Your rover just lost RTK fix under a bridge overpass, and your layout points are 200 feet down the roadway. GNSS can’t see satellites through steel and concrete, so you’re stuck waiting or walking back to clear sky. An Inertial Navigation System (INS) wouldn’t blink, it measures position through motion sensors, not satellites. But INS drifts over time, accumulating errors that make it useless for long-duration surveys without correction. The real question is understanding what each system does well, where each one fails, and when combining them makes sense.
What GNSS Actually Does
How Inertial Navigation Systems Work
The Drift Problem with Pure INS
GNSS/INS Integration: Covering Each Other’s Weaknesses
When GNSS Alone Gets the Job Done
Where INS Integration Becomes Essential
Understanding IMU Quality Grades
When Hybrid Approaches Make More Sense
Positioning Technology That Matches Your Workflow
GNSS positioning relies on satellite signals. Your receiver tracks GPS, GLONASS, Galileo, and BeiDou satellites, measures signal travel time from each, and calculates position based on geometry. In RTK mode with corrections from a base station, GNSS delivers centimeter-level accuracy anywhere satellites are visible.
The strength of GNSS is absolute positioning. It doesn’t accumulate error over time, each measurement is independent. Whether you’re on a point for 10 seconds or 10 hours, accuracy stays consistent. GNSS systems also provide precise time synchronization and velocity measurements as byproducts of the positioning calculation.
Standard GNSS (autonomous mode) gets you 2-10 meter accuracy. RTK GPS systems push that to 1-2 centimeters horizontally and 2-3 centimeters vertically. For surveying, construction layout, and machine control, RTK GNSS is the baseline technology that everything else compares against.
The limitation is simple: GNSS needs a clear sky view. Satellites orbit 20,000 kilometers overhead, and their signals are weak by the time they reach Earth. Buildings, bridges, tree canopy, tunnels, anything that blocks line of sight to four or more satellites kills your position solution.
INS takes a completely different approach. Instead of external reference signals, an inertial navigation system measures acceleration and rotation using accelerometers and gyroscopes. It starts from a known position and velocity, then integrates sensor measurements to calculate how far you’ve moved and in what direction.
Think of it as dead reckoning with modern sensors. The system knows where it started, measures every acceleration and turn, and updates position continuously based on that motion. No satellites required, no external signals, no dependency on infrastructure.
High-quality INS units update at 100-200 Hz, capturing motion changes far faster than GNSS (which typically updates at 1-20 Hz). That high-frequency data tracks dynamic motion, sharp turns, rapid elevation changes, vibration, with precision GNSS can’t match in short time windows.
The advantage shows up in GNSS-denied environments: tunnels, under bridge decks, inside buildings, and other GNSS denied areas. INS keeps positioning through obstructions where GNSS drops out entirely. For autonomous vehicles, marine navigation, or aircraft, that continuity matters more than absolute accuracy.
INS sounds perfect until you understand error accumulation. Every sensor has bias, a tiny constant error that adds up over time. Gyroscopes drift, accelerometers have zero-point offset, and integrating those small errors twice (once for velocity, again for position) compounds the problem fast.
A navigation-grade INS might drift 1-2 meters per hour. That sounds acceptable until you realize survey-grade work needs centimeter accuracy over 8-hour days. Consumer-grade MEMS sensors drift far worse, 10-100 meters per hour depending on quality.
Pure INS works for short durations. Aircraft use it for approach and landing where GPS isn’t reliable. Submarines navigate underwater for hours. But for continuous positioning over long periods, INS needs external corrections to reset accumulated drift.
Integrated GNSS/INS systems combine both technologies in a single unit. GNSS provides absolute position when satellites are visible. INS fills gaps when GNSS drops out. The fusion algorithm continuously corrects INS drift using GNSS measurements, and INS smooths GNSS noise and maintains position through brief signal loss.
The result is better than either system alone:
For surveying and construction, the practical benefit is fewer dropouts and faster recovery. You don’t lose fix every time you walk near a building or under tree branches. The system rides through brief obstructions on INS and reacquires GNSS seamlessly.
