LCVSTP: LiDAR-Camera Fusion Simulation Platform

LCVSTP is a lightweight interactive research simulator for studying how LiDAR geometry and camera observations can be combined to track surgical instruments with high precision. It makes synchronization error, calibration drift, sensor disagreement, and recovery behavior visible without requiring physical hardware.

Research Focus

• Cross-modal sensing: Compare complementary camera and LiDAR observations throughout the fusion pipeline.
• Reliability monitoring: Observe residual growth, safety lockout, recalibration, and recovery when measurements diverge.
• Reproducible evaluation: Repeat seeded scenarios and compare system behavior under controlled timing and drift conditions.

Validation Scenarios

• TC-A: Drift and Loop examines drift detection, safety lockout, recalibration, and recovery.
• TC-B: Temporal Skew compares behavior with asynchronous camera and LiDAR streams.
• TC-C: Multi-Target evaluates deterministic tracking across a sequence of task targets.

Evidence and Scope

Sessions expose fusion consistency, pre- and post-fusion residuals, estimated drift, task-state transitions, and contact error. Seeded runs can be exported as JSON, CSV, and LaTeX tables for repeatable analysis.

LCVSTP is a research and teaching simulation. It is not a clinical device and is not intended for patient diagnosis or treatment.