NVIDIA and didi chuxing announced today that didi will use NVIDIA gpus and other technologies to develop autonomous driving and cloud computing solutions.
Didi will use NVIDIA gpus to train machine learning algorithms in data centers and NVIDIA DRIVE to provide reasoning capabilities for its l4-level self-driving cars. Didi upgraded its autonomous driving unit to a separate company in August and has been working extensively with industrial chain partners.
As part of didi's self-driving AI process, NVIDIA DRIVE USES a deep neural network to integrate data from various sensors (cameras, lidar, radar, etc.) to gain a 360-degree understanding of the surrounding environment of a car and map out a safe route.
Rishi Dhall, vice President of autonomous vehicles at NVIDIA, said: "end-to-end AI is essential to developing secure autonomous driving technology, whether in the cloud or in the car. With NVIDIA's AI technology, didi will be able to develop safer and more efficient transportation systems and provide rich cloud services."
To train these deep neural networks, didi will use NVIDIA GPU data center servers. In cloud computing, didi will also build a leading AI infrastructure and launch computing, rendering and gaming vGPU cloud servers.
Didi cloud will adopt the new vGPU license model, aiming to provide users with better experience, richer application scenarios, more efficient, more innovative and flexible GPU computing cloud services. Currently, didi cloud has joined hands with NVIDIA and other industry partners to serve transportation, AI, graphics rendering, video games, education and training and other fields.