TECHNOLOGY

技术

Image-based Biometric Recognition Technology

Image-based Biometric Recognition Technology
The face recognition scheme based on deep learning can accurately recognize face information, extract image features, compare template data and confirm driver status.

Recognition and Tracking of 3D Objects

Recognition and Tracking of 3D Objects
The gesture tracking data is processed, and the gesture motion capture is realized by optics and sensors. The algorithm of gesture recognition speculation is realized by statistical sample features and deep learning neural network.

Eye-Tracking Technology

Eye-Tracking Technology
Based on the eye tracking technology to simulate the three-dimensional model of human eye, provide the real data basis for eye movement test, and remind the driver at the right time.

Extravehicular Perception Technology

Extravehicular Perception Technology
External environment perception in the field of computer vision, real-time monitoring and early warning of road environment outside the driving warehouse.

Autopilot on Highway

Deep learning based, with multi-sensor fusion perception, combined with deep learning and rule-based hybrid decision-making algorithms;
Innovative architecture design and component development mode, platformization of L1-L4 application development, expansion and empowerment of OEMs’ software-defined cars;
Meet the requirements of high computing power and high real-time performance for different modules of autonomous driving; 

Software Architecture Supporting L3/L4 Autonomous Driving 

Support L3/L4 autopilot software architecture, provide hard real-time operating system on MCU and soft real-time operating system on SoC transplantation and optimization on specific hardware platform; provide middleware technology suitable for autonomous driving; well-designed support The application framework of L2/L3 autonomous driving serves as a container for various algorithms to accelerate the implementation of autonomous driving functions

Multi-Sensor Fusion 

Multi-sensor or multi-source information and data are analyzed and integrated under certain criteria to complete the required target and road detection, estimation and prediction

Planning and Control Algorithm 

The automatic driving system establishes the global driving path of the vehicle driving task through dynamic path finding, and predicts the behavior of the surrounding vehicles based on the correct perception of the surrounding environment, and then plans the behavior and path of the vehicle, and finally achieves correctness, reasonableness and accuracy control to the vehicle.