Pony.ai, a global leader in autonomous driving, today announced its Level 4 (L4) automotive-grade autonomous driving (AD) domain controller—deployed in the company’s seventh-generation Robotaxi fleet—has surpassed 2 million kilometers of on-road testing. This achievement reflects the system’s exceptional hardware reliability and software adaptability, meeting the rigorous demands of L4 autonomous driving.
Unveiled at the Shanghai Auto Show in April as a production-ready solution for L4 autonomy, Pony.ai’s seventh-generation AD system is built entirely with automotive-grade components. Designed for a lifecycle of 10 years or 600,000 kilometers, the full AD stack—including the domain controller—represents a major leap in performance, durability, and cost-efficiency. Through extensive algorithm optimization and system engineering, Pony.ai has reduced the domain controller’s cost by 80% compared to the previous generation, while also extending its lifespan—substantially lowering the total lifecycle cost of the system.
Developed entirely in-house, the world-first L4 automotive-grade domain controller utilizes three NVIDIA OrinX chips in the main unit with a fourth OrinX chip dedicated to system redundancy, delivering 1,016 TOPS of computing power. Through a co-designed hardware-software architecture, Pony.ai custom-developed the domain controller to meet the comprehensive system and functional requirements of fully driverless Robotaxis, as well as the performance and safety demands of L4 autonomous driving algorithms. As a result, the controller achieves a 50% to 80% reduction in size, weight, power consumption, and cost—significantly optimizing hardware expenses. With continuous algorithm iteration and highly efficient inference optimization, Pony.ai has successfully delivered L4 autonomous driving capabilities and safety assurance using mass-produced, cost-effective, and mature automotive-grade chips.
Functioning as the central computing hub, this controller integrates a wide array of sensors, essential functions and supporting controllers required for Robotaxi operation. These include system power management, interactive display, hazard warning lights, gateways, and Global Navigation Satellite System (GNSS). By integrating these features into a single unit, Pony.ai has reduced wiring complexity and the number of individual components, streamlining vehicle layout, lowering manufacturing costs, and minimizing potential failure points. Furthermore, it simplifies assembly line processes, making mass production faster and more efficient. The domain controller supports both liquid cooling and passive cooling. When the liquid cooling system encounters malfunction, the redundant system ensures safe pull-over with passive cooling.
The system incorporates a multi-layered safety architecture and degradation strategies, delivering “Fail Operational” capability. In the event of a primary system failure, the controller can seamlessly switch to the redundant system or the Minimum Risk Condition Controller (MRCC) to ensure safe vehicle control. Even if the main system’s power or chassis communication fails, the redundant system can still maintain critical perception (including blind spot coverage) and safe driving capabilities. This allows the vehicle to navigate intersections or ramps and safely pull over, minimizing the risk of traffic disruption or collisions.
At the core of Pony.ai’s system is a “safety first” philosophy, realized through full-vehicle redundancy and a high degree of algorithmic intelligence. Leveraging its PonyWorld foundation model and reinforcement learning, the company delivers driving performance that consistently surpasses human-level capabilities, while continuously improving its system’s response to complex real-world scenarios.
With seventh-generation Robotaxis now in mass production and actively undergoing road testing, this safe, robust, and cost-effective solution is set to significantly expand Pony.ai’s driverless fleet and enhance user mobility experiences. 2025 marks Pony.ai’s first year of mass production, with a planned fleet of 1,000 autonomous vehicles—a key step toward achieving economies of scale and operational excellence.
SOURCE: Pony.ai
