LiDAR Trajectory Precision vs. LiDAR Point Cloud Accuracy

1. Introduction

One of the most common misconceptions in LiDAR projects is assuming that the trajectory quality reported by the GNSS/IMU processing software directly represents the final accuracy of the LiDAR point cloud.

❓ "If my POSPac report shows trajectory accuracies better than 2–3 centimeters, does that mean my LiDAR point cloud is also accurate to 2–3 centimeters?"

The short answer is:

🔴 No.

While trajectory precision and point cloud accuracy are closely related, they are fundamentally different metrics that describe different aspects of the LiDAR workflow.

Understanding this distinction is essential when evaluating data quality and validating the final deliverables of a LiDAR survey.

2. What is Trajectory Processing? 

🛰️ Trajectory precision refers to the estimated quality of the GNSS/IMU solution used to determine the position and orientation of the LiDAR sensor during data acquisition. 

During post-processing in POSPac, a Smoothed Best Estimate of Trajectory (SBET) is generated. The SBET represents the best estimate of the platform's position and attitude throughout the survey.

The POSPac Quality Report includes the Smoothed Performance Metrics, which provide RMS uncertainty estimates for:

• North
• East
• Down

For example:

These values indicate the estimated uncertainty of the trajectory solution and demonstrate the quality of the GNSS/IMU processing.

However, these metrics describe only the trajectory and should not be interpreted as the final accuracy of the LiDAR point cloud.

3. What is Point Cloud Accuracy? 

🎯 Point cloud accuracy describes how closely the LiDAR points represent their true positions in the real world.   

Unlike trajectory precision, point cloud accuracy cannot be determined solely from the GNSS/IMU solution. Instead, it must be evaluated using independent reference measurements collected in the field.

Common reference data include:

• Ground Control Points (GCPs)
• Checkpoints
• Survey monuments
• Accuracy Stars and calibration targets

By comparing the LiDAR point cloud against these independently surveyed locations, accuracy metrics such as the following can be calculated:

• RMSE X
• RMSE Y
• RMSE Z
• Horizontal RMSE
• Vertical RMSE

These metrics quantify the actual positional accuracy of the final LiDAR product.

4. Why High Trajectory Precision Does Not Guarantee High Point Cloud Accuracy

The trajectory solution is only one component of the direct georeferencing process. Several additional factors influence the final position of each LiDAR point.

⚙️ System Calibration

Errors in boresight angles or lever-arm measurements can introduce systematic offsets into the point cloud even when trajectory precision is excellent.

🌎 Coordinate Reference System Configuration

Incorrect datums, geoid models, projections, or reference epochs can create significant offsets between the point cloud and ground control.

🎛️Sensor Performance

IMU performance, GNSS signal quality, flight dynamics, and environmental conditions can all affect the final point positions.

🔄Processing Workflow

Additional processing steps such as trajectory generation, strip adjustment, and calibration refinement also contribute to the final accuracy of the dataset.

For these reasons, a trajectory with centimeter-level precision does not automatically produce a point cloud with centimeter-level accuracy.

5. Practical Example

Consider a survey with the following POSPac trajectory statistics:

Based on the trajectory report alone, the GNSS/IMU solution appears to have sub-2-centimeter precision.

However, when the resulting LiDAR point cloud is compared against independent checkpoints, the following results are obtained:

The final point cloud accuracy is therefore approximately:

• 4 cm horizontal
• 5 cm vertical

Although the trajectory quality is excellent, the point cloud accuracy reflects the cumulative effect of all components within the georeferencing and processing workflow.

6. Evaluating Point Cloud Accuracy in LP360 🔍

LP360 provides tools for validating LiDAR datasets against independent control points and checkpoints.

These workflows allow users to:

• Identify horizontal and vertical offsets.
• Detect systematic biases.
• Calculate RMSE statistics.
• Validate final deliverables against project specifications.

If discrepancies are identified, additional investigation can be performed to determine whether the source is related to calibration, georeferencing, coordinate systems, or processing parameters.

For troubleshooting workflows involving control points and LiDAR data, refer to:

Control Point Dialog

Troubleshooting Error/Offset between GCPs and Point Cloud

7. Conclusion ✅

Trajectory precision and point cloud accuracy are not interchangeable metrics.

Trajectory precision describes the estimated uncertainty of the GNSS/IMU solution used to position and orient the sensor. Point cloud accuracy describes how well the final LiDAR data matches independently surveyed ground truth.

While high-quality trajectory metrics are an important indicator of successful GNSS/IMU processing, they should not be used as a substitute for point cloud accuracy validation.

The true accuracy of a LiDAR dataset can only be determined through comparison with independent control measurements and appropriate statistical analysis of the resulting errors.