Development of a Reference Architecture for streaming of Cloud infotainment system to In-Car Thin clients

Abstract

The gaming industry is rolling out new gaming solutions, where games are hosted and rendered in the cloud and streamed to thin clients in 60 fps with 4K resolution. This technology enables new possibilities in extending the lifetime of existing hardware, simplifying software deployment to multiple clients, and enabling data analytics. These are topics that are of great interest to the automotive industry, where vehicles are typically in use for ten years or more. By providing solutions similar to the gaming industry, the automotive sector could address these issues. The expectation is that car users should be able to get the same gaming experience in an infotainment system as with playing it on the phone. But, it is not possible currently due to certain limitations. The video streaming from the phone to the car head unit requires devices in the head unit. Even having such a device in head-unit will not guarantee a high-quality, low latency video gaming experience because of lag issues, as when you stream your phone screen, a lot of operations are involved. Using the design science research methodology, we elicited the functional and non-functional requirements required to develop and design the architecture through semi-structured interviews. Based on the data collected from the interview and the literature study, AGL’s technical feasibility for real-time low latency content streaming was reviewed. Architectural patterns and tactics were used to design the AGL architecture, which supports streaming in both online and offline mode. Architecture evaluation using a semi-structured interview was adopted to evaluate the functional and quality attributes identified for the architecture developed.