User Guide: Creating a Trajectory in LiDAR Boost Post Process

Prerequisites

• An RD file (.rd) containing GNSS and IMU raw data.
• A base station file (RINEX format) for GNSS corrections if RTK not present in real time.
• One or more .pcap LiDAR data files.
• A full LiDAR calibration (LIP and LIR offsets).

Step 1: Input Files

Open LiDAR Boost Post Process and click the Input Files tab.

1. In the RD* field, select your RD file (.rd).
2. In the Base Stations field, select your RINEX base station file.
3. Click the + button next to LiDARs. The Lidar Configuration dialog opens. Configure the LiDAR:
   • Select your sensor from the Select LiDAR dropdown.
   • In the PCAP* field, select your LiDAR data file.
   • Set Use only vertical offset to No if a full LIP/LIR is known, then load your LIR and LIP files. If a full LIP/LIR offset is not known, you must first run a boresight to obtain the full calibration: see the boresight guide.
   • Set RPM to match your LiDAR. The default is 600.
   • Click Accept.
4. Click Continue.

Step 2: Processing Options

Click the Processing Options tab.

1. Under Process type, select Trajectory Generation.
2. Configure the Trajectory Settings:
   • Processing type: Combined blends forward and backward processing runs for the highest accuracy. Simulated runs a single forward pass, which is faster but less accurate.
   • Output coordinate frame for positional information: Ellipsoidal heights are referenced to the ellipsoid. Geoidal heights are referenced to the geoid (approximate mean sea level).
   • Use smoothing: Smoothing refines the solution using the full dataset, which improves accuracy across the trajectory.
   • GNSS recovery: This controls how many unexpected GNSS updates are ignored before the engine is forced to use them. No applies the default behavior. Yes lets you set the threshold manually. GNSS updates run at 4 Hz and velocity at 10 Hz.
   • Use angular GAD updates (Beta): This adds LiDAR angular rate updates to the standard velocity updates. It is a beta feature, and the performance benefit is not yet confirmed.
   • Use ZVU (Zero Velocity Updates): ZVU detects when the system is stationary and constrains the solution to zero motion. This reduces drift during a GNSS outage. The GNSS rejection duration, Packets before activation, and Speed threshold fields tune when ZVU activates.
3. Under LiDAR Boost Settings, choose your algorithm:
   • Odometry provides a velocity update to the INS.
   • Map Matching uses a pre-created map to provide position updates to the INS.
   • SLAM provides position updates to the INS using loop closure. Select SLAM if any part of the data is fully indoors with no GNSS access. This gives the most accurate result and avoids double vision in the output.

Step 3: Process

1. Click Process. The software computes a GNSS/INS trajectory and displays it on the map. Select the data to use in the Data Overlap timeline. Drag the red handles to set the start and end of the segment. Target the sections with GNSS-denied or poor coverage. These sections benefit most from LiDAR Boost. Data with good GNSS coverage is already accurate and does not need LiDAR Boost.
2. Click Accept Overlap. The remaining stages run automatically:
   • Odometry runs the odometry stage, then a final trajectory.
   • Map Matching runs the map matching stage, then a final trajectory.
   • SLAM runs odometry, then a trajectory, then SLAM, then a final trajectory.

Step 4: Trajectory processing complete

When processing finishes, the outputs are saved to a timestamped folder inside your output folder.

The processed trajectory is in a sub-folder named after the processing type. For example, aided_cmb contains the LiDAR odometry-aided trajectory from Combined processing.