Blockset models autonomous driving functions
To develop function for automated driving, dSpace introduces the ADASIS v3 Horizon Reconstructor Blockset to access map data in Simulink.
The blockset is based on the EB robinos reconstructor from Elektrobit, which is suitable for series production. It provides the electronic horizon in HD quality via the standardised ADASIS v3 protocol, making exact route and position data available to applications in Simulink. In contrast to conventional vehicles, highly automated and autonomous vehicles must be able to recognise where they are with a precision of only a few centimetres to be able to move safely through traffic. With the ADASIS v3 blockset, alternative vehicle positions can now be determined to the centimetre including their probabilities of occurrence. Detailed environment data, eg, for intersections with individual lanes, enable precise manoeuvre calculations. Developers also benefit from the new map management, which lets them manage large volumes of data in ADASIS v3, says dSpace. The horizon reconstructor makes only required data available and automatically deletes superfluous data.
The blockset is based on the EB robinos reconstructor for ADASIS v3 by Elektrobit, which means that developers do not have to implement their own horizon reconstructor or know the details of the ADASIS protocol. Instead they can concentrate on application development and experience the results in the vehicle for a smooth development cycle from prototyping and transfer to the production-intent engine control unit (ECU).
The structure of the Horizon Reconstructor Blockset enables a quick and easy connection to the application in the vehicle, explains dSpace. For example, adding only two blocks to the existing Simulink model makes it possible to access profiles of a vehicle trajectory.
The blockset is another building block in dSpace’s end-to-end tool chain for autonomous driving. It supports the development of map-based ADAS applications and functions for autonomous driving on dSpace prototyping systems and dSpace’s PC-based simulation platform VEOS.