The rapid acceleration of autonomous vehicle development is reshaping the future of mobility, with investment, regulation and consumer demand all pushing the industry forward. As OEMs move beyond advanced driver assistance toward higher levels of automation, the industry faces a defining moment. Continued progress will depend on a broader and more innovative vision to overcome the limitations that still constrain real-world deployment. Xavier Banque, strategic marketing manager, automotive and IoT at Trimble, discusses the issue
At the center of this challenge is positioning. Vehicles must be able to determine their exact location and velocity at all times with a level of precision and reliability that supports safe decision-making. Without that capability, even the most advanced perception systems cannot consistently enable the transition from L2++ to L3 and beyond.
Unifying relative and absolute intelligence
Modern autonomous systems rely on a perception stack built on vision-based cameras, lidar and radar, which provides detailed environmental awareness but remains inherently local relative to the vehicle’s surroundings.
Scaling autonomy requires anchoring this data to a global absolute location so vehicles can determine their position within the broader road network. Advanced positioning provides that reference by aligning perception outputs with high-definition (HD) maps and route-level intelligence, while also supporting physical AI implementation. This emerging technology relies on geotag metadata to provide precise geolocation information for vision-based sensors to identify an object such as a traffic light.
Combining relative sensing with absolute positioning creates a more complete understanding of the driving environment, essential for automated and connected driving scenarios such as multilane highways, complex interchanges and dense urban environments.
High-precision GNSS as the spatial backbone
High-precision GNSS provides the global reference required to enable large-scale deployment of cooperative and connected autonomous mobility (CCAM). CCAM takes autonomy to the next level by connecting vehicles to each other, to other mobility stakeholders (e.g. bicyclists) and to transportation infrastructure. With decimeter-level accuracy, precise GNSS allows vehicles to accurately understand their environment and predict scenarios beyond line-of-sight objects, which is currently out of scope for relative sensors.
CCAM supports vehicle-to-everything (V2X) use cases such as collision avoidance in urban intersections. The common coordinate framework anchors all the involved elements regardless of line-of-sight. In this context, positioning becomes central to autonomy, enabling vehicles to anticipate upcoming maneuvers, follow optimal trajectories and maintain performance when visual inputs degrade.
Satellite-based positioning is subject to inherent constraints from atmospheric conditions, signal reflections and challenging environments such as tunnels or urban canyons. Raw accuracy is typically limited to several meters. Correction services reduce these errors to the centimeter level required for lane-level autonomy.
Maintaining integrity through GNSS and INS fusion
Even with correction services, GNSS alone cannot meet reliability requirements in all environments. To address this, it is tightly integrated with inertial navigation systems (INS). Sensors such as inertial measurement units (IMU) and wheel odometry provide continuous motion data, while GNSS delivers long-term accuracy and calibration.
The interaction of these technologies is complementary. GNSS measurements continuously correct drift, while inertial sensors preserve continuity during temporary signal loss. Together, they provide a resilient positioning solution that supports safety-critical operations.
In the future, advanced driver assistance systems (ADAS) and automated driving systems (ADS) will require more than just precise navigation. Autonomy algorithms must also incorporate integrity information to assess the trustworthiness of positioning. This is addressed through the protection level concept, which defines with very high confidence a statistical bound within which the true receiver position lies. To ensure positioning integrity for lane-level navigation in line with the SOTIF approach, statistical forecasting methods are combined with extensive drive-testing data across diverse regions and environments.
Flexible, scalable, cost-effective positioning architectures
As vehicle architectures evolve toward software-defined platforms, positioning is increasingly delivered as a centralized software service rather than a fixed hardware function. As a result, location intelligence becomes an adaptable layer within the broader vehicle system, enabling a more flexible and scalable approach.
A modern architecture aggregates sensor data, applies correction services and fusion algorithms, and distributes a unified position estimate across the vehicle. Perception, planning and control systems operate from a shared spatial reference, while in-vehicle infotainment (IVI) solutions, including touchscreens, audio systems and connectivity modules, improve the overall driver experience.
Decoupling software from hardware allows OEMs to adopt best-in-class components without full system redesigns while maintaining compatibility with existing architectures. The result is reduced complexity, faster deployment and lower barriers to entry for OEMs and Tier 1 suppliers. High-precision positioning can therefore be introduced without major hardware changes, making it accessible across a broader range of vehicle platforms.
Scalability also depends on cost efficiency. A software-centric approach enables incremental deployment, allowing capabilities to expand over time and be offered as optional or subscription-based features.
By combining hardware flexibility, scalable software and cost-efficient integration, positioning evolves into a widely accessible foundation for autonomous driving. Lucid Gravity is on track to be the first electric vehicle to fully integrate Trimble’s positioning solution as part of its standard navigation equipment.
Synchronizing positioning with HD maps
HD maps play a central role in advanced autonomy, providing detailed information about road geometry, lane structure and infrastructure. But their effectiveness depends entirely on accurate vehicle localization.
Alignment between location data and map content allows vehicles to navigate with confidence, supporting safe guidance and predictive path planning. Even small discrepancies can lead to incorrect decisions, making both accuracy and integrity essential. If a vehicle has a detailed, up-to-date map but bad positioning, it won’t know what lane it is in. Alternatively, precise positioning without an accurate map is useless.
Positioning systems must therefore deliver clear confidence indicators that inform system behavior. Reliable localization enables vehicles to adapt in real time, maintaining safe operation in complex or changing environments.
Leveraging multilayer data
Future positioning systems will integrate additional data sources to improve accuracy and resilience. Vision-based localization, ultra-wideband and emerging satellite constellations will complement GNSS and inertial navigation.
To increase the reliability and accuracy of satellite-based positioning, low earth orbit (LEO) constellations such as Xona’s Pulsar will provide new, enhanced capabilities along with high levels of uptime. The integration of Trimble and Pulsar will benefit users in areas without reliable cell coverage, or with limited sky visibility and other challenging geographies.
Combining these inputs within advanced fusion frameworks will enable consistent performance and support the increasing demands of autonomous driving.
Positioning as the foundation of scalable autonomy
Bridging the gap between relative sensing and absolute positioning remains a key challenge facing the autonomy industry. Linking environmental awareness with a precise global reference enables vehicles to operate with the accuracy, reliability and trust required for universal deployment.
Positioning forms the foundation for scalable autonomy. It allows vehicles not only to perceive their surroundings but also to understand exactly where they are within them. That capability will ultimately determine how quickly and effectively autonomous driving can move from controlled environments to widespread adoption.