A cost efficient multi remote driver selection for remote operated vehicles

Abstract

Technological, social and economic forces are changing the notion of mobility as culture is exerting strong pressure on demand for a “shared,” “reliable” and “on-demand” economy. In an era of the next mobility revolution, autonomous vehicles (AV) still fail to overcome edge cases that can stop an AV in their tracks. Remote driving has, however, emerged as a practical and promising solution to help AV in edge cases, such as AVs passing through bad weather or near a construction site. Remote driving relies on communication network infrastructure, which presents many challenges, both in terms of latency and performance, in connectivity between road vehicles and the remote driver. In addition, governments are passing laws to test AV technology without a safety driver, which has sparked the need for a remote driver who can take vehicle control if it encounters an edge case. In this paper, we enhance our previous work and present algorithms to select multiple remote drivers under user-defined objectives, to minimize the distance between the vehicle and remoter drivers. To evaluate the performance of the proposed algorithm, we perform extensive simulations both in a simulated environment using synthetic data and in a real environment with real-world data. We compare the performance of our proposed algorithms with baseline algorithms. The results show that the proposed heuristic algorithms perform similar to the optimal baseline line algorithms with much shorter computation time.