Variable Traffic Sign Recognition

The Task

Detecting and interpreting variable message signs and traffic lights in virtual traffic scenarios to test the controllers for highly automated and autonomous driving by means of simulation at an early stage.

The Challenge

Defining virtual traffic scenarios with variable message signs and traffic lights that can influence the allowed speed or control the traffic in an adaptive manner, for example. Detecting and interpreting the variable message signs and traffic lights to test the algorithms for traffic sign recognition and to control the vehicle under test.

The Solution

With ModelDesk you can add variable message signs and traffic lights to virtual traffic scenarios and define their behavior using a convenient graphical user interface. ModelDesk provides variable traffic signs that can display numerical or color information. The ASM Traffic Model includes a traffic sign sensor that detects the signs even if they are attached to a gantry, interprets them, and assigns them to the relevant lanes. The signals of the traffic sign sensor are then available to be processed further by the traffic sign recognition algorithm.