Internship project description
Intelligent Speed Assist (ISA) concerns communication of the location-dependent maximum
speed to the driver or, optionally, automatic implementation of that maximum speed. Although
ISA is a mandatory feature in new road vehicles, the system is not flawless, being subject to
wrong or missed detection of speed signs.
This internship project will contribute to a test setup for ISA involving both real-life
implementation in a vehicle and setting up a so-called digital twin, i.e., a virtual test
environment. The short-term aim is to present this test setup as a demonstrator for the Digital
Infrastructure for Future Mobility (DITM) project. The demonstrator will be co-developed by
Siemens, TNO, and TomTom. The internship project focuses in particular on building the virtual
Want more jobs like this?
Get jobs in Eindhoven, Netherlands delivered to your inbox every week.
test setup for the ISA use case. The planned test setup includes the Siemens simulation tool
(Simcenter Prescan) and the ISA functions to be tested. Figure 1 shows a simplified schematic of
the test setup.
The ISA functions include feature-based localization and a speed advice module. The
localization module detects features based on simulated camera and LIDAR data, and matches
them to the map to estimate the position of the vehicle in the map. The speed limit is then
obtained from the map and combined with traffic sign recognition from a physical Mobileye
camera which is part of the simulation loop. The ISA functions will be provided as ROS nodes,
deployed within a docker container.
The internship project will have two main contributions to the demonstrator:
1. Interfacing the outputs of the simulation, i.e., camera and LIDAR data, to the ISA
functions.
2. Configuring the sensor simulation to meet the requirements of the feature detector (part
of the localization module). The feature detector is developed to be used on a physical
vehicle in real-driving conditions. The simulation may need to adapt to the feature
detector, e.g., with respect to the way features are represented/rendered in simulation or
regarding the configuration of sensor models.