GNSS and Autonomous Driving - Precision Under Pressure

The development of autonomous driving is rapidly reshaping the global automotive industry. By 2026, the market for autonomous vehicles (AVs) is expected to reach $557 billion, highlighting the massive growth potential of this technology. More than 500 companies around the world are currently involved in the development of autonomous vehicle technology, with major players based in the United States, China, and Germany. The number of autonomous vehicles on the road is expected to grow from 8.5 million in 2019 to more than 20 million by 2030.

In addition, the industry will introduce Level 3 (L3) autonomous driving, which allows hands-off driving under certain conditions, with premium automakers planning to introduce such features in flagship models by 2024. From 2023, regulatory frameworks supporting these advances are evolving, particularly in North America, Europe and Asia, paving the way for wider adoption.

Navigating the Future: Understanding Autonomous Driving Levels

The future of transportation is being fueled by autonomous driving technology. Using advanced computing, artificial intelligence, mapping, and sensor systems, autonomous vehicles aim to achieve higher levels of safety and efficiency than human drivers. The Society of Automotive Engineers (SAE) has defined six levels of driving automation, ranging from Level 0 (no automation) to Level 5 (full automation).

Today, most vehicles equipped with autonomous driving systems fall into the Level 2 (L2) category. These systems, often referred to as Advanced Driver Assistance Systems (ADAS), provide features such as lane-keeping and adaptive cruise control. L2 autonomous driving enhances the driving experience in scenarios such as highway driving but cannot autonomously handle complex road conditions. The driver still needs to pay constant attention to road information and operate the vehicle.

Level 3 (L3) automation, also known as "conditional automation," is gradually moving out of the lab and into our daily lives. At this level, vehicles can perform most driving tasks autonomously, including navigation, lane changing, and responding to traffic conditions. Although the driver still needs to monitor the vehicle's work to prevent accidents, the number of operations required is greatly reduced.

The Role of GNSS in Autonomous Driving Ecosystems

Level 2 and higher autonomous driving systems require a comprehensive sensor ecosystem to provide the environmental information and positioning data needed for decision making. This ecosystem typically includes:

Situational Awareness Sensors

• RGB Cameras: Provide visual input.
• LiDAR (Light Detection and Ranging): Offers precise 3D mapping.
• Millimeter-Wave Radar: Detects objects and measures speed.
• Ultrasonic Sensors: Used for short-range obstacle detection.

Positioning and Navigation

• GNSS/INS Systems: Deliver accurate location and timing information essential for autonomous navigation.

The GNSS/INS combination is fundamental as it provides the backbone for vehicle situational awareness and decision making. This technology enables vehicles to accurately determine their position on the road, navigate complex road networks, make informed decisions about routing and maneuvering, and synchronize with other vehicles and infrastructure.

GNSS/INS Sensors: Overcoming Challenges in Autonomous Vehicle Navigation

Unlike GNSS receivers used in surveying applications, those designed for autonomous driving face a unique set of challenges:

• Higher Positioning Accuracy: Autonomous vehicles require sub-meter and even centimeter accuracy to determine which lane the vehicle is in. Tall buildings in city centers can block satellite signals, reducing the number of satellites visible to the receiver and potentially disrupting real-time kinematic (RTK) connections. In addition, multipath caused by signal reflections off buildings can affect positioning accuracy.
• Environmental Adaptability: GNSS receivers in vehicles must contend with constant vibration and extreme temperature variations. Beyond physical reliability, maintaining consistent positioning accuracy is crucial for safe driving.
• Real-Time Precise Positioning: Autonomous vehicles need to maintain accurate positioning data in real-time, even in challenging environments like urban canyons, under overpasses, and in tunnels where GNSS signal loss and RTK connection interruptions are common.

Tightly Coupled GNSS + INS Systems: The Backbone of Reliable Autonomous Navigation

A tightly coupled GNSS (Global Navigation Satellite System) + INS (Inertial Navigation System) receiver is an advanced technology that significantly improves navigation accuracy, reliability and robustness. This system integrates satellite-based positioning with inertial measurements, making it essential for applications that require precise navigation, such as autonomous vehicles, aerospace, and precision agriculture.