Most survey and construction work doesn’t need INS integration. If you’re working in open sites, subdivisions, highway layout, agricultural fields, open-pit mining, GNSS maintains clear sky view consistently. RTK GPS systems deliver centimeter accuracy without the added cost and complexity of inertial sensors.
GNSS-only setups make sense when:
The Hemisphere S631 and similar RTK receivers handle most survey applications without inertial assistance. Boundary surveys, topographic mapping, construction staking, and machine control all run on GNSS alone as long as you’re working under open sky.
GNSS/INS systems justify their cost in specific scenarios where continuous positioning through obstructions is non-negotiable.
Mobile mapping and LiDAR. Vehicle-mounted mapping systems travel through urban corridors, under overpasses, and past buildings constantly. GNSS drops out every few seconds in dense cities. INS maintains position and attitude between GNSS fixes, keeping LiDAR point clouds georeferenced accurately.
Corridor surveys and right-of-way mapping. Utility corridors, pipeline routes, and transmission line surveys often work in mixed environments, open fields transitioning to forest canopy or urban areas. INS fills coverage gaps where GNSS struggles.
Marine and hydrographic surveying. Boats pitch, roll, and heave in waves. INS measures vessel attitude continuously, which matters for multibeam sonar and positioning antennas above the waterline. Combined GNSS/INS systems deliver stable position and orientation data through motion.
Autonomous vehicle control. Self-driving construction equipment and agricultural machinery need continuous, high-frequency position updates. GNSS/INS integration provides redundancy and maintains control through brief signal loss.
Aviation and drone mapping. Aircraft and UAVs move fast and cover varied terrain. INS provides high-rate position and attitude for camera triggering and direct georeferencing of imagery.
Not all inertial sensors perform equally. IMU quality directly affects how long the system maintains accuracy during GNSS outages.
MEMS (consumer/tactical grade) sensors are small, inexpensive, and drift quickly. Expect 10-50 meters of position error after 60 seconds of GNSS loss. Fine for brief dropouts but not for extended outages.
Tactical-grade IMUs improve drift performance to 1-5 meters per minute. These work for mobile mapping and most integrated survey applications where GNSS outages last seconds, not minutes.
Navigation-grade IMUs maintain sub-meter accuracy for minutes during GNSS loss. Aircraft and marine vessels use these for long-duration autonomous navigation. Overkill and expensive for land survey work.
For survey and construction applications, tactical-grade IMUs hit the sweet spot between performance and cost. They handle brief obstructions without breaking budgets.
Before committing to full GNSS/INS integration, consider whether simpler solutions solve your actual problem.
Improve GNSS performance first. Multi-frequency receivers (L1/L2/L5) maintain lock in partial obstructions better than single-frequency units. Better antennas with improved multipath rejection help. Sometimes upgrading your GNSS receiver solves 90% of dropout issues without needing INS.
Use tilt compensation instead. Many modern RTK poles include tilt sensors that correct for pole angle, letting you measure points without perfect verticality. This isn’t INS, it’s a simple inclinometer, but it improves productivity in ways that overlap with what people think they need INS for.
Combine GNSS with total stations. For projects mixing open and obstructed areas, running RTK in the open and switching to a robotic total station under canopy or near buildings often costs less than integrated GNSS/INS.
GNSS and INS solve different problems. GNSS delivers absolute accuracy anywhere satellites are visible. INS maintains positioning through brief obstructions but drifts over time. Integrated systems combine both, trading cost and complexity for continuous coverage.
Bench-Mark equips survey crews and contractors across the U.S. with positioning systems ranging from standalone RTK receivers to full GNSS/INS integration. Because the best positioning system is the one that solves your specific problems without breaking your budget.
For most survey and construction work in North America, RTK GPS systems handle positioning requirements without inertial assistance. Open job sites, rural mapping, and standard layout don’t need the continuous positioning that justifies GNSS/INS investment. Save the money and buy better GNSS receivers, antennas, or additional rovers instead.
Integrated systems make sense for mobile mapping, marine surveying, and applications where GNSS outages happen frequently enough to impact productivity or data quality. If you’re regularly working in urban canyons, under heavy canopy, or on dynamic platforms, GNSS/INS delivers value worth the premium.
