Real World Results: Swift Navigation’s Latest Test Outcomes
This month, we’re diving deep into real world test results. From corrections delivery via LEO satellites to nation-wide drive tests and development of a novel robotics test platform, read on for the latest updates.
Low Earth Orbit (LEO) satellite networks—most prominently Starlink—present a new alternative to L-Band in regions without cellular coverage. We tested our SkylarkTM Precise Positioning Service on Starlink and achieved terrestrial-class precision with low latency and at a fraction of the cost of L-Band.
Read the blog post for detailed test results, a side-by-side comparison of LEO vs L-Band, and practice tips for implementation.
We partnered with Korea Telecom (KT) to evaluate the performance of Skylark across a 2,526 km drive spanning cities, highways, tunnels, and mountains. The test confirmed Skylark's ability to maintain accuracy through even the most challenging environments: it achieved 21x greater accuracy than uncorrected GNSS across the entire test, reducing error from 6.28 m to 29 cm.
Sometimes testing GNSS performance for a particular use case can be challenging due to the specific conditions of the use case. This was a problem we ran into for small outdoor robots. Testing GNSS performance requires the use of a ground truth reference system, but these are quite difficult to install on small robots because they cannot bear the additional weight and often operate in GNSS-constrained environments.
We developed an innovative solution to measure ground truth for small robots that leverages surveying equipment to precisely track the location of our robotic test platform as it moves around an area.
Read more about how we designed our ground truth system and see the test results.
