MATRIX Positioning
Algorithmic Magic to Enhance Precision, Reliability, and Ease, for an Exceptional User Experience.
MATRIX is PRECISE's positioning algorithm engine, built from 500+ algorithm modules and 3,000+ algorithm parameters that work together to resolve a fixed position and keep it fixed, across every product line that depends on GNSS positioning.
How It Works
A Data-Driven Positioning Philosophy
MATRIX is built on a data-driven philosophy: rather than applying one fixed correction model to every positioning scenario, it draws on a large modular library and selects and tunes the combination that fits the current signal environment, satellite geometry, and motion state.
Efficient
Optimizing how fast the engine iterates toward a solution, so a fix is reached and re-validated quickly rather than through brute-force computation.
Comprehensive
Drawing on the breadth of the module and parameter library to handle the range of conditions a receiver actually encounters in the field, rather than a narrow set of "ideal" scenarios.
Precise
Evaluating the final output against the accuracy the application actually needs, whether that's centimeter-level real-time positioning or millimeter-level post-processed results.
Dynamic Mode
Continuous, real-time positioning for moving platforms: autonomous driving systems, drones, and any application where the receiver is in motion and needs an unbroken position stream.
Static Mode
Single-point precision positioning for stationary or slow-moving work: real-time surveying and post-processed monitoring, where the priority is maximum achievable accuracy at a known location rather than continuous tracking through motion.
Core Architecture
The Three Core Algorithm Modules
RTK Algorithm Module
Uses machine learning to address problems that trip up conventional RTK processing: automatic recognition of the current scene so the engine can adapt its approach, AI-driven satellite selection to weight the signals most likely to produce a reliable fix, and ambiguity validation to confirm a resolved position before reporting it. The module resolves high-precision position from carrier-phase observations shared between the base station and the rover, using the same base-rover RTK principle found across GNSS surveying with an adaptive, learning-based layer added on top.
PVT Algorithm Module
Handles Position-Velocity-Time computation using multi-frequency, non-combined signal updates, processing each frequency's raw observations directly rather than only working from a combined/differenced signal, which preserves more of the available information. The module combines prior information with posterior information to make fuller use of available data, and uses INS (inertial) data to assist GNSS from multiple directions, letting the module adapt its parameters automatically to whatever scenario the receiver is operating in.
Integrated Algorithm Module (GNSS+IMU)
Fuses GNSS and IMU data through a multi-layered fusion architecture, using robust filtering design and adaptive sensor fusion with built-in fault diagnosis, to produce a combined estimate of position, velocity, and attitude. This is the module that underlies PRECISE products that need orientation and motion data alongside position.
The Real Numbers
What MATRIX Actually Delivers
99.9%
Average Fix Rate
Under normal field conditions, only 1 in 1,000 positioning attempts fails to reach a fixed solution.
+18%
Fix Verification Success
Improvement from the AI Data Correction Algorithm Module, applied to real-time positioning data.
500+
Algorithm Modules
The scale of the underlying library the RTK, PVT, and Integrated modules draw on and tune from.
3,000+
Algorithm Parameters
Tuned combinations selected to fit the current signal environment, satellite geometry, and motion state.
Fix rate and accuracy can be affected by external conditions such as multipath, obstacles, satellite geometry, and atmospheric conditions. See the FAQ below for how the AI Data Correction and Partial Ambiguity Resolution modules produce the +18% and 99.9% figures above.
Where MATRIX Runs
The Shared Positioning Core
MATRIX is the shared positioning core across PRECISE's product line, not a feature exclusive to one device.
RTK Receiver
PRECISE X7
MATRIX processes multi-frequency, multi-constellation GNSS signals with scene-aware satellite selection, ambiguity validation, and INS-assisted optimization, delivering a 99.9% fix rate under normal field conditions and typically converging to a fix in under 5 seconds.
Learn moreAdditional product links to be added as pages go live: X7 Lite, X, BASE2, T3, T3 Lite, C, A2, A Pro, E2.
FAQ
Common Questions
MATRIX is the software layer inside PRECISE products that processes raw GNSS (and, where applicable, IMU) signals into a resolved position. It's built from a library of 500+ algorithm modules and 3,000+ parameters, organized into three core modules (RTK, PVT, and Integrated GNSS+IMU), and it runs underneath every PRECISE product that reports RTK positioning or a fix rate.
It means that, on average, only 1 out of every 1,000 positioning attempts fails to resolve into a fixed solution under normal conditions. It's a reliability statistic about how often MATRIX successfully reaches a fix, not a claim that every fix is perfectly accurate regardless of conditions. Fix rate and accuracy can still be affected by multipath, obstacles, satellite geometry, and atmospheric conditions.
It applies an XGBoost machine-learning model, trained on a large pre-training dataset, to generate a correction function for real-time positioning data. This improves the real-time fixed-solution verification success rate by at least 18%, meaning fewer valid fixes are incorrectly rejected or left unresolved.
It's the module responsible for resolving carrier-phase ambiguity into a final fixed position, using the LAMBDA method. When signal conditions are imperfect, it can perform up to ten intelligent satellite-exclusion operations, progressively removing the weakest or most compromised satellite signals, so the engine can still reach a reliable fix instead of failing outright because of a few bad signals.
Each targets a different output: RTK resolves a high-precision position from base-rover carrier-phase data, PVT computes position, velocity, and time together from multi-frequency, INS-assisted updates, and Integrated (GNSS+IMU) fuses both sensor types into one estimate of position, velocity, and attitude for products where orientation and motion matter, not just position. See the three core modules above for what makes each one adaptive.
The underlying engine is shared, but how it's applied depends on the product. In RTK receivers like X7, MATRIX drives the primary position fix. In 3D scanners like S7 and S2, MATRIX runs the onboard RTK module used to georeference SLAM-generated point clouds, playing a supporting role to the scanner's own positioning system rather than serving as its core tracking method.
See MATRIX at work in the product built for your job.
Whether you need a survey-grade RTK receiver, a machine guidance system, or a 3D scanner with georeferenced output, MATRIX is the positioning engine behind the fix. Talk to a PRECISE engineer about which product configuration fits your project.