When we started working on the next-generation RT-series of GNSS-aided inertial navigation systems, development was driven by an underlying ethos of simplification and enhanced user experience. Thanks to feedback from our automotive customers, we know that ADAS engineers are under constant pressure to perform increasingly complex tests on time and with minimal fuss. All the while, the real-time acquisition of precise, robust and reliable position data – both ground-truth reference and relative position – has remained paramount.
When the next-gen RT-series launched, the biggest headline feature was the integration of the RT-Range Hunter processing engine within the RT3000 v3 itself. More than anything else, the built-in relative positioning functionality of the RT-Range Hunter represents those key aims of simplification and enhanced user experience.
First, a bit of background. The RT-Range was conceived as an extension product to our GNSS-aided inertial navigation systems. Its function has always been to produce real-time relative position, orientation and motion measurements between the vehicle under test and other targets, such as fixed street furniture, parked or moving vehicles, pedestrians and lane markings.
Those measurements could then be used to monitor real-time data, such as time to collision and even visibility within a defined field of view. The RT-Range soon established itself as a key tool for the testing and validation of ADAS technology to global NCAP, IIHS and NHTSA standards. Application scenarios include NCAP’s autonomous emergency braking City, Inter-Urban and VRU protocols and the NHTSA’s forward collision warning protocols, along with blind spot detection, lane departure warning and adaptive cruise control and traffic sign recognition validation, among others. The RT-Range has also become popular with driving robot manufacturers for its unparalleled accuracy and repeatability.
The hunter and the target
The RT-Range is composed of two parts: the RT-Range Hunter and the RT-Range Target. The RT-Range Hunter was installed in the vehicle under test with an RT, while the much simpler RT-Range Target was installed in the target vehicle, along with another RT unit to provide position and heading data which was sent back to the Hunter for processing.
For all its standard-setting functionality, the RT-Range did offer an added layer of logistical complication in terms of installation and set-up. The cabin of a vehicle under test is becoming a crowded place as test engineers compete for space with the sheer volume of hardware that’s necessary for today’s complex ADAS evaluation scenarios. The RT-Range Hunter unit required installation close to the RT3000 GNSS-INS, to which it was connected via ethernet cable, which presented a potential point of failure.
The RT-Range Target, meanwhile, also needed installing in the target vehicles, again connected to an RT unit via ethernet cable.
All in one box
Now, though, the RT-Range functionality is incorporated within the RT3000 v3 (which is itself smaller and 1 kg lighter than the v2 it replaces). The new RT can perform the Range calculation and the RT3000 v3 can now be used as a Target box by connecting an XLAN to the new PoE port, making the test engineer’s job that much easier – and their workspace a that less cramped.
In fact, we’d go so far as to say it’s a significant step forward in terms of time-saving simplicity and functionality: set-up and configuration time are reduced and both space and weight are saved in the vehicle under test and in target vehicles, and all without compromise to the RT’s and RT-Range’s functionality.
Greater performance, less space
The inclusion of the RT-Range within the RT unit has been made possible thanks to the next-gen RT’s new and more powerful processor. The RT3000 v3’s upgraded CPU has enough processing capacity to perform the RT-Range’s relative position measurement functions while also carrying out the unit’s primary role as a GNSS-aided INS without compromising on performance.
With less hardware to install, reduced cabling complexity and less time spent troubleshooting, even the most complex multiple-target ADAS tests can now be performed more quickly and with minimal fuss.