According to Mark Winton, Director of Product Development in the GNSS Division at Quectel Wireless Solutions, geolocation is maturing as new use cases are enabled by the emergence of cost-effective systems that harness the power of connectivity along with adjacent advances in data processing and artificial intelligence. (Amnesty International). In an interview with EEWeb, Mark Winton provided important information about geolocation. The demand for increased adoption of geolocation has led to the introduction of several technical developments to enable the effective management of data within applications such as logistics. At the same time, artificial intelligence and machine learning are gaining popularity as key elements for analyzing group patterns. Many human activities create distinct geographic patterns that can be analyzed to improve operations. Quectel highlights 4 important considerations: accuracy, and the adoption of real-time kinematic positioning (RTK) in GNSS units routinely achieves accuracy down to centimeter levels; Low cost opportunities Ultra low power consumption; and high-performance challenges with the growing demand for RTK-capable GNSS for use in precision agriculture. “Improved accuracy is essential for ADAS and DMS applications. High-performance sensors need high accuracy and refresh rates to operate across wide-temperature automotive environments. Low power is required because many vehicle systems need low power thresholds defined when the vehicle is turned off. Low cost is not the engine The largest in the automotive sector where high accuracy and performance are critical priorities.However, we are very competitive in the automotive field, without compromising on quality.We offer price flexibility through our ability to offer industrial grade IMU sensors on some of our automotive products Winton said. EEWeb: Looking at the geolocation market, what are the key business trends that you think are developing today? Mark Winton: Major business trends are unfolding due to some important technical developments that support the increased uptake of geolocation. Specifically, this is the confluence of geodatabases and the addition of rich metadata to GNSS data streams. This powerful data environment enables tagging of applications such as frozen cargo temperatures at their GPS location, closing the loop in secured cold chain logistics, and other automotive-related functions. At the same time, artificial intelligence and machine learning are deployed, enabling the analysis of cluster patterns. Many human activities create distinct geographic patterns that can be analyzed to improve operations. Cluster patterns of road accident locations can be derived, for example, from ambulance tailgate hatches that were geolocated to location information from the Global Navigation Satellite System (GNSS). These improvements are making our roads safer with smart technology that makes the world a better place. EEWeb: What are the “new sectors” for geolocation? Mark Winton: We’re seeing a dangerous rise in precision farming and farming equipment automation. This isn’t a completely new concept, but this vertical has been an area of expensive gear. The combination of improved accuracy, significantly lower pricing, and adoption by carriers and SAS providers for public NTRIP networks has opened up opportunities to track more equipment and processes that previously could only be reserved for a small portion of the most critical equipment. EEWeb: What future products and technologies are on your roadmap? Mark Winton: We’re currently focusing on the RTK and Dead account. We are in the process of releasing several sub-versions of our products and alternative chip versions of our products. This has allowed us to bypass the current shortage of components much easier than many of our traditional competitors. Flexibility with our clients also allowed us to move toward assignments to key markets. EEWeb: In which particular market do you think AI and machine learning would be useful in geolocation? Mark Winton: Automotive yard (ADAS and DMS) is the clearest published market, where GNSS is combined with AI. However, we are seeing small farmers using RTK GNSS along with machine learning and vision-based AI to automatically detect crop problems. This allows the control to be highly targeted, highly effective, and fully automated. We’re seeing very clear human-recognizable geographic patterns, which once combined with machine learning and artificial intelligence become very powerful tools. This is now widely used in mining applications, where bottlenecks in production processes are tied to specific geographic areas. It uses AI to identify constraints and run scenarios before implementing process changes to accurately model and optimize material flow through a complex plant. EEWeb: What are the challenges in RTK technology? In what specific agricultural application can this be advantageously adopted? Mark Winton: The main challenges for RTK are the stringent antenna requirements and compatibility standards used across different networks. It is not good to have a sky full of satellite constellations with only a few constellations supported. RTK works very well in open skies. There are limits to how small the antennas can be and where to install the device to get the best performance. The accuracy depends on the distance from the nearest upload point, and the NTRIP network density will need to be optimized. Large-scale, low-cost adoption is new. I think the challenges will disappear as the networks reap the revenue and the return on investment in the benchmarks allows for higher density. We are seeing the adoption of RTK in small greenhouses. We also have several projects with lawn mowers and standalone irrigation systems. .