2022 | OriginalPaper | Buchkapitel
Low-latency Probabilistic Collision Detection Method for C-V2X Applications
verfasst von : Ryu Yachikojima, Shin’ichi Arakawa, Takeshi Kitahara, Nagao Ogino, Go Hasegawa, Masayuki Murata
Erschienen in: Commercial Vehicle Technology 2022
Verlag: Springer Fachmedien Wiesbaden
Aktivieren Sie unsere intelligente Suche, um passende Fachinhalte oder Patente zu finden.
Wählen Sie Textabschnitte aus um mit Künstlicher Intelligenz passenden Patente zu finden. powered by
Markieren Sie Textabschnitte, um KI-gestützt weitere passende Inhalte zu finden. powered by
A key to achieving advanced safety in autonomous vehicles is to perceive and understand the surrounding the environment using many sensors equipped in the vehicle. However, because the sensors rely on the visibility of the vehicle, there are limitations to understanding the environment. Therefore, dynamic information sharing among vehicles is important to achieve sophisticated safety, in addition to the self-perception of the vehicle. Cellular vehicle-to-everything (C-V2X) using multi-access edge computing has attracted attention as a method to share information among vehicles and provide centralized safety validation of traffic. Particularly, in safety uses, such as collision detection at an intersection, it is essential to predict the uncertain position of vehicles with a probabilistic process, such as a Markov chain. However, the calculation time of the existing method is too large to meet real-time requirements. In this paper, we developed a probabilistic collision detection method for an edge computing environment with a cellular system. For this purpose, we modified the existing probabilistic collision detection method. We reduced the calculation time to predict the probability distribution of a vehicle state by utilizing the prediction error between the current state and predicted state.