en.Wedoany.com Reported - At the 7th China International Automotive Ethernet Summit, Real-Time Innovations (RTI), a company specializing in intelligent distributed system data flow platforms, demonstrated the capability of its Connext Drive platform to provide deterministic, real-time communication for software-defined vehicles (SDVs). At the summit, RTI showcased the underlying logic of Data Distribution Service (DDS) working in coordination with Time-Sensitive Networking (TSN). This was demonstrated through real-time video processing via a regional AI node, with Connext Drive distributing perception data across the entire system. The demonstration showed that even under simulated network congestion, the system maintained stable, uninterrupted deterministic data transmission.
In an exclusive interview, Kelvin Hor, Sales Director for RTI Asia Pacific, discussed the company's development history, technological advantages, and its strategic layout in the automotive industry. This software middleware company, with over 30 years of history, focuses on high-performance, highly reliable communication platforms. Its core product, Connext Drive, has become one of the key technologies in the transformation of intelligent vehicle electronic and electrical architectures.
DDS originated in the military and aerospace sectors. RTI's founding team drove its development into an open standard under the Object Management Group (OMG), aiming to prevent enterprises from being locked into a single vendor. The automotive business is currently a significant growth area for RTI. With the advancement of vehicle intelligence and electrification, traditional CAN and LIN buses can no longer meet the demands of massive data volumes and high real-time requirements. Through its Quality of Service (QoS) mechanism, DDS can prioritize different data streams and ensure real-time performance within the same network, thereby efficiently managing vehicle-wide communication. Connext Drive builds a data infrastructure based on the DDS standard, enabling OEMs to integrate high-performance applications including advanced driver-assistance systems, zonal architectures, and vehicle-to-everything (V2X) connectivity, while ensuring deterministic system performance.
Addressing the competition between high-bandwidth AI data and entertainment information sharing the Ethernet network, Kelvin proposed that the combination of DDS and TSN represents a complete engineering methodology. RTI's approach starts from the data flows at the business layer, grouping various types of in-vehicle data and analyzing attributes such as data volume, real-time requirements, and transmission frequency. After grouping, the data is divided into different topics at the DDS layer, with corresponding QoS policies configured for each topic. At the network layer, for extreme scenarios with high concurrency or near-saturation bandwidth, TSN mechanisms are overlaid to achieve enhanced network scheduling. RTI provides an integrated design methodology spanning from business requirements to the DDS layer and then to the network layer, covering the entire engineering lifecycle from design, verification, and implementation to mass production and upgrades.
Globally, over 2 million vehicles are equipped with RTI's DDS solution, with collaborations involving more than 25 OEMs, including both domestic "new forces" and traditional automakers. Taking XPeng as an example, RTI partnered with it years ago in the autonomous driving domain. This solution has now evolved into a unified vehicle communication platform, supporting XPeng's transition from a distributed architecture to a zonal architecture. Unlike traditional Service-Oriented Architecture (SOA), RTI's DDS solution is data-centric, employing a finer-grained data streaming model that allows OEMs to flexibly subscribe to required data based on new business needs. Currently, some leading international OEMs are further diminishing the concept of "domains," moving more decision-making functions to high-performance computing (HPC) nodes. This architecture imposes higher demands on the real-time performance and processing capability of the communication layer.
In the practical implementation process, Kelvin pointed out that challenges extend beyond technical aspects to include cross-departmental collaboration and process optimization within OEMs. RTI needs to communicate with multiple teams to help all parties understand the overall value of a unified communication platform. Facing competitors in the DDS market, in-house solutions, or open-source implementations, RTI positions itself as providing highly reliable, mass-production-verified communication components with a robust ecosystem. It also offers extensive toolchain integration capabilities (e.g., with Vector, ETAS, dSPACE, etc.) and specialized information security services. Looking ahead, DDS will continue to evolve, including supporting more transmission mechanisms, better enabling AI application deployment, and exploring the integration of 5G/6G for external "external brain" assisted decision-making. Kelvin stated that the Chinese market is at the forefront of intelligence globally, and RTI will continue to deepen its engagement with Chinese OEMs, collaborating with leading companies to validate cutting-edge technologies.
