Autobrains, a developer of AI-powered automotive applications, recently introduced its new localization technology Air2Road. This innovation aims to improve advanced driver assistance systems (ADAS) and autonomous driving. Based on Autobrains’ Liquid AI technology, Air2Road provides precise, real-time vehicle location for safe and efficient navigation, even in difficult conditions such as occlusions, bad weather or complex urban environments.
The technology combines signatures from two independent modalities: aerial photography and ground-based perception. This ensures multimodal redundancy and high accuracy. The signatures are sparse binary hyperdimensional representations of both the driving environment and the objects within it. This innovative approach overcomes many limitations of traditional systems.
In contrast to traditional localization systems, which are often based on global navigation satellite systems (GNSS) or crowdsourced maps, Air2Road offers several advantages: It works reliably even with poor GNSS reception, uses aerial images for additional data layers and enables immediate, precise localization without prior notice Data collection. The system is cost-effective as it is compatible with affordable aerial imagery and large areas can be stored directly in the vehicle.
Air2Road also supports environmentally conscious route planning by using real-time data on road conditions, traffic flow and parking spaces. The technology filters out irrelevant data and focuses on critical elements such as roadsides and buildings, significantly improving localization accuracy.
Overall, Air2Road represents a significant advance in vehicle localization and promises to advance the development of autonomous vehicles and advanced driver assistance systems. The technology provides a scalable, efficient and accurate solution to the challenges of vehicle localization in various environments and conditions.