AliOS In-Vehicle Operating System

Definition and Technical Positioning

AliOS is an intelligent automotive operating system independently developed by Banma. Initiated in 2014, it first entered series production in 2016 on the Roewe RX5—the world’s first internet car—pioneering an entirely new vehicle segment. As China’s earliest and largest-scale proprietary automotive OS in production, AliOS features over 300 million lines of code, more than 2,500 OS-related patents, and a cumulative R&D investment exceeding RMB 10 billion. Ranked alongside QNX, Android, Linux, and HarmonyOS as one of the world’s top five mainstream automotive operating systems, AliOS powers the largest number of production models among China’s proprietary intelligent automotive OS platforms.

Core Technical Architecture and Product Evolution

AliOS adopts a cloud-integrated architecture, covering three key tiers: intelligent in-vehicle systems, intelligent cockpits, and full-vehicle operating systems. Its core technical capabilities are defined across three key dimensions:

Dual-Core Architecture: Tailored for intelligent driving scenarios, AliOS Drive utilises a "microkernel + monolithic kernel" dual-core setup. The safety domain runs on the AliOS RTOS real-time microkernel, certified to the highest ISO 26262 ASIL-D functional safety standard. The performance domain is powered by AliOS Safety Linux, engineered with real-time safety enhancements to handle high-performance computing tasks with scheduling latency kept under 10 microseconds.

Distributed and Multi-Domain Fusion: AliOS natively supports a Service-Oriented Architecture (SOA) via its Fusion framework, enabling seamless communication across different OS environments, vehicle domains, and vehicle-to-cloud platforms. It standardises, modularises, and converts vehicle hardware capabilities into services while supporting industry protocols such as SOME/IP and DDS. Its multi-core heterogeneous distributed architecture ensures robust functional isolation and safety compliance across diverse domains, including the digital instrument cluster and infotainment system.

Multi-Chipset Platform Compatibility: The system offers extensive compatibility with mainstream automotive System-on-Chips (SoCs), including the Qualcomm Snapdragon 8155/8295, SemiDrive X9 series, MediaTek, Horizon Journey 5, Black Sesame A1000, and NVIDIA Orin.

On the technological front, AliOS pioneered the integration of large AI models into the automotive sector in 2021 with the launch of its "YuanShen AI" agent architecture. Encompassing multi-modal sensory fusion, cognitive evolution, and intelligent service dispatch, the system boosted human-machine interaction (HMI) response speeds by more than fivefold.

Industry Application and Ecosystem Partnerships

Deployed via the Banma system, AliOS has been rolled out across more than 10 automotive brands, over 50 vehicle models, and nearly 3 million production smart vehicles, reaching 5 million connected car users. Key production models featuring the OS include the IM L7 (underpinning the IMOS system), All-New Roewe RX5 MAX, MG ONE, and Maxus MIFA 9.

Within its ecosystem, AliOS integrates deeply at the system level with the Unity engine, enabling multi-screen casting, 3D vehicle control interfaces, and immersive in-cabin experiences—exemplified by the full-screen, cross-display teamLab visual features in the IM L7. The platform remains compatible with Android and Linux applications while supporting major cockpit development frameworks such as Unity, Unreal Engine, Kanzi, and Flutter. In 2025, Banma introduced the industry's first dual-OS (AliOS + HarmonyOS) smartphone-to-vehicle connectivity solution, breaking HiCar's reliance on Android via an AliOS Linux SDK adaptation to deliver seamless one-touch handover for navigation, music streaming, and connected services.

Technical Limitations

While AliOS stands as a flagship proprietary operating system, its commercial ecosystem continues to face competitive headwinds. Amid the shift towards Software-Defined Vehicles (SDVs), QNX maintains a dominant footprint in digital instrument clusters and ADAS/autonomous driving domains thanks to its rigorous functional safety certifications and mature toolchains. Meanwhile, Android Automotive OS is aggressively expanding its cockpit footprint backed by an expansive app ecosystem, exerting dual pressure on AliOS. Additionally, with AliOS currently concentrated on the cockpit and intelligent driving domains, further validation and development are needed to achieve a truly unified full-vehicle OS architecture and seamless cross-domain integration.