How to Reach Higher Automation Levels with AI-based Sensor Fusion.
As more complex ADAS, automated driving and self-driving technology is developed, a fundamental element is a need for engineers to create systems that allow a vehicle to perceive their environment and process vast amounts of data effectively.
Volkswagen subsidiary CARIAD develops ADAS/AD projects to introduce Level 4 automated driving for use in individual mobility. The company is transitioning from classic, object-based sensor fusion to the more advanced approach of AI-based sensor fusion.
Sensor fusion brings together information from various sensors and generates the first step of the so-called environment model. This model provides a comprehensive view of the vehicle’s surroundings. Therefore, the algorithms used in sensor fusion should ensure both spatial and temporal consistency.
To power sensor fusion, inputs are required. These inputs are typically sensors, such as cameras, lidar, radar and ultrasonic sensors, but can also include information from the car-to-car communication system.
CARIAD has released a Youtube video from GTC22 demonstrating how to reach higher automation levels with AI-based sensor fusion. The video is presented by Dr Cristina Rica Garcia - Head of Sensor Fusion at CARIAD who provides an excellent overview.
Watch the whole video: