V-Spline: An Adaptive Smoothing Spline for Trajectory Reconstruction.

Area of Science:

• * Computational geometry
• * Applied mathematics
• * Kinematics

Background:

• * Trajectory reconstruction is essential for understanding object motion between observations.
• * Existing methods struggle with irregularly sampled data and noisy velocity measurements.
• * Smoothing splines are commonly used but may not optimally incorporate all available information.

Purpose of the Study:

• * To introduce a novel smoothing spline, the V-spline, for enhanced trajectory reconstruction.
• * To develop an adaptive V-spline to address challenges of irregular sampling and noisy velocity data.
• * To evaluate the V-spline's performance against existing trajectory reconstruction techniques.

Main Methods:

• * Proposed a V-spline incorporating position, velocity, and an acceleration-controlling penalty term.
• * Introduced an adaptive V-spline variant for handling irregular data and measurement noise.
• * Developed a cross-validation scheme for V-spline parameter estimation.
• * Applied the V-spline to two-dimensional vehicle trajectory reconstruction, with adaptable penalty terms.

Main Results:

• * The V-spline demonstrated superior performance compared to existing methods in simulation studies.
• * The adaptive V-spline effectively mitigated issues from irregularly sampled observations and noisy velocity data.
• * The method showed successful application in two-dimensional vehicle trajectory reconstruction.

Conclusions:

• * The V-spline offers a robust and accurate solution for trajectory reconstruction.
• * The adaptive V-spline provides a significant advancement for handling real-world observational data.
• * The V-spline framework is versatile and can be extended for specific applications like vehicle dynamics.