According to dSpace, Understand.ai possesses valuable expertise in field of AI-based automated data analysis, data annotation and extraction of simulation scenarios for autonomous vehicles. Understand.ai offers with "semantic segmentation" a precise method for data annotation.
In the future, these key technologies will complement the dSpace portfolio in order to offer customers in the automotive industry an integrated development and test solution for autonomous driving. Under the umbrella of the dSpace group, Understand.ai is to invest in the use of artificial intelligence (AI) and cloud-based "tools" and further develop its existing products as an integrated component of the dSpace solution offering.
In the development and introduction of autonomous vehicles, it is crucial to perceive the environment of the car "flawlessly" and realistically. Other road users, traffic signs, lanes, the static edge development and open spaces must be reliably recognized. For this purpose, machine learning algorithms, in particular deep neural networks (DNNs) based on artificial intelligence, are used in cars. These algorithms must be trained and validated efficiently. For this purpose, enormous amounts of recorded (camera, lidar and radar) sensor data have to be analyzed, annotated and anonymized.
The quantity, quality and diversity of these training and validation data determine the quality of the resulting DNNs. The annotation process, also called labeling, is required for the classification of objects as a reference for machine learning. Today, this process is predominantly time-consuming and not always carried out manually at the best quality level.
understand.ai’s expertise enables this process to be widely automated The company also uses self-learning algorithms to prepare high-quality training and validation data. The underlying key technology is also based on AI.
More information: https://www.dspace.com/