Dynamic Dissemination Method for Collective Perception

In future, collective perception might be essential to gather the best possible environmental awareness for an automated driving vehicle. Collective perception means that every partner shares information describing its perceived environment by means of Vehicle-to-X (V2X) communication technology. In order to avoid package loss, channel utilization shall be minimized, which is typically achieved by regulating the transmission frequency. The present work addresses the trade-off concerning optional information and message size. The conducted analyzes investigates the impact of including optional information such as correlation or higher order derivatives on the fusion accuracy. It shows that the impact of correlation on the accuracy is negligible. In contrast, the awareness of higher order derivatives such as acceleration or yaw rate improves the accuracy particularly during accelerating, respectively turning. We propose as trade-off to include higher order derivatives dependent on the estimated mode of the object. The evaluation shows that the proposed trade-off achieves comparable accuracy as disseminating the full state space, while reducing the network load.