Vehicle PNT testing for SAE Level 3 and above: What to look for in a GNSS simulator

There may only be a single number difference, but the jump from Level 2 to Level 3 vehicle automation presents many challenges for developers of automotive positioning, navigation and timing (PNT) systems.

At Level 3 in the SAE’s framework, the vehicle is no longer supporting the human driver, but is itself in control—observing the environment and making autonomous decisions around lane position, steering, braking and collision avoidance.

While a human driver must be present and prepared to take over when requested, the aim for automotive OEMs is to provide an experience akin to having a chauffeur. The less the vehicle requests human intervention, the more luxurious the experience.

The ability to simulate an infinite variety of realistic conditions is essential

Automotive PNT systems developers must therefore ensure the system is prepared for the infinite variety of conditions it may encounter in the real world. Simulation is the fastest, most rigorous and most cost-effective way to run test scenarios at scale, but the safety-criticality of the PNT system means simulation must reflect real-world conditions as closely as possible.

That makes the choice of PNT simulation equipment a critical one for automotive OEMs, Tier 1 and Tier 2 suppliers. Particular capabilities to consider include:

• GNSS signal emulation: A simulator must be able to faithfully emulate all of the global navigation satellite system (GNSS) signals, on all of the relevant frequencies, that will be used by the system’s GNSS receiver. For early R&D on future autonomous vehicle systems, the ability to emulate GNSS and LEO signals that are in plan but not yet live will also be valuable.

• Augmentation signals: It’s likely that a safety-critical PNT system will also make use of space-based augmentation systems (SBAS) for more accurate and more precise positioning, especially as more services start to launch in Low Earth Orbit (LEO). The ability to faithfully emulate SBAS and other LEO PNT signals will therefore be a key consideration.

• Error correction messages: GNSS error correction messages, such as those provided by Real Time Kinematic (RTK) and Precise Point Positioning (PPP) services, can play a key role in lane-keeping and lane-changing precision. A good simulator will offer error correction message simulation as standard.

• Sensor outputs: Automated PNT systems rely on sensor fusion algorithms to determine the absolute position of the vehicle and its position relative to objects around it. While the exact mix of sensors may vary between vehicle designs, most PNT systems will rely on some combination of GNSS, inertial measurement units (IMUs) like accelerometers and wheel-tick sensors, camera vision, LiDAR and radar. A simulator that can emulate the output of micro-electronical systems (MEMS) sensors, along with environmental factors like vibration and temperature-induced fluctuations, can significantly enhance the realism of the test and cut out expensive and unrepeatable field testing of integrated positioning engines.

• Spoofing and meaconing: Whether accidental or malicious, the illicit transmission of fake or recorded GNSS signals represents one of the biggest threats to vehicle PNT systems. Anti-spoofing capabilities will be critical to meeting functional safety standards at Level 3 and higher. The simulator must be capable of realistically simulating different types of spoofing attack, including navigation data spoofing, trajectory spoofing and meaconing (rebroadcast of recorded signals).

• RF interference: Another threat to automotive PNT systems comes in the form of radio frequency interference (RFI) in the vehicle’s environment. This may be caused by illegal signal jamming using a purpose-built jammer, or by radio equipment transmitting in or close to the GNSS frequency bands, potentially due to faulty installation. As RFI may manifest in many ways, the ability to simulate a wide range of realistic waveforms and environmental conditions will be key, as will the ability to simulate the antenna in static and dynamic scenarios and in a variety of placements on the vehicle.

• Multipath and obscuration: Buildings and other structures can reflect, block or attenuate GNSS signals, potentially causing loss of positioning accuracy in urban areas and areas with either dense foliage or natural structures. Understanding PNT system performance in these environments requires the ability to simulate a wide range of variables, from realistic renditions of satellite geometry, buildings and materials, to accurate modelling of signal paths—including their angle of arrival—together with vehicle trajectory and antenna placement.

Choosing a simulator for hardware in the loop (HIL) testing

Further considerations apply when choosing a GNSS simulator for integration into a hardware in the loop (HIL) test setup. Here, additional key criteria include:

• System latency: The speed with which a simulator can process commands and produce the corresponding RF is a key factor in reducing error and uncertainty in results. There are a number of factors that can influence latency. You can find out more about simulator latency in our blog.

• API integration: Out-of-the box integration with leading brands of automotive test equipment can save time and budget in setting up the test environment. A powerful application programming interface (API) combined with low system latency can ensure a rich, realistic, and reliable RF signal environment for the DUT.

Spirent GSS7000: Realistic automotive PNT testing in an infinite range of scenarios

While today’s in-vehicle navigation systems (IVNS) are a useful driving aid, at Level 3 and above the PNT system is critical both for safety and for the customer experience. The ability to create infinite numbers of realistic test scenarios in the lab becomes essential, and the choice of simulation equipment is crucial to achieving the requisite levels of realism.

The Spirent GSS7000 RF signal simulator offers automotive PNT developers all of the functionality needed to create and run ultra-realistic test scenarios. Its industry-leading four-millisecond latency and out-of-the-box integration with leading automotive HIL test equipment providers means developers can get started fast and achieve deterministic results.

If you’d like to know more, watch our six-minute demo video to get a flavor of the GSS7000’s performance and capabilities, or get in touch to request a demo of your own.