Multicoin Capital: Why is Geodnet a key piece of the physical AI era?

Original: Shayon Sengupta, Partner at Multicoin Capital

Compilation: Yuliya, PANews

Cryptocurrency investment firm Multicoin announced on February 25 that it has strategically acquired $8 million worth of GEOD tokens from the Geodnet Foundation. Geodnet is a precision network that provides critical positioning services for drones, autonomous vehicles, and potentially emerging robots in the future through the DePIN( decentralized physical infrastructure network) economic model.

The era of physical AI is coming

Although we have not yet reached the future scenario depicted in Blade Runner 2049 in which humanoid robots live together with humans, in 2025, professional robots have been quietly integrated into all aspects of modern life. At present, language models are commoditizing intelligence, unleashing new creativity and changing the shape of knowledge work. However, the evolution of AI is not limited to text, images, and video, but is extending to the physical world, and robotics will be the next important frontier.

With the improvement of AI model capabilities, their ability to understand and interact with the physical environment will be significantly enhanced. This transformation will give birth to new autonomous systems, from autonomous driving cars, humanoid robots to drones and industrial automation equipment, these systems will operate with unprecedented intelligence and autonomy. And in order for these robots to perform their functions, they all need to answer a fundamental question: where am I?

Location: A $5 billion dollar puzzle

When tens of millions of independent and semi-independent devices are deployed, they all face the same challenge: spatial awareness. Drone delivery, autonomous vehicles navigating in dense urban environments, precision seeding by agricultural machinery, etc., all require sub-centimeter-level positioning accuracy to operate safely and efficiently.

To achieve such high precision, these systems rely on sensor fusion, combining multiple positioning methods.

Lidar (LiDAR) provide high-resolution depth mapping, but it doesn't perform well in foggy or rainy days (and has problems with heavy weight, high power consumption, and high cost). Radar(RADAR) provides robust distance measurements but lacks fine accuracy. Vision-based SLAM technology supports real-time mapping, but performance is significantly degraded in low-visibility environments.

Among the existing positioning solutions, the positioning system based on GNSS (Global Navigation Satellite System) combined with RTK (Real-Time Kinematic) base station correction is recognized as one of the most reliable solutions.

Standard GNSS positioning has an inherent accuracy deviation of 5-10 meters due to atmospheric interference and multipath errors. RTK solves this problem by using a fixed base station to compare satellite signals with known positions and transmit real-time correction data, achieving centimeter-level accuracy—which is crucial for physical AI applications. Currently, mainstream robotics and automation companies have integrated RTK into their systems:

• DJI is used for high-precision drones;
• John Deere is used for autonomous driving tractors;
• Tesla is used for self-driving cars;
• Boston Dynamics used for industrial inspection robots.


Source: GNSS Store

However, RTK is currently facing cost challenges. The construction and maintenance of RTK base stations require a large amount of capital investment, including hardware, site acquisition, and maintenance, among other investments. Traditional suppliers such as Trimble, Hexagon, and Topcon build their own base stations and charge thousands of dollars for each tracking device, which limits their accessibility and deployment speed. The coverage often focuses on profitable areas, leading to many areas still being unable to access high-precision positioning services.

) Reconstruct the cost structure with DePIN

Based on the contributor ownership model of token incentives, DePIN fundamentally reconstructs the cost structure of global hardware-intensive networks. As the world's largest pinpoint network, Geodnet leverages this principle to incentivize independent operators to install RTK base stations and provide correction data in order to earn token rewards.

By shifting the cost burden from a single operator to a distributed network, Geodnet is able to deploy production-level RTK base stations at a cost much lower than traditional suppliers, mainly by eliminating the two major costs of network construction: land and labor. This enables Geodnet to expand coverage more quickly, ensure redundancy, and provide GNSS correction services that are more reliable than existing service providers.

This model has created an organically expanding global distributed network, significantly reducing end-user costs while ensuring service quality. Geodnet has been rapidly expanding since last year, growing from 1,400 base stations in November 2022 to 7,800 in June 2024, and as of January 2025, it has over 13,000 base stations, covering 4,377 cities in more than 142 countries.

! [Multicoin Capital: Why is Geodnet a key part of the physical AI era?] ]###https://img.gateio.im/social/moments-32c273db5128cc5091b7c6b4000ebdf7(

At present, the Geodnet network has achieved "threshold scale", and the service capacity covers more than 60% of the GNSS correction market. Notably, the network has been extremely capital-efficient in obtaining this supply – only 11% of the tokens have been distributed to contributors over the past three years.

)

Geodnet is a good example of the DePIN model in practice. Not only is the acquisition and deployment cost of Geodnet base stations an order of magnitude lower ($700 for consumer miners vs. $12,000 for enterprise RTK base stations), but the annual pricing is also significantly reduced due to denser consumer-grade network deployments. Geodnet is undoubtedly the most scalable and cost-competitive positioning solution in the world today.

At present, Geodnet has generated about $3 million in annualized on-chain revenue by providing services to leading drone companies such as Propeller, DroneDeploy, and Quectel, as well as government agencies such as the USDA (United States Department of Agriculture), showing the fastest demand-side growth in the DePIN field.

! [Multicoin Capital: Why is Geodnet a key part of the physical AI era?] ](https://img.gateio.im/social/moments-8d8e6fb701696ae7be5e4f11d53ae9b3)

In early conversations with Geodnet founder Mike Horton, when asked how to win the world's most influential autonomous driving, drone, and agricultural robot customers, his answer has always been consistent: the core structural cost advantage brought by the DePIN mode.

( The turning point has arrived

In the next ten years, robotic technology will enter a period of explosive growth. From industrial manufacturing to daily life, tens of millions of robots are expected to be deployed in various fields. Drones will inspect power lines and pipeline facilities in the air, self-driving cars will reshape the pattern of freight and transportation services, intelligent warehousing robots will largely replace manual operation, and household humanoid robots will gradually enter thousands of households to assist with daily tasks.

Precise positioning capabilities are the cornerstone of this robotics revolution. Without reliable positioning data, it will be difficult for any autonomous robot to function as it should. In the case of autonomous vehicles, for example, in urban canyons with weak GPS signals, RTK-enhanced GNSS systems must be relied upon to navigate safely; Delivery drones also require centimeter-level accuracy to land accurately at the designated charging station location. As a result, RTK positioning has become more than just a complement to sensors such as lidar and radar, but has become a critical infrastructure for robot operation.

Currently, many robot companies pay high fees for traditional GNSS correction services every year, but still face many problems such as limited coverage and unstable accuracy. In contrast, Geodnet, through its innovative DePIN mode, not only significantly reduces the cost of services, but also provides higher-quality positioning services. This breakthrough solution is attracting more and more enterprise-level customers, providing reliable positioning guarantee for current and future autonomous robots.

The AI-driven robot revolution has become an irreversible trend, and the key lies in the timing of the full-scale explosion. The robotics industry is accelerating, and Geodnet's precise positioning network will ensure that these robots can operate safely and efficiently. In the coming era of robots, Geodnet's location services will play an irreplaceable role in providing accurate and reliable location information support for robots around the world.