The automatic driving platform was invented with the ability to read boards and follow the voice commands

Chinese company Deeproute.ai has launched its latest smart driving platform, Deeproute IO 2.0. The company says Deeproute IO 2.0, which works with its VLA (Vision-Language-Action) model, is an important step towards providing smarter, safer and more human driving experiences.

The platform has a flexible design with multiple chips and a few sensors, supporting both Lidar -based configurations and pure vision -based configurations that allow perfect integration in various car and automakers. The company has signed five partnership contracts with the original equipment manufacturers (OEM) to implement Deeproute IO 2.0, and the first manufacturing vehicles equipped with the technology are expected to enter the market later this year.

Maxwell Zhou, CEO of Deeproute.ai, announced that Deeproute IO 2.0 uses integrated VLA model with a large language model that has two main advantages: chain reasoning and extensive knowledge base. Zhou says that chain reasoning allows the system to analyze complex traffic scenarios through rational and human -like reasoning and increase the transparency and reliability of driving decisions.

The extensive knowledge database also allows the model to use the accumulated driving experience in a variety of conditions and adapt to real situations. Deeproute.ai says it validates the Deeproute IO 2.0 platform and VLA model in real urban environments to ensure their reliable performance in a variety of situations. The company continues to work with the World OEM partners to accelerate the establishment of smart driving technology in mass -manufacturing cars around the world.