In a tightly coupled system, the GNSS and INS sensors work in close coordination. The GNSS component provides global positioning data by receiving signals from multiple satellites, while the INS uses accelerometers and gyroscopes to track the receiver's position, velocity, and orientation relative to a known starting point. Unlike a loosely coupled system, where the GNSS and INS operate largely independently and integration occurs at the output level, a tightly coupled system fuses the raw data from both sensors at a much earlier stage. This fusion typically occurs within a Kalman filter framework that continuously processes GNSS pseudorange and Doppler measurements along with raw acceleration and rotation rate data from the INS.

The key advantage of a tightly coupled system is its ability to maintain an accurate navigation solution even in challenging environments where GNSS signals may be weak or obstructed. The INS provides continuous navigation data that fills gaps in GNSS coverage, while the GNSS data helps correct the inherent drift of the INS over time. This synergy results in a more reliable and accurate navigation solution than a loosely coupled system that may struggle during GNSS signal outages.

CHCNAV's Advanced Solution: The CGI GNSS+INS Sensor Series

CHCNAV understands the importance of high-precision positioning in autonomous driving. Our CGI tightly coupled GNSS+INS sensors are specifically designed for precise positioning and navigation in a variety of vehicles. For example, the CGI-610 GNSS/INS sensor addresses these challenges through several innovative approaches:

• Centimeter-Level Accuracy: The CGI-610 delivers centimeter-level positioning accuracy, ensuring precise vehicle localization for safe autonomous operation.
• Robust Positioning: By leveraging multiple GNSS constellations, the CGI-610 provides robust positioning even in challenging urban environments.
• Advanced Algorithms: These enable quick initialization and re-convergence, minimizing downtime and enhancing reliability.
• Durability: The CGI-610 is built to perform in diverse conditions, from extreme temperatures to high-vibration environments.

Proving Excellence: Field Testing the CGI-610 in Real-World Scenarios

To validate the performance of the CGI-610, CHCNAV conducted field tests in complex urban environments in Japan. These tests evaluated the system's accuracy and usability under real-world conditions, including challenging scenarios such as urban environments, elevated roads, highways, and tunnels.

The CGI-610 was mounted in a light truck and connected to an external antenna and GNSS RTK correction network. The test results showed remarkable performance:

• Urban Central Business District Road Scenarios: The CGI-610 maintained positioning accuracy with an RMS value within 0.7m and heading accuracy around 0.1°.
• Open Urban Road Scenarios: Positioning accuracy improved to within 0.1m RMS, with heading accuracy remaining around 0.1°.
• Tall Buildings Central Business District Complex Scenarios: The CGI-610 maintained positioning accuracy within 0.5-1m RMS and heading accuracy around 0.1°.
• Tunnels: Positioning errors increased proportionally to tunnel length but remained within 1.5‰ at maximum. Overall RMS values in tunnels were maintained within 0.5m, with heading accuracy around 0.1°.

Pushing the Boundaries: The Future of GNSS+INS Technology in Autonomous Driving

The development of autonomous driving technology faces significant technical challenges. One of the most pressing is ensuring the reliability of real-time environmental sensing systems. These sensors must accurately detect and interpret data from diverse and complex driving conditions, including weather variations and urban environments, placing high demands on component manufacturers and automotive integrators.

As a leader in GNSS+INS technology, CHCNAV is committed to addressing these needs and pushing the boundaries of GNSS technology for autonomous driving. CHCNAV's new generation receivers, the CGI-830 and CGI-230, build on the strengths of the CGI-610 and offer significant improvements in INS accuracy and POS support.

As autonomous vehicle technology continues to evolve, the demand for highly accurate and dependable GNSS+INS solutions will only increase. CHCNAV remains at the forefront of this technology, developing innovative solutions to meet the demanding needs of the autonomous driving industry.

About CHC Navigation

CHC Navigation (CHCNAV) creates innovative mapping, navigation, and positioning solutions to make customers' work more efficient. CHCNAV products and solutions cover multiple industries such as geospatial, construction, agriculture, and marine. With a presence across the globe, distributors in more than 130 countries, and more than 1,900 employees, CHC Navigation is recognized as one of the fastest-growing companies in geomatics technologies. For more information about CHC Navigation [Huace:300627.SZ ], please visit: www.chcnav.com .