he hesitation to include GNSS in ADAS stacks is historical. Traditionally, this technology was unreliable, especially in
dense, urban environments, where satellite signals bounce off tall buildings, misleading the receiver. That reputation has influenced how vehicles are engineered, with many automakers choosing to rely more on cameras, LiDAR and other sensors.
Cameras can identify lane markings, traffic signs and objects, while LiDAR can build highly detailed 3D maps of the vehicle’s surroundings. Both sensors provide important navigational data, but they can only describe a car’s location relative to its immediate environment. With no reliable source of absolute location, these relative measurements can’t confirm the vehicle’s exact place in the world – information that is critical for safe navigation. It’s easy to see why cameras, LiDAR and
radar have become the backbone of many ADAS stacks. They observe the world directly and can detect lane markings, vehicles, pedestrians, kerbs, signs and roadside features. They therefore create a live environmental picture that underpins vehicular functions like lane keeping, adaptive cruise control and automated emergency braking. However, the engineering challenge is that
these sensors do not fail neatly. Cameras are information-rich but are still optical systems. Glare, low sun, heavy rain, fog, spray and night driving can all reduce contrast and make features harder to extract. Dirt on the lens or windscreen can have the same effect. Even when the camera still ‘sees’, its confidence can drop without obvious warning. LiDAR has different strengths. It can build
a 3D representation of the environment and measure distance to objects. However, it is also affected by adverse weather and can be compromised by occlusion and contamination. Meanwhile, radar is robust in low-visibility settings and can measure range and relative velocity. Yet, it can struggle to classify complex scenes and is easily influenced by reflections and multipath effects, particularly in areas with dense infrastructure. These are not reasons to avoid cameras,
44
LiDARorradar–theyare reasons to design for uncertainty. In practice, ADAS performance depends on how well these sensors complement each other, and on how fast the system can detect an unreliable reading.
The common thread across these sensing technologies is that they describe the vehicle’s position relative to what they can observe. A camera watching lane markings might help a car stay in a lane, but it won’t help choose the correct one, and may run into trouble when junctions overlap, road markings are worn or roadworks introduce temporary changes. Similarly, LiDAR can produce an accurate
local map of obstacles and boundaries, but without a stable global frame of reference, anchoring that map to the road network becomes more challenging. Radar can reliably track objects, but it doesn’t solve the question of absolute position on its own. This matters because modern ADAS stacks
increasingly depend on sensor fusion and map alignment. Inertial measurement units (IMUs) and wheel-speed sensors can provide a smooth short-term motion estimate, but they can drift over time. HD maps offer another useful layer, but if the global position estimate is wrong, the map can be misaligned with the real world, undermining the value of both.
Ensuring designs featuring reliable GNSS
strengthens this entire chain. GNSS doesn’t replace environmental sensors, but it helps constrain drift, improve map matching and provide a consistent, absolute reference that the rest of the stack can check itself against. In higher levels of automation, that cross-check becomes an important additional safety feature.
Increasingly, the importance of GNSS in ADAS stacks is being recognised. As automotive production moves toward Level 3 (L3) automation and beyond, the demand for better positioning increases, along with the need for safe, layered sensing. GNSS, alongside cameras, LiDAR and radar, can help automakers improve navigational resilience without reinventing vehicle architecture. Having a global frame of reference helps
ensure that the relative data from other sensors is grounded in the correct place. For automakers, the next step is recognising that GNSS can improve safety and trust in ADAS stacks, supporting the transition toward autonomous driving.
Even with GNSS integrated into an autonomous vehicle’s sensors, challenges remain. Urban
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48 |
Page 49 |
Page 50 |
Page 51 |
Page 52
