Connected Vehicles: Can Network Slicing Make it Possible?

The 5G Automotive Association (5GAA) is creating a cross-industry platform to promote and harmonize 5G connectivity and use cases that include connected vehicles, driverless/autonomous cars, and intelligent transportation for smart cities. This newly formed ecosystem, that includes leading automakers, technology giants, and mobility service providers, is aiming to start an era of fully automated and connected vehicles. At the same time, telcos and telco vendors are considering 5G to eventually become a unifying automotive connectivity standard, and are working on designing diversified relevant automotive use cases. For example, AT&T has announced collaborations with more than 20 automotive manufacturers, including BMW, Audi, Chevrolet, and GMC, to develop and test connected and autonomous vehicles.

But questions remain: How can these collaborations drive the growth of advanced automotive technologies? When can we expect autonomous vehicles on the road? Will car manufacturers wait for 5G and network slicing to appear or will they use telco networks as a utility?

Telcos, vendors, and automakers have different requirements and opinions when discussing autonomous vehicle technology. If the market is to reach a critical mass, it will eventually be critical to harmonize these discussions and bridge the gaps between these distinct technical interpretations of the same concept. Driverless cars will require single-digit-millisecond latencies and, in theory, only 5G with network slicing can make this use case work effectively by partitioning a single network into multiple virtual networks tailored to each individual use case.

The acceptance rate for driverless cars is expected to be very low initially because of concerns about safety, but as standardization bodies (such as EAMA, 3GPP, ETSI ITS, and GSMA) have initiated issuing the standards for autonomous vehicles, the trend is expected to go mainstream within the next 4 to 5 years. Recently, in August 2017, BMW Group announced plans to build a fleet of 100 fully autonomous cars by the end of 2017 in collaboration with Intel and Intel-owned Mobileye, and other leading automakers are making similar announcements. It is clear that self-driving connected cars will reshape the automotive, parking, and insurance industries, and this revolution will be enabled by the integration of various technologies, such as edge computing, artificial intelligence, 5G, IoT (for V2X), and short-range wireless communications.

As we continue down the path of autonomous vehicles using the potential of 5G connectivity and network slices, expect drastic technological progression to bring the paradigm shift that can be obtained by designing the following automotive-grade capabilities:

• V2X for positioning accuracy and cooperative mobility; and cloud/Internet connectivity for smoothing out data transfer and communication between connected internal/external devices.
• Ultra-Low Latency and Faster Data Transmission Speeds for ensuring timely responsiveness and interoperation between all automobile system components for both mission-critical and regular functions.
• Mission-Critical Reliability attained by leveraging cellular and Wi-Fi hyper-connectivity applications for supporting uninterrupted infotainment, security, and driving assistance systems.
• Car-to-Cloud Vehicle Sensor Data Crowdsourcing for real-time and future analysis of granular-level mobility data to understand the responsiveness of elements in automated vehicle ecosystems.

Today’s automotive ecosystem partners, which include automakers, telcos, vendors, tech giants, automotive SDOs, and mobility providers, need to work in congruence to ensure the sustainable growth of autonomous and connected automobile technologies. Many areas need to be addressed before this market becomes mainstream, including the following:

• Coverage: Ensure the availability of ultra-reliable networks in both urban and remote areas.
• Technology: Position, implementation, and penetration of “network slicing” using NFV and SDN for network optimization and flexibility.  
• Analytics: Enhancing data management and analytics capabilities to manage the flood of data (from automotive cameras, LiDAR, radar, GPS, and sensors) from autonomous vehicles worldwide.

In order to fully utilize the potential of autonomous vehicle technology, eventually (along with the cars) there will be self-driving trucks and buses, and roads will need to have more streamlined traffic control in smart cities and highways. For example, Embark, a self-driving trucking startup, received approval to test its self-driving semi-trucks in Nevada. ABI Research’s recent Industry Survey: Transformative Technology Adoption and Attitudes – Automotive (PT-1954) conducted in June 2017 polled decision makers and influencers on their opinions about transformative technology adoption. The result of this survey stated that more than 60% of respondents in this sector have plans for assessing the potential of autonomous driving technology, and many of them have already entered the research and operations planning process for launching connected and autonomous vehicles. During the next 3 to 5 years, ABI Research believes that network slicing will fuel the growth of the autonomous and connected vehicle paradigm by strengthening V2X communication and providing full network support while ensuring high security and reliability.