AI-based system warns drivers of obstacles
With the help of artificial intelligence (AI) and machine learning (ML), the HERE Live Sense SDK transforms devices with front cameras such as smartphones, dashcams or vehicle cameras into smart vehicle sensors. These can permanently capture their surroundings. They can identify objects such as other vehicles, pedestrians or cyclists as well as traffic lights and traffic signs and potential hazards such as potholes, road closures or construction sites.
Based on the collected data, the HERE Life Sense SDK then informs drivers via audio and image notifications so that they can adapt their driving behaviour accordingly. This is done in real time on the device itself, without the need for cloud data processing or an Internet connection. The SDK is also capable of comparing changes in the route with the navigation system map and informing the driver of current changes, such as changes in speed limits.
The Live Sense SDK uses multiple AI/ML detection models to detect and classify objects on the road. These include the detection of unexpected changing conditions such as road closures not yet shown on the map, construction sites and temporary road works. The software can also recognize and classify traffic signs.
The SDK is aimed at automobile manufacturers, fleet operators, ride hailing providers and app developers. They can thus integrate the AI/ML detection models into applications or devices and make AI functionalities available to their customers. The SDK is currently available in a beta version for Android and will soon also be available for iOS. In addition, the platform provider is already working with customers on initial commercial implementations.
More information: https://developer.here.com/products/live-sense-sdk
Related articles
Deep Learning makes static map data dynamic
Tomtom taps AI to create street map for autonomous driving
Intel buys into map services provider Here
BMW, Mobileye feed HD map data to service cloud
If you enjoyed this article, you will like the following ones: don't miss them by subscribing to :
