Application of adaptive streaming technology in remotely driven electric vehicles

Abstract

The need for live video transfer across networks is increasingly crucial for modern day applications, ranging from video-based surveillance systems to disarming explosives in a minefield using robots. Some of these applications, for e.g., a tele-operated electric vehicle (EV) as focused in this paper, involve real-time video streaming across wireless networks, which is bandwidth-intensive and delay-sensitive at the same time. This necessitates the need for a mechanism to adapt the data rate in order to suit the network condition and to ensure hassle-free functioning of the intended application. This paper proposes an adaptive streaming mechanism which involves bandwidth estimation based on transmission feedback and systematic adaptation of frame rate and frame quality for a safe drive in a remotely driven EV. The algorithm is implemented in a physical tele-operated EV and experiments are carried out to verify its adaptation performance.