimotive_dataset}.

Michal Čáp, Szabolcs Vajna, Emilio Frazzoli

Vehicle sharing system consists of a fleet of vehicles (usually bikes or cars) that can be rented at one station and returned at another station. We study how to achieve guaranteed service availability in such systems. Specifically, we are interested in determining a) the fleet size and initial allocation of vehicles to stations and b) the minimum capacity of each station needed to guarantee that a) every customer will find an available vehicle at the origin station and b) the customer will find a free parking spot at the destination station. We model the evolution of number of vehicles at each station as a stochastic process and prove that the relevant probabilities in the system can be approximated from above using a computationally-tractable decoupled model. This property can be exploited to efficiently determine the size of fleet, initial distribution of vehicles to stations, and station capacities that are sufficient to achieve the desired service level. The applicability of the method is demonstrated by computing the initial vehicle stock and the capacity of each station that would be needed to avoid service failures in Boston's bike sharing system "The Hubway". Our simulation shows that the proposed method is able to find more efficient design parameters than the naive approach and consequently it can achieve the equivalent quality-of-service level with half of the vehicle fleet and half of the parking capacity.