Autonomous devices require highly accurate sources of positioning to safely and successfully complete the jobs for which they are designed.
By Aaron Nathan, Founder and CEO at Point One Navigation.
Autonomous vehicles, drones, and robots are becoming popular solutions for a wide range of applications. According to a report from Data Bridge Market Research, this market is expanding at a compound annual growth rate (CAGR) of 19.7%, reaching $11.6 billion by 2030. The robot revolution has arrived!
As these autonomous devices become more commonplace and pervasive in our world, they will need to be able to navigate their surroundings with greater independence, accuracy, and safety.
Take, for example, the agriculture industry— autonomous tractors, seeding machines, herbicide drones, and harvesters must carefully make their way through crop fields while performing their repetitive functions in the same locations each time. Autonomous lawn mowers must follow a highly prescribed path every time they set out to do their job. Across a wide range of applications, autonomous devices require highly accurate sources of positioning to safely and successfully complete the jobs for which they are designed.
Accomplishing such precision requires two sets of location capabilities. One is for the autonomous device to understand its relative position to other objects. This provides critical input to understand the world around it and, in the most obvious case, avoid obstacles that are both stationary and under motion. This dynamic maneuvering requires an extensive stack of navigational sensors like cameras, radar, lidar (Light Detection and Ranging), and the supporting software to process these signals and give real time direction to the autonomous device.
The second set of capabilities is for the autonomous device to understand its precise physical location (or absolute location) in the world so it can repeatedly navigate, over the course of days, weeks, months, or years, a path that was programmed into the device. This requires a different set of navigational capabilities, starting with Global Navigation Satellite System (GNSS) technology. Augmenting GNSS are corrections capabilities like Real-Time Kinematic (RTK) and State Space Representation (SSR) that help drive 100x higher precision than GNSS alone for “open-sky” applications where GNSS is available, and Inertial Measurement Units (IMUs) combined with sensor fusion software for navigating where GNSS is not available (dead reckoning).
The risks of ineffective use of relative and absolute positioning can be severe. Imagine, for instance, the costs involved if an agricultural robot sprays herbicides in the wrong spot, ruining an entire section of crops. Safety is also a concern. Consider the consequences of a construction robot accidentally puncturing a gas pipeline or a self- driving forklift depositing materials in the middle of an aisle, causing a hazard for human workers and other machines.
Worker injuries, substantial product losses, and costly delays are all likely without the combination of highly accurate relative and absolute positioning. Essentially, this means that anywhere a mistake of more than a few centimeters may result in significant performance or safety concerns, the autonomous device needs both relative and absolute positioning.
GNSS is not enough
The baseline technology used for absolute positioning starts with GNSS (colloquially known as the Global Positioning System or GPS). Given that GNSS is affected by atmospheric conditions and satellite inconsistencies, it can give a reading that is off by many meters. For autonomous devices that require more precise navigation, this isn’t good enough—thus leading to the emergence of a field known as “corrections,” which narrows this error down to as low as one centimeter.
Two of the main “open sky” corrections technologies that can be employed where GNSS is available are RTK and SSR, which may be summarized as follows:
• RTK uses a network of base stations with known positions as reference points for correcting GNSS receiver location estimates. So long as the device is within 50km of a base station, RTK can reliably provide 1- to 2-centimeter accuracy.
• SSR leverages information from the base station network but—instead of sending corrections directly from a local base station—it models the errors across a wide geographical area. The result is broader coverage and no requirement to be within 50km of a base station. Accuracy broadens slightly to within 3 to 10 centimeters.
However, many autonomous vehicles may operate in (or move into) areas where there is an obstruction between the GNSS receiver on an autonomous vehicle or an autonomous mobile robot (AMR) and the sky. This can happen inside factories and storehouses, and in tunnels, parking garages, tree covered areas, and urban environments. This is where Inertial Navigation Systems (INS) come into play with their IMUs and Sensor Fusion software.
• IMUs combine accelerometers, gyroscopes, and (sometimes) magnetometers to measure a system’s linear acceleration, angular velocity, and magnetic field strength, respectively. This is crucial data that enables an INS to determine the position, velocity, and orientation of an object relative to a starting point in real time.
• Sensor Fusion software is responsible for combining data from the IMU, along with the data from other sensors, to create a cohesive and accurate understanding of an AMR’s absolute location when GNSS is not available. It essentially “fills in the gaps” in real time between when the GNSS signal is dropped and when it is picked back up again by the AMR. The accuracy of sensor fusion software depends on several factors, including the quality and calibration of the sensors involved, the algorithms used for fusion, and the specific application or environment in which the system is deployed.
Choosing the Best RTK Network
RTK for GNSS provides a highly accurate source of absolute location for open-sky applications. Consistently reliable RTK corrections require a highly dense network of base stations, so receivers are always within close enough range for accurate error corrections. The larger the network, the easier it is to get corrections from anywhere.
For example, Point One’s Polaris network features more than 1,700 base stations across the globe. There is already ample coverage across the
U.S., Canada, Puerto Rico, Europe, Australia, and South Korea, and Polaris continues to add more base stations every month, even allowing customers to contribute to this base station growth by hosting one or more on their own site.
This is a critical time for robotic and vehicle automation development. Today, sensors of all kinds can provide the essential data to drive this technology forward. However, without a reliable source of absolute positioning, the sensor stack for AVs and ADAS is incomplete.
By using RTK receivers and networks, GNSS receivers can provide a reliable, centimeter- accurate estimate of the vehicle’s position in the real world. And, by adding IMU data via sensor fusion, autonomous vehicles can keep close tabs on their position in real time, even when GNSS signals are blocked or unavailable.
Developing comprehensive sensor stacks for vehicle automation is the most effective way to overcome the obstacles that remain between current technology and a future where we can safely hand over more control to self-driving vehicles and robots.
