A Closed-Form Solution to Planar Feature-Based Registration of LiDAR Point Clouds

Since pairwise registration is a necessary step for the seamless fusion of point clouds from neighboring stations, closed-form solution to planar feature-based LiDAR (Light Detection and Ranging) proposed in this paper. Based on Plücker coordinate-based representation linear features three-dimensional space, quad tuple-based introduced, which makes it possible directly determine difference between any two features. Dual quaternions are employed represent spatial transformation operations dual given, with an error norm constructed. L2-norm-minimization, detailed derivations explained by step. Two experiments were designed simulated data real both used verify correctness feasibility solution. With data, calculated results consistent pre-established parameters, verifies presented iterative methods. Conclusions can be drawn experiments: (1) The does not require initial estimates unknown parameters advance, assures stability robustness solution; (2) Using greatly reduces additional constraints estimation process.