A Distributed Remote Driver Selection for Cost Efficient and Safe Driver Handover

Advances in autonomous or self-driving automobile can revolutionize intelligent transportation. But their safety is widely open to debates as failures are anticipated, because the technology is nascent. In this paper we address this concern by remote driving. A process which brings safety mechanism, allowing autonomous automobiles to be monitored and controlled by a remote driver. Remote driving relies on real-time transfer of data from vehicle to the remote driver by utilizing communication network infrastructure, prone to latency. Under ideal conditions, with the use of multiple remote drivers, many latency related factors between driver and vehicle for safe control can be reduced. However, frequent remote driver change or handover may cause vulnerability due to possible conflict. In this paper, we propose greedy algorithms to select minimum number of remote driver(s) at a minimum distance away from automobile under given distance constraint. This reduces the number of driver hand-overs, thus increasing the reliability of remote operation and safety of automobile. Through extensive simulations, we show that the proposed algorithms perform similar to the optimal exhaustive search algorithm, with much faster computation time.