Nullmax launches pure-vision multimodal autonomous driving tech ‘Nullmax Intelligence’

Beijing (Gasgoo)- On July 16, autonomous driving tech company Nullmax officially launched its next-generation autonomous driving technology, Nullmax Intelligence (NI). The new technology focuses on developing full-scenario autonomous driving applications with pure vision, true HD map-less, and multimodal capabilities to drive the evolution of automotive intelligence.

Photo credit: Nullmax

NI integrates innovative multimodal models and a safety brain, enhancing vehicles with the sensory abilities to see, hear, and read, and generating visual outputs, scenario descriptions, and driving behaviors. Based on the NI solution, Nullmax plans to achieve full-scenario intelligent driving applications by 2025 and expand AI capabilities—such as visual observation, cognitive processing, and interaction with the physical world—to passenger transport, cargo delivery, and robotics.

Nullmax Intelligence addresses the challenges of autonomous driving with a smarter, more human-like approach. Beyond visual input, NI supports auditory, textual, and gestural inputs, utilizing a multimodal end-to-end model for task inference.

Photo credit: Nullmax

The standout feature of Nullmax Intelligence is its support for pure vision, true HD map-less, and multimodal full-scenario autonomous driving applications. Without relying on LiDAR or stereo cameras, Nullmax achieves precise obstacle detection and 3D reconstruction through pure vision approach, generating local maps in real-time independent of HD maps.

This multimodal approach, centered on vision, outputs various information including static and dynamic perception, scene descriptions, and driving behaviors. These features grant Nullmax’s solutions exceptional generalization capabilities for full-scenario applications with lower computing power demands, achieving full-scenario navigation assistance with sparse computing power under 100T. The reduced sensor and chip costs also enhance the cost-effectiveness, making Nullmax’s solution suitable for widespread adoption.

Nullmax’s algorithm platform supports multiple vehicle types, adapts to various sensor quantities, and is easy to deploy. These algorithm capabilities can be extended to robotics for tasks such as visual navigation, learning human behaviors, and interacting with the physical world.

This means Nullmax Intelligence not only upgrades Nullmax's comprehensive intelligent driving solutions, enhancing the humanization and intelligence of high, mid, and low-end vehicle configurations, and accelerating the adoption of advanced intelligent driving but also promotes full-scenario passenger and cargo applications, achieving valuable large-scale autonomous driving. Additionally, the reusable AI capabilities can be applied to broader embodied intelligence applications.