The standard approach used by perception platforms in the ADAS and AD market is object fusion, where information about object detection performed by each type of sensor is brought together to support the driving decision-making.
The main limitation of this approach is that no single sensor on its own possesses sufficient information to support all possible driving scenarios.
LeddarVision™ provides a different, innovative approach to understanding the vehicle’s changing environment with raw-data fusion. Through a leveraged algorithmic approach, the software solution encompasses 3D reconstruction, artificial intelligence (AI), and computer vision to turn sparse data into a most precise dense 3D environmental model, contributing to improving the perception system’s performance.
This e-book presents the main features and benefits related to this advanced, cost-effective sensor fusion and perception solution, which requires low computational power and makes optimal use of available sensor suites